<%BANNER%>

The Impact of Invasive Upland Plants on the Recreational Value of Natural Areas

Permanent Link: http://ufdc.ufl.edu/UFE0021470/00001

Material Information

Title: The Impact of Invasive Upland Plants on the Recreational Value of Natural Areas The Case of Wooded Parks in Florida
Physical Description: 1 online resource (116 p.)
Language: english
Creator: Bwenge, Anafrida N
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: attributes, outdoor, recreation, utility
Food and Resource Economics -- Dissertations, Academic -- UF
Genre: Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Invasive plants can have impacts on the quality and quantity of recreational activities such as hunting, wildlife viewing and hiking because they negatively affect environmental attributes that are important in supporting recreation. This study examined the relationships between invasive plants and recreational values of Florida?s wooded parks using a Multi-Attribute Utility Model. Surveys were electronically distributed to Florida residents to examine preferences for these attributes; invasive species, park fee, species diversity and facilities. A conditional logit model predicted the respondents? choice behavior and quantified the relationship between utility derived from recreation and these attributes. Results indicate that invasive species have a negative impact to recreational utility. Florida residents have a marginal willingness to pay (MWTP) to reduce invasive species of up to $7.15 which is higher than the MWTP to improve facilities or increase species diversity. MWTP to reduce invasive species was even higher with invasive species knowledge ($19.25 for experts). Residents' willingness to pay to control these species ranged from $29.1 to $108.7 million per year with knowledge level differences. Our findings suggest that an invasive species educational program could increase Florida residents? MWTP to control invasive species in natural areas.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Anafrida N Bwenge.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Lee, Donna J.
Local: Co-adviser: Larkin, Sherry L.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021470:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021470/00001

Material Information

Title: The Impact of Invasive Upland Plants on the Recreational Value of Natural Areas The Case of Wooded Parks in Florida
Physical Description: 1 online resource (116 p.)
Language: english
Creator: Bwenge, Anafrida N
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: attributes, outdoor, recreation, utility
Food and Resource Economics -- Dissertations, Academic -- UF
Genre: Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Invasive plants can have impacts on the quality and quantity of recreational activities such as hunting, wildlife viewing and hiking because they negatively affect environmental attributes that are important in supporting recreation. This study examined the relationships between invasive plants and recreational values of Florida?s wooded parks using a Multi-Attribute Utility Model. Surveys were electronically distributed to Florida residents to examine preferences for these attributes; invasive species, park fee, species diversity and facilities. A conditional logit model predicted the respondents? choice behavior and quantified the relationship between utility derived from recreation and these attributes. Results indicate that invasive species have a negative impact to recreational utility. Florida residents have a marginal willingness to pay (MWTP) to reduce invasive species of up to $7.15 which is higher than the MWTP to improve facilities or increase species diversity. MWTP to reduce invasive species was even higher with invasive species knowledge ($19.25 for experts). Residents' willingness to pay to control these species ranged from $29.1 to $108.7 million per year with knowledge level differences. Our findings suggest that an invasive species educational program could increase Florida residents? MWTP to control invasive species in natural areas.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Anafrida N Bwenge.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Lee, Donna J.
Local: Co-adviser: Larkin, Sherry L.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021470:00001


This item has the following downloads:


Full Text





THE IMPACT OF INVASIVE UPLAND PLANTS ON THE RECREATIONAL VALUE
OF NATURAL AREAS: THE CASE OF WOODED PARKS IN FLORIDA




















By

ANAFRIDA BWENGE


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2007

































2007 Anafrida N. Bwenge

































(To Kisha for understanding when mom was too busy with school)









ACKNOWLEDGMENTS

This thesis could not have been completed without the help, encouragement and support of

all my supervisory committee members. First, I wish to express my sincere thanks and

appreciation to my committee chair, Dr. Donna Lee who guided me in writing this thesis and

also made sure I had all the financial help I needed to acquire this degree. I would also like to

thank my other committee members, Dr. Sherry Larkin and Dr. Janaki Alavalapati whose

devoted support helped me to complete this thesis.

I have so much appreciation for my fellow student, Santiago Bucaramu, who provided the

much needed technical and mathematical expertise to design, implement and analyze this study.

I would also like to express my gratitude to my friend Jennifer Nunez, who offered to read my

thesis and helped me with both technical advice and corrections to my English grammar. Finally,

I would like to acknowledge and give special thanks to my husband, Dr. Charles Bwenge, and

my girls for all their help and support during my time in school.









TABLE OF CONTENTS

page

A CK N O W LED G M EN T S ................................................................. ........... ............. .....

L IST O F T A B L E S .................................................. ........................ 7

LIST OF FIGURES .................................. .. ..... ..... ................. .9

L IST O F A B B R E V IA T IO N S ....................... ......... .................................................................. 10

A B S T R A C T ......... ....................... ............................................................ 1 1

CHAPTER

1 INTRODUCTION ............... .......................................................... 12

B ack g rou n d ......................................................................................................12
R recreation in Florida N natural A reas.......................................................................... ... ... 13
Upland Invasive Plants in Florida Natural Areas ........................................ ............... 15
Problem Statem ent ............................................................... .... ..... ........ 18
S tu dy O bjectiv e s ....................................................... ................ 19
H ypotheses......... ................................ ................................................ 19

2 L ITE R A TU R E R E V IE W ........................................................................ .. .......................23

V valuation of N on-M market Goods ................................................... ........ ............... .23
Selection of A attributes and L evels............................................................................. ............. 25
S u rv ey M eth o d ......................................................................... 2 8
W eb S u rv ey s ....................................................... ................ 2 9
A advantages of w eb surveys............................... ..................................... ..............30
W weaknesses of w eb surveys ............................................. ............................ 30

3 THE MULTI-ATTRIBUTE UTILITY MODEL ........................................ .....................33

4 DATA COLLECTION AND RESEARCH METHODS ................................................. 37

S u rv e y s .................................................................................. 3 7
Park M managers' Survey ............... .............. ........... .......... .... ............38
Florida Residents: Invasive Species Knowledge Survey .........................................39
Florida Residents: Attributes Selection Survey.......................................................41
Building the M ulti-Attribute Utility Survey ..................................................................... 42

5 SU RVEY RESU LTS ..... ...... ............ ............ .......... .......... ..... .... ............ .. .. .... 1

Introduction ......... .. ............................... ....... .. ..... ................ 51
Respondents' Profiles ................................................... ........ .............................. 51









Invasive Species K now ledge ......................................................................... ...................52

6 MODEL SPECIFICATION AND EMPIRICAL RESULTS ...........................................59

M odel V ariab les ........................................................ ................ 59
N on-R response E errors T testing ....................................................................... .................... 59
H ypothesized Signs on Param eters.................................................. ............ ............... 60
Statistically Significant Individual Specific Variables.........................................................61
S p ecificatio n ........................................................................6 2
R e su lts ........................... ..... ............................................................................................... 6 3
O v erall M odel F it ................................................................63
A tribute V ariables ........................................................ .... ..... ........ ........63
Probabilities and Marginal Effects for the Models............................... ...............64
M arginal W willingness to Pay M easures...................................................... ..................65
Marginal willingness to pay and individual specific variables .............................66
M marginal willingness to pay by regions ....................................... ............... 66
R elative W fighting of A ttributes..................................................................................67
Choices R anking by U utility Scores............................................ ........... ............... 67
R results for the C om bined M odel......................................................................... ...... 68
H ypotheses T testing ........... .............................................................................. .... .......69

7 INTERPRETATION OF THE RESULTS FOR INVASIVE PLANT MANAGEMENT.....81

Invasive Species and O outdoor R creation ................................................. ........ .............81
Estimations of Willingness to Pay to Reduce Invasive Species...........................................81
Marginal Willingness to Pay and the Invasive Species Attitude................ .............. ....84
The Importance of Invasive Species Knowledge ..................... ...................................86

8 SUMMARY AND CONCLUSIONS......................................................... ............... 90

APPENDIX

A MAIN SURVEY.............. .... .......... .. .. .... ..... ..................93

B PARK RAN K IN G BY U TILITY .............................................................. ....... ........106

C WOODED PARK CLASSIFICATION BY REGIONS................................................... 108

L IST O F R E F E R E N C E S ............................................... ................................... ..................... 110

B IO G R A PH IC A L SK E T C H ......................................................................... ... ... ............ 116









LIST OF TABLES


Table page

1-1 Ten m ost unw anted plants in Florida.......................................... ........................... 22

4-1 Reasons for participating in outdoor recreational activities ...........................................48

5-1 Demographic profiles for surveys compared to Florida profiles................. ................54

6-1 Basic m odels independent variables ............................................................................ 71

6-2 Socioeconomic and invasive species attitude variables.........................................71

6-3 X2 Tests statistics of first 50 and last 50 respondent characteristics ..............................71

6-4 Significance test for attribute interaction with individual specific variables-WPS ..........72

6-5 Significance test for attribute interaction with individual specific variables-WAS ..........72

6-6 Overall fit for the m ulti-attribute m odels................................... ...................... .. .......... 72

6-7 Coefficient estimate for the multi-attribute models .......................................................73

6-8 Probabilities for park choices-WPS ........................ .. ......................... ............... 75

6-9 M marginal effects for the m models .............................................................. .....................75

6-10 Comparison of implicit prices estimates for recreational attributes ..................................75

6-11 Implicit prices to reduce invasive species with attitude variables ...................................76

6-12 Regional marginal willingness to pay for the attributes ..........................................76

6-13 Impact of change in attribute level on the total utility ..................................................79

6-14 Coefficient estimates for the attributes (combined model).................. ...... .............79

7-1 Annual WTP to control invasive plants in Florida wooded parks (million $)................... 88

7-2 Annual regional WTP to reduce invasive plants levels -WPS............... .................. 88

7-3 WTP per visit to reduce Melaleuca in recreational areas by Florida residents .................88

7-4 Annual WTP to control invasive species- FL residents with knowledge (million $)........ 88

7-5 Coefficient estimate for the attributes (invasive plants experts).................... ........ 88

B-l Park ranking by utility -W PS ....................................................................... 106









B -2 P ark ranking by utility-W A S ............................................ ......................................... 107

C- Park classification by regions ................................................ .............................. 108




















































8









LIST OF FIGURES


Figure page

1-1 Annual attendance at Florida state parks: 1995-2006........................................................21

1-2 Florida's estimated demand for wooded park activities: 1997-2010..............................21

4-1 Level of invasive species knowledge in Florida...........................................................47

4-2 Perceived impact of invasive species on enjoyment and other aspects-(preliminary).......47

4-3 Natural areas outdoor activities participation ........................................ ............... 48

4-4 Outdoor recreation attributes selection ..................................................... ... .......... 49

4-5 Suggested park attributes for the state to improve .......................................................49

4-6 Example of the MAUM questions A) plant species B) animal species.............................50

5-1 Location profiles of samples and Florida residents ................................. ............... 54

5-2 Gender profiles of samples and Florida residents..........................................................55

5-3 Age profiles of samples and Florida residents ......................................... .............55

5-4 Income profiles of samples and Florida residents................................... ............... 56

5-5 Education profiles of samples and Florida residents ............................... ............... .56

5-6 Comparison of knowledge of invasive species in three surveys ......................................57

5-7 Perceived impact of invasive species on enjoyment and other aspects-(main) .................57

5-8 Respondents taking action against invasive species ................... ......................... 58

5-9 P ark visit frequencies........ ........................................................................ ....... .. .... 58

6-1 Im plicit prices for recreational attributes................................................ ........ ....... 76

6-2 Regional marginal willingness to pay A) plant species B) animal species......................77

6-3 Relative weights of attributes A) plant species B) animal species ..................................78

6-4 Relative weights of the attributes for the combined model .......................................80

7-1 Relative weights of attributes for invasive plants experts ..............................................89











APHIS

BIPM

CE

CV

FACT

FLDEP

FLEPPC

FLSCORP

MAU

MAUM

MWTP

NISC

OTA

RUM

UF-IRB

USDA

WAS

WPS

WTA

WTP


LIST OF ABBREVIATIONS

Animal and Plant Health Inspection Service

Bureau of Invasive Plant Management

Choice Experiment

Contingent Valuation

Florida Assessment of Coastal Trend

Florida Department of Environmental Protection

Florida Exotic Plant Pest Council

Florida Statewide Comprehensive Outdoor Recreation Plan

Multi-Attribute Utility

Multi-Attribute Utility Model

Marginal Willingness to Pay

National Invasive Species Council

Office of Technology Assessment

Random Utility Maximization

University of Florida Institutional Review Board

United States Department of Agriculture

Wooded Park Animal Species

Wooded Park Plant Species

Willingness to Accept

Willingness to Pay









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

THE IMPACT OF INVASIVE UPLAND PLANTS ON THE RECREATIONAL VALUE OF
NATURAL AREAS: THE CASE OF WOODED PARKS IN FLORIDA

By

Anafrida N. Bwenge

August 2006

Chair: Donna Lee
Co chair: Sherry Larkin
Major: Food and Resource Economics

Invasive plants can have impacts on the quality and quantity of recreational activities such

as hunting, wildlife viewing and hiking because they negatively affect environmental attributes

that are important in supporting recreation. This study examined the relationships between

invasive plants and recreational values of Florida's wooded parks using a Multi-Attribute Utility

Model. Surveys were electronically distributed to Florida residents to examine preferences for

these attributes; invasive species, park fee, species diversity and facilities. A conditional logit

model predicted the respondents' choice behavior and quantified the relationship between utility

derived from recreation and these attributes.

Results indicate that invasive species have a negative impact to recreational utility. Florida

residents have a marginal willingness to pay (MWTP) to reduce invasive species of up to $7.15

which is higher than the MWTP to improve facilities or increase species diversity. MWTP to

reduce invasive species was even higher with invasive species knowledge ($19.25 for experts).

Residents' willingness to pay to control these species ranged from $29.1 to $108.7 million per

year with knowledge level differences. Our findings suggest that an invasive species educational

program could increase Florida residents' MWTP to control invasive species in natural areas.









CHAPTER 1
INTRODUCTION

Background

Invasive plant species are defined as non-indigenous species with the ability to establish

self-sustaining and expanding populations within plant communities and may cause economic or

environmental harm (NISC, 2001). The United States' natural ecosystems have been invaded by

over 5,000 non-indigenous plant species, which compete with approximately 17,000 native plant

species for space and resources. When invasive plants successfully invade natural areas they tend

to displace native species and associated wildlife, degrade upland habitats and cause loss of

biodiversity (Olson 1999). This is because they possess many weedy characteristics which

enable them to spread rapidly and effectively in the new environment.

Some of these non-native plants are responsible for $25 billion in damages to the Unites

States' food and horticultural crops and $10 million in losses to natural ecosystems each year

(Pimentel, 2002). The annual total cost from damages and controlling invasive plants in the

agriculture and horticulture sectors is $34.5 billion and an additional $159.5 million is spent on

managing invasive plants in natural systems (Pimentel, 2002). Nationally, invasive species are

the second greatest threat to endangered species after habitat destruction and cost the country

over $138 billion each year in environmental damages and crop losses (Pimentel, 2002).

While the invasive species problem has become a global and national concern, the state of

Florida has been the most affected in the United States (OTA, 1993). The favorable climate,

geographical location and environmental conditions that foster the state's high level of plant

diversity have consequently made Florida's land susceptible to invasive plant species. To date,

more than 1,300 exotic plant species have been introduced and established throughout the state

with 124 species destructive to the biological diversity of natural areas (FLEPPC, 2006). In 2005









upland weeds such as the Australian pine, Brazilian pepper and climbing ferns infested over 1.65

million acres of Florida's 11 million acres of public conservation lands (FLDEP, 2005). These

plants have also affected millions of acres of Florida's privately owned land.

Non-native invasive plants can also have substantial impact on recreational activities such

as hunting, hiking, wildlife viewing and water-based recreation (Eiswerth et al., 2005). Olson

(1999) suggests that this is because invasive weeds negatively affect a wide range of

environmental attributes that are important in supporting recreation such as plant and animal

diversity and abundance. In a study about the economic impacts of weeds on outdoor recreation

in the riparian areas of Nevada, Eiswerth et al., (2005) found that non-native weeds had a

recreational losses impact over the five years period ranging from $30 million to $40 million.

Recreation in Florida Natural Areas

Florida natural areas play a significant role in the states' economy by providing

recreational activities for residents and visitors. Ecotourism recreational activities such as hiking,

camping, sightseeing, and wildlife viewing in Florida's natural areas have an estimated economic

contribution of $8 billion per year (FLDEP, 2005).

Outdoor recreation is one of the state's main attractions and the Florida's state park

system is one of the largest in the country with 159 parks covering 723,852 acres of land and 100

miles of sandy beach (FLDEP, 2006). Florida's state parks offer year-around outdoor activities

for all ages. The park system is comprised of beaches, rivers and lakes and wooded parks

offering diversified activities in each park. There are some parks like for example, Oleta River

state park where visitors can enjoy beach activities, engage in kayaking, mountain biking,

camping, swimming, fishing, trail walking, horse riding, wildlife viewing and even hiking, all

activities in one place. Over 100 parks offer wooded park activities like hiking, nature trails and

horse riding. Over 50 parks offer both river and lake (boating, fishing, and kayaking) activities









and wooded park activities while more than 40 parks offer beach activities like swimming,

sunning, surfing and wooded park activities.

A good number of visitors come to Florida for the primary purpose of viewing wildlife.

According to the Fish and Wildlife Service, approximately 800,000 visitors came to Florida in

1996 primarily for the purpose of viewing wildlife, and over 40 percent of Florida's residents

participated in some form of wildlife viewing. In this same year, Florida ranked second in the

nation behind Texas for wildlife-related recreation expenditures. In the 2001 Fish and Wildlife

survey nearly 4.9 million Florida residents and nonresidents over 16 years of age fished, hunted,

or watched wildlife in Florida parks. Over 65% of this total number participated in wildlife-

watching activities, including observing, feeding, and photographing. More than half of Florida

visitors engage in some type of nature-based activity during their visit, and 19 to 33 percent of all

travel and tourism in the southern United States is linked to outdoor recreation (Hodges, 2006).

Between 1995 and 2004, the state park system's economic impact on local economies

throughout the state grew from $189 million to over $600 million (FLDEP, 2005) with the

annual park attendance growing from 12 to 18.5 million visitors (Figure 1-1). In the last fiscal

year, 2005-2006 over 18 million people visited Florida state parks spending over $442 million in

the state (FLDEP, 2006). It is expected that in the next five years the need for public outdoor

recreation land and parks in Florida will increase greatly as the state's population is growing

(FLDEP, 2006).

The demand for wooded park activities like biking, horse riding and wildlife viewing has

also been on the rise and expected to continue. The estimated demand for selected wooded park

activities are presented in Figure 1-2. On Florida's scenic trails, a growing number of people are









undertaking longer day and overnight hikes while horseback riding participation, relative to other

forms of outdoor recreation, has been steady (FLSCORP, 2000).

The state has responded to the growing outdoor recreation demand by investing over $3

billion to expand conservation lands and recreational opportunities over the past decade (FLDEP,

2006). The focus has been on making natural areas more accessible to the public through

restoration including management and removal of non-native plants.

Upland Invasive Plants in Florida Natural Areas

Upland invasive plants are terrestrial (vs. aquatic) invasive exotic plants. Invasive plants

displace native plants and associated wildlife, including endangered species and can alter natural

process such as fire and water flow. Exotic plants were brought to the U.S. to be grown for

various reasons like food, feed, fiber, and ornamental purposes but some have become invasive

and have proven to be a challenge to keep under control. The problem of exotic invasive species

in Florida parks has been cited as one of the greatest threats to park resources (Glisson, 1994).

According to the Florida's Recreation and Parks Division, the most troublesome invasive

plants in the state parks are Brazilian pepper, Australian pine, Chinese tallow, Cogongrass, Air

potato and Japanese climbing fern. Based on the total acres treated in projects dealing with

upland weeds management in the state in 2005, the "ten most unwanted" upland invasive exotic

plants in Florida are listed in Table 1-1.

Florida Exotic Plant Pest Council (FLEPPC) compiles invasive species lists that are

revised every two years. Invasive exotic plants are termed Category I invasive when they are

altering native plant communities by displacing native species, changing community structures

or ecological functions, or hybridizing with natives. Category II invasive exotics are invasive

plants that have increased in abundance or frequency but have not yet altered Florida plant

communities to the extent shown by Category I species. They may become category I if they









demonstrate ecological damage. The plants on the list of the ten most unwanted in Florida are

currently all listed as category I invasive by the FLEPPC. They are scattered everywhere mostly

from Central Florida, along the East and West coasts towards South Florida.

Among the list, Australian pine, Brazilian pepper and Malaleuca are the most widely

spread in Central and South Florida. These plants, like other invasive species, have a tendency to

crowd out native plants and animals. Australian pine invades coastal areas interfering with

nesting of endangered sea turtles and American crocodiles. Brazilian pepper grows in dense

monocultures and reduces nesting sites for bird, amphibian and reptile populations. It is believed

to have displaced some populations of rare listed species, such as the Beach Jacquemontia

(Jacquemontia reclinata) and Beach Star (Remirea maritima) in South Florida (Doren, 2002).

Cogongrass invades pinelands, scrub and prairie also threatening rare plants and interfering with

fire patterns. In places invaded by Cogongrass, wildfires can be more frequent and intense

(FLDEP). Old world climbing fern is naturalized in southern and western Florida, and the

Japanese climbing fern is frequently naturalized in north and west Florida.

South Florida's upland environments are the part of the state most heavily invaded by non

native species (FLDEP). The reason for South Florida's heavy invasion is said to be high

importation activity in the area, highly disturbed landscapes and a climate conducive to growth

of subtropical plants.

The plants mentioned above are also restricted by the federal government along with other

plants that are known to interfere with agro-ecosystems, native ecosystems, the management of

ecosystems, or to cause injury to public health. The USDA Animal and Plant Health Inspection

Service (APHIS) runs the Federal noxious weed program designed to prevent the introduction

and the spread of newly introduced non-indigenous invasive plants in the United States by









excluding, detecting and eradicating introduced weeds that pose the highest risk to agriculture or

the environment.

Controlling and management of invasive species in natural areas is one of the state's

priorities but this exercise has been costly. Private expenditures for controlling invasive plants in

Florida's agriculture and forest industries are estimated at $265 million per year (Lee and Kim,

2005). Overall the state spends $103 million per year on prevention and control of invasive

plants (FLDEP, 2006). In 2005 the state exceeded the $6.3 million annual estimate by spending

$8.7 million managing just upland invasive exotic species (FLDEP, 2005).

The Upland Plant Management Program responsible for managing invasive exotic plants in

the state's public lands is under the Florida Department of Environmental Protection (FLDEP) in

the Bureau of Invasive Plant Management (BIPM). The program works in eleven regions within

the state to develop strategies to address upland invasive plant management issues locally

through regional working groups.

Since the inception of the Upland Program in 1997, BIPM has spent nearly $40 million to

bring over 300,000 acres of upland weeds under maintenance control (FLDEP, 2005).

Maintenance control is a control method of invasive exotic plants in which control techniques are

utilized in a coordinated manner on a continuous basis in order to maintain the plant population

at the lowest feasible level. As stated in the FLDEP Agency Strategic Plan, the long-term

program goal is to reduce infestations of upland invasive exotic plants on public lands by twenty-

five percent by 2010, based on estimated 1995 levels of 1.5 million acres infested with invasive

weeds. In 2005 over 22% of affected public land was under maintenance control (FLDEP, 2005).

The program treated about 100 different invasive species at 144 publicly managed areas. The









control program utilizes a variety of methods including chemical, mechanical, and biological

techniques.

Problem Statement

Millions of residents and tourist who participate in outdoor activities derive satisfaction

from various attributes of natural areas. The most general attributes specific to the parks include

parks sanitation and safety, extent and condition of facilities, the quality of the natural

environment and accessible scenic trails for a variety of nature-based recreational activities (e.g.,

hiking, camping, sightseeing and wildlife viewing).

Invasive plants may have a negative impact on the quality or quantity of outdoor

recreational activities. Through altering ecosystems, invasive species can negatively change the

supply and composition of environmental amenities that are important for recreation and

adversely affect recreational service flows. Plants like the Old World climbing fern (Lygodium

microphyllum) and Japanese climbing fern (Lygodiumjaponicum), both on the list of the "ten

most unwanted" plants in Florida, grow with thick climbing and twining fronds, preventing easy

access to natural areas. Likewise, the Air potato (Dioscorea bulbifera L.) grows vigorously,

twining and forming a dense blanket that engulfs surrounding plants, which can limit access for

navigation and recreational activities. The presence of dense twining invasive plants in natural

areas may also obstruct wildlife viewing and access to scenic trails making hiking and biking

difficult in wooded parks. Some park visitors may be bothered by Catlaw mimosa (Mimosapigra

L.) or Tropical soda apple (Solanum viarum) due to their hairy stem and prickly nature,

respectively.

Studies on the consequences of invasive plants in Florida natural areas have focused on

management costs (Lee and Kim, 2005; Doren, 2002; Harding and Thomas, 2003) and

ecological impact (Mazzoti, 1981; Gordon, 1998). In order to fully understand how invasive









species affect outdoor recreation, it is essential to know how the user's enjoyment and use of

natural areas is affected by these exotic species. However, studies that estimate the relationship

between recreational utility and invasive species in natural areas do not exist at the national or

state level. Given the significance of outdoor recreation in natural areas to the citizens'

enjoyment and the state's economy this knowledge could aid in planning statewide programs

aiming to control invasive species in Florida. Therefore, this study proposes to examine the

relationships between upland invasive plants and the recreational value of Florida natural areas.

Study Objectives

The general objective of this study is to examine the relationship between invasive upland

plants and the implicit value of recreational activities in natural areas specifically, in Florida

wooded parks. This was achieved through the following objectives:

* Objective 1: Quantify the relationship between invasive plants and recreational value in
natural areas using a Multi-attribute Utility Model (MAUM);

* Objective 2: Determine the relative importance of invasive species in relation to other
attributes of a natural recreation experience;

* Objective 3: Determine the marginal willingness to pay for fewer invasive species in
recreational areas, specifically in Florida wooded parks;

Hypotheses

The basic premise in this study is that invasive plants are undesirable in wooded parks and

recreational places such that recreation satisfaction in these areas will be reduced by the presence

of invasive species. For this reason natural area users will be willing to pay more for a

recreational exercise at a wooded park that has fewer invasive species in outdoor recreation

areas.

It is anticipated however that the willingness to pay more for fewer invasive species will

not only be due to the effect of invasive plants on recreational utility but also because of the









perceived impact of these species on the environment1. Therefore, users of natural areas who

have a higher level of knowledge of invasive plants and a higher level of environmental

consciousness are expected to be willing to pay more for fewer invasive species.

Socioeconomic variables such as income, age, and education are expected to influence

individual's perception about the relationship between invasive plants and the recreational value.

Furthermore, since Florida regions are affected at different levels by invasive plants, willingness

to pay will likely be different between regions and perhaps higher for the most affected region.



































1 It is common for public to express their willingness to pay for passive or non use values. For example people want
to pay to make sure blue whales are conserved even if they may not see them in their lifetime.














25.00

20.00

15.00

10.00

5.00


0.00


94-95* 95-96* 96-97* 97-98* 98-99* 2003-04 2004-05 2005-06
Year

Figure 1-1. Annual attendance at Florida state parks: 1995-2006
Source: *FACT, 2000, FLDEP, 2004-2006






18.00
16.00
S14.00
S12.00 -
:|o 10--- Hiking
E 10.00
0 -- Horse Riding
a- 8.00
a Nature Study
.2 6.00
I 4.00
2.00
0.00


1997 2000


2005* 2010*


Year
Figure 1-2. Florida's estimated demand for wooded park activities: 1997-2010
Source: FLSCORP 2000










Table 1-1. Ten most unwanted plants in Florida
Common name Scientific Name Acres Treated
Australian pine Casuarina species 436
Caesar's weed Urena lobata 749
Old world climbing fern Ligod!hiim microphyllum 3,728
Melaleuca Melaleuca quinquenervia 46,498
Skunk vine Paederiafoetida 1,021
Brazilian pepper, Schinus terebinthifolius 7,830
Chinese tallow Sapium sebiferum 667
Japanese climbing fern Ligodninm japonicum 771
Cogongrass Imperata cylindrica 1,212
Tropical soda apple Solanum viarum 1,023
Source FLDEP: Upland Exotic Plant Management Program, Annual report FY 2004-2005









CHAPTER 2
LITERATURE REVIEW

Valuation of Non-Market Goods

This study attempts to determine the non-market value of recreation attributes in natural

areas with special focus on invasive plants. A non-market good or service is something that is

not bought or sold directly, therefore, does not have an observable price (i.e. market value). Non-

market values can be categorized as use values and non-use values. The use value of a good is

the value to an individual from the active use of the asset like recreational fishing, swimming or

bird watching and the non-use values reflect the value to the society or future generation (Letson

and Milon, 2002).

Although participation in outdoor recreation through activities like camping and hiking

can be categorized as the use value for natural areas and may generate market based economic

activity, use of a natural area is often not allocated by markets (Swanson and Loomis, 1996).

Therefore the impact of invasive species on natural areas will not fully be captured by market

goods and services.

Where there is no price available for non-market goods, it may be possible to use the prices

of related market goods or prices from hypothetical markets to estimate the value (Letson and

Milon, 2002; Longo, 2007). A number of methods have been devised for valuation of an

environmental good or service where market value is not evident. These methods include

hedonic price, travel cost, the contingent valuation and the conjoint choice experiment. The

hedonic pricing and travel cost method are called revealed preferences because they measure

preferences for non-market goods based on observing people's choice behavior on other related

goods. For example hedonic pricing method would use the property sales information to assess

the monetary value of a cultural tourism attraction. Travel cost method would measure the value









of a tourism attraction by using the money spent on that attraction as a proxy of the value that

users attach to the place. Contingent valuation (CV) and choice experiment (CE) are referred to

as stated preferences because they ask individuals about their preferences through surveys.

Typically the CV approach asks people to directly report their willingness to pay (WTP) for a

specific good, or their willingness to accept (WTA) compensation for a good rather than

inferring them from observed behavior in regular market like in revealed preference methods

(Longo, 2007).

In the past, most studies on valuation of non-market environmental goods like natural areas

have used CV (Bennet et al., 1997; Tsuge and Washida, 2003; Berrens et al., 2004). This method

has also been used in valuing cultural heritage destinations (Alberini et al., 2005). However, CV

has been criticized for its weakness in valuing goods when most of the value of the good is

derived from non use value. It is believed that the CV provides an incomplete view of the value

of the good because this value is multidimensional and may not be easily expressed by

qualitative or quantitative scales (Throsby, 2003). Pearce et al. (1994) summarized some of the

CV method issues as problems of reliability, bias and validity.

The CE has been successfully used in marketing and transportation research. Following

this success, the CE methodology has been increasingly preferred over the CV in valuation of

environmental recourses (Holmes et al., 2003). CE has been used to assess public preferences for

restoration goals (Milon et al., 1999; MacDonald et al., 2005,) and valuing environmental

amenities by Adamowicz and others (1997) who argue that the advantage of CE over CV is that

this method could help reduce strategic bias as the attributes change in choice sets.

When a non-market good's non-use values are impacted, only stated preference models

can capture the impact. One such method is the Multi-attribute Utility Model (MAUM) which is









a choice experiment type of method (Louviere and Woodworth, 1983; Milon et al., 1999;

Louviere et al., 2000). There are two types of MAUM; one is preference based where

respondents are asked to rate or rank the provided alternatives and the other is choice based

where respondents are asked to choose their preferred alternative from the provided choice

package.

Blamey et al. (1997) describe the CE as a technique in which individuals are typically

presented with six to ten 'choice sets', each containing a base option and several alternatives and

asked to indicate their preferred option in each set. Although there is no definitive number of

choices to be presented, it is recommended to use not more than eight to avoid respondents being

fatigued (Holmes et al., 2003). The respondent chooses from a set of hypothetical "goods or

services" that require an evaluation of the trade-offs between attribute levels, with the concept

that the choice is based on the individual's utility maximization. Attributes trade off is presented

in the value elicitation process such that reduction in one attribute may be compensated by an

increase in another. The MAUM research approach involves prior identification of attributes and

levels for building the choices or alternatives to present to the respondent.

Selection of Attributes and Levels

The initial step in developing a choice experiment survey is to determine attributes of the

good which has to be valued. The attributes must be relevant to the decision problem and each

attribute should reflect independent dimensions to the degree possible to avoid redundancy

(Loviere, 1988). The valuation method requires that one of the attribute be the price or cost of

the good to the respondent (Longo, 2007).

In previous studies researchers have used focus groups or structured conversation with

people who broadly represent the population to be sampled to determine the attributes (Holmes

et al., 2003; Milon et al., 1999). When using focus group the researcher would ask the group a









series of questions aimed at identifying the important attributes of the good in question as they

are considered by the population which is to be surveyed. Some researchers have used

information from the existing literature about the subjects to determine desired attributes

(Makokha et al., 2006) and some have used both methods depending on the research budget and

time constraints.

After identifying attributes the range of each attribute can be determined through the same

procedures as the attributes by using the focus groups, subject expert interviews or literature

reviews. The levels also have to be realistic and should cover the range over which respondents

can have preferences.

When the task of establishing the attributes and levels is completed, presenting all the

combinations of the attributes becomes complicated. Sometimes only a sub-set of attributes

choice combinations has to be presented. The sub-set of combinations needs to be presented in a

statistically representative form to ensure that the maximum amount of information is revealed

by the study without bias. A variety of orthogonal factorial experimental design is available to

reduce and create a balanced sample of possible attribute combinations.

In a full factorial experimental design, a study with five attributes with three levels each

would have 35 or 243 possible combination. With a pair wise choice this full factorial design will

require more that 120 choice decision which is impractical to manage for both the researcher and

the respondent. Researchers have employed software packages for the construction of optimized

fractional factorial experimental designs such as SAS "Factex" procedure (SAS institute) to

identify subsets of possible combinations of attributes and levels that will best represent attribute

preferences with a manageable size (Milon et al., 1999). This technology provides a significant

saving in time and resources while allowing the estimation of all the main effects.









Once the experimental design is created, the next step is to construct the choice sets which

may consist of two or more alternatives. Most CE surveys present three; two alternatives plus the

status quo (Milon et al., 1999; MacDonald et al., 2005; Alberini et al., 2006).The status quo

should be included in the choice set if the purpose of the study is to estimate the willingness to

pay for a policy option (Longo, 2007). The last step in the development of the experimental

design is to choose the number of choice sets to be presented to the respondent and develop the

questionnaire. Complexity of CE surveys is said to increase with the number of choices, the

number of attributes and levels and the number of alternatives in a choice set (Swait and

Adamowicz, 2001). Complicated CE surveys may be tiring to respondents and may lead to poor

quality data. When the survey is completed, the study sample size is given by a number of

respondents receiving the questionnaire times the number of choice sets presented in the

questionnaire.

While each of the mentioned types of MAUM (rank, rate and choice) offers distinct

advantages for measuring preferences, the choice based method has the advantage of reflecting

an actual consumer behavior (Green and Srinivasan, 1978). With choice models it is possible to

partition utility into parts allowing the estimation of the value of individual attributes that make

up the good rather than considering the whole good (Adamowicz et al., 1998). In addition, the

task of choosing the preferred bundle of attributes levels does not require as much effort by the

respondent as do ranking and rating methods (Longo, 2007).

Choice model studies can be achieved with smaller samples relative to contingent

valuation. Studies have been completed with less than 110 interviews (Halbrendt et al., 2007;

Snowball and Willis, 2006) with most studies being completed with interviews between 250 and

500 (Bateman et al., 2002). Furthermore, providing different choice alternatives to the









respondent, which can be accomplished through the attribute based choice modeling methods, is

believed to provide richer information for policy maker's at the end of the research.

Estimation of the attribute coefficients in past studies of CE models was done using

multinomial logit (MNL), conditional logit or probit models (Siikamaki, 2001; Milon et al.,

1999; Makokha et al., 2006). The conditional logit model is useful when the choice probabilities

are functions of the choice characteristics (Maddala, 1983), for instance, when the probability

that the individual chooses to visit a particular park is affected by the characteristics of that park.

The difference between the multinomial logit and conditional logit models is that conditional

logit considers the effects of choice characteristics on the determinants of choice probabilities,

while MNL makes the choice probabilities dependent on individual characteristics only

(Maddala, 1983).

Today, there is an extensive use of choice models, including binary logit/probit, censored

probit, conditional logit, finite mixture logit, group logit, random effects and random parameter

models, nested logit, mixed logit and multivariate probit. Limdep or Nlogit, Stata, Gauss, and

SAS, in that order, are some of the most frequently cited software used in the econometric

estimation of choice models (Zapata et al., 2007).

Survey Method

Available literature suggests that CE questionnaire can be administered using different

modes: mail, telephone, in person, Internet or a combination. Survey modes differ in cost, time,

quality of data, sample control and the quality and amount of information that can be presented

to the respondent. Mail surveys have dominated the data collection in CE studies (Adamowicz,

1994; 1998; MacDonald et al., 2005). This survey mode is inexpensive but has a low response

rate when compared to telephone and in person interviews and it is limited in the amount of

information presented to respondents. Complex scenarios of CE have also been implemented









through in person interviews (Milon et al., 1999; Hanley et al., 1998) or computer-based

questionnaires (Alberini et al., 2006) but the web survey mode has not been widely used.

Web Surveys

Web surveys have recently been recognized as a valuable instrument for collecting data

(Dillman, 2000). The more widespread use of the Internet now makes it possible for more people

to access surveys online. As the Internet access widens to include more representation of the

adult population, the applications for web-based surveys may increase. The rapid development of

web surveys is leading some to argue that soon web surveys will replace traditional methods of

survey data collection (Couper, 2000).

Web surveys are presented in two main formats; interactively or passively. Interactive

surveys are presented screen-by-screen. When the respondent clicks a button like "next", it

allows the data from the question to be immediately transmitted to the surveyor such that

partially completed surveys may be received. Interactive web surveys also allow for

customization as subsequent questions may depend on the answer from the previous question.

However, this format may create difficulties for respondents to review or change their answers

(Couper, 2000).

The passive survey designs involve presenting the entire survey with data transmitted once

the respondent completes the survey and submits the answers. Here, survey respondents can

easily browse through questions and review their responses before submitting. These types of

web surveys are also easy to produce and easy to access with less technical difficulties.

While web surveys may offer some positive opportunities in data collection, its strengths

and weaknesses are still being debated. The strengths and weaknesses of web surveys need to be

recognized to ensure that they are designed appropriately and results are considered carefully.









Advantages of web surveys

In terms of survey administration, web surveys offer several advantages relative to the

telephone and face-to-face interviews. Web surveys have relatively low marginal costs compared

to the other two survey modes which involve the time of interviewers and supervisors. Lower

marginal costs of distributing web surveys and receiving responses makes it possible to have

larger samples for a given research budget. Web-based surveys are self-administered, allowing

respondents to complete the survey at their convenience. This process, besides making Internet

surveying relatively low cost, reduces data entry requirements and eliminates the possibility of

data entry errors (Alvares et al., 2003).

Compared to other survey modes including mail survey, web surveys allow rapid

turnaround, allow access to a vast geographically diverse pool of potential respondents, and has

the superior capability of providing complex and varied information to respondents (Alvares et

al., 2003). Tracking respondents' utilization of information, which is only possible with web

surveys can provide basis for assessing the degree of respondent effort devoted to the survey

(Berrens, et al., 2004).

Weaknesses of web surveys

Making general statements about large populations based on Internet survey results is

currently problematic as this survey mode faces important methodological issues (Alvares et al.,

2003). It is widely agreed that the major sources of error in web surveys include sampling,

coverage, non- response bias and estimation errors. The nature of the samples that it can provide

is questionable because it is difficult to draw representative samples from among Internet users.

Coverage error is the deviation between the sampling frame and the target population.

When surveying a large group, the coverage error becomes a major concern because not all









population members have Internet access and also there is no list of email addresses for the

population.

Non-response bias also occurs in web surveys because some members of the selected

sample are unable or unwilling to complete the survey, but this is not unique to Internet surveys.

However, the potential problem is said to be severe for web-based surveys due to low response

rates (or inability to calculate response rates) and non-random recruitment procedures. In

addition, non-response errors for web surveys may be higher because potential respondents may

encounter technological difficulties with Internet if they have no basic computer literacy skills.

Furthermore, technological hurdles, such as browser incompatibility and slow Internet access

will influence whether a potential respondent completes a survey (Couper et al., 2001).

Although coverage error and non-response bias are a concern for web-based surveys, some

web surveys have performed better than telephone surveys. On the objective measure of election

forecasting in the 2000 presidential election, the Harris Interactive web poll did better than

similar telephone surveys at predicting state level presidential votes (Berrens et al., 2003).

In one study to find out if Internet samples produce estimates of willingness-to-pay

functions comparable to those from the telephone survey, researches presented the results of

parallel telephone and Internet surveys to investigate their comparability. The Internet samples

produced relational inferences quite similar to the telephone sample (Berrens et al., 2003). It was

concluded that all survey methods involve errors and the appropriate question should not be

whether Internet replaces telephone as the mode of survey but rather, under what circumstances

the use of Internet surveys is appropriate.

Two trends that are likely to increase the use of Internet surveys in the future are the

increasing difficulty of doing valid telephone surveys and the increasing representation of the









Internet users. Internet use in the US and around the globe has been growing rapidly, and is

becoming more demographically representative of the population. In 1995, only about 10 percent

of U.S households had access to the Internet but in 2003 about 55% of households had access to

Internet (U.S Census Bureau, 2005). But still the population of adult Internet users in the U.S has

different demographic characteristics than the general population. According to the Census

Bureau (2005), it is on average, younger, more educated, has more males, has people in higher

income groups, and is disproportionately white or Asian.

An additional advantage of internet surveys is its recent use in splits within the Internet

sample to investigate methodological issues. Researchers have used the Internet to find out if the

provision of extensive information related to the policy being evaluated when conducting survey

could influence the research outcome. The generation of large sample sizes in web surveys

permits the investigation of methodological questions within the same survey by comparing the

split-sample treatment effect when estimating willingness to pay (Berrens et al., 2004).









CHAPTER 3
THE MULTI-ATTRIBUTE UTILITY MODEL

This chapter summarizes the theoretical and methodological concepts used in the study.

The Multi-attribute Utility Model (MAUM) is used to determine the relationship between

invasive species and the recreational value in Florida natural areas. MAUM is a choice modeling

method based on the random utility maximization (RUM) theory. The model finds its origins

with Lancaster (1966) who built the conceptual framework for conjoint analysis by clarifying

that utility is gained from the attributes of a good rather than the good itself. For example, the

actual source of the individual's utility when engaged in recreation at a state park are the park

attributes like facilities and activities provided at the park and the variety of animals and plant

species available to see at the park among other things. If an individual is presented with a

number of alternatives to choose from, it is assumed that utility is linear in parameters such that:

Uj = Ikp=lkXk +j (3-1)
where Uj is the indirect utility associated with alternative, pk is the preference parameter

associated with attribute k, Xjk is the attribute k in choice and e is the random error term.

In a multi-attribute choice setting, a person compares the attributes of the alternatives and

would select the alternative that provides the maximum utility. Suppose an individual was faced

with a pair wise attribute setting to choose alternative A or B. If an individual choose A with its

set of attributes over B, then to that individual utility from choice A, is greater than utility from

choice B; in symbolic terms:

U(XA) > U(XB) (3-2)

RUM models assume that utility is the sum of the deterministic component v(.) and a

random component s, such that:

U(X) = v(X) + s (3-3)









The error term is introduced for estimation purposes and it is assumed that it comes from

omissions of explanatory variables, random preferences and errors in measuring the dependent

variable. The error term in the model allows probabilistic statements to be made about the choice

behavior. Assumptions made on how the error term is distributed result in different choice

models.

In a pair wise choice setting for any given respondent i, the probability that the respondent

will choose XA over XB equals the probability that the difference between the deterministic

components exceeds the difference between the random components.

P (A) = P[vi(XA)- vi(XB) > (iB CiA)] (3-4)

From the above foundation, we use attributes which are the basic components of an

individual's indirect utility or preference function. Since the attributes are built with levels, a

preference function relates the level of each attribute to utility as independent dummy variable

making a part worth utility model. So as the respondents make their choices between the

alternatives, the utility associated with changes in levels of specific attributes can be estimated.

The additive linear function produces the main effects of the model and the effects indicates how

utility is affected by the level of the attribute when it is separated from all other attributes.

From the utility function in equation 3-1, the choice behaviors for a respondent from any

set of choices are predicted with the conditional logit model (McFadden, 1974) and the

preference parameters in the equation are estimated. With the conditional logit model the

probability values are estimated under the assumptions that E is independently and normally

distributed2



2 When the choice sets has more than two alternatives, it also assumes that the ratio of probabilities between any two
alternatives is unaffected by other alternatives in the choice set.









For each set of alternatives, the probability that an individual chooses A over the other

alternatives is expressed as:

Pe ( p 'X) (3-5)
P (A) = e e(8 X
J
For a pair of alternatives A and B the probability that an individual chooses A over B is

expressed as a logit function:


log A = 8='(XA B ) (3-6)

where XA is the vector of values of the attributes in the choice A and XB is the vector of values

of the attributes in the choice B.

The other dimension to the model is to evaluate if socioeconomic characteristics like age,

income and education influence the attribute weights or preference variation within the

population. In choice models, these variables are not examined directly because they do not vary

across alternatives (Holmes et al., 2003). In order to account for their differences on preferences,

socioeconomic characteristics are incorporated in the model through interaction terms with the

attribute level variables:

U,= P'X,, + a',S, + -,j (3-7)
where U,, is the indirect utility for an individual i associated with alternative, S, are individuals

socioeconomic variables interacted with attribute levels 's; X, is the vector of values of the

attributes in thejth alternative, /'and a, are vectors of coefficients to be estimated. The random

utility model estimates the probability that utility of the ith individual derived from thejth

alternative is greater than the utility from the other alternatives. From equation (3-5), with the

social economic variables, the probability that the ith individual makes thejth choice will be

estimated as:









(f 'X,+a' s,)
P (3-8)
J I (p 'X k+a'k S,)
k= 1

With the utility function defined, we can model the choice as the relative differences in

utility (Darby, 2006). The difference between choice A and choice B with the social economic

variables included is:

dUIAB PAX + aAS + AB (3-9)
where dU'B is the utility difference between choice A and choice B; AS = Si (xjA XjB); Ax ( XjA

- XjB); and SAB = (eA- eB).

Since each alternative has the price or cost as one of the attribute, the marginal willingness to pay

for each attribute can be defined as:


MWTPk (3-10)
P,
Where MWTPk is the marginal willingness to pay for the attribute k, 8k is the preference

parameter for the attribute k and ,8 is the price or cost parameter.









CHAPTER 4
DATA COLLECTION AND RESEARCH METHODS

Surveys

This chapter summarizes the survey design and the Multi-attribute Utility (MAU) survey

development procedure. Since this study was part of a broader study which was to determine the

effect of invasive plants in both aquatic and upland parks, initial preparations and development

of the MAU survey was generalized to include three types of parks. The park types were ocean

and beach, river and lakes, and wooded parks. After the MAU survey instrument was developed,

the surveys were administered separately for each of the three park types. Data analysis was also

conducted by park type and this part of the study is only dealing with wooded parks. We used

interactive online surveys to gather the data from Florida residents.

The study started in fall of 2006 with a survey of Florida state park managers. Then based

on park managers' responses, two groups of Florida residents were surveyed to determine natural

areas relevant recreational attributes and the level of knowledge regarding invasive species in the

state. Finally the MAU survey was developed using the information from these three surveys as

well as from literature review and expert interviews. The Florida residents' questionnaires for the

two preliminary surveys were designed using Survey Monkey software package and sent through

emails with Expedite Marketing Survey Company.

Before being administered, the questionnaires were reviewed and approved by the

University of Florida Institutional Review Board (UF-IRB) for compliance with ethical standards

for human subject research. The surveys included a short statement on the purpose of the

questionnaire, the importance of providing information, completing instructions, confidentiality

guarantee and contact information in case the respondents had any questions.









The knowledge assessment survey was sent to 40,000 and the recreational attribute survey

was sent to 80,000 Florida residents. Response rates for the two surveys were less than 1%. The

invasive species knowledge survey had a 0.82% while the attribute survey had a 0.37% response

rate. Despite low response rates, the number of responses received was sufficient to make the

required analysis. The details for each of the mentioned surveys are presented below.

Park Managers' Survey

First, questions were sent through email to 159 Florida state park managers asking for

important recreational attributes and the situation of invasive species in Florida parks.

Specifically, managers were asked the type of their park (wooded, ocean and beach or river and

lake) and what were the valuable characteristics of the park. If they had any invasive plants

problems they were asked to name the species. Managers were also asked to indicate any

complaint that park visitors had related to park attributes and if any were related to invasive

plants. In order to determine the park priorities indirectly, park managers were asked if they had

$200,000 to spend, how they would spend it to improve the park. Managers were also asked if

they felt that invasive species had impaired natural areas or diminished its recreational use. This

survey revealed that park visitors care most about the following attributes:

* To be able to see native plants in the park

* To be able to see a variety of animals in the park

* Visitors are bothered by congestion in the parks ( if there are too many visitors)

* Park visitors care about services and facilities availability and condition

* Cost for the trip

* Distance from home

From this survey it was indicated that the most important upland invasive species in the

parks are Brazilian pepper (Schinus terebinthifolius), Cogongrass (Imperata cylindrica),









Australian pine (Casuarina species), Chinese tallow (Sapium sebiferum) and Japanese climbing

fern (Lygodiumjaponicum). Since the MAU survey was to be developed for both the upland and

aquatic recreational parks, invasive aquatic plants were also mentioned. They included Hydrilla

(Hydrilla verticillata) and Water Hyacinth (Eichhornia crassipes).

Park managers also revealed that generally invasive species do not seem to affect the

satisfaction of park visitors and some visitors even like invasive species like Australian pine for

shade. According to park managers only few people who are educated about invasive species are

concerned about these species. Park Managers know about the environmental impact of invasive

plants on natural areas but control of these species is not their priority. If given some extra funds,

most managers' priority was to improve park facilities arguing that the state government is

already taking care of invasive species in their respective parks.

The six recreational attributes identified from this survey were used in constructing the

attribute selection for the Florida residents' surveys. The type of invasive species reported in the

parks and the managers' indication of low knowledge of invasive plants among Florida residents

were used in the development of an invasive species knowledge assessment.

Florida Residents: Invasive Species Knowledge Survey

Based on the invasive plants information above, a survey was created to determine the

level of awareness and knowledge for invasive species among Florida residents. With the study

objective focusing on invasive species and recreation, the level of knowledge of invasive species

is crucial in designing the background information for the MAU survey and Model.

In the survey, respondents were asked how they characterize their knowledge on invasive

species on a 1-5 Likert scale where 1= None, 2= Little, 3=Modest 4= Well versed and 5= Expert.

Out of 319 respondents, a total of 146 (46%) could be classified as having no knowledge, 132

(41%) had moderate knowledge and 41 (13%) were very knowledgeable about invasive species.









In order to determine whether residents had the actual knowledge on invasive species, a

simple test was given to respondents. For respondents who said they had the knowledge, a

twelve plant quiz with pictures and names was administered and they were asked to identify

which plants were invasive.

The knowledge level was analyzed using a scoring method. Scoring was based on the

correct answers out of twelve questions. Those scoring > 8 were categorized as experts. Scores 6

and 7 were grouped in moderate knowledge and scores <5 were grouped into the no knowledge

category. A total of 254 respondents participated in the test. From the results, 161 (63.4%) had

no knowledge on invasive species, 64 (25.2%) had moderate knowledge on invasive species and

only 29 (11.4%) were very knowledgeable about invasive plants. By combining the number of

people with moderate knowledge with the number of those very knowledgeable about invasive

species, we arrived at the conclusion that 36.6% of Florida residents have knowledge on invasive

species. Results on invasive species knowledge among Florida residents are presented in Figure

4-1. From the test we determined that the actual knowledge respondents had was different from

what the respondents said they knew. For respondents who said they were well versed or experts

on invasive species, there was not much difference between what they said and the test score

results (13% vs. 11.4%). There was, however a big difference for the moderate knowledge group

(41% vs. 25.2%) and the no knowledge group (46% vs. 63.4%) with respect to what they said

and the test results.

In three separate questions respondents were also asked to indicate whether they felt that

invasive species had affected natural areas, their enjoyment in outdoor recreation, or if these

species may influence the choice of destination for outdoor recreation. The answers in each of

the three questions was to choose from Yes, No or Not sure. Results for the perception of









invasive species in the three aspects are presented in Figure 4-2. Overwhelmingly, 82% of the

respondents were aware that invasive species affected natural areas and the environment but less

than 30% indicated that the presence of these species in a park would interfere with their outdoor

activity enjoyment or influence their choice on which park to visit. This observation could be

related to park managers' indication that visitors are not bothered by invasive plants when

participating in park recreational activities.

The last section of the survey gathered demographic information about the respondent. In

addition, respondents were asked to what extent they were environmentally conscious on a 1-5

Likert scale where 1= Not at all, 3= Moderately conscious and 5= Extremely conscious. Over

50% of the respondents said they were moderately environmentally conscious.

Florida Residents: Attributes Selection Survey

The main objective for this survey was to determine important recreational attributes for

Florida natural area users. In the survey, respondents were asked which nature related outdoor

activities they participated in during the past twelve months. The activities choices included bird

watching, backpacking, camping, hunting walking, hiking and many more. These activities and

the level of participation for the respondents are presented in Figure 4-3. According to the

survey, most people participated in walking, hiking and running, followed by swimming in the

ocean then nature watching or observation.

Respondents were also asked their reasons to participate in outdoor recreation from a list of

suggested reasons in order to find features which contribute to utility in outdoor recreation. The

suggested reasons are listed on Table 4-1 as they were ranked by the respondents. The top

reason was to experience nature, followed by exercise and enjoying with friends and families.

In order to make the MAU survey manageable not all park attributes could be included in

the model, so we had to select the most important. Respondents graded the importance of six









recreational attributes when visiting wooded parks. The attributes graded in this survey were

determined through park manager's survey results. They included park fees, facilities' condition,

congestion at the parks, animal and plant species diversity and travel distance to the park.

Results are shown in Figure 4-4 revealing that the three most important attributes in

Florida upland natural areas recreation were: Plant Species diversity, Animal Species diversity

and Facilities condition. However, all the six presented attributes were important as each one of

them was chosen by over 50% of the respondents.

An indirect evaluation of important attributes of outdoor recreation was included by asking

respondents how should the state government invest more or make some improvement in Florida

parks. Again, results from this question reflected that important attributes were animal and plant

species diversity and facilities (Figure 4-5).

Building the Multi-Attribute Utility Survey

As mentioned before, choice Modeling or Multi-attribute Utility Model development

required the identification of relevant attributes and levels. This was the reason for the above

preliminary surveys. From the knowledge survey, it was determined that only 36.6% of Florida

residents have knowledge about invasive species. From the attribute selection survey, the three

most important attributes were animal species diversity, native plant species diversity and

facilities condition. In addition to these three attributes, the extent of invasive species in the

parks and park fees were included as attributes. The fee attribute was added for the purpose of

determining implicit prices or marginal willingness to pay for the attributes. The invasive species

attribute was added because the main purpose of the study was to determine its relationship with

outdoor recreational utility.

Park facilities condition was defined to include parking lots, boat docks and ramps, picnic

tables, restroom, showers among other things. Diversity of animal species was defined to include









wild birds, animals and fish. Fees included fees for admission, parking, camping among others.

Presence of invasive species was defined as all non-native plants known to disrupt ecosystem

process. Diversity of native plants was defined to include all the plants which are indigenous to

Florida.

After deciding on the attributes, levels were determined for each attribute. Levels for plant

and animal species diversity were low, moderate and high. Levels for facilities were minimal,

adequate, and excellent. The levels for the invasive species were none, few and dispersed,

numerous and dense. The fee levels were $0 (free), $10, and $20. The levels were determined

from the survey information both from park managers and residents. For example, park

managers indicated that invasive species in most parks are few and dispersed and over 53%

residents indicated that they spent less than $10 on park fees for each visit at a wooded park.

Park fees were also reviewed from the Florida Division of Recreation and Parks web site for the

159 state parks. From this information the levels were created to provide distinct variations in

order to identify the preference parameters.

Combinations of attributes and levels were to be grouped by selecting one level from each

attribute and combining across attributes. We had five attributes at three levels each resulting in

35 = 243 profiles. As already mentioned in chapter two, survey with this many profiles would be

complicated for both the respondents and the researcher. Therefore, the study was limited to four

attributes with three levels each, which form 34 = 81 profiles. In order to confine the attributes to

only four, two surveys were developed to separate the plant and animal species attributes. The

rest of the attributes were the same in each of the two surveys. The other reason for separating

animal and plant species diversity is that these two attributes could be highly correlated. If two

correlated attributes are treated as independent in a choice experiment, respondents might









become confused and fail to answer the questions (Holmes et al., 2003). So it is advised to select

attributes that represent separate dimensions of valuation problem. Eighty one profiles were

reduced by using a fractional factorial design, selecting a sample of attribute levels from a full

factorial design using SAS Factex procedure. The samples for the MAU questions for both the

plant (WPS) and animal species (WAS) surveys are shown in Figure 4-6. A choice of the park

included four attributes at one of the three levels in each attribute. In the end, each of the animal

and plant species diversity surveys had seven choice questions each composed by two park

options.

It is believed that choice alternatives should include the "neither" or the "status quo"

option because in most real world choice situations individuals are not in a" forced choice

situation" (Holmes et al., 2003; Blamey et al., 1997). However, this study presents one of the

circumstances when these options are considered not to be realistic choices. This is because our

research is interested in the trade-off of attributes mainly in invasive species as an attribute in

outdoor recreation and how the trade-off is made between the invasive species and other

recreational attributes (Snowball and Willis, 2006; Alberini et al., 2003). As previously

mentioned, we are not attempting to analyze policy options or estimate welfare changes due to

policy changes. The estimation of statewide willingness to pay to control invasive species will be

a component of this study only for the purpose of assessing the attribute trade-offs. The

willingness to pay to reduce invasive plants in natural areas will be compared to the willingness

to pay to improve other park attributes. This information will provide an insight on the value that

respondents place on each attribute.









Furthermore, in Florida there are 159 state parks, over 7,700 big lakes, 2,276 miles of

shoreline, over 663 miles of beaches (FLDEP, 2006). It will be unrealistic to generalize the state

of rivers and lakes, ocean and beaches and wooded parks to create a status quo option.

The survey instrument (Appendix A) had three sections: (1) the first section presented the

MAU choices after eliminating respondents who were not Florida residents; (2) the next section

asked questions about attitudes and personal experience with invasive species; (3) the last section

asked questions to elicit some demographic information.

The MAU survey included a brief description of the study, potential problems with

specific invasive plants from the biological/ecosystem perspective and photos depicting invasive

plants in natural areas (Australian pine, Brazilian pepper). Because it was determined that more

than 60% of Florida residents had no knowledge on invasive species, this information was

necessary to ensure common interpretation and understanding of the subject among the

respondents. The survey also included park pictures and the activities a respondent could find at

the park. For The MAU section, we asked the respondents to assume that each of the two park

choices are (1) the only alternatives (2) the same distance from the respondents home and (3)

both parks offer the same described activities and facilities. The attributes were defined in details

to make sure they were well understood by the respondents.

Before sending the survey to respondents, it was pre-tested through a series of trials among

242 UF students and four academic staff to identify any problems with the length, content and if

it will be easily understood by the respondents. Appropriate corrections were made accordingly

following the given comments or suggestions.

Based on the available literature that choice analysis studies can be conducted with

relatively small samples (Snowball and Willis, 2006; Bateman et al., 2002), it was decided that a









sample size of at least 400 complete responses for each survey would be sufficient. The surveys

were electronically distributed at www.surveymonkey.com to Florida residents in May 2007

through a marketing company (www.zoomerang.com) that had a good reputation with web

surveys. For each survey, 6,655 emails were sent to solicit participant. With this company,

responses were even higher than anticipated as each survey had over 700 responses only a few

days after the survey was launched. The survey hosting company provided results for each

respondent in a database which could be accessed anytime for analysis.

Preference parameters ac and 3 in the model equations were estimated with the conditional

logit model for the samples separately. The study estimated the basic models first for both plant

and animal species diversity surveys. By basic models we mean that the estimations were made

for the attributes without the social characteristics interactions. These models were specified with

and without the intercept parameters to test if there was an order bias in respondents' choice

between alternative parks (Milon et al., 1999).

After estimating the basic models, the study estimated the demographic characteristics and

invasive species attitude variables as they interact with the defined recreational attributes. Using

the estimated parameters, Marginal Willingness to Pay (MWTP) and the relative weights of the

four attributes were determined. MWTP for the attributes were also calculated for the three

Florida regions: North, Central and South Florida.

We also conducted a brief analysis of the combined data for the animal and plant species

diversity only with recreational attributes (without social interactions). This was done to compare

if the results for the separated models had any relationship with the results when the two models

were combined. We estimated the preference parameters, the MWTP for the attributes and

determined the relative weights of the attributes for combined data.














Expert


0,


0
c Moderate
4-
0


-1
None


20% 40% 60%

Percentage of respondents


mWhat they know n=254
mWhat they say n=319






80%


Figure 4-1. Level of invasive species knowledge in Florida


Enjoyment


n= 284


Choice of
Location



Natural Area



0% 20% 40% 60% 80%
Percentage of Respondents


o Not Sure
SNo
* Yes


100%


Figure 4-2. Perceived impact of invasive species on enjoyment and other aspects-(preliminary)











Hike & walk
Sunning
Nature watching
Photographing
Boating
Swimming
River & Lake fishing
Bird watching
Wake bord
Surfing n=282

Camping
Tubing
Biking
Sports
None
Hunting
Off-roading

0% 10% 20% 30% 40% 50% 60% 70%
Percentage of participants

Figure 4-3. Natural areas outdoor activities participation



Table 4-1. Reasons for participating in outdoor recreational activities
Reason n=240 Response %
To enjoy and experience nature 87.50%
To spend time with family 55.00%
To meet friends 41.20%
To be with people with similar interests 24.60%
To get exercise and improve health 74.60%
To meet new people 12.10%
To share knowledge and skills 18.30%
To be engaged in thrill situations 13.80%
Other 8.30%












n=254


10%
0%
Animals Facilities Plants Congestion
Attributes
Figure 4-4. Outdoor recreation attributes selection


100%
90%
80%
70%
60%
50%
40%
30%


n=251


Animals Facilities


Plants Congestion Fee
attributess


Figure 4-5. Suggested park attributes for the state to improve


Distance


40% -
35%
30%
25%
20%
15%
10%
5%
0%


I~


I












Minimal Adequate A


Moderate High

Numerous and
Few and dispersed e n
dense

$10 $20

Which of the two parks do you prefer?
O Park A O Park B





Minimal Excellent B


Low High


None Few and dispersed


Free $20

Which of the two parks do you prefer?
O Park A O Park B

Figure 4-6. Example of the MAUM questions A) plant species B) animal species









CHAPTER 5
SURVEY RESULTS

Introduction

The primary purpose of the survey was to determine the implicit value of invasive plants

on the value of outdoor recreation. The survey also aimed to collect general demographic and

invasive species attitude information to better understand the factors influencing choice

preferences and the WTP to reduce invasive species in natural areas.

The two surveys differentiated by animal and plant species diversity had the same format.

The animal species sample (WAS) had 640 respondents and the plant species sample (WPS) had

648 respondents. These complete responses were about 89% for each survey after leaving out the

incomplete responses. About 10% of the respondents skipped the MAU questionnaire and 12%

skipped the demographic questionnaire with most of them skipping the household income

question (13.5%). The surveys had high response rates for an Internet survey, which was 8.49%

for animal species and 8.69% for plant species. Responses mean the number of people in each

sample and we have observations which refer to the total number of choice experiments

completed. Since we had seven choice pairs for each survey we multiply the number of

responses by seven leading to a total of 4480 observations for animal species and 4536 for plant

species survey.

Respondents' Profiles

Overall, the surveys had over 55% respondents from Central Florida, about 24% from

South Florida and only 20% from North Florida. Social economic and demographic profile of

respondents differed from Florida state demographics (Table 5-1). A comparison of these

characteristics with the 2000 US census revealed some potential non-coverage bias3. Relatively,


3 Non coverage bias is the deviation between the sampling frame and the population.









the preliminary surveys samples represented the state's demographic characteristics better than

the main survey except for education and the high income groups.

Selected characteristics of the sample population compared to the state profiles are

presented in Figure 5-1 to 5-5. In relation to location, one would think that because we used

Internet survey the samples would be skewed towards more urban residents. Surprisingly, the

samples were skewed in a different direction with more people from suburban and rural areas

above the state representation (Figure 5-1). In both samples 55% of respondents were from sub

urban population.

Respondents were overwhelmingly female, with 66% females in animal species survey and

64% in plant species compared to the state statistics which is 51% female (Figure 5-2). Majority

of the respondents were also mature people between age 46 and 65 who accounted for over 50%

of respondents compared to about 11% of people under 35 years old (Figure 5-3).

About 30% had an annual taxable household income between $35,000 and $60, 000. A

total 6% had less than $15,000. The survey was an exact representation of people with over

$150,000 taxable income in the state which is 4% of the population (Figure 5-4).

The samples had more highly educated people (Bachelors degree and above about 40%

respondents) compared to the state (22%). The population with at least some college classes was

well represented compared to people with high school education and college degrees (Fig 5-5).

Invasive Species Knowledge

Respondents were asked to indicate how much knowledge they had on invasive species

prior to our survey. This knowledge assessment results revealed the results close to the

preliminary survey (11% vs. 13%) for people who said they were experts on invasive species

(Figure 5-6). The moderate knowledge group results in this survey were higher and the no

knowledge group number was lower compared to the preliminary survey results on knowledge.









Respondents were asked to give their opinion as to whether invasive species have effect on

the parks and would affect their choice, enjoyment and frequency of visit to parks. This was done

on a Likert scale of 1-5, where Strongly agree=l, Undecided=3, and Strongly disagree =5.

Results for the two samples were almost the same with less than 30% (cumulative of strongly

agree and somewhat agree) of the respondents agreeing to the effects. On the same cumulative

basis, over 15% respondents said that invasive species could benefit Florida parks (Figure 5-7).

Respondents were asked to indicate whether they have taken any action in response to

invasive species threat. Examples of actions against invasive species were defined to the

respondent .They included helping to remove invasive species from natural areas, travel farther

to visit an alternative location with fewer invasive species and donating money or supplies to

help remove invasive plants in natural areas. Less than 16% of residents in the state have taken

some action against invasive species (Figure 5-8). For the respondents who had taken action,

over 50% participated in removing invasive plants in natural areas and some respondents said

that they have participated in removing invasive species in their communities. Some

communities have organized volunteer sessions to combat invasive plants. Here in Gainesville,

one such event is organized annually and this year it took place on January 27. Over 1200

volunteers gathered to collect the exotic invasive plant, Air potato, (Dioscorea bulbifera) which

is threatening native plants in the area (The alligator 01/29/2007).

Respondents were also asked how frequently they have participated in nature related

outdoor activities in the last twelve months (Figure 5-9). Although more than 35% of the

respondents did not visit the park at all, over 15% visited the state parks at least once a year.

Cumulatively, over 65% of the respondents visited Florida state parks last year. On average,

survey respondents visited the park at least once a month (11.11 times a year).










Table 5-1. Demographic profiles for surveys compared to Florida profiles


Survey
Urban
Suburban
Rural
Male
Female
18 25 years
26 35 years
36 45 years
46 55 years
56 65 years
More than 65 years
High School or less
Associate or some college
Bachelor's degree
Advanced degree beyond bachelor's
Less than $14,999
$15,000 $34,999
$35,000 $59,999
$60,000 $74,999
$75,000 $99,999
$100,000 $149,999
More than $150,000
DUS Census 2000


Preliminary Surveys
Knowledge Attributes
49% 47.2%
41.7% 44.1%
9.3% 8.7%
55.3% 49.0%
44.7% 51.0%
15.3% 13.2%
14.3% 12.4%
19.6% 18.8%
29.9% 24.1%
15.6% 17.3%
5.0% 14.3%
4.0% 18.1%
25.7% 17.0%
18.3% 22.4%
52.0% 42.7%
15.1% 22.8%
14.1% 13.1%
13.7% 11.0%
12.3% 10.1%
13.0% 12.7%
12.0% 10.5%
19.7% 19.8%


Main Surveys


WAS
27.2%
55.5%
17.3%
34.5%
65.5%
2.0%
9.7%
20.0%
25.2%
25.3%
17.8%
32.5%
26.1%
24.2%
17.2%
6.1%
18.9%
29.2%
13.3%
17.0%
11.4%
4.1%


WPS
28.0%
55.0%
17.0%
36.0%
64.0%
1.2%
11.2%
19.0%
27.3%
24.8%
16.5%
39.1%
27.6%
19.8%
13.5%
5.9%
21.5%
31.5%
14.0%
13.8%
9.4%
3.9%


60% -

50%Z

40%-

30%- N WAS n=640
*WPS n=648
O Florida
10%

0%
Urban Suburban Rural
Location
Figure 5-1. Location profiles of samples and Florida residents


Florida"
47.0%
44.0%
9.0%
48.8%
51.2%
7.8%
16.9%
20.1%
16.8%
12.6%
25.9%
48.9%
28.8%
14.3%
8.0%
16.3%
28.7%
24.8%
11.1%
8.7%
6.3%
4.1%












100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%


WAS WPS Florida


Figure 5-2. Gender profiles of samples and Florida residents


0%-1--1___ _
18-25 26-35 36-45 46-55 56 65 Over 65
Age
Figure 5-3. Age profiles of samples and Florida residents


SWAS
SWPS
O Florida


m Female
m Male


WAS n=640
WPS n=648


30%

25%

20%

15%

10%

5%













* WAS
*WPS
O Florida


30% .. .
WPS n=648
25% -/

20% -
15%
10%

5%
0%
<$14.9 $15-34.9 $35-59.9 $60-74.9 $75-99.9 $100- >$150,000
149.9
Income
Figure 5-4. Income profiles of samples and Florida residents


50%
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%


SWAS
r WPS
o Florida


High Some Bachelors Graduate
School College

Education Level


Figure 5-5. Education profiles of samples and Florida residents












100%
90%
80%
70%
60%
50% vo None
40% m Moderate
30% m Expert
20%
10% WAS n=640
10%
0% WPS n=648
WPS WAS Preliminary
Surveys

Figure 5-6. Comparison of knowledge of invasive species in three surveys


90%
80%
70%
60%
50%-
40%
30%-
20%-
10%_
0% I
Enjoyment Frequency Choice Beneficial
Effects of invasive plants


WAS Yes
WAS No
o WPS Yes
o WPS No

WAS n=640
WPS n=648


Figure 5-7. Perceived impact of invasive species on enjoyment and other aspects-(main)












90%
80%
S- 70%
0 60%

50%
40% Yes
30% No

20%
10%- WAS n=640
0% WPS n=648
WAS WPS
Surveys

Figure 5-8. Respondents taking action against invasive species






Not at all
Once every 7 to 12 months
Once every 4 to 6 months
Once every 2 to 3 months
Monthly
Weekly n=640
Daily
0% 10% 20% 30% 40%
Percentage of respondents

Figure 5-9. Park visit frequencies









CHAPTER 6
MODEL SPECIFICATION AND EMPIRICAL RESULTS

Model Variables

This chapter presents the empirical results estimated using the conditional logit models as

described in chapter 3. In the model the dependent variable is the respondents choice which takes

a value of Y=I if park A is chosen and Y=0 otherwise. The independent variables include the

park attributes and fee and the respondents' socioeconomic and invasive species attitude

characteristics as presented in Tables 6-1 and 6-2. Socioeconomic variables and invasive species

attitude variables are together referred to as individual specific variables. Invasive species

attitude variables include knowledge of invasive species, if affected or have taken action against

invasive species and if the respondent think that invasive species could be beneficial.

Non-Response Errors Testing

Since we used internet as the mode of data collection, we believe that problems like non

response errors which are common when data is collected on the web, may exist. We checked for

the presence of non-response bias by testing for statistical significance between the distribution

of demographic characteristics in sub-samples of the early and late respondents (Armstrong and

Overton, 1997). The early respondents are the first fifty and the late respondents are the last

fifty. We conducted a chi-square distribution of age, income, education, gender and location for

early and late responses to determine whether the characteristics of early survey occur in the

same proportion as those who answered later. Late comers are thought to resemble-non

respondents so a statistically significant difference between the sub samples will give an

indication for non response bias, which would need to be corrected. Results from these sub-

samples are presented in Table 6-3. The chi square test indicated no statistical difference

between the early and late respondents sub-samples at the 95% confidence level, therefore, no









evidence that non-response bias exists. The ratio of males to females for the early and late sub-

samples was statistically significant at 99% confidence level in the plant species model. The

early respondent group had 40% male while the late respondent group had only 26% males.

Hypothesized Signs on Parameters

The other specification requirements of the model are hypothesized relationships between

the independent variables and the probability of choosing a certain park alternative. Based on

general economic theory, we hypothesize a negative sign to the fee parameter. It is expected that

as fee increases the probability of choosing a particular park would decrease. While invasive

species can provide some positive utility to some individuals, such as in their own yards, this

study sought to aid managers in making decisions regarding ongoing and planned control and

mitigation programs. As such, this study focused on the negative impacts of invasive species. For

this reason it was assumed that invasive plants have negative impact on utility. We hypothesize

that respondents will be more likely to choose parks with less invasive plants giving a

hypothesized negative sign to the invasive species attribute. Therefore, the negative coefficient

indicates a willingness to pay to reduce invasive species. We also hypothesize a positive sign to

facilities, plant and animal species diversity. The interaction variables were not assigned specific

signs. In summary, the models specification for plant species follows:

Y = p0 + 8,Afa + 2Aps 3Ais- 4fe aS, (6-1)

The model specification for animal species:

Y = /, + /Afa + f2Aas- /Ais f4Afe aS, (6-2)
where Y is the respondents choice, Pi and aj are preference parameters to be estimated, Si

represent the individual specific variable interacted with alternative specific attribute Xi (fa, as,

ps, is and fe). The signs on parameters in the specified models indicate the hypothesized sign.









Parameters are estimated with the maximum likelihood procedure for logit model with the

STATA statistical package version 9.

Statistically Significant Individual Specific Variables

Part of the process of selecting a model is to compile the individual estimated coefficients.

A good model will include estimated coefficients that are statistically different from zero

(McDonald et al., 2005). For this reason individual specific variables which play a role in

preference variability had to be determined. Socioeconomic and invasive species attitude

variables were interacted with park attributes and the parameters were estimated. Table 6-4

shows the results of test of significance for these variables when interacted with park attributes in

the plant species model. Variables that were tested and found not significant display an "ns" and

if the interaction was not tested it is indicated by a double dash. The coefficients for the

interaction terms which were found significant in this process are indicated by asterisks (*) for

significance at the 5% confidence level.

In order to maintain consistency, the same model was used for each of the sub-samples.

The same process was completed for the animal species survey and interestingly, the interaction

terms had relatively similar results as the plant species sample (Table 6-5). A complete list of

parameter estimates for the attributes, social economic and invasive species attitude variables is

presented in Table 6-7.

Generally, for both models, the socioeconomic variables were not significant. The invasive

species attribute when interacted with species attitude characteristics were significant and

therefore, included in the final model. The interaction with invasive species resulted into

Knowledge*Invasive species where knowledge was classified as Expert=l, Moderate = 2 and

None =3; Affected*Invasive species where affected =1 and not affected = 0. Taken









action*Invasive species where action =1 and no action =0; Benefit*Invasive species where

benefit =1 and no benefit = 0.

Specification

Returning to the theoretical model described in chapter 3 and the four parks attributes, the

utility functions for the two surveys are:

UWPS =fo + Pfa + j2ps + A3is + ffe + E (6-3)

UWAS = fo+ ffa + 2as + P3is + f4fe + (6-4)

For each survey the utility difference between choices with invasive species attitude included:

dUAB-= PAX+ aAS + 4AB (6-5)
where dU'A = the utility difference between choice A and choice B; AS = Si (xjA xjB);

Ax ( XjA XjB); and EAB = (e,A- eB). Ax can be defined as changes in attribute levels as follows:


AFacilities = faA-faB (fa ranging from 0-2)

ANative Plant diversity = pSA-pSB (ps ranging from 0-2)

AAnimal species diversity = aSA-aSB (as ranging from 0-2)

AInvasive species =isA-iSB (is ranging from 0-2)

AFee =feA-feB ( fe ranging from $0-$20)

where A and B represents park choices for example fa =0 if park A has minimal facilities

and fa=l if the facilities were adequate as shown in Table 6-1. An Example of AS will be Si(isA-

isB) for change in invasive species level as it interacts with one of the invasive species attitude

variables from the respondent for example invasive species knowledge. The determined utility

difference goes into the logit function for the estimation of preference parameters.









Results


Overall Model Fit

The estimated models based on the animal and plant species samples are presented in

Table 6-6. The focus is on the probability of choosing an option with the underlying theory that

maximum individual's utility will determine which option they choose. The probability of

choosing an option is a function of facilities condition, species diversity ( both with positive

relationship), invasive species and fee (both with negative relationship).

In models of choice, R2 for good models ranges from 0.2 to 0.4 (Louviere et al., 2000).

From the likelihood ratio index (Pseudo R2 or McFadden's R2), both models do not appear to fit

the data well as the McFadden's R2 was 0.0346 for the WPS model and 0.0226 for the WAS

model. However, for large numbers of observations, a model can fail goodness of fit test but still

be adequate for practical purposes and it is possible to focus on estimation rather than hypothesis

testing (Agresti, 1996). So the analysis was conducted on this basis because we had about 4500

observations for each sample. The plant species model relatively has a higher prediction rate.

This might be an indication that there is less preference variability within the respondents about

plant specie diversity in natural areas than it is for animal species.

Attribute Variables

The parameter estimates for the variables in the two empirical models are included in

Table 6-7. The conditional logit models indicate that each of the attribute parameter estimates

were significant with the hypothesized sign. Fee and invasive species had negative signs and

facilities and species diversity had positive signs. The intercept was not significant so the results

interpreted are from the models repeated without a constant.

Comparison of the parameter estimates across the models allows us to draw conclusions

about the relative importance of these attributes for the two samples. For both samples the









invasive species variable shows a negative coefficient that is larger in magnitude than any other

attribute. This indicates that in choosing a park, respondents were more sensitive to changes in

invasive species in natural areas than any other factor in the models.

Probabilities and Marginal Effects for the Models

The sensitivity for choice preference can be further explained by the marginal effects of the

attributes to the probabilities. Using the plant species results as an example, we calculated the

probabilities of choosing park A for the seven choices which were presented to the respondents.

The choice with the highest level of invasive plants and fee had the lowest probability while the

choice with the lowest level of invasive plants but free has the highest probability as shown in

Table 6-8.

It is recommended that parameter estimates from choice models should be transformed to

yield estimates of the marginal effect, that is, the change in predicted probability associated with

changes in the explanatory variables (Green, 2003). The marginal effects are nonlinear functions

of the parameter estimates and the levels of the explanatory variables, so they cannot generally

be inferred directly from the parameter estimates. The marginal effect in this study as they

compare to the actual parameter estimates are presented in Table 6-9.

The plant species model shows a marginal effect of -0.127 for invasive species compared

to 0.079 plant species diversity. In the animal species model, the marginal effect of invasive

species is -0.114 compared to 0.094 for animal species diversity. These attributes were valued

more than facilities and fee. Based on estimated coefficients and marginal effects, overall fee

was the least important attribute in impacting the probabilities but it plays an important role in

determining the willingness to pay as will be reflected in the next section.









Marginal Willingness to Pay Measures

Willingness to pay was another goal of this study and its correct interpretation from the

choice experiment could benefit the state in planning invasive species management programs.

However, the CE in this study presented the respondent with hypothetical choice sets without the

"Neither" or "status quo" option. This forced choice exercise is said to limit the usefulness of the

CE methodology as we can only estimate the marginal price of the attribute but not the net

willingness to pay for different choices (Longo, 2007). Therefore, the MWTP for attributes in

this study are simply the implicit prices and the choices are only ranked by using the utility

scores.

The study had four attributes in each survey and the MWTP or the implicit prices were

determined for each attribute for the two surveys. The implicit price of the attribute was made on

the basis of "ceteris paribus"4 assumptions. By factoring out the marginal utility of money

(Holmes et al., 2003), the coefficients for the attributes can be translated into the implicit prices

for attribute (Table 6-10) as follows:


Implicit Price attnbute (6-6)
fifee

The marginal utility of money is simply the negative of the fee coefficient. Since we

assumed linear relationship in attribute levels, with three levels of invasive species, a change

from numerous and dense to few and dispersed in the WPS survey is valued at $6.85 which is the

same value for moving from few and dispersed to none.








4 With all the other attributes remaining at the mean levels.









The implicit price was the highest for invasive species in both models followed by species

diversity (Figure 6-1). Implicit prices were slightly higher for all the attributes in the animal

species model. Facilities implicit prices ($4.49 for WAS and $4.13 for WPS) were the lowest.

Marginal willingness to pay and individual specific variables

Although demographic characteristics per see (gender, education, age, income) overall did

not cause variations in choice preferences when interacted with attributes, some special

characteristics of the respondent had effects on estimated parameters and MWTP for the invasive

species attribute. These characteristics included knowledge of invasive species, if the person has

been personally affected or participated in taking action against invasive plants. People with

these characteristics were willing to pay more to reduce invasive species. On the other hand,

respondents who claimed that invasive plants were beneficial for the natural areas were willing

to pay less compared to those who did not think these plants were beneficial (Table 6-11).

Marginal willingness to pay by regions

The MWTP or implicit price for invasive species was further analyzed by regions for

North, Central and South Florida. We asked each respondent to indicate which County they

reside. Using the results from this question we grouped the 68 state counties in their respective

regions and generated MWTP estimates for attributes by regions. South Florida had the highest

implicit prices for all the attributes in both the samples (Table 6-12). For the WAS sample, South

Florida was followed by North Florida and Central Florida last (Figure 6-2B). For the WPS

sample, South Florida had the highest price for invasive plants but Central and North Florida

switched places. In the WPS there are no visible variations between regions for the implicit

prices for facilities and plant species diversity but the price difference for invasive species is very

evident (Figure 6-2A). This implies that South Florida region have a relatively higher willingness

to pay to reduce invasive species than the other two regions. In addition, within regions in the









WPS sample respondents appeared to value species diversity and facilities about the same.

Respondents in three regions were practically similar in their preferences. They all had invasive

species with the highest implicit price, followed by species diversity and then facilities.

Relative Weighting of Attributes

The importance of each of the model attributes had to be expressed in relation to all the

other attributes in the model and how they impact recreational utility. The normalized weights

for attributes were calculated for this purpose (Figure 6-3 A and B)5. Respondents gave negative

weights to fee and invasive species and positive weights for facilities and species diversity. From

these weights the most important attribute is the fee followed by invasive species. By looking at

Figure 6-3A, the weights for plant species and facilities in the WPS model look the same. Since

native plant species diversity and facilities condition have the same mean level, a test statistic

was conducted for this sample and we failed to reject the null hypothesis that P1 = 32. P1 and 32

represents the attribute coefficients for facilities condition and native plant diversity,

respectively. The same test was applied to the WAS sample for facilities and animal species

diversity attributes but this time we rejected the null hypothesis. Therefore, the ranking for the

attributes based on the relative weights in both models, fee is the most important followed by

invasive species. Native plant species are tied with facilities in the WPS model while animal

species diversity is ranked number three followed by facilities in the WAS model.

Choices Ranking by Utility Scores

As previously mentioned ranking in this study could only be done by utility score, not the

net willingness to pay. The most preferred attribute levels were excellent facilities, high diversity

in plant and animal species, no invasive plants and free. The least preferred combination of


5 The normalized weight were computed by multiplying each attribute coefficient with its mean level and then
multiplying the product times ten (Milon et al., 1999)









attribute levels was the minimal facilities, low diversity in plant and animal species, numerous

and dense invasive species at a $20 fee. Park choices according to utility for all 81 possible

alternatives were calculated. The choices with positive derived utility for both the plants and

animal species samples are presented in Appendix B. Utility scores are 1.33 and 1.246 for the

most preferred and -2.194 and -2.498 for the least preferred for the animal and plant species

samples, respectively. The largest contributor to the utility in the alternative is fee as reflected by

high utility for the alternatives which were free. This is clearly presented in Table 6-13 by

changing just one attribute level in a park choice and calculating the change in utility as a result

of the level change. For example in the WPS model, a change in fee from free to $10 will reduce

the utility by 0.74 units while a change in invasive species from none to few and dispersed will

reduce the utility by 0.51 units.

Results for the Combined Model

Since there was not much differences in the estimated coefficients and MWTP for the two

samples, at the end of the analysis we decided to combine the WAS and WPS samples to get the

results for one model. The MAU survey for the animal species had no native plants attribute and

vice versa for the plant species survey. We made an assumption that for each survey the missing

attribute was the same in the two alternatives. For example, in WPS when a respondent is

presented with a choice set of park A and park B the animal species level is the same in both

parks. Therefore, asA-asB =0 for the plant species sample and psA-psB =0 for the animal species

sample. We estimated the model only for the attributes and the results are presented in Table 6-

14. For the attributes which were in WAS and WPS, their estimated coefficients and MWTP

were between the values for the separated model. For example, the MWTP to reduce invasive

species in the combined model was $6.99 which is between $6.85 and $7.15, the values of

MWTP in WPS and WAS, respectively. The ranking of the attributes based on relative weights









turned out exactly as they were when the models were separated (Figure 6-4). Fee is the highest

ranked followed by invasive species, then animal species diversity. Native plant species and

facilities received the same weight.

Hypotheses Testing

The basic hypothesis in this study is that invasive plants are undesirable and are considered

unsightly in parks and recreational places such that recreation satisfaction in these areas will be

reduced by the presence of invasive species. For this reason natural area users will be willing to

pay to reduce invasive species in outdoor recreation areas. We fail to reject this hypothesis.

Invasive species had a statistically significant negative coefficient indicating a willingness to pay

to reduce invasive species. The MWTP to reduce invasive species was $6.85 and $7.15 per

person for the plants and animal species samples, respectively.

It was anticipated that the willingness to pay for invasive species will not only be because

of the effect of invasive plants on recreational utility but also because of the perceived impact of

these species on the environment. We fail to reject this hypothesis. Our estimates suggest that

invasive species attitude characteristics like the level of knowledge of invasive species and

whether people had taken action against invasive species has an impact on MWTP. For example

people with higher knowledge on invasive plants were willing to pay more to reduce invasive

species than people with less knowledge.

Some socioeconomic variables such as income, age, and education were expected to be

significant determinants of the impact of invasive plants on the recreational value. We reject this

hypothesis. Socio-economic characteristics had no statistically significant impact on park

preferences and MWTP when interacted with the invasive species attribute.

It was also anticipated that since Florida regions are affected at different levels by invasive

plants, willingness to pay will likely be different between regions and perhaps higher for the









most affected region. We fail to reject this hypothesis. MWTP for invasive species differed

significantly between regions and South Florida which is the most affected region had the

highest MWTP to reduce invasive species in natural areas.











Table 6-1. Basic models independent variables
Attribute Levels
Minimal


Facilities Condition


Native Plants Diversity


Animal Species Diversity


Presence of Invasive Species


Fee


Scores
0


Adequate
Excellent
Low
Moderate
High
Low
Moderate
High
None
Few & dispersed
Numerous &dense
$0
$10
$20


Table 6-2. Socioeconomic and invasive species attitude variables


Meaning
Age of the respondent (years)
Gender
Attained level of Education
Employment status
Annual Gross household income
Membership in Environmental org.
Location where respondent live
Region in Florida
Marital status
Took action against Invasive plants
Invasive plants are beneficial
Knowledge of invasive plants
Affected by invasive plants


Response/categories
1-6 (six categories)
1,2 (male, female)
1-4 (four categories)
1-5 (five categories)
1-7 (seven categories)
1,2 (yes, no)
1,2,3 (urban, suburban, rural)
1,2,3 (Central, South, North)
1-4 (four categories)
1,0 (yes, no)
1,0 (yes, no)
1,2,3 (expert, moderate, none)
1,0 (yes, no)


Table 6-3. X2 Tests statistics of first 50 and last 50 respondent characteristics
Location(dfJ2) Gender(df 1) Education(dfJ3) Income(df=4)
WPS X2 4.964 7.040*** 1.974 0.616
p-value 0.084 0.008 0.579 0.961


WAS X2
p-value


0.891
0.641


1.630
0.202


1.827
0.608


6.163
0.188


rI Yates chi-square test used instead of the Pearson chi-square when df= 1.
*** Significant at the 99% level of confidence


Variable


Variable
Age
Gen
Edu
Emp
Inco
Env
Loc
Reg
Mar
Act
Ben
K
Aff


Type
Continuous
Categorical
Categorical
Categorical
Continuous
Categorical
Categorical
Categorical
Categorical
Categorical
Categorical
Categorical
Categorical


Age (dfJ3)
1.896
0.593
5.055
0.168












Table 6-4. Significance test for attribute interaction with individual specific variables-WPS
Variable Fee Plant Sp. Facilities Invasive Sp.
Knowledge ns ns *
Income ns ns
Education ns ns ns ns
Affected ns ns *
Benefit ns ns ns *
Action ns ns ns *
Location ns ns ns ns
Gender ns ns ns ns
Age ns ns ns
Marital status -- -- ns
Environment -- ns ns ns
Employment -- -- -- --
* Significant at 5% confidence level

Table 6-5. Significance test for attribute interaction with individual specific variables-WAS
Variable Fee Animal Sp. Facilities Invasive Sp.
Knowledge ns ns ns *
Income ns
Education ns ns ns ns
Affected ns ns *
Benefit ns ns *
Action ns N* ns *
Location ns ns ns ns
Gender ns ns ns ns
Age ns ns ns
Marital status ns ns -- ns
Environment ns ns ns ns
Employment ns ns ns
* Significant at 5% confidence level

Table 6-6. Overall fit for the multi-attribute models
WPS WAS
No. of Observations 4536 4480
Likelihood function value -2994 -3027
Pseudo R2 0.0346 0.0226
Sensitivity (actual Is correctly predicted) 66.2% 64.3%
Specificity (actual Os correctly predicted) 53.6% 51%












Table 6-7. Coefficient estimate for the multi-attribute models
Plant Species Model (WPS) Animal Species Model (WAS)
Variable Coefficient Std. Error Variable Coefficient
Facilities 0.306* 0.038 Facilities 0.287*
Native Plant Species 0.308* 0.038 Animal Species Diversity 0.379*
Invasive Species -0.497* 0.040 Invasive Species -0.458*
Fee -0.074* 0.006 Fee -0.064*
constant 0.031 0.034 constant -0.002


Facilities
Native Plant Species
Invasive Species
Fee


age*fa
age*ps
age*Is
age*fe

gen*fa
gen*ps
gen*Is
gen*fe

edu*fa
edu*ps
edu*Is
edu*fe

inco*fa
inco*ps
inco*Is
inco*fe

env*fa
env*ps
env*Is
env*fe

loc*fa
loc*ps
loc*Is
loc*fe


0.307*
0.316*
-0.509*
-0.074*


-0.005
0.022
-0.089
0.004

0.016
0.022
-0.002
0.002

0.007
0.025
0.022
0.000

0.040*
0.021
-0.003
0.005*

-0.139
-0.079
0.124
0.002


0.002
0.012
-0.031
0.001


0.038
0.037
0.038
0.006


0.015
0.043
0.016
0.002

0.039
0.043
0.042
0.005

0.017
0.019
0.019
0.002

0.012
0.013
0.013
0.001

0.081
0.090
0.091
0.005

0.028
0.031
0.031
0.003


Facilities
Animal Species Diversity
Invasive Species
Fee


age*fa
age*as
age*Is
age*fe

gen*fa
gen*as
gen*Is
gen*fe

edu*fa
edu*as
edu*Is
edu*fe

inco*fa
inco*as
inco*Is
inco*fe

env*fa
env*as
env*Is
env*fe

loc*fa
loc*as
loc*Is
loc*fe


0.033 0.051 ac*fa


Std. Error
0.038
0.039
0.040
0.006
0.034


0.287*
0.378*
-0.457*
-0.064*


0.028
-0.009
-0.087
0.008

0.039
-0.022
-0.104
0.011

-0.006
0.009
0.001
-0.001

0.065*
0.053*
-0.014
0.009*

0.172
0.165
0.153
0.016

0.010
-0.013
-0.053
0.005


0.038
0.038
0.038
0.006


0.014
0.015
0.016
0.002

0.039
0.043
0.042
0.005

0.016
0.018
0.018
0.002

0.012
0.013
0.013
0.001

0.078
0.085
0.086
0.009

0.028
0.031
0.031
0.003


ac*fa


-0.090 0.052











Table 6-7. Continued
Plant Species Model (WPS
Variable Coefficient
ac*ps -0.066
ac*Is -0.169*
ac*fe 0.010


aff*fa
aff*ps
aff*Is
aff*fe

ben*fa
ben*ps
ben*Is
ben*fe


k*fa
k*ps
k*Is
k*fe


Central Florida
Facilities
Native Plant Species
Invasive Species
Fee


South Florida
Facilities
Native Plant Species
Invasive Species
Fee
North Florida
Facilities
Native Plant Species
Invasive Species
Fee


0.085*
-0.023
-0.237*
0.023*

-0.043
0.090
0.120*
-0.008

-0.021
0.019
0.168
-0.013



0.289*
0.306*
-0.509*
-0.072*



0.287*
0.279*
-0.454*
-0.059*


0.397*
0.402*
-0.592*
-0.102*


Std. Error
0.056
0.057
0.006

0.040
0.044
0.044
0.005

0.052
0.057
0.056
0.006

0.029
0.032
0.032




0.050
0.049
0.050
0.008


0.076
0.076
0.076
0.012


0.089
0.087
0.088
0.015


Variable
ac*as
ac*Is
ac*fe

aff*fa
aff*as
aff*Is
aff*fe

ben*fa
ben*as
ben*Is
ben*fe


k*fa
k*as
k*Is
k*fe


Animal Species Model (WAS)
Coefficient
-0.217*
-0.424*
0.004


0.068
-0.064
-0.359*
0.029

0.057
0.221*
0.242*
0.003

-0.009
-0.008
0.123*
-0.016

-0.024
-0.058
-0.056
0.000


mar*fa
mar*as
mar*Is
mar*fe


Central Florida
Facilities
Animal Species Diversity
Invasive Species
Fee


South Florida
Facilities
Animal Species Diversity
Invasive Species
Fee
North Florida
Facilities
Animal Species Diversity
Invasive Species
Fee


0.260*
0.380*
-0.444*
-0.066*



0.279*
0.355*
-0.390*
-0.048*


0.374*
0.408*
-0.583*
-0.079*


*Indicates Significance at the 0.05 level
Bold Indicates Variables in the Final Model


Std. Error
0.058
0.061
0.006

0.039
0.043
0.044
0.005

0.047
0.052
0.052
0.006

0.004
0.035
0.035
0.031

0.025
0.028
0.028
0.003


0.051
0.051
0.051
0.008



0.076
0.078
0.077
0.012


0.085
0.085
0.086
0.014










Table 6-8. Probabilities for park choices-WPS


CHOICE
PARK A
PARK B

PARK A
PARK B


FA
Minimal
Adequate

Minimal
Excellent


PARK A Excellent
PARK B Adequate


PARK A
PARK B

PARK A
PARK B

PARK A
PARK B


Minimal
Excellent

Adequate
Excellent

Excellent
Minimal


PS
Moderate
High

Low
High

High
Low

High
Moderate

Moderate
High

Moderate
High


PARK A Excellent High
PARK B Minimal Low


IS
Few and dispersed
Numerous and dense

None
Few and dispersed

None
Numerous and dense

Few and dispersed
None

None
Numerous and dense

Few
Numerous and dense

Numerous and dense
None


Table 6-9. Marginal effects for the models
Animal Species (WAS) Plant Species (WPS)
Parameter Coefficient Marginal Effects Coefficient Marginal effects
Facilities 0.287 0.071 0.307 0.076
Species Diversity 0.378 0.094 0.316 0.079
Invasive Species -0.457 -0.114 -0.509 -0.127
Fee -0.064 -0.016 -0.074 -0.019



Table 6-10. Comparison of implicit prices estimates for recreational attributes
Animal Species(WAS) Plant Species (WPS)
Marginal utility of money $0.064 $0.074
Invasive Species $(7.15) $(6.85)
Facilities $4.49 $4.13
Diversity $5.92 $4.25


FE
$10
$20

Free
$20

$20
Free

$10
$20

$10
Free

$10
Free


P(Yi=A)
0.653



0.679



0.615


0.484



0.414



0.516



0.374











$8.00 -
$7.00 -
$6.00 -
$5.00 -
$4.00
$3.00
$2.00
$1.00
$0.00
Facilities Sp. Diversity Invasive Sp
Attributes
Figure 6-1. Implicit prices for recreational attributes


- WAS
SWPS


Table 6-11. Implicit prices to reduce invasive species with attitude variables
Variable WAS WPS


Knowledge
Expert
Moderate
None
Affected
Yes
No
Actions
Yes
No
Benefits
Yes
No


9.43
7.52
5.60

10.86
5.35

12.76
6.22

4.09
7.84


9.63
7.39
5.15

9.00
5.84

8.76
6.49

5.48
7.09


Table 6-12. Regional marginal willingness to pay for the attributes
Animal Species Model (WAS) Plant Species Model (WPS)
North FL Central FL South FL North FL Central FL South FL
Facilities $4.76 $3.93 $5.83 $3.88 $3.99 $4.83
Sp.Diversity $5.19 $5.75 $7.40 $3.93 $4.24 $4.69
Invasive Sp. $(7.42) $(6.72) $(8.15) $(5.78) $(7.05) $(7.63)












$8.00
$7.00
$6.00
$5.00
$4.00
$3.00
$2.00
$1.00
$0.00
North FL
Central
South FL
FL





$9.00
$8.00
$7.00
$6.00
$5.00
$4.00
$3.00
$2.00
$1.00
$0.00
North FL
Central
South FL
FL


m Facilities
m Plant Sp
o Invasive Sp.


r Invasive Sp.
F Plant Sp
Facilities


m Facilities
m Animal Sp.
o Invasive Sp.


A Invasive Sp.
SAnimal Sp.
Facilities


Figure 6-2. Regional marginal willingness to pay A) plant species B) animal species











Fee

Invasive Sp.

Plant Sp.

Facilities

-8


Fee

Invasive Sp.

Animal Sp.

Facilities


miiui


-6 -4 -2 0 2 4
Relative Weights


'I-I-I."


U


2 4


Figure 6-3. Relative weights of attributes A) plant species B) animal species


-8 -6 -4 -2 C
Relative Weights


F- -1


m-












Table 6-13. Impact of change in attribute level on the total utility


Diversity
High
High
High
Moderate
High
High
High
High

High
High
High
Moderate
High
High
High
High


Animal Species
Invasive species Fee
None Free
None Free
None Free
None Free
None Free
Few and dispersed Free
None Free
None $10
Plant Species
None Free
None Free
None Free
None Free
None Free
Few and dispersed Free
None Free
None $10


Utility
1.33
1.043
1.33
0.952
1.33
0.873
1.33
0.69

1.246
0.939
1.246
0.93
1.246
0.737
1.246
0.506


A Attribute
Facilities

Diversity

Invasive Sp.

Fee


Facilities

Diversity

Invasive Sp.

Fee


A Utility
0.287


0.378

0.457


0.64


Facilities
Excellent
Adequate
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent

Excellent
Adequate
Excellent
Excellent
Excellent
Excellent
Excellent
Excellent


0.74


Table 6-14. Coefficient estimates for the attributes (combined model)
Variable Coefficient Std.Err MWTP Rank
Facilities 0.297 0.027 $4.30 4
Animal Species Diversity 0.410 0.030 $5.94 3
Native Plant Species 0.284 0.030 $4.11 4
Invasive Species -0.483 0.027 ($-6.99) 2
Fee -0.069 0.004 1


Choices
1
2
1
2
1
2
1
2


0.307

0.316

0.509










fee

invasive species

plant species

animal species

facilities


-I-I-I."


I


I



U


-8.00 -6.00 -4.00 -2.00 0.00 2.00 4.00 E
Relative weights

Figure 6-4. Relative weights of the attributes for the combined model


00
5.00









CHAPTER 7
INTERPRETATION OF THE RESULTS FOR INVASIVE PLANT MANAGEMENT

Invasive Species and Outdoor Recreation

Regarding the invasive species attribute, based on respondents' interviews, our results are

consistent with some studies that while Florida residents know that invasive plants have a

negative impact to natural areas, the plants do not seem to affect people's enjoyment in outdoor

recreation (Wirth and White, 2006; Finn, 2006). Our preliminary surveys while searching for

attributes also indicated that people do not care about invasive species when engaged in outdoor

recreation in natural areas. However, a study by Adams and Lee (2006) found that the presence

of hydrilla in Florida lakes had a negative influence on angler's decisions to use certain lakes.

But there is no indication of this happening with recreation in wooded parks.

Surprisingly, in our study both the animal and the plant species models found high invasive

species sensitivity in choosing a wooded park for recreation. One explanation is that no previous

work has measured the effect of invasive plants in recreation using a survey based methodology

and analysis. The other studies were not designed to capture the choice behavior of the

respondent as in this study. These past studies analyzed the relationship between park enjoyment

and the presence of invasive species based on what people said and not on how they behaved in

choosing a park with different levels of invasive species.

Estimations of Willingness to Pay to Reduce Invasive Species

Based on the determined implicit price for the invasive species attribute, the statewide

willingness to pay to control invasive species in state parks was estimated. The attribute levels

for invasive plants were "None," "Few and Dispersed," and "Numerous and Dense." As

respondents chose their preferred park, they revealed their average willingness to pay for each

attribute. This included the willingness to pay to reduce invasive species from "Numerous and









Dense" to "Few and Dispersed" and from "Few and Dispersed" to "None." Given a known

current condition of state parks with respect to invasive plants, and with the assumption we made

that the average value associated with each level changes is the same, we estimated the total

willingness to pay to implement an invasive plant control policy that achieves reductions along

these levels.

From an online survey with the park managers to determine the current conditions of

invasive plants in state parks, we can presume that the levels of invasive species in the majority

of Florida state parks are currently "Few and Dispersed"; there were 4% responses indicating no

presence of invasive species, 50% indicated few and dispersed, 33% indicated an obvious

presence and 13% of the park managers indicated (in an open ended question) that the invasive

species problem is being taken care of in their respective parks. Florida has various types of

parks; wooded, ocean and beach and river and lakes. From the 159 state parks, 116 parks offer

hiking, nature trails, biking and horse trails activities and we considered them to be "wooded

parks".

We applied the MWTP estimates using the "Few and Dispersed" level as the status quo

and calculated the range of WTP to prevent invasive species from becoming "Numerous and

Dense" for wooded parks for the entire state. We applied the MWTP measures with the annual

estimated park attendance for the fiscal year 2005-2006 (FLDEP, 2006); the state parks received

about 18.2 million visitors last year with the 116 wooded parks taking 16.1 million visitors. The

number of visitors per year has been fluctuating around 18 million for Florida state parks with

the local visitors estimated at 35% of total park visitors (FLDEP, 2004). For the "low" estimate,

we multiplied the invasive species MWTP by 35% of annual park attendance; which was 5.6









million visitors for the wooded parks. For the "high" estimate, we multiply MWTP by the total

annual park attendance; which was 16.1 million visitors for the wooded parks.

The total amount that Florida residents would be willing to pay to keep invasive plants

from becoming "Numerous and Dense" in Florida wooded parks amounts to $38.7 million per

year in the WPS sample. With the application of the MWTP estimates to 100% of 2005/2006

wooded park attendance for high WTP estimates, we arrive at $110.48 million per year (Table 7-

1). Since the MWTP measures for the WAS survey was relatively higher, the estimates ranges

from $40.4 to $115.3 per year for this model.

We compared the WTP ranges for the plant species model with ranges of WTP for the

other attributes in the same model. Using the same calculation method, the WTP to improve

facilities from adequate to the excellent level ranged from $23.3 to $66.6 million per year while

the WTP to improve native plants diversity from the moderate to high level ranged from $23.9 to

$68.5 million per year. For the animal species model, the annual statewide WTP ranged from

$25.3 to $72.4 million to improve facilities and $33.4 to $95.5 million dollars to improve animal

species diversity. Taking care of the invasive species has the highest range of total WTP than to

improve the other two park attributes.

Since the money could be collected through an increased park entrance fee, we assume that

non-Florida residents have the same MWTP as Florida residents and interpret these amounts as

the ranges of annual WTP by state park visitors to prevent invasive species from becoming

"Numerous and Dense" and to improve facilities and species diversity in the wooded parks.

Also, since Florida's regions are affected at different levels by invasive plants and not all

the state parks have invasive plants, it will be practical to estimate the willingness to pay based

on this reality. From the status quo determination with the park managers, we can assume that









4% of the state parks have no invasive plants, 63% (50% + 13%) of the parks have few and

dispersed invasive plants and 33% have numerous and dense invasive plants. The MWTP also

differed between regions, ranging from $5.78 to $7.63 for invasive species in the WPS sample.

The Florida recreation and park system is divided in five districts which can be classified into the

three regions of the state as follows: the Northwest and Northeast districts as "North Florida", the

Central district as "Central Florida" and the Southwest and Southeast districts as "South Florida"

(Appendix C).

The MWTP from each level of invasive species was calculated by multiplying last years

park attendance for the regions with the regional invasive species implicit price and the

percentage of parks assumed for each level of invasive plants. South Florida, with the second

largest number of parks but the highest attendance and highest MWTP has the willingness to pay

of $20.8 million per year to improve its parks from "Numerous and Dense" invasive species to

the "Few and Dispersed". North Florida, with the highest number of parks but low MWTP has

only $9.2 million WTP per year to achieve the same improvement. The remaining values by

region are in Table 7-2.

Marginal Willingness to Pay and the Invasive Species Attitude

A study using data collected from the National Survey on Recreation and the

Environment reported that 96.7% of the respondents supported user fees or a combination of user

fees and taxes to fund the services (Bowker et al.,1 999). However, the implementation of an

additional park fee for the purpose of invasive management should be considered carefully.

Although people have indicated some willingness to pay to reduce invasive plants through an

increased park usage fee, the amounts they are willing to pay are very low even less than one

dollar per visit. An example of low willingness to pay to reduce invasive species in outdoor

recreation areas in Florida from Finn (2006) is presented in Table 7-3. On average people were









willing to pay less than $4 per visit to reduce Malaleuca in recreational areas. Compared to these

figures our estimate for invasive plants MWTP is relatively high.

An important observation is that although the overall MWTP for invasive species was up

to over $7 dollars for this study, the ranking of park preferences based on utility gives a different

outcome. In both samples, the top preferences that gave the respondent utility greater than

"zero," over 84% of the choices are "free" and only one park was chosen for the $20 fee. This

might be an indication that the increase in park fee may not be received positively by Florida

state park users.

In this study the respondent's invasive species attitude had an impact on the MWTP to

reduce invasive plants. These characteristics included the level of knowledge of invasive species,

if the respondent was affected by invasive species and people who have taken action against

invasive plants. Thus the state could improve the awareness and knowledge of invasive species

among Florida residents to increase the MWTP values. There have been some indications that

Florida residents have interest in receiving information on invasive species (Wirth and White,

2006).

As we calculated the statewide WTP to control invasive plants, we calculated the statewide

WTP changes for a change in knowledge level. We had three levels of knowledge; none,

moderate and expert. Since it was determined that 60% of Florida residents have no knowledge

of invasive species, it could be reasonable for the state to implement an educational campaign

that could give Florida residents at least the moderate level of knowledge and raise the awareness

of the impact of invasive plants. The marginal willingness to pay to reduce invasive species in

WPS was $5.15 and $7.39 for the "none" and "moderate" knowledge, respectively. The different

MWTP were multiplied with the 2006 annual park attendance of local visitors to determine the









change of WPT with the change in invasive species knowledge level. From the WPS model, an

educational program that could increase the knowledge from "none" to "moderate" level would

increase statewide WTP by $12.6 million per year (Table 7-4).

In summary, invasive plant management programs in Florida could achieve the goal of

reducing invasive species through education campaigns on invasive species to Florida residents

in order to increase park users' willingness to spend on park fees, which may in turn be used to

control invasive plants in state parks. In addition, the knowledge and awareness of the impact of

invasive species among Florida residents may lead to increased participation in invasive species

management voluntary activities like the Gainesville community annual event of collecting the

invasive Air potato which was mentioned in chapter four.

The Importance of Invasive Species Knowledge

To assess the importance of knowledge of invasive plants among the state residents on

invasive species management programs, we conducted a small survey on 27 invasive plants

experts as park visitors in Florida at the end of the study. We used the same MAU questionnaires

for this small survey. From this survey, two attributes (facilities and fees) were not significant

while the invasive species attribute was statistically significant (Table 7-5). The invasive plant

experts did not give importance to the park fee and facilities but they ranked the invasive species

attribute the highest among the four attributes affecting recreational utility (Figure 7-1).

In the overall models (WAS and WPS), fee was the highest ranked in impacting

recreational utility. The MWTP to reduce invasive species for the experts was almost three times

as much as the MWTP for the general population ($19.25 compared to $6.85 in the plant species

model). The MWTP to reduce invasive species for these proven experts was twice as much as

the MWTP for respondents who claimed to be invasive species experts ($19.25 vs. $9.63). The

WTP to reduce invasive species from Florida residents who visit state parks varies between









$29.1 million and $108.7 million per year with invasive species knowledge level differences.

This is the comparison of WTP between residents who have no invasive species knowledge at all

and residents who are invasive species experts. From this observation, it is evident that

improving invasive species knowledge could significantly increase willingness to pay for

invasive species management.












Table 7-1. Annual WTP to control invasive plants in Florida wooded parks (million $)
WPS WAS
Attribute Low High Low High
Facilities $23.3 $66.6 $25.3 $72.4
Species Diversity $23.9 $68.5 $33.4 $95.5
Invasive Species $38.7 $110.5 $40.4 $115.3




Table 7-2. Annual regional WTP to reduce invasive plants levels -WPS
Regional WTP to reduce invasive plants
Number Few and Numerous and
Region of parks Attendance MWTP Total Dispersed Dense

North FL 52 4,820,594 $5.78 $27,863,033 $17,553,711 $9,194,801

Central FL 24 3,032,471 $7.05 $21,378,921 $13,468,720 $7,055,044

South FL 40 8,275,978 $7.63 $63,145,712 $39,781,799 $20,838,085


Table 7-3. WTP per visit to reduce Melaleuca in recreational areas by Florida residents
Expense Range Number Percent
$0 275 44
$0-$1 48 8
$1-$4 171 27
$5-$9 58 9
$10-$15 42 7
$16-$25 23 4
$25 + 9 1
Source: Finn, (2005)


Table 7-4. Annual WTP to control invasive species- FL residents with knowledge (million $)
Level of knowledge WPS WAS
None $29.1 $31.6
Moderate $41.7 $42.5
Experts (self declared) $54.3 $53.2
Real Experts $108.7 $108.7


Table 7-5. Coefficient estimate for the attributes (invasive plants experts)
Variable Coefficient Std. Error P>lzl Relative Importance MWTP$
Facilities 0.170 0.230 0.460 4 3.82
Species Diversity 0.513 0.279 0.066 2 11.52
Invasive Species -0.856 0.280 0.002 1 -19.25
Fee -0.044 0.038 0.244 3
















Fee


Invasive Sp.


Sp. Diversity


Facilities

-10 -8 -6 -4 -2 0 2 4 6

Relative Weights

Figure 7-1. Relative weights of attributes for invasive plants experts











CHAPTER 8
SUMMARY AND CONCLUSIONS

The determination of the impact of invasive plants in outdoor recreation utility in Florida

parks is important for proper decision-making about invasive plant management. The revenues

and satisfaction realized from state parks are important to many residents' quality of life and are

an important element of the state's economy.

This study provides information about residents' preferences for state parks based on park

attributes including invasive plants. A Multi-attribute model was designed to assess these

preferences and was implemented through the Internet to Florida residents. The model consisted

of seven different choices sets each with two alternatives for each of the two surveys separated

by animal and plant species diversity. The plant species survey contained four attributes;

facilities condition, native plant diversity, presence of invasive species and the park fee. The

animal species survey had the same attributes as the plants survey but with animal species

diversity instead of the native plant diversity attribute.

Responses to the Multi-attribute surveys were analyzed using the conditional logit model

to determine which factors significantly influenced park choices, the implied ranking of the

attributes and the alternatives based on the utility and the marginal willingness to pay for the

attributes. All the attributes were significant but demographic characteristics like age, income

and education had no influence on the relationship between recreational utility and invasive

species. The invasive species attribute had the most impact on preference probabilities while the

park fee impacted utility the most. Invasive species attitude characteristics (which included

invasive species knowledge, if a person have been affected by the species or have taken action

against invasive plants and if the person think invasive plants are beneficial) had a significant









impact on choice preference and, therefore, willingness to pay for the invasive species attribute.

To the contrary being a member of an environmental club did not have any impact, most likely

because we had a negligible number of environmentally conscious respondents (6%).

The MWTP for the various attribute levels ranged from $4.13 to $7.15. The MWTP to

reduce invasive species was higher than the MWTP to improve facilities or increase native plants

and animals. Estimated annual statewide WTP based on fiscal year 2005-2006 park attendance,

ranged from $38.7 to $110.5 to reduce invasive plants, $23.3 to $66.6 million to improve

facilities and $23.9 to $68.5 million dollars to improve plant species diversity. For the animal

species model, the annual WTP ranged from $40.4 to $115.3 to reduce invasive plants, $25.3 to

$72.4 million to improve facilities and $33.4 to $95.5 million dollars to improve animal species

diversity. The WTP to reduce invasive species ranged from $29.1-$108.7 million per year with

different levels of invasive species knowledge among Florida residents.

Since the results from the two surveys were relatively similar, a quick analysis was done at

the end of the study combining the two surveys. Results for the combined model for all the

attributes are between the plant species and the animal species model results since overall the

animal species model had relatively higher estimates.

MWTP were calculated for three regions in order to capture any differences in WTP and

determine if they were related to the extent of invasive species currently in the region. South

Florida, which is relatively highly affected by invasive plants, had the highest MWTP for all the

attributes including the invasive species attribute.

The study determined the extent to which the park preferences and MWTP varied based on

individual specific factors. These results allow us to identify various aspects within the

population, which offers the opportunity for policy makers to capture the aspects to focus on in









dealing with invasive plants. The realized effect of knowledge and perception of invasive

species on choice preference and MWTP for the invasive species attribute could give some

direction on how to deal with invasive species in the state more specifically in recreational parks.

Since our MWTP values were higher than in past studies and free parks were preferred the

most, increasing park fee may not be a feasible option. An educational program would be an

appropriate approach towards the solution to invasive species in Florida parks. Residents could

be educated on the overall impact of invasive species both to the state economy and destruction

to the environment to improve their MWTP to reduce invasive species. Since demographic

characteristics when interacted with the invasive species attribute did not play an important role

in choice preference, the question is not who should be targeted to eradicate invasive species but

rather how the populations' knowledge and awareness should be improved.

The next question is how the invasive species knowledge was acquired for respondents

who had the knowledge. The answer to this question will determine how the education program

on invasive species could be implemented. It is therefore recommended that future studies find

out how residents acquire knowledge about invasive species and use these means to educate the

public about invasive species.

One limitation of this study is that it was too broad considering the various types of parks

in the state's park system. It would also have been best to gather research data through face to

face interviews at the parks with the actual park users rather than gathering the information from

any Florida resident through Internet.

Finally, caution should be taken when referring to the statewide willingness to pay

estimates because the MWTP for the attributes made inference to out of state park visitors who

were not part of the analyzed samples.












APPENDIX A


MAIN SURVEY

Survey Cover Letter


Dear Florida Resident,

We are requesting your participation in a University of Florida survey on Recreation and Invasive
Plants in Florida's State Parks (the link to the survey webpage is located at the bottom of this letter).
You have been selected as a part of a small sample of Florida residents who are being asked to complete
this online questionnaire. Please take a few minutes to complete the survey.

This survey is divided in three parts. In the first part you will be asked to provide different valuations
about a specific natural area and a second one of your choice, which is optional. In the second part you
will be asked to give your opinion about what effects invasive species have had in your decision of which
location to attend and enjoyment when engaging in outdoor recreational activities. Finally, we will ask
you to give us some socio-economic information for our analysis.

Remember that to participate in this survey you must be 18 years or older. Participation is voluntary. You
do not have to answer any questions you do not wish to answer. You are free to stop the questionnaire at
any time. There are no anticipated risks, compensation, or other direct benefits to you as a participant in
this study. You may be assured of complete confidentiality. You will not be identified or connected with
the questionnaire in any way and participation is totally anonymous. Results will only be reported as
summarized data. The information gathered in this study may be published in professional journals or
presented at scientific meetings, but will not be accessible as individual data.

The survey is funded by the Florida Department of Environmental Protection and is administered by the
University of Florida and the Institute of Food and Agricultural Sciences. For questions about this study,
please feel free to contact graduate student investigators Santiago Bucaram (santibu@ufl.edu) or Frida
Bwenge (fbwenge@ufl.edu). For questions about your rights as a research participant, please contact the
University of Florida Institutional Review Board (PO Box 112250, Gainesville, Fl 32611, telephone 352-
392-0433).

Please remember that your answers to this survey are extremely important and may impact your future
enjoyment of Florida's state parks.

Thank you for your cooperation.




WEB SURVEY LINK: http://www.surveymonkey.com/s.asp?u=864193701263











Do you live in Florida? QUESTIONNAIRE (WPS)
QYes
Q No


What is the county of your primary residence in Florida? (choose from the menu
below)
I


How frequently have you participated in nature related outdoor activities at
each of the following locations during the past 12 months?
Once Once Once
Daily Weekly Monthly every 2 to every 4 to every 7 to Not at all
3 months 6 months 12 months
OCEAN AND BEACH 0 0 0 0 0 0 0
RIVER AND LAKE 0 0 0 0 0 0 0
WOODED PARK 0 0 0 0 0 0 0
















We would like to know more about how invasive plants affect your recreation decisions and your enjoyment of Florida parks.

In the questions to follow we would like you to:
(1) Compare pairs of "WOODED" parks
based on the 4 features show n in the table on the right
(2) Indicate your preference by choosing ONE park
(3) Do this 7 times

This part of the survey should take no more than 5 minutes


W EiOODED PA


About the two Wooded parks:


(1) The two parks are your only alternatives
(2) Each park is the same distance from your home
(3) Both parks offer the following activities and facilities


1.- PARK FACILITIES CONDITION: Park facilities include
parking lots, boat docks, boat ramps, picnic tables,
restrooms, showers, among others
2.- DIVERSITY OF PLANT SPECIES: Include all the plants
which are natural or indigenous to Florida
3.- FEES: Include fees for admission, parking, camping
among others
4.- PRESENCE OF INVASIVE SPECIES: All non-native
plants known to disrupt ecosystem processes
















Minimal Adequate


Moderate High

Numerous and
Few and dispersed dense
dense

510 $20


Which of the two parks do you prefer?
C Park A


C Park B


Minimal Excellent


Low High


None Few and dispersed


Free $20


Which of the two parks do you prefer?
0 Park A


O Park B




























Which of the two parks do you prefer?
c Park A


Excellent Adequate


High Low

Numerous and
None
dense

$20 Free


C Park B


Minimal Excellent


High Moderate


Few and dispersed None


$10 $20


Which of the two parks do you prefer?
O Park A


O Park B
















Adequate Excellent


Moderate High

Numerous and
None
dense

S10 Free


Which of the two parks do you prefer?
c Park A


C Park B


Excellent Minimal


Moderate High

Numerous and
Few and dispersed dense
Frdense

$10 Free


Which of the two parks do you prefer?
O Park A


O Park B
















Excellent Minimal


High Low

Numerous and
None
dense

$20 $10


Which of the two parks do you prefer?
O Park A


C Park B


There are two other types of parks that are highly impacted by invasive plants.
Of the two, which one would you answer more questions about?
SOcean and Beach
G River and Lake
O Neither. I would like to proceed to other questions






















Excellent Minimal



High Low


Numerous and
None
dense


S$20 $10



Sample for WAS


Which of the two parks do you prefer?

C Park A


( Park B


Please indicate your knowledge of invasive plants prior to this survey.

r I knew a lot about invasive plants
E I knew a little about invasive plants
r I knew nothing about invasive plants






Indicate your agreement or disagreement with the following statements:


Somewhat
Strongly agree
agree
Invasive plants have affected my enjoyment of r r
outdoor recreation activities in State Parks
Invasive plants have affected the number of r r
my visits to State Parks
Invasive plants have affected which State r r
Parks I attend
Invasive plants can also provide benefits to r
Florida's parks


Somewhat
Neutral
disagree
r r


Strongly
disagree
r





r















Have you taken any personal actions in response to invasive plants In Florida?
Q Yes
O No


Please indicate whether you have done any of the following in response to
invasive plants:
D I helped remove invasive plants from natural (public) areas
D I made a personal contribution (money or supplies) to help remove invasive plants from natural (public) areas.
] I have driven to farther parks just to avoid invasive species plants
D Other (please specify)


Examples of actions against invasive species are:
To become active to help remove Invasive plants from
natural areas;
To drive or travel farther to visit an alternative location with
fewer invasive plants;
To donate money or supplies to help remove invasive
plants from natural areas; among others












Please indicate the area that best describes where you live
O Urban Area city or town
O Suburban Area- within 5 miles of a city center or town
SRural Area more than 5 miles from a city center or town

Please indicate your gender
O Male
O Female

Please indicate your age
0 18 -24
25 34
0 35 44
S45 -54
0 55 64
O 65 or older












Please indicate your marital status
O Single, never married
O Married
Q Divorced
Q Widowed

How many people including yourself occupy the residence where you live?
01
02
03
04
0s
Q more than 5

How many people under age 18 live with you?
O None
O1
02
03
04
0 mo5re
0 more than 5











Indicate the highest level of education you have completed
Q Some high school
Q High school graduate
O Associate (AA) or 2 year technical degree
O Bachelor (BA, BS, or other 4 year degree)
O Advanced or Professional training beyond a bachelor degree

Indicate your race or ethnic background
Q White/Caucasian
Q Black/Afican -American
Q Hispanic, Latino, Chicano
O Asian or Pacific Islander
SNative American

Is anyone in your household affiliated with an environmental organization?
Q Yes
0NO



What is your employment status? (Check only one answer)
O Employed
O Not employed, but seeking work
O Not employed and not seeking work
O Student
O Retired

What is your annual household income before taxes? (Check only one answer)
Q Less than $14,999
S $15,000 $34,999
S$35,000 $59,999
Q $60,000 $74,999
Q $75,000 $99,999
Q $100,000 $149,999
0 More than $150,000
















Thank you for participating in this study. The information you provided is important. For questions about this
study, please contact graduate research assistants Santiago Bucaram (sanriou@ufl.eau) or Frida Bwenge
(fbwenge@ufl.edui. For questions about your rights as a research participant, contact the University of
Florida Institutional Review Board (PO Box 112250, Gainesville, Fl 32611, telephone 352-392-0433). Click here
to Qualify for your incentive



Thank you for your time. This study was developed exclusively for Florida residents.

For questions about this study, please contact graduate research assistants Santiago Bucaram (santibu@ufl.edu) or Frida
Bwenge (fbwenge@ufl.edu).

For questions about your rights as a research participant, contact the University of Florida Institutional Review Board (PO Box
112250, Gainesville, FI 32611, telephone 352-392-0433).




THANK YOUI










APPENDIX B
PARK RANKING BY UTILITY

Table B-1. Park ranking by utility -WPS
FA PS IS FE FA PS IS
2 2 0 0 Excellent High None
1 2 0 0 Adequate High None
2 1 0 0 Excellent Moderate None
2 2 1 0 Excellent High Few and dispersed
0 2 0 0 Minimal High None
1 1 0 0 Adequate Moderate None
2 0 0 0 Excellent Low None
2 2 0 10 Excellent High None
1 2 1 0 Adequate High Few and dispersed
2 1 1 0 Excellent Moderate Few and dispersed
0 1 0 0 Minimal Moderate None
1 0 0 0 Adequate Low None
2 2 2 0 Excellent High Numerous and dense
1 2 0 10 Adequate High None
2 1 0 10 Excellent Moderate None
0 2 1 0 Minimal High Few and dispersed
1 1 1 0 Adequate Moderate Few and dispersed
2 0 1 0 Excellent Low Few and dispersed
0 0 0 0 Minimal Low None


FE Utility
Free 1.246
Free 0.939
Free 0.930
Free 0.737
Free 0.632
Free 0.623
Free 0.614
$10 0.506
Free 0.430
Free 0.421
Free 0.316
Free 0.307
Free 0.228
$10 0.199
$10 0.190
Free 0.123
Free 0.114
Free 0.105
Free 0.000










Table B-2. Park ranking by utility-WAS
FA AS IS FE FA AS IS
2 2 0 0 Excellent High None
1 2 0 0 Adequate High None
2 1 0 0 Excellent Moderate None
2 2 1 0 Excellent High Few and dispersed
0 2 0 0 Minimal High None
2 2 0 10 Excellent High None
1 1 0 0 Adequate Moderate None
1 2 1 0 Adequate High Few and dispersed
2 0 0 0 Excellent Low None
2 1 1 0 Excellent Moderate Few and dispersed
2 2 2 0 Excellent High Numerous and dense
1 2 0 10 Adequate High None
0 1 0 0 Minimal Moderate None
2 1 0 10 Excellent Moderate None
0 2 1 0 Minimal High Few and dispersed
1 0 0 0 Adequate Low None
2 2 1 10 Excellent High Few and dispersed
1 1 1 0 Adequate Moderate Few and dispersed
1 2 2 0 Adequate High Numerous and dense
2 0 1 0 Excellent Low Few and dispersed
0 2 0 10 Minimal High None
2 2 0 20 Excellent High None
2 1 2 0 Excellent Moderate Numerous and dense
1 1 0 10 Adequate Moderate None
0 0 0 0 Minimal Low None


FE Utility
Free 1.330
Free 1.043
Free 0.952
Free 0.873
Free 0.756
$10 0.690
Free 0.665
Free 0.586
Free 0.574
Free 0.495
Free 0.416
$10 0.403
Free 0.378
$10 0.312
Free 0.299
Free 0.287
$10 0.233
Free 0.208
Free 0.129
Free 0.117
$10 0.116
$20 0.050
Free 0.038
$10 0.025
Free 0.000










APPENDIX C
WOODED PARK CLASSIFICATION BY REGIONS

Table C-1. Park classification by regions


NORTH FLORIDA
ALFRED B. MACLAY
BALD POINT
BIG LAGOON
BLACKWATER RIVER
CAMP HELEN
DEER LAKE
ECONFINA RIVER
FALLING WATERS
FLORIDA CAVERNS
GRAYTON BEACH
HENDERSON BEACH
LAKE JACKSON MOUNDS
LAKE TALQUIN
LETCHWORTH MOUNDS
OCHLOCKONEE RIVER
PONCE DE LEON SPRINGS
ROCKY BAYOU
SAN MARCOS DE APALACHE
ST. ANDREWS
ST. GEORGE ISLAND
ST. JOSEPH PENINSULA
TARKILN BAYOU
THREE RIVERS
TOPSAIL HILL
TORREYA
WAKULLA SPRINGS


NORTH FLORIDA
BIG SHOALS
BIG TALBOT ISLAND
CEDAR KEY
CEDAR KEY SCRUB
CRYSTAL RIVER PRESERVE
DEVIL'S MILLHOPPER
DUDLEY FARM
FANNING SPRINGS
FORT CLINCH
FORT COOPER
FORT GEORGE ISLAND
GOLD HEAD BRANCH
HOMOSASSA SPRINGS
ICHETUCKNEE SPRINGS
LITTLE TALBOT ISLAND
MANATEE SPRINGS
MARJORIE KINNAN RAWLINGS
O'LENO
OLUSTEE BATTLEFIELD
PAYNES PRAIRIE
PEACOCK SPRINGS
RAINBOW SPRINGS
SAN FELASCO HAMMOCK
STEPHEN FOSTER
SUWANNEE RIVER
TROY SPRINGS


CENTRAL FLORIDA
ANASTASIA
BLUE SPRING
BULOW CREEK
BULOW PLANTATION RUINS
CATFISH CREEK
DE LEON SPRINGS
DUNN'S CREEK
FAVER-DYKES
FORT MOSE
GAMBLE ROGERS
HONTOON ISLAND
KISSIMMEE PRAIRIE
LAKE KISSIMMEE
LAKE LOUISA
LOWER WEKIVA RIVER
NORTH PENINSULA
RAVINE
ROCK SPRINGS RUN
SEBASTIAN INLET
SILVER RIVER
ST.SEBASTIAN RIVER
TOMOKA
WASHINGTON OAKS
WEKIWA SPRINGS










Table C-1. Continued.


Table C-1. Continued.


SOUTH FLORIDA
ALAFIA RIVER / CYTEC
ANCLOTE KEY
AVALON
CALADESI ISLAND
CAYO COSTA
COLLIER-SEMINOLE
DADE BATTLEFIELD
DON PEDRO ISLAND
EGMONT KEY
FAKAHATCHEE STRAND
GASPARILLA ISLAND
HIGHLANDS HAMMOCK
HILLSBOROUGH RIVER
HONEYMOON ISLAND
KORESHAN
LAKE JUNE
LOVERS KEY
MADIRA BICKEL MOUND
MOUND KEY
MYAKKA RIVER
OSCAR SCHERER


SOUTH FLORIDA
PAYNES CREEK
WERNER-BOYCE
BAHIA HONDA
CAPE FLORIDA
CORAL REEF
CURRY HAMMOCK
FORT PIERCE INLET
FORT ZACHARY TAYLOR
HUGH TAYLOR BIRCH
INDIAN KEY
JONATHAN DICKINSON
KEY LARGO HAMMOCK
LLOYD BEACH
LONG KEY
MACARTHUR BEACH
OLETA RIVER
SAVANNAS
SEABRANCH
ST. LUCIE INLET
WINDLEY KEY FOSSIL REEF









LIST OF REFERENCES


Adamowicz, W., J. J. Louviere, and M. Williams. 1994. "Combining Stated and Revealed
Preference Methods for Valuing Environmental Amenities." Journal ofEnvironmental
Economics andManagement 26:271-292.

Adamowicz, W., J. Swait, P. Boxall, J. J. Louviere, and M. Williams. 1997. "Perceptions Versus
Objective Measures of Environmental Quality in Combined Revealed and Stated Preference
Models of Environmental Valuation." Journal ofEnvironmental Economics and
Management 32:65-84.

Adamowicz, W., P. Boxall, M. Williams, and J.J. Louviere. 1998. "Stated Preference
Approaches for Measuring Passive Use Values: Choice Experiments and Contingent
Valuation." American Journal ofAgricultural Economics 80:64-75.

Adams, D.C., and D.J. Lee. 2006. "Statewide Bioeconomic Model of the Invasive Aquatic Plants
Hydrilla, Water Hyacinth, and Water Lettuce." Draft report to FLDEP-BIPM.
Unpublished, University of Florida.

Agresti, A. 1996. An Introduction to Categorical Data Analysis. New York John: Wiley & Sons,
Inc.

Alberini, A., P. Rigati, and A. Longo. 2003. "Can People Value the Aesthetic and Use Services
of Urban Sites? Evidence from a Survey of Belfast Residents." Journal of Cultural
Economics 27(3-4): 193-213.

Alberini, A., P. Rosato, A. Longo, and V. Zanatta. 2005. "Information and Willingness to Pay in
Contingent Valuation Study: The Value of S. Erasmo in the Venice Lagoon." Journal of
Environmental Planning and Management 48:155-175.

Alberini, A., A. Longo, and P. Rigati. 2006. "Using Survey to Compare the Public's and
Decision makers' Preferences for Urban Regeneration: The Venice Arsenale." Working
Paper 137.06, FEEM, Milan.

Alvarez, R.M, R.P. Sherman, and C. VanBeselaere. 2003. "Subject Acquisition for Web-Based
Surveys." Political Analysis 11(1):23-43.

Armstrong, J.S., and T.S. Overton. 1977. "Estimating Non-response Bias in Mail Surveys."
Journal ofMarketing Research 14:396-402.

Bateman, I.J., R.T. Carson, B. Day, M. Hanemann, N. Hanley, T. Hett, M. Jones-Lee, G.
Loomes, S. Maurato, E. Ozdemiroglu, D.W. Pearce, R. Sugden, and J. Swanson. 2002.
Economic Valuation iilh Stated Preference Techniques: A manual. Cheltenham: Edward
Edgar, Ltd.

Bennett, J., R. Blamey, and M. Morrison. 1997. "Valuing Damage to South Australian Wetlands
Using the Contingent Valuation Method." Occasional Paper 13/97, LWRRDC, Canberra.









Berrens, R.P., A.K. Bohara, H.C. Jenkins-Smith, C.L. Sivla, and D.L. Weimer. 2003. "Advent of
Internet Surveys for Political Research: Comparison of Telephone and Internet Samples."
Political Analysis 11(1): 1-22.

Berrens, R.P., A.K. Bohara, H.C. Jenkins-Smith, C.L. Sivla, and D.L. Weimer. 2004.
"Information and Effort in Contingent Valuation Surveys: Application to Global Climate
Change Using National Internet Samples." Journal ofEnvironmental Economics and
Management 47:331-363.

Blamey, R.K., J.C. Rolfe, J.W. Bennett, and M.D. Morrison. 1997. "Environmental Choice
Modeling: Issues and Qualitative Insights." Choice Modeling Research Report No. 4.
University College, The University of New South Wales, Canberra.

Bowker, J.M, H.K. Cordell, and C.Y. Johnson. 1999. "User Fee for Recreation Services on
Public Lands: A national Assessment." Journal of Parks and Recreation Administration
17(3)1-14.

Couper, M. P. 2000. "Web surveys: A Review of Issues and Approaches." Public Opinion
Quarterly 64:464-494.

Couper, M. P., M. Traugott, and M. Lamias. 2001. "Web Survey Design and Administration."
Public Opinion Quarterly 65:230-253.

Darby, K.J., 2006. "Consumer Preferences for Locally-Grown Berries: A Discrete Choice Model
Estimating Willingness-to-Pay." MS Thesis Ohio state university, Columbus.

Dillman, D.A. 2000. Mail and Internet Surveys: The Tailored Design Methods, 2nd Ed. New York:
John Wiley and Sons, Inc.

Doren. R.F., A. Ferriter, and H. Hastings (eds). 2002. "Weeds Won't Wait! An Assessment of
Invasive Exotic Plants in Florida." A Report to the South Florida Ecosystem Restoration
Task Force and Working Group 305/348.1665, South Florida Ecosystem Restoration Task
Force, Miami, Florida.

Eiswerth, M.E., T.D. Darden, W.S. Johnson, J. Agapoff, and T.R. Harris. 2005. "Input -Output
Modeling, Outdoor Recreation and the Economic Impacts of Weeds." WeedScience 53:130-
137.

Finn, K. 2006. "Social Economic Impact of Controlling Melaleuca in South Florida." MS Thesis,
University of Florida, Gainesville.

Florida Department of Community Affairs. 2000. Florida Assessment of Coastal Trends (FACT).
Florida Coastal Management Program, Tallahassee, Florida.

FLDEP. 2000. Outdoor Recreation in Florida 2000: Florida's Statewide Comprehensive outdoor
Recreation Plan (SCORP). Division of Recreation and Parks, Tallahassee, Florida.









FLDEP. The Environmental Impact and Regional Differences of Invasive Plants in Florida.
Bureau of Invasive Plant Management, Circular 10, downloaded on 04/20/07 from
www.dep.state.fl.us/lands/invaspec/Circular

FLDEP. 2004. FY 2003/2004 Florida State Park System Economic Impact Assessment. Division of
Recreation and Parks, Tallahassee, Florida.

FLDEP. 2004. Upland invasive exotic plant management program, Annual Report FY 2002-3.
Bureau of Invasive Plant Management, Tallahassee, Florida.

FLDEP. 2005. Upland invasive exotic plant management program, Annual Report FY 2004-5.
Bureau of Invasive Plant Management, Tallahassee, Florida.

FLDEP. 2006. Final Long Range Program Plan for FY 2007-2008 through 2011-2012.
Department of Environmental Protection, Florida.

FLEPPC. 2006. Florida Exotic Pest Plant Council. Fort Lauderdale, FL, http://www.fleppc.org.
visited on 09/16/06.

FLEDP. 2006. FY 2005/2006 Florida State Park System Economic Impact Assessment. Division of
Recreation and Parks, Tallahassee, Florida.

FLDEP. Division of Recreation and Parks Website http://www.dep.state.fl.us/parks/ visited
between September 06 and May 07.

Glisson, M. W. 1994. "Invasive Non-indigenous Species in Florida's State Parks." In An
Assessment of Invasive Non-indigenous Species in Florida's Public Lands. Technical
Report No. TS S-94-100, Florida Department of Environmental Protection, Tallahassee,
Florida.

Gordon, D.R. 1998. "Effects of Invasive, Non- Idigenous Specied on Ecosystem Process:
Lessons from Florida." Ecological Applications 8(4):975-989.

Green P.E., and V. Srinivasan. 1978. "Conjoint Analysis in Consumer Research: Issues and
Outlook." Journal of Consumer Research 5:103-123.

Greene, W.H. 2003. Econometric Analysis, 5th ed. Upper Saddle River, NJ: Prentice-Hall.

Grosz, A. 2007. "Volunteers Scour the City for Invasive Air Potatoes." The alligator January
29, pp 3, University of Florida.

Halbrendit, C.C., F. Yang, L. Thomas, and K. Krisnankumar. 2007. "Analysis of Farm
Household Preferences in Management of Invasive Species: The case of Miconia in
Hawaii." Paper Presented at the 17th Annual World Food and Agribusiness Forum
Symposium, Parma Italy, 23-26 June.

Hanley, N., R. Wright, and W. Adamowicz. 1998. "Using Choice Experiments to Value the
Environment." Environmental and Resource Economics 11:413-428.











Harding, D., and M. Thomas. 2003. The Economics of Selected Florida Wildlife Management
Areas. Tallahassee, Florida: Florida Fish and wildlife conservation Commission. Rep. 21,
September.

Hodges, A. 2006. "Valuing Environmental Services for Forest Ecosystems: Implications for
Policy and Outreach." Paper presented at a forum on conserving ecosystems: A
cooperative research and outreach approach, Gainesville, Florida, 05/01/2006.

Holmes, T.P., and W.L.Adamowicz. 2003. "Attribute-Based Methods." In P.A. Champ, K.J.
Boyle, and T.C. Brown, eds. The Economics ofNon- Market Goods and Resources: A
primer on Nonmarket Valuation. Dordrecht: Kruwer Academic Publishers, pp.171-219.

Lancaster, K. 1966. "A New Approach to Consumer Theory." Journal of Political Economy
74:132-157.

Lee, D. J. and C.S. Kim. 2005. "Managing Upland Invasive Plants on Florida Public Lands."
Paper presented at USDA-TSTAR Annual Meeting, San Juan Puerto Rico, 01/06/05.

Letson, D. and J.W. Milon.(eds). 2002. Florida Coastal Environmental Resources: A Guide to
Economic Valuation andImpact Analysis. Gainesville: Florida Sea Grant College
Program.

Longo, A. 2007. "The Use of Conjoint Choice Experiments in Valuing Cultural Tourism
Programs." Working paper, Pro-active management of the Impact of Cultural Tourism
Upon Urban Resource and Economies (PICTURE), European Union Sixth Framework
Programme.

Louviere, J. J., and G. Woodworth. 1983. "Design and Analysis of Simulated Consumer Choice
or Allocation Experiments: An Approach Based on Aggregate Data." Journal ofMarketing
Research 20 (4):350-367.

Louviere, J. J. 1988. Analyzing Decision making: Metric Conjoint Analysis. Newbury Park, CA: Sage
Publications.

Louviere, J.J, D. A., Hensher, and J. D. Swait. 2000. Stated Choice Methods Analysis and
Application. Cambridge: Cambridge University Press.

MacDonald, D.,. and M. Morrison. 2005. "The value of Habitat and Agriculture." Client Report,
CSIRO Land and Water, Australia.

Maddala, G.S. 1983. Limited-Dependent and Qualitative Variables in Econometrics. New York:
Cambridge University Press.

Makokha, S.N, J.Karugia, S.Staal, and 0. Kosura. 2006. "Valuation of Cow Attributes by
Conjoin Analysis: A case Study in Western Kenya." Paper Presented at the International
Association of Agricultural Economists conference, Gold Coast, Australia, 12-18 August.









Mazzotti, F.J., W. Ostrenko and A.T. Smith. 1981. "Effects of the Exotic Plants Malaleuca
quinquenervia and Casuaria equisetifolia on Small mammal Populations in Eastern
Florida Everglades." Quarterly Journal of the Florida Academy of Sciences Spring 1981,
Vol. 44 No. 2.

McFadden, D. 1974. "Conditional Logit Analysis of Qualitative Choice Behavior." In P. Zarembka,
ed. Frontiers in Econometrics. New York: Academic Press, pp. 105-142.

Milon, J.W., A.W. Hodges, A. Rimal, C.F. Kiker, and F. Casey. 1999. "Public Preferences and
Economic Values for Restoration of the Everglades /South Florida Ecosystem." Economics
Report 99-1, Food and Resource Economics Department, University of Florida, Gainesville.

National Invasive Species Council. 2001. Meeting the Invasive Species Challenge: National
Invasive Species Management Plan. http://www.invasivespecies.gov/council/main.shtml
visited 11/20/06.

Office of Technology Assessment, U.S. Congress. 1993. Two Case Studies: Non-indigenous
Species in Hawaii and Florida. In Indigenous Species of the United States. OTA-F-565,
U.S. Government Printing Office, Washington, DC., pp. 233-266.

Olson, B.E. 1999. "Impacts of Noxious Weeds on Ecological and Economic Systems." In R.L.
Sheley, and J.K. Petroff, eds. Biology and Management of Noxious Rangeland Weeds.
Corvallis, OR: Oregon State University Press pp, 4-18.

Pearce, D., and D. Moran. 1994. The economic value ofBiodiversity. IUCN- the World
Conservation Union: Earthcan Publications.

Pimentel, D., L. Lach, R. Zuniga, and D. Morrison. 2000. "Environmental and Economic Costs
of Non-indigenous Species in the United States." Bioscience 50:53-65.

Pimentel, D. 2002. Biological Invasions. Boca Raton, FL: CRC Press.

SAS Institute. 1996. "The FACTEX Procedure" SAS Procedure Guide. Cary, N.C, pp.209-235

Snowball, J.D., and K.G. Willis. 2006. "Estimating the Marginal Utility of Different Sections of
Art Festival: The case of Visitors to South African national Arts Festival." Leisure Studies
25(1):43-56.

Siikamaki, J. 2001. "Valuing Benefits of Finnish Forest Biodiversity Conservation-Logit Models
for Pooled Contingent Valuation and Contingent Rating/Ranking Survey Data."
Department of Agricultural and Resource Economics Seminar, UC Davis,
http://are.berkeley.edu/courses/envres_seminar/s2001/siikamaki.doc

Swanson, C.S., and J. B. Loomis. 1996. Role ofNon-market Economic Values in Benefit-Cost
Analysis ofPublic Forest Management. Portland, Oregon, USDA, Forest Service, Gen
Tech Report PNW GTR-361, March.









Swait, J., and W. Adamowics. 2001. "The influence of Task Complexity on Consumer Choice: A
Latent Class Model of Decision Strategy Switching." Journal of Consumer Research
28:135-148.

Throsby, D. 2003. "Determining the Value of Cultural Goods: How much (or How Little) does
Contingent Valuation Tell Us?" Journal of Cultural Economics 27: 275-285.

Tsuge, T., and T. Washida. 2003. "Economic Valuation of the Seto Inland Sea by Using an
Internet CV Survey." Marine pollution bulletin 47:230-236.

U.S. Department of Commerce, Census Bureau. 2001. U.S. Census 2000 American Fact Finder,
U.S.. http://www.census.gov

U.S. Department of Commerce, Census Bureau. 2005. Computer and nternet Use in the United
States: 2003, Special studies. Economics and Statistics Administration, Washington DC,
October http://www.census.gov

U.S. Department of the Interior, Fish and Wildlife Service and United States Department of
Commerce, Bureau of the Census. 1998. 1996 National Survey ofFishing, Hunting, and
Wildlife-Associated Recreation (Florida). Washington, D.C.

U.S. Department of the Interior, Fish and Wildlife Service, and United States Department of
Commerce, Bureau of the Census. 2002. 2001 National Survey ofFishing, Hunting, and
Wildlife-Associated Recreation (Florida). Washington, D.C.

Wirh, F.F., and J.K.White. 2006. "Florida Residents' Perception and Willingness to Pay for
Invasive Plants Management in Public Lands." Report No SL49-063 Submitted to FLDEP
Tallahassee, Florida.

Zapata, H.O., P.R. Sambidi, and E.A. Dufour. 2007. "Choice Models in Policy Analysis"
Selected Paper prepared for presentation at the Southern Agricultural Economics
Association Annual Meetings, Mobile, Alabama, 3-6 February.









BIOGRAPHICAL SKETCH

Friday Bwenge is from Tanzania. She received her Bachelor of Science degree in

Agriculture from Sokoine University in Tanzania in 1989 with specialization of rural economy.

After graduation, Frida worked with the Tanzanian Ministry of Agriculture in the Planning

Division as an agricultural economist. While working with the ministry of agriculture, she went

to the United Kingdom and received her Master of Science degree in national development and

project planning from the University of Bradford in 1992.

She began another Masters of Science program in food and resource economics in August,

2005. Friday plans to work for an international organization in development planning and poverty

eradication in her home continent, Africa, after completing her M.S. degree requirements at the

University of Florida.





PAGE 1

1 THE IMPACT OF INVASIVE UPLAND PLANTS ON THE RECREATIONAL VALUE OF NATURAL AREAS: THE CASE OF WOODED PARKS IN FLORIDA By ANAFRIDA BWENGE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Anafrida N. Bwenge

PAGE 3

3 (To Kisha for understanding when mom was too busy with school)

PAGE 4

4 ACKNOWLEDGMENTS This thesis could not have been completed wi thout the help, encouragement and support of all my supervisory committee members. First, I wish to express my sincere thanks and appreciation to my committee ch air, Dr. Donna Lee who guided me in writing this thesis and also made sure I had all the financial help I need ed to acquire this degree. I would also like to thank my other committee members, Dr. Sherry Larkin and Dr. Janaki Alavalapati whose devoted support helped me to complete this thesis. I have so much appreciation for my fellow student, Santiago Bucar amu, who provided the much needed technical and mathematical expertis e to design, implement and analyze this study. I would also like to express my gratitude to my friend Jennifer Nunez, who offered to read my thesis and helped me with both technical advice and corrections to my English grammar. Finally, I would like to acknowledge and give special thanks to my husband, Dr. Charles Bwenge, and my girls for all their help a nd support during my time in school.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .........9 LIST OF ABBREVIATIONS........................................................................................................10 ABSTRACT....................................................................................................................... ............11 CHAPTER 1 INTRODUCTION..................................................................................................................12 Background..................................................................................................................... ........12 Recreation in Florida Natural Areas.......................................................................................13 Upland Invasive Plants in Florida Natural Areas...................................................................15 Problem Statement.............................................................................................................. ....18 Study Objectives............................................................................................................... ......19 Hypotheses..................................................................................................................... .........19 2 LITERATURE REVIEW.......................................................................................................23 Valuation of Non-Market Goods............................................................................................23 Selection of Attributes and Levels..........................................................................................25 Survey Method.................................................................................................................. ......28 Web Surveys....................................................................................................................29 Advantages of web surveys......................................................................................30 Weaknesses of web surveys.....................................................................................30 3 THE MULTI-ATTRIBUTE UTILITY MODEL...................................................................33 4 DATA COLLECTION AND RE SEARCH METHODS.......................................................37 Surveys........................................................................................................................ ...........37 Park Managers Survey...................................................................................................38 Florida Residents: Invasive Species Knowledge Survey................................................39 Florida Residents: Attrib utes Selection Survey...............................................................41 Building the Multi-Attribute Utility Survey...........................................................................42 5 SURVEY RESULTS..............................................................................................................51 Introduction................................................................................................................... ..........51 Respondents Profiles.......................................................................................................... ...51

PAGE 6

6 Invasive Species Knowledge..................................................................................................52 6 MODEL SPECIFICATION AND EMPIRICAL RESULTS.................................................59 Model Variables................................................................................................................ ......59 Non-Response Errors Testing.................................................................................................59 Hypothesized Signs on Parameters.........................................................................................60 Statistically Significant Indi vidual Specific Variables...........................................................61 Specification.................................................................................................................. .........62 Results........................................................................................................................ .............63 Overall Model Fit............................................................................................................63 Attribute Variables..........................................................................................................63 Probabilities and Marginal Effects for the Models..........................................................64 Marginal Willingness to Pay Measures...........................................................................65 Marginal willingness to pay and individual specific variables................................66 Marginal willingness to pay by regions...................................................................66 Relative Weighting of Attributes.....................................................................................67 Choices Ranking by Utility Scores..................................................................................67 Results for the Combined Model.....................................................................................68 Hypotheses Testing............................................................................................................. ....69 7 INTERPRETATION OF THE RESULTS FOR INVASIVE PLANT MANAGEMENT.....81 Invasive Species and Outdoor Recreation..............................................................................81 Estimations of Willingness to Pay to Reduce Invasive Species.............................................81 Marginal Willingness to Pay and the Invasive Species Attitude............................................84 The Importance of Invasive Species Knowledge...................................................................86 8 SUMMARY AND CONCLUSIONS.....................................................................................90 APPENDIX A MAIN SURVEY.................................................................................................................... .93 B PARK RANKING BY UTILITY.........................................................................................106 C WOODED PARK CLASSIFI CATION BY REGIONS.......................................................108 LIST OF REFERENCES.............................................................................................................110 BIOGRAPHICAL SKETCH.......................................................................................................116

PAGE 7

7 LIST OF TABLES Table page 1-1 Ten most unwanted plants in Florida.................................................................................22 4-1 Reasons for participating in outdoor recreati onal activities..............................................48 5-1 Demographic profiles for survey s compared to Florida profiles.......................................54 6-1 Basic models independent variables..................................................................................71 6-2 Socioeconomic and invasive species attitude variables.....................................................71 6-3 2 Tests statistics of first 50 and last 50 respondent characteristics..................................71 6-4 Significance test for attribute interact ion with individual specific variables-WPS...........72 6-5 Significance test for attribute interact ion with individual specific variables-WAS..........72 6-6 Overall fit for the multi-attribute models...........................................................................72 6-7 Coefficient estimate for the multi-attribute models...........................................................73 6-8 Probabilities for park choices-WPS...................................................................................75 6-9 Marginal effects for the models.........................................................................................75 6-10 Comparison of implicit prices es timates for recreational attributes..................................75 6-11 Implicit prices to reduce invasive species with attitude variables.....................................76 6-12 Regional marginal willingness to pay for the attributes....................................................76 6-13 Impact of change in attri bute level on the total utility.......................................................79 6-14 Coefficient estimates for th e attributes (combined model)................................................79 7-1 Annual WTP to control invasive plan ts in Florida wooded parks (million $)...................88 7-2 Annual regional WTP to reduce invasive plants levels -WPS..........................................88 7-3 WTP per visit to reduce Melaleuca in recreational areas by Florida residents.................88 7-4 Annual WTP to control invasive speciesFL residents with knowledge (million $)........88 7-5 Coefficient estimate for the attr ibutes (invasive plants experts)........................................88 B-1 Park ranking by utility -WPS...........................................................................................106

PAGE 8

8 B-2 Park ranking by utility-WAS...........................................................................................107 C-1 Park classification by regions..........................................................................................108

PAGE 9

9 LIST OF FIGURES Figure page 1-1 Annual attendance at Florida state parks: 1995-2006........................................................21 1-2 Floridas estimated demand for wooded park activities: 1997-2010.................................21 4-1 Level of invasive species knowledge in Florida................................................................47 4-2 Perceived impact of invasive species on enjoyment and other aspects-(preliminary).......47 4-3 Natural areas outdoor ac tivities participation....................................................................48 4-4 Outdoor recreation attributes selection..............................................................................49 4-5 Suggested park attributes for the state to improve.............................................................49 4-6 Example of the MAUM questions A) plant species B) animal species.............................50 5-1 Location profiles of samples and Florida residents...........................................................54 5-2 Gender profiles of samples and Florida residents..............................................................55 5-3 Age profiles of sample s and Florida residents...................................................................55 5-4 Income profiles of samples and Florida residents..............................................................56 5-5 Education profiles of samp les and Florida residents.........................................................56 5-6 Comparison of knowledge of inva sive species in three surveys........................................57 5-7 Perceived impact of invasive specie s on enjoyment and other aspects-(main).................57 5-8 Respondents taking action against invasive species..........................................................58 5-9 Park visit frequencies..................................................................................................... ....58 6-1 Implicit prices for recreational attributes...........................................................................76 6-2 Regional marginal willingness to pay A) plant species B) animal species........................77 6-3 Relative weights of attributes A) plant species B) animal species....................................78 6-4 Relative weights of the attri butes for the combined model...............................................80 7-1 Relative weights of attributes for invasive plants experts.................................................89

PAGE 10

10 LIST OF ABBREVIATIONS APHIS Animal and Plant Health Inspection Service BIPM Bureau of Invasive Plant Management CE Choice Experiment CV Contingent Valuation FACT Florida Assessment of Coastal Trend FLDEP Florida Department of Environmental Protection FLEPPC Florida Exotic Plant Pest Council FLSCORP Florida Statewide Compre hensive Outdoor Recreation Plan MAU Multi-Attribute Utility MAUM Multi-Attribute Utility Model MWTP Marginal Willingness to Pay NISC National Invasive Species Council OTA Office of Technology Assessment RUM Random Utility Maximization UF-IRB University of Florida Institutional Review Board USDA United States Department of Agriculture WAS Wooded Park Animal Species WPS Wooded Park Plant Species WTA Willingness to Accept WTP Willingness to Pay

PAGE 11

11 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science THE IMPACT OF INVASIVE UPLAND PLANTS ON THE RECREATIONAL VALUE OF NATURAL AREAS: THE CASE OF WOODED PARKS IN FLORIDA By Anafrida N. Bwenge August 2006 Chair: Donna Lee Co chair: Sherry Larkin Major: Food and Resource Economics Invasive plants can have impacts on the quality and quantity of recrea tional activities such as hunting, wildlife viewing and hiking because they negatively affect environmental attributes that are important in supporting recreation. Th is study examined the relationships between invasive plants and recreational values of Flor idas wooded parks using a Multi-Attribute Utility Model. Surveys were electronically distributed to Florida residents to examine preferences for these attributes; invasive species, park fee, sp ecies diversity and facilities. A conditional logit model predicted the respondents choice behavior and quantified the relationship between utility derived from recreation and these attributes. Results indicate that invasive species have a negative impact to recreational utility. Florida residents have a marginal willingness to pay (M WTP) to reduce invasive species of up to $7.15 which is higher than the MWTP to improve faci lities or increase spec ies diversity. MWTP to reduce invasive species was even higher with invasive species knowledge ($19.25 for experts). Residents willingness to pay to control th ese species ranged from $29.1 to $108.7 million per year with knowledge level differences. Our findings suggest that an invasive species educational program could increase Florida residents MWTP to control invasive species in natural areas.

PAGE 12

12 CHAPTER 1 INTRODUCTION Background Invasive plant species are defi ned as non-indigenous species with the ability to establish self-sustaining and expanding populations within plant communities and may cause economic or environmental harm (NISC, 2001). The United Stat es natural ecosystems have been invaded by over 5,000 non-indigenous plant species, which compete with approximately 17,000 native plant species for space and resources. When invasive plan ts successfully invade natural areas they tend to displace native species and a ssociated wildlife, degrade upla nd habitats and cause loss of biodiversity (Olson 1999). This is because they possess many weedy characteristics which enable them to spread rapidly and effectively in the new environment. Some of these non-native plants are responsib le for $25 billion in damages to the Unites States food and horticultural crops and $10 million in losses to natural ecosystems each year (Pimentel, 2002). The annual total cost from dama ges and controlling invasive plants in the agriculture and horticulture sectors is $34.5 bill ion and an additional $159.5 million is spent on managing invasive plants in natural systems (P imentel, 2002). Nationall y, invasive species are the second greatest threat to endangered species after habitat destructi on and cost the country over $138 billion each year in environmental damages and crop losses (Pimentel, 2002). While the invasive species problem has become a global and national concern, the state of Florida has been the most affected in the Un ited States (OTA, 1993). The favorable climate, geographical location and environm ental conditions that foster the states high level of plant diversity have consequently made Floridas land susceptible to i nvasive plant species. To date, more than 1,300 exotic plant species have been introduced and establis hed throughout the state with 124 species destructive to the biological diversity of natu ral areas (FLEPPC, 2006). In 2005

PAGE 13

13 upland weeds such as the Australian pine, Braz ilian pepper and climbing ferns infested over 1.65 million acres of Floridas 11 million acres of public conservation lands (FLDEP, 2005). These plants have also affected millions of acres of Floridas privately owned land. Non-native invasive plants can also have subs tantial impact on recreational activities such as hunting, hiking, wildlife view ing and water-based recreation (Eiswerth et al., 2005). Olson (1999) suggests that this is because invasive weeds negatively affect a wide range of environmental attributes that are important in supporting recreation such as plant and animal diversity and abundance. In a study about the economic impacts of w eeds on outdoor recreation in the riparian areas of Nevada, Eiswerth et al., (2005) found that non-native weeds had a recreational losses impact over the five years period ranging from $30 million to $40 million. Recreation in Florida Natural Areas Florida natural areas play a significant role in the states economy by providing recreational activities for resident s and visitors. Ecotourism recrea tional activities such as hiking, camping, sightseeing, and wildlife viewing in Florid as natural areas have an estimated economic contribution of $8 billion per year (FLDEP, 2005). Outdoor recreation is one of the states ma in attractions and the Floridas state park system is one of the largest in the country with 159 parks covering 723,852 acres of land and 100 miles of sandy beach (FLDEP, 2006). Floridas state parks offer year-around outdoor activities for all ages. The park system is comprised of beaches, rivers and lakes and wooded parks offering diversified activities in each park. Ther e are some parks like for example, Oleta River state park where visitors can enjoy beach ac tivities, engage in kayaking, mountain biking, camping, swimming, fishing, trail walking, horse ri ding, wildlife viewing and even hiking, all activities in one place. Over 100 parks offer wooded park activities like hi king, nature trails and horse riding. Over 50 parks offer both river and lake (boating, fishing, a nd kayaking) activities

PAGE 14

14 and wooded park activities while more than 40 parks offer beach activities like swimming, sunning, surfing and wooded park activities. A good number of visitors come to Florida for the primary purpose of viewing wildlife. According to the Fish and Wild life Service, approximately 800,000 visitors came to Florida in 1996 primarily for the purpose of viewing wildlif e, and over 40 percent of Floridas residents participated in some form of wildlife viewing. In this same year, Florida ranked second in the nation behind Texas for wildlife-related recreatio n expenditures. In the 2001 Fish and Wildlife survey nearly 4.9 million Florida residents and n onresidents over 16 years of age fished, hunted, or watched wildlife in Florida pa rks. Over 65% of this total number participated in wildlifewatching activities, including observing, feeding, and photographing. More than half of Florida visitors engage in some type of nature-based ac tivity during their visit, and 19 to 33 percent of all travel and tourism in the southern United Stat es is linked to outdoor recreation (Hodges, 2006). Between 1995 and 2004, the state park system s economic impact on local economies throughout the state grew from $189 million to over $600 million (FLDEP, 2005) with the annual park attendance growing from 12 to 18.5 milli on visitors (Figure 1-1). In the last fiscal year, 2005-2006 over 18 million people visited Florida state parks spending over $442 million in the state (FLDEP, 2006). It is expe cted that in the next five ye ars the need for public outdoor recreation land and parks in Florida will increase greatly as the states population is growing (FLDEP, 2006). The demand for wooded park activities like bi king, horse riding and wildlife viewing has also been on the rise and expected to conti nue. The estimated demand for selected wooded park activities are presented in Figure 1-2. On Floridas scenic trails, a growing number of people are

PAGE 15

15 undertaking longer day and overnight hikes while horseback riding pa rticipation, rela tive to other forms of outdoor recreation, has been steady (FLSCORP, 2000). The state has responded to the growing out door recreation demand by investing over $3 billion to expand conservation la nds and recreational opportunities over the past decade (FLDEP, 2006). The focus has been on making natural ar eas more accessible to the public through restoration including management a nd removal of non-native plants. Upland Invasive Plants in Florida Natural Areas Upland invasive plants are terrestrial (vs. aqua tic) invasive exotic pl ants. Invasive plants displace native plants and associated wildlife, in cluding endangered species and can alter natural process such as fire and water flow. Exotic plants were brought to the U.S. to be grown for various reasons like food, feed, fi ber, and ornamental purposes but some have become invasive and have proven to be a challenge to keep under c ontrol. The problem of exotic invasive species in Florida parks has been cited as one of the greatest threats to park resources (Glisson, 1994). According to the Floridas Recreation and Pa rks Division, the most troublesome invasive plants in the state parks are Brazilian pepper, Australian pine, Chinese tallow, Cogongrass, Air potato and Japanese climbing fern. Based on the total acres treated in projects dealing with upland weeds management in the state in 2005, the ten most unwanted upland invasive exotic plants in Florida are listed in Table 1-1. Florida Exotic Plant Pest Council (FLEPPC) compiles invasive spec ies lists that are revised every two years. Invasive exotic plants are termed Cate gory I invasive when they are altering native plant communities by displacing native species, changing community structures or ecological functions, or hybrid izing with natives. Category II i nvasive exotics are invasive plants that have increased in abundance or fr equency but have not ye t altered Florida plant communities to the extent shown by Category I sp ecies. They may become category I if they

PAGE 16

16 demonstrate ecological damage. The plants on the list of the ten most unwanted in Florida are currently all listed as category I invasive by the FLEPPC. They ar e scattered everywhere mostly from Central Florida, along the East and West coasts towards South Florida. Among the list, Australian pine, Brazilian pepper and Malaleuca are the most widely spread in Central and South Flor ida. These plants, like other inva sive species, have a tendency to crowd out native plants and animals. Australian pine invades coastal areas interfering with nesting of endangered sea turtle s and American crocodiles. Br azilian pepper grows in dense monocultures and reduces nesting sites for bird, am phibian and reptile populations. It is believed to have displaced some populations of rare li sted species, such as the Beach Jacquemontia ( Jacquemontia reclinata) and Beach Star ( Remirea maritima ) in South Florida (Doren, 2002). Cogongrass invades pinelands, scrub a nd prairie also threatening rare plants and interfering with fire patterns. In places invaded by Cogongrass, wildfires can be more frequent and intense (FLDEP). Old world climbing fern is naturalized in southern and west ern Florida, and the Japanese climbing fern is frequently naturalized in north and west Florida. South Floridas upland environmen ts are the part of the stat e most heavily invaded by non native species (FLDEP). The reason for South Fl oridas heavy invasion is said to be high importation activity in the area, highly disturbed landscapes and a climate conducive to growth of subtropical plants. The plants mentioned above are also restrict ed by the federal government along with other plants that are known to interfer e with agro-ecosystems, native ecosystems, the management of ecosystems, or to cause injury to public hea lth. The USDA Animal and Plant Health Inspection Service (APHIS) runs the Fede ral noxious weed program desi gned to prevent the introduction and the spread of newly introduced non-indigeno us invasive plants in the United States by

PAGE 17

17 excluding, detecting and eradicating introduced weeds that pose the highest risk to agriculture or the environment. Controlling and management of invasive spec ies in natural areas is one of the states priorities but this exercise has been costly. Priv ate expenditures for contro lling invasive plants in Floridas agriculture and forest industries are estimated at $265 million per year (Lee and Kim, 2005). Overall the state spends $103 million per year on preventi on and control of invasive plants (FLDEP, 2006). In 2005 the state exceeded the $6.3 million annual estimate by spending $8.7 million managing just upland invasive exotic species (FLDEP, 2005). The Upland Plant Management Program responsible for managing invasive exotic plants in the states public lands is under the Florida Depa rtment of Environmental Protection (FLDEP) in the Bureau of Invasive Plant Management (BIPM) The program works in eleven regions within the state to develop strategies to address upland invasive pl ant management issues locally through regional working groups. Since the inception of the Upland Program in 1997, BIPM has spent nearly $40 million to bring over 300,000 acres of upland weeds under maintenance control (FLDEP, 2005). Maintenance control is a control method of invasive exotic plants in which contro l techniques are utilized in a coordinated manner on a continuous basis in order to maintain the plant population at the lowest feasible level. As stated in the FLDEP Agency Strategic Plan, the long-term program goal is to reduce infestations of upland invasive exotic plants on public lands by twentyfive percent by 2010, based on estimated 1995 levels of 1.5 million acres infested with invasive weeds. In 2005 over 22% of af fected public land was under main tenance control (FLDEP, 2005). The program treated about 100 different invasive species at 144 publicly managed areas. The

PAGE 18

18 control program utilizes a variety of methods including chemical, mechanical, and biological techniques. Problem Statement Millions of residents and tourist who partic ipate in outdoor activitie s derive satisfaction from various attributes of natural areas. The most general attributes specif ic to the parks include parks sanitation and safety, extent and conditi on of facilities, the qua lity of the natural environment and accessible scenic trails for a vari ety of nature-based recr eational activities (e.g., hiking, camping, sightseeing and wildlife viewing). Invasive plants may have a negative im pact on the quality or quantity of outdoor recreational activities. Through al tering ecosystems, invasive speci es can negatively change the supply and composition of environmental amen ities that are important for recreation and adversely affect recreational service flows. Plants like th e Old World climbing fern ( Lygodium microphyllum ) and Japanese climbing fern ( Lygodium japonicum ), both on the list of the ten most unwanted plants in Florida, grow with thick climbing and twining fronds, preventing easy access to natural areas. Likewise, the Air potato ( Dioscorea bulbifera L .) grows vigorously, twining and forming a dense blanket that engul fs surrounding plants, which can limit access for navigation and recreational activitie s. The presence of dense twini ng invasive plants in natural areas may also obstruct wildlife viewing and access to scenic trails making hiking and biking difficult in wooded parks. Some park visi tors may be bothered by Catlaw mimosa ( Mimosa pigra L .) or Tropical soda apple ( Solanum viarum ) due to their hairy st em and prickly nature, respectively. Studies on the consequences of invasive plan ts in Florida natural areas have focused on management costs (Lee and Kim, 2005; Doren, 2002; Harding and Thomas, 2003) and ecological impact (Mazzoti, 1981; Gordon, 1998). In order to fully understand how invasive

PAGE 19

19 species affect outdoor recreation, it is essential to know how th e users enjoyment and use of natural areas is affected by these exotic species However, studies that estimate the relationship between recreational utility and i nvasive species in natural areas do not exist at the national or state level. Given the significance of outdoor re creation in natural ar eas to the citizens enjoyment and the states economy this knowledg e could aid in planning statewide programs aiming to control invasive species in Florida. Therefore, this study proposes to examine the relationships between upland invasi ve plants and the recreational value of Florida natural areas. Study Objectives The general objective of this study is to examine the relatio nship between invasive upland plants and the implicit value of recreational activ ities in natural areas specifically, in Florida wooded parks. This was achieved through the following objectives: Objective 1: Quantify the relationship between invasi ve plants and recreational value in natural areas using a Multi-attribute Utility Model (MAUM); Objective 2: Determine the relative importance of i nvasive species in relation to other attributes of a natural recreation experience; Objective 3: Determine the marginal willingness to pay for fewer invasive species in recreational areas, specifically in Florida wooded parks; Hypotheses The basic premise in this study is that invasi ve plants are undesirable in wooded parks and recreational places such that recr eation satisfaction in these areas will be reduced by the presence of invasive species. For this reason natural area users will be willing to pay more for a recreational exercise at a wooded park that ha s fewer invasive species in outdoor recreation areas. It is anticipated however that the willingness to pay more for fewer invasive species will not only be due to the effect of invasive plants on recreational utility but also because of the

PAGE 20

20 perceived impact of these species on the environment1. Therefore, users of natural areas who have a higher level of knowledge of invasive plants and a higher level of environmental consciousness are expected to be willing to pay more for fewer invasive species. Socioeconomic variables such as income, age, and education are expected to influence individuals perception about the relationship between invasive plan ts and the recreational value. Furthermore, since Florida regions are affected at different levels by invasive plants, willingness to pay will likely be different between regions and perhaps higher for the most affected region. 1 It is common for public to express their willingness to pay for passive or non use values. For example people want to pay to make sure blue whales are conserved ev en if they may not see them in their lifetime.

PAGE 21

21 0.00 5.00 10.00 15.00 20.00 25.00 94-95*95-96*96-97*97-98*98-99*2003-042004-052005-06 YearMillion Visitors Figure 1-1. Annual attendance at Florida state parks: 1995-2006 Source: *FACT, 2000, FLDEP, 2004-2006 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 199720002005*2010* YearMillion Participants Hiking Horse Riding Nature Study Figure 1-2. Floridas estimated dema nd for wooded park activities: 1997-2010 Source: FLSCORP 2000

PAGE 22

22 Table 1-1. Ten most unwa nted plants in Florida Common name Scientific Name Acres Treated Australian pine Casuarina species 436 Caesars weed Urena lobata 749 Old world climbing fern Lygodium microphyllum 3,728 Melaleuca Melaleuca quinquenervia 46,498 Skunk vine Paederia foetida 1,021 Brazilian pepper, Schinus terebinthifolius 7,830 Chinese tallow Sapium sebiferum 667 Japanese climbing fern Lygodium japonicum 771 Cogongrass Imperata cylindrica 1,212 Tropical soda apple Solanum viarum 1,023 Source FLDEP: Upland Exotic Plant Manage ment Program, Annual report FY 2004-2005

PAGE 23

23 CHAPTER 2 LITERATURE REVIEW Valuation of Non-Market Goods This study attempts to determine the non-market value of recreation attributes in natural areas with special focus on invasi ve plants. A non-market good or service is something that is not bought or sold directly, theref ore, does not have an observable price (i.e. market value). Nonmarket values can be categorized as use valu es and non-use values. The use value of a good is the value to an individual from the active use of the asset lik e recreational fishing, swimming or bird watching and the non-use valu es reflect the value to the society or future generation (Letson and Milon, 2002). Although participation in out door recreation thro ugh activities like camping and hiking can be categorized as the use value for natura l areas and may generate market based economic activity, use of a natu ral area is often not allocated by markets (Swanson and Loomis, 1996). Therefore the impact of invasive species on natu ral areas will not fully be captured by market goods and services. Where there is no price available for non-market goods, it may be possible to use the prices of related market goods or prices from hypotheti cal markets to estimate the value (Letson and Milon, 2002; Longo, 2007). A number of methods ha ve been devised for valuation of an environmental good or service where market va lue is not evident. These methods include hedonic price, travel cost, the contingent valu ation and the conjoint choice experiment. The hedonic pricing and travel cost method are calle d revealed preferences because they measure preferences for non-market goods based on observi ng peoples choice behavior on other related goods. For example hedonic pricing method would us e the property sales information to assess the monetary value of a cultura l tourism attraction. Tr avel cost method would measure the value

PAGE 24

24 of a tourism attraction by using the money spent on that attraction as a proxy of the value that users attach to the place. Contingent valuation (CV) and choi ce experiment (CE) are referred to as stated preferences because they ask indivi duals about their prefer ences through surveys. Typically the CV approach asks people to dire ctly report their willingness to pay (WTP) for a specific good, or their willingness to accept (W TA) compensation for a good rather than inferring them from observed behavior in regula r market like in reve aled preference methods (Longo, 2007). In the past, most studies on valuation of nonmarket environmental goods like natural areas have used CV (Bennet et al., 1997; Tsuge and Washida, 2003; Berrens et al., 2004). This method has also been used in valuing cultural heritage destinations (Alberini et al., 2005). However, CV has been criticized for its w eakness in valuing goods when mo st of the value of the good is derived from non use value. It is believed that the CV provides an incomplete view of the value of the good because this value is multidimen sional and may not be easily expressed by qualitative or quantitative scales (Throsby, 2003). Pearce et al. (1994) summarized some of the CV method issues as problems of reliability, bias and validity. The CE has been successfully used in mark eting and transportation research. Following this success, the CE methodology has been increa singly preferred over th e CV in valuation of environmental recourses (Holmes et al., 2003). CE ha s been used to assess public preferences for restoration goals (Milon et al., 1999; MacDonald et al., 2005, ) and valuing environmental amenities by Adamowicz and others (1997) who argue that the advantage of CE over CV is that this method could help reduce strategic bias as the attributes change in choice sets. When a non-market goods non-use values are impacted, only stated preference models can capture the impact. One such method is the Multi-attribute Utility M odel (MAUM) which is

PAGE 25

25 a choice experiment type of method (Louvier e and Woodworth, 1983; Milon et al., 1999; Louviere et al., 2000). There ar e two types of MAUM; one is preference based where respondents are asked to rate or rank the provid ed alternatives and th e other is choice based where respondents are asked to choose their preferred alternative fr om the provided choice package. Blamey et al. (1997) describe the CE as a technique in which indi viduals are typically presented with six to ten choice sets, each cont aining a base option and several alternatives and asked to indicate their preferred option in each set. Although there is no definitive number of choices to be presented, it is r ecommended to use not more than ei ght to avoid respondents being fatigued (Holmes et al., 2003). The respondent ch ooses from a set of hypothetical goods or services that require an evalua tion of the trade-offs between at tribute levels, with the concept that the choice is based on the individuals uti lity maximization. Attributes trade off is presented in the value elicitation process such that reduc tion in one attribute may be compensated by an increase in another. The MAUM research approach involves prior identifica tion of attributes and levels for building the choices or alte rnatives to present to the respondent. Selection of Attributes and Levels The initial step in developing a choice experiment survey is to determine attributes of the good which has to be valued. The attributes must be relevant to the decision problem and each attribute should reflect independent dimensi ons to the degree possibl e to avoid redundancy (Loviere, 1988). The valuation method requires that one of the attribute be the price or cost of the good to the respondent (Longo, 2007). In previous studies researcher s have used focus groups or structured conversation with people who broadly represent the population to be sampled to determine the attributes (Holmes et al., 2003; Milon et al., 1999). When using focus group the researcher would ask the group a

PAGE 26

26 series of questions aimed at identifying the impor tant attributes of the good in question as they are considered by the population which is to be surveyed. Some researchers have used information from the existing literature about th e subjects to determine desired attributes (Makokha et al., 2006) and some have used bot h methods depending on the research budget and time constraints. After identifying attributes the range of each attribute can be determined through the same procedures as the attributes by using the focus groups, subject expert inte rviews or literature reviews. The levels also have to be realistic and should cover the range over which respondents can have preferences. When the task of establishing the attributes and levels is completed, presenting all the combinations of the attributes becomes compli cated. Sometimes only a su b-set of attributes choice combinations has to be presented. The sub-se t of combinations needs to be presented in a statistically representative form to ensure that the maximum am ount of information is revealed by the study without bias. A vari ety of orthogonal factorial experi mental design is available to reduce and create a balanced sample of possible attribute combinations. In a full factorial experimental design, a study wi th five attributes w ith three levels each would have 35 or 243 possible combination. With a pair wi se choice this full f actorial design will require more that 120 choice decision which is impr actical to manage for both the researcher and the respondent. Researchers have employed software packages fo r the construction of optimized fractional factorial experimental designs such as SAS Factex procedure (SAS institute) to identify subsets of possible combinations of attribut es and levels that will best represent attribute preferences with a manageable size (Milon et al., 1999). This technology provides a significant saving in time and resources while allowing th e estimation of all the main effects.

PAGE 27

27 Once the experimental design is created, the next step is to construct the choice sets which may consist of two or more alternatives. Most CE surveys present three; two alternatives plus the status quo (Milon et al., 1999; MacDonald et al., 2005; Alberini et al., 2006).The status quo should be included in the choice se t if the purpose of th e study is to estimat e the willingness to pay for a policy option (Longo, 2007). The last step in the development of the experimental design is to choose the number of choice sets to be presented to the respondent and develop the questionnaire. Complexity of CE surveys is said to increase with the number of choices, the number of attributes and levels and the numbe r of alternatives in a choice set (Swait and Adamowicz, 2001). Complicated CE surveys may be tiring to respondents and may lead to poor quality data. When the survey is completed, the study sample size is given by a number of respondents receiving the questi onnaire times the number of c hoice sets presented in the questionnaire. While each of the mentioned types of MAUM (rank, rate and choice) offers distinct advantages for measuring preferences, the choi ce based method has the advantage of reflecting an actual consumer behavior (G reen and Srinivasan, 1978). With choice models it is possible to partition utility into parts allowing the estimation of the value of individual attributes that make up the good rather than considering the whole good (Adamowicz et al., 1998). In addition, the task of choosing the preferred bundle of attributes levels does not require as much effort by the respondent as do ranking a nd rating methods (Longo, 2007). Choice model studies can be achieved with smaller samples relative to contingent valuation. Studies have been completed with le ss than 110 interviews (Halbrendt et al., 2007; Snowball and Willis, 2006) with most studies be ing completed with interviews between 250 and 500 (Bateman et al., 2002). Furthermore, provi ding different choice alternatives to the

PAGE 28

28 respondent, which can be accomplished through the attribute based choice modeling methods, is believed to provide richer information for policy makers at the end of the research. Estimation of the attribute coefficients in past studies of CE models was done using multinomial logit (MNL), conditional logit or probit models (Siikamki, 2001; Milon et al., 1999; Makokha et al., 2006). The conditional logit model is useful when the choice probabilities are functions of the choice characteristics (Mad dala, 1983), for instance, when the probability that the individual chooses to visit a particular park is affected by the characteristics of that park. The difference between the multinomial logit and conditional logit models is that conditional logit considers the effects of c hoice characteristics on the determ inants of choice probabilities, while MNL makes the choice probabilities de pendent on individual characteristics only (Maddala, 1983). Today, there is an extensive use of choice models, including binary logit/probit, censored probit, conditional logit, finite mixture logit, group logit, rand om effects and random parameter models, nested logit, mixed logit and multivariate probit. Limdep or Nlogit, Stata, Gauss, and SAS, in that order, are some of the most fr equently cited software used in the econometric estimation of choice models (Zapata et al., 2007). Survey Method Available literature suggests that CE questi onnaire can be administ ered using different modes: mail, telephone, in person, Internet or a combination. Surv ey modes differ in cost, time, quality of data, sample control and the quality and am ount of information that can be presented to the respondent. Mail surveys have dominated the data collect ion in CE studies (Adamowicz, 1994; 1998; MacDonald et al., 2005). This survey mode is ine xpensive but has a low response rate when compared to telephone and in person interviews and it is limited in the amount of information presented to respondents. Complex s cenarios of CE have also been implemented

PAGE 29

29 through in person interviews (Milon et al., 1999 ; Hanley et al., 1998) or computer-based questionnaires (Alberini et al ., 2006) but the web survey mode has not been widely used. Web Surveys Web surveys have recently been recognized as a valuable instrument for collecting data (Dillman, 2000). The more widespread use of the Internet now makes it possible for more people to access surveys online. As the In ternet access widens to include more representation of the adult population, the ap plications for web-based surveys may increase. The rapid development of web surveys is leading some to argue that s oon web surveys will replace traditional methods of survey data collection (Couper, 2000). Web surveys are presented in two main format s; interactively or passively. Interactive surveys are presented screen-by-screen. When the respondent clicks a button like next, it allows the data from the question to be immedi ately transmitted to the surveyor such that partially completed surveys may be receive d. Interactive web surveys also allow for customization as subsequent questions may depe nd on the answer from the previous question. However, this format may create difficulties fo r respondents to review or change their answers (Couper, 2000). The passive survey designs involve presenting the entire survey with data transmitted once the respondent completes the survey and submits the answers. Here, survey respondents can easily browse through questions and review thei r responses before submitting. These types of web surveys are also easy to produce and easy to access with less technical difficulties. While web surveys may offer some positive oppo rtunities in data collection, its strengths and weaknesses are still being de bated. The strengths and weaknesses of web surveys need to be recognized to ensure that they are designed appr opriately and results ar e considered carefully.

PAGE 30

30 Advantages of web surveys In terms of survey administration, web survey s offer several advantages relative to the telephone and face-to-face intervie ws. Web surveys have relatively low marginal costs compared to the other two survey modes which involve th e time of interviewers and supervisors. Lower marginal costs of distributing web surveys a nd receiving responses makes it possible to have larger samples for a given research budget. We b-based surveys are self -administered, allowing respondents to complete the survey at their convenience. This pr ocess, besides making Internet surveying relatively low cost, re duces data entry requirements and eliminates the possibility of data entry errors (A lvares et al., 2003). Compared to other survey modes includi ng mail survey, web surveys allow rapid turnaround, allow access to a vast geographically diverse pool of potential respondents, and has the superior capability of provi ding complex and varied informa tion to respondents (Alvares et al., 2003). Tracking respondents ut ilization of information, which is only possible with web surveys can provide basis for assessing the degree of respondent effort devoted to the survey (Berrens, et al., 2004). Weaknesses of web surveys Making general statements about large populatio ns based on Internet survey results is currently problematic as this survey mode faces important methodological issues (Alvares et al., 2003). It is widely agreed that the major sour ces of error in web surveys include sampling, coverage, nonresponse bias and estimation errors. The nature of the samples that it can provide is questionable because it is difficult to draw re presentative samples from among Internet users. Coverage error is the deviation between the sampling frame and the target population. When surveying a large group, the coverage e rror becomes a major concern because not all

PAGE 31

31 population members have Internet access and also there is no lis t of email addresses for the population. Non-response bias also occurs in web surv eys because some members of the selected sample are unable or unwilling to complete the surv ey, but this is not unique to Internet surveys. However, the potential problem is said to be severe for web-based surveys due to low response rates (or inability to calculate response ra tes) and non-random recruitment procedures. In addition, non-response errors for web surveys ma y be higher because potential respondents may encounter technological difficulties w ith Internet if they have no ba sic computer literacy skills. Furthermore, technological hurdles, such as br owser incompatibility an d slow Internet access will influence whether a potential respondent completes a survey (Couper et al., 2001). Although coverage error and non-response bias are a concern for web-based surveys, some web surveys have performed better than telephone surveys. On the objective measure of election forecasting in the 2000 presidential election, the Harris Interactive web poll did better than similar telephone surveys at pr edicting state level pr esidential votes (Ber rens et al., 2003). In one study to find out if Internet sa mples produce estimates of willingness-to-pay functions comparable to those from the telephone survey, researches pr esented the results of parallel telephone and Internet surveys to invest igate their comparability The Internet samples produced relational inferences quite similar to th e telephone sample (Berrens et al., 2003). It was concluded that all survey met hods involve errors and the appr opriate question should not be whether Internet replaces telephon e as the mode of survey but rather, under what circumstances the use of Internet surveys is appropriate. Two trends that are likely to increase the use of Internet surveys in the future are the increasing difficulty of doing valid telephone surv eys and the increasing representation of the

PAGE 32

32 Internet users. Internet use in the US and around the globe has been growing rapidly, and is becoming more demographically representative of the population. In 1995, only about 10 percent of U.S households had access to the Internet but in 2003 about 55% of households had access to Internet (U.S Census Bureau, 2005) But still the populati on of adult Internet users in the U.S has different demographic characteristics than th e general population. Acco rding to the Census Bureau (2005), it is on average, younger, more educated, has more males, has people in higher income groups, and is disproportionately white or Asian. An additional advantage of internet surveys is its recent use in splits within the Internet sample to investigate methodological issues. Research ers have used the Intern et to find out if the provision of extensive informati on related to the policy being ev aluated when conducting survey could influence the research outcome. The gene ration of large sample sizes in web surveys permits the investigation of me thodological questions within the same survey by comparing the split-sample treatment effect when estimating willingness to pay (Berrens et al., 2004).

PAGE 33

33 CHAPTER 3 THE MULTI-ATTRIBUTE UTILITY MODEL This chapter summarizes the theoretical and methodological concepts used in the study. The Multi-attribute Utility Model (MAUM) is used to determine the relationship between invasive species and the recreational value in Fl orida natural areas. MAUM is a choice modeling method based on the random utility maximization (RUM) theory. The model finds its origins with Lancaster (1966) who built the conceptual framework for conjoint analysis by clarifying that utility is gained from the attributes of a good rather than the good itself. For example, the actual source of the individuals utility when enga ged in recreation at a st ate park are the park attributes like facilities and ac tivities provided at the park and the variety of animals and plant species available to see at the park among other things. If an individual is presented with a number of alternatives to choose from, it is assumed that utility is linear in parameters such that: j jk k k jx U 1 (3-1) where Uj is the indirect utility a ssociated with alternative j k is the preference parameter associated with attribute k xjk is the attribute k in choice j and is the random error term. In a multi-attribute choice setting, a person comp ares the attributes of the alternatives and would select the alternative that provides the maximum utility. Suppose an individual was faced with a pair wise attribute setti ng to choose alternative A or B. If an individual choose A with its set of attributes over B, then to that individual ut ility from choice A, is greater than utility from choice B; in symbolic terms: U(XA) > U(XB) (3-2) RUM models assume that utility is the sum of the deterministic component v(.) and a random component such that: U(X) = v(X) + (3-3)

PAGE 34

34 The error term is introduced for estimation pur poses and it is assumed that it comes from omissions of explanatory variab les, random preferences and erro rs in measuring the dependent variable. The error te rm in the model allows probabilistic st atements to be made about the choice behavior. Assumptions made on how the error te rm is distributed result in different choice models. In a pair wise choice setti ng for any given respondent i the probability that the respondent will choose XA over XB equals the probability that the difference between the deterministic components exceeds the difference between the random components. P (A) = P[vi(XA)vi(XB) > ( iB iA)] (3-4) From the above foundation, we use attributes which are the basic components of an individuals indirect utility or preference functio n. Since the attributes are built with levels, a preference function relates the leve l of each attribute to utility as independent dummy variable making a part worth utility model. So as th e respondents make their choices between the alternatives, the utility as sociated with changes in levels of specific attributes can be estimated. The additive linear function produces the main effects of the mode l and the effects indicates how utility is affected by the level of the attribute when it is separated from all other attributes. From the utility function in equation 3-1, the choice behaviors for a respondent from any set of choices are predicte d with the conditional logit model (McFadden, 1974) and the preference parameters in the equation are es timated. With the conditional logit model the probability values are estimated under the assumptions that is independently and normally distributed2. 2 When the choice sets has more than two alternatives, it also assumes that the ratio of probabilities between any two alternatives is unaffected by other alternatives in the choice set.

PAGE 35

35 For each set of alternatives, th e probability that an individual chooses A over the other alternatives is expressed as: (3-5) For a pair of alternatives A and B the proba bility that an individual chooses A over B is expressed as a logit function: iB iA iB B iA Ax x x x log (3-6) where XiA is the vector of values of the attributes in the choice A and XiB is the vector of values of the attributes in the choice B. The other dimension to the model is to evalua te if socioeconomic char acteristics like age, income and education influence the attribute weights or preference variation within the population. In choice models, these variables are not examined directly because they do not vary across alternatives (Holmes et al., 2003). In order to account for their diffe rences on preferences, socioeconomic characteristics ar e incorporated in the model th rough interaction terms with the attribute level variables: Uij = Xij + jSi + ij (3-7) where Uij is the indirect utility for an individual i associated with alternative j Si are individuals socioeconomic variables interacted with attribute level j s; Xij is the vector of values of the attributes in the jth alternative, and jare vectors of coefficients to be estimated. The random utility model estimates the probability that utility of the ith individual derived from the jth alternative is greater than the ut ility from the other alternatives From equation (3-5), with the social economic variables, the probability that the ith individual makes the jth choice will be estimated as: j Xij Xe e P(A)iA) ( ) (

PAGE 36

36 j k X Xi S k ik i S j ije e1 ) ( ) ( ij' 'P (3-8) With the utility function defined, we can mode l the choice as the relative differences in utility (Darby, 2006). The difference between ch oice A and choice B with the social economic variables included is: dUi AB = X + S + AB (3-9) where dUi AB is the utility difference between choice A and choice B; S = Si (xjA xj B); x= ( xjA xj B); and AB = ( eiAeiB). Since each alternative has the price or cost as one of the attribute, the marginal willingness to pay for each attribute can be defined as: p k kMWTP (3-10) Where MWTPk is the marginal willingness to pay for the attribute k k is the preference parameter for the attribute k and p is the price or cost parameter.

PAGE 37

37 CHAPTER 4 DATA COLLECTION AND RESEARCH METHODS Surveys This chapter summarizes the survey design a nd the Multi-attribute Utility (MAU) survey development procedure. Since this study was part of a broader study which was to determine the effect of invasive plants in both aquatic and upland parks, ini tial preparations and development of the MAU survey was generalized to include three types of parks. The park types were ocean and beach, river and lakes, and wooded parks. After the MAU survey instrument was developed, the surveys were administered separately for each of the three park types. Data analysis was also conducted by park type and this part of the st udy is only dealing with wooded parks. We used interactive online surveys to gather the data from Florida residents. The study started in fall of 2006 w ith a survey of Florida state park managers. Then based on park managers responses, two groups of Florid a residents were surveyed to determine natural areas relevant recreational attri butes and the level of knowledge re garding invasive species in the state. Finally the MAU survey was developed usi ng the information from these three surveys as well as from literature review a nd expert interviews. The Florida residents questionnaires for the two preliminary surveys were designed using Su rvey Monkey software package and sent through emails with Expedite Marketing Survey Company. Before being administered, the questionna ires were reviewed and approved by the University of Florida Institutional Review Board (UF-IRB) for compliance with ethical standards for human subject research. The surveys incl uded a short statement on the purpose of the questionnaire, the importance of providing inform ation, completing instructions, confidentiality guarantee and contact information in case the respondents had any questions.

PAGE 38

38 The knowledge assessment survey was sent to 40,000 and the recreational attribute survey was sent to 80,000 Florida reside nts. Response rates for the two surveys were less than 1%. The invasive species knowledge survey had a 0.82% wh ile the attribute survey had a 0.37% response rate. Despite low response rate s, the number of responses rece ived was sufficient to make the required analysis. The details for each of th e mentioned surveys are presented below. Park Managers Survey First, questions were sent through email to 159 Florida state park managers asking for important recreational attributes and the situat ion of invasive species in Florida parks. Specifically, managers were asked the type of th eir park (wooded, ocean and beach or river and lake) and what were the valuable characteristics of the park. If they had any invasive plants problems they were asked to name the species Managers were also asked to indicate any complaint that park visitors had related to park attributes and if any were related to invasive plants. In order to determine the park priorities indirectly, park managers were asked if they had $200,000 to spend, how they would spend it to impr ove the park. Managers were also asked if they felt that invasive species had impaired natura l areas or diminished its recreational use. This survey revealed that park visitors ca re most about the fo llowing attributes: To be able to see native plants in the park To be able to see a variety of animals in the park Visitors are bothered by congestion in the parks ( if there are too many visitors) Park visitors care about services a nd facilities availability and condition Cost for the trip Distance from home From this survey it was indicated that the most important upland i nvasive species in the parks are Brazilian pepper ( Schinus terebinthifolius) Cogongrass ( Imperata cylindrica)

PAGE 39

39 Australian pine ( Casuarina species), Chinese tallow ( Sapium sebiferum) and Japanese climbing fern ( Lygodium japonicum) Since the MAU survey was to be developed for both the upland and aquatic recreational parks, invasive aquatic plants were also men tioned. They included Hydrilla ( Hydrilla verticillata) and Water Hyacinth ( Eichhornia crassipes) Park managers also revealed that generally invasive species do not seem to affect the satisfaction of park visitors and some visitors even like invasive species like Australian pine for shade. According to park managers only few peopl e who are educated about invasive species are concerned about these species. Park Managers know about the envi ronmental impact of invasive plants on natural areas but control of these species is not their priority. If given some extra funds, most managers priority was to improve park facilities arguing that the state government is already taking care of invasive spec ies in their respective parks. The six recreational attributes identified from this survey were used in constructing the attribute selection for the Florida residents surveys. The type of invasive species reported in the parks and the managers indication of low knowledge of invasive plants among Florida residents were used in the development of an invasive species knowledge assessment. Florida Residents: Invasive Species Knowledge Survey Based on the invasive plants information a bove, a survey was crea ted to determine the level of awareness and knowledge for invasive species among Florida re sidents. With the study objective focusing on invasive species and recreati on, the level of knowledge of invasive species is crucial in designi ng the background information for the MAU survey and Model. In the survey, respondents were asked how th ey characterize their knowledge on invasive species on a 1-5 Likert scale wher e 1= None, 2= Little, 3=Modest 4= Well versed and 5= Expert. Out of 319 respondents, a total of 146 (46%) co uld be classified as having no knowledge, 132 (41%) had moderate knowledge and 41 (13%) were very knowledgeable about invasive species.

PAGE 40

40 In order to determine whether residents had the actual knowledge on invasive species, a simple test was given to respondents. For respondents who said they had the knowledge, a twelve plant quiz with pictures and names was administered and they were asked to identify which plants were invasive. The knowledge level was analyzed using a scoring method. Scoring was based on the correct answers out of twel ve questions. Those scoring 8 were categorized as experts. Scores 6 and 7 were grouped in modera te knowledge and scores 5 were grouped into the no knowledge category. A total of 254 respondents participated in the test. From the results, 161 (63.4%) had no knowledge on invasive species, 64 (25.2%) ha d moderate knowledge on invasive species and only 29 (11.4%) were very knowledg eable about invasive plants. By combining the number of people with moderate knowledge with the numbe r of those very knowledg eable about invasive species, we arrived at the conclu sion that 36.6% of Florida resi dents have knowledge on invasive species. Results on invasive species knowledge among Florida residents are presented in Figure 4-1. From the test we determined that the ac tual knowledge respondents had was different from what the respondents said they knew. For responde nts who said they were well versed or experts on invasive species, there was not much difference between what they said and the test score results (13% vs.11.4%). There was, however a big difference for the moderate knowledge group (41% vs. 25.2%) and the no knowledge group (46% vs 63.4%) with respect to what they said and the test results. In three separate questions respondents were al so asked to indicate whether they felt that invasive species had affected natural areas, their enjoyment in outdoor recreation, or if these species may influence the choice of destination for outdoor recreation. The answers in each of the three questions was to choose from Yes, No or Not sure. Results for the perception of

PAGE 41

41 invasive species in the three aspects are presen ted in Figure 4-2. Overwhelmingly, 82% of the respondents were aware that invasi ve species affected natural ar eas and the environment but less than 30% indicated that the presen ce of these species in a park w ould interfere with their outdoor activity enjoyment or influence their choice on whic h park to visit. This observation could be related to park managers indicat ion that visitors are not both ered by invasive plants when participating in park recreational activities. The last section of the survey gathered demographic inform ation about the respondent. In addition, respondents were asked to what extent they were en vironmentally conscious on a 1-5 Likert scale where 1= Not at all, 3= Moderate ly conscious and 5= Ex tremely conscious. Over 50% of the respondents said they were moderately environmentally conscious. Florida Residents: Attr ibutes Selection Survey The main objective for this survey was to de termine important recreational attributes for Florida natural area users. In the survey, res pondents were asked whic h nature related outdoor activities they participated in during the past tw elve months. The activitie s choices included bird watching, backpacking, camping, hunting walkin g, hiking and many more. These activities and the level of participation for the respondents ar e presented in Figure 4-3. According to the survey, most people participated in walking, hiking and running, followed by swimming in the ocean then nature watching or observation. Respondents were also asked thei r reasons to participate in ou tdoor recreation from a list of suggested reasons in order to fi nd features which contri bute to utility in outdoor recreation. The suggested reasons are listed on Table 4-1 as th ey were ranked by the respondents. The top reason was to experience nature, followed by exercise and enjoying with friends and families. In order to make the MAU survey manageable not all park attributes could be included in the model, so we had to select the most im portant. Respondents graded the importance of six

PAGE 42

42 recreational attributes when visi ting wooded parks. The attributes graded in this survey were determined through park managers survey results. They included park fees, facilities condition, congestion at the parks, animal and plant spec ies diversity and travel distance to the park. Results are shown in Figure 4-4 revealing that the three most important attributes in Florida upland natural areas recreation were: Plan t Species diversity, Animal Species diversity and Facilities condition. However, all the six presented attribut es were important as each one of them was chosen by over 50% of the respondents. An indirect evaluation of im portant attributes of outdoor recreation was included by asking respondents how should the state government invest more or make some improvement in Florida parks. Again, results from this question reflected that important attributes were animal and plant species diversity and facilities (Figure 4-5). Building the Multi-Attribute Utility Survey As mentioned before, choice Modeling or Mu lti-attribute Utility Model development required the identification of rele vant attributes and levels. Th is was the reason for the above preliminary surveys. From the knowledge survey, it was determined that only 36.6% of Florida residents have knowledge about invasive species. From the attribute selection survey, the three most important attributes were animal specie s diversity, native plant species diversity and facilities condition. In addition to these three attr ibutes, the extent of i nvasive species in the parks and park fees were include d as attributes. The fee attribut e was added for the purpose of determining implicit prices or ma rginal willingness to pay for the attributes. The invasive species attribute was added because the main purpose of the study was to determine its relationship with outdoor recreational utility. Park facilities condition was defined to include parking lots, boat docks and ramps, picnic tables, restroom, showers among other things. Divers ity of animal species was defined to include

PAGE 43

43 wild birds, animals and fish. Fees included fees for admission, parking, camping among others. Presence of invasive species was defined as a ll non-native plants known to disrupt ecosystem process. Diversity of native plants was defined to include all the plants which are indigenous to Florida. After deciding on the attributes, levels were dete rmined for each attribute. Levels for plant and animal species diversity were low, moderate and high. Levels for facilities were minimal, adequate, and excellent. The leve ls for the invasive species were none, few and dispersed, numerous and dense. The fee levels were $0 ( free), $10, and $20. The levels were determined from the survey information both from park managers and residents. For example, park managers indicated that invasi ve species in most parks are few and dispersed and over 53% residents indicated that they spent less than $10 on park fees for each visit at a wooded park. Park fees were also reviewed from the Florida Division of Recreation and Parks web site for the 159 state parks. From this information the levels were created to provide distinct variations in order to identify the pr eference parameters. Combinations of attributes and levels were to be grouped by selecting one level from each attribute and combining across attributes. We had fi ve attributes at three levels each resulting in 35 = 243 profiles. As already mentioned in chapter two, survey with this many profiles would be complicated for both the respondents and the resear cher. Therefore, the st udy was limited to four attributes with three levels each, which form 34 = 81 profiles. In order to confine the attributes to only four, two surveys were developed to separate the plant and animal species attributes. The rest of the attributes were the same in each of the two surveys. The other reason for separating animal and plant species diversity is that these two attributes could be highly correlated. If two correlated attributes are treated as independe nt in a choice experi ment, respondents might

PAGE 44

44 become confused and fail to answer the questions (Holmes et al., 2003). So it is advised to select attributes that represent separa te dimensions of valuation problem. Eighty one profiles were reduced by using a fractional factor ial design, selecting a sample of attribute levels from a full factorial design using SAS Factex procedure. The samples for the MAU questions for both the plant (WPS) and animal species (WAS) surveys ar e shown in Figure 4-6. A choice of the park included four attributes at one of the three levels in each attribute. In the end, each of the animal and plant species diversity surveys had seven choice questions each composed by two park options. It is believed that choice alternatives s hould include the neither or the status quo option because in most real world choice situat ions individuals are not in a forced choice situation (Holmes et al., 2003; Blamey et al., 1997). However, this study presents one of the circumstances when these options are considered not to be realistic choices. This is because our research is interested in the trade-off of attribut es mainly in invasive species as an attribute in outdoor recreation and how the trade-off is ma de between the invasive species and other recreational attributes (Snowball and Willis, 200 6; Alberini et al., 2003). As previously mentioned, we are not attempting to analyze poli cy options or estimate welfare changes due to policy changes. The estimation of statewide willingness to pay to c ontrol invasive species will be a component of this study only for the purpose of assessing the attribute trade-offs. The willingness to pay to reduce invasive plants in natural areas will be compared to the willingness to pay to improve other park attr ibutes. This information will provi de an insight on the value that respondents place on each attribute.

PAGE 45

45 Furthermore, in Florida there are 159 state parks, over 7,700 big lakes, 2,276 miles of shoreline, over 663 miles of beaches (FLDEP, 2006). It will be unrealistic to generalize the state of rivers and lakes, ocean and beaches and wooded parks to create a status quo option. The survey instrument (Appendix A) had three sections: (1) the first section presented the MAU choices after eliminating respondents who were not Florida residents; (2) the next section asked questions about attitudes a nd personal experience with invasi ve species; (3) the last section asked questions to elicit some demographic information. The MAU survey included a brief descrip tion of the study, pote ntial problems with specific invasive plants from the biological/ecosy stem perspective and photos depicting invasive plants in natural areas (Australian pine, Brazili an pepper). Because it was determined that more than 60% of Florida residents had no knowledge on invasive species, this information was necessary to ensure common interpretation and understanding of the subject among the respondents. The survey also incl uded park pictures and the activ ities a respondent could find at the park. For The MAU section, we asked the re spondents to assume that each of the two park choices are (1) the only altern atives (2) the same distance from the respondents home and (3) both parks offer the same described activities and facilities. The attributes were defined in details to make sure they were well understood by the respondents. Before sending the survey to respondents, it wa s pre-tested through a se ries of trials among 242 UF students and four academic staff to identify any problems with the length, content and if it will be easily understood by the respondents. Appropriate corrections were made accordingly following the given comments or suggestions. Based on the available literature that choice analysis studies can be conducted with relatively small samples (Snowball and Willis, 2006; Bateman et al., 2002), it was decided that a

PAGE 46

46 sample size of at least 400 complete responses for each survey would be sufficient. The surveys were electronically distributed at www.surveymonkey.com to Florida residents in May 2007 through a marketing company (www.zoomerang.com ) that had a good reputation with web surveys. For each survey, 6,655 emails were sent to solicit participant. With this company, responses were even higher than anticipated as each survey had over 700 responses only a few days after the survey was launched. The su rvey hosting company provided results for each respondent in a database which coul d be accessed anytime for analysis. Preference parameters and in the model equations were estimated with the conditional logit model for the samples separately. The study estimated the basic models first for both plant and animal species diversity surveys. By basic models we mean that the estimations were made for the attributes without the social characteristic s interactions. These models were specified with and without the intercept paramete rs to test if there was an or der bias in respondents choice between alternative pa rks (Milon et al., 1999). After estimating the basic models, the study es timated the demographic characteristics and invasive species attitude variables as they intera ct with the defined recrea tional attributes. Using the estimated parameters, Marginal Willingness to Pay (MWTP) and the re lative weights of the four attributes were determined. MWTP for the attributes were also ca lculated for the three Florida regions: North, Cent ral and South Florida. We also conducted a brief analysis of the combined data for the animal and plant species diversity only with recreational at tributes (without social interactions). This was done to compare if the results for the separated models had any re lationship with the results when the two models were combined. We estimated the preference pa rameters, the MWTP for the attributes and determined the relative weights of the attributes for combined data.

PAGE 47

47 Figure 4-1. Level of invasive species knowledge in Florida Figure 4-2. Perceived impact of invasive species on enjoyment and other aspects-(preliminary) 0% 20%40%60%80%100% Natural Area Choice o f Location Enjoymen t Percentage of Respondents Not Sure No Yes n= 284

PAGE 48

48 Figure 4-3. Natural areas out door activities participation Table 4-1. Reasons for participatin g in outdoor recrea tional activities Reason n=240 Response % To enjoy and experience nature 87.50% To spend time with family 55.00% To meet friends 41.20% To be with people with similar interests 24.60% To get exercise and improve health 74.60% To meet new people 12.10% To share knowledge and skills 18.30% To be engaged in thrill situations 13.80% Other 8.30% 0% 10%20%30%40%50%60%70% Off-roading Hunting None Sports Biking Tubing Camping Surfing Wake bord Bird watching River & Lake fishing Swimming Boating Photographing Nature watching Sunning Hike & walkPercentage of participants n=282

PAGE 49

49 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% AnimalsFacilitiesPlantsCongestionDistanceFee A ttributes Figure 4-4. Outdoor recreat ion attributes selection Figure 4-5. Suggested park attri butes for the state to improve 0% 5% 10% 15% 20% 25% 30% 35% 40% A nimalsFacilitiesPlantsCongestionFee Attributes n=254 n=251

PAGE 50

50 Which of the two parks do you prefer? Park A Park B Which of the two parks do you prefer? Park A Park B Figure 4-6. Example of the MAUM questions A) plant species B) animal species A B

PAGE 51

51 CHAPTER 5 SURVEY RESULTS Introduction The primary purpose of the survey was to dete rmine the implicit value of invasive plants on the value of outdoor recreation. The survey al so aimed to collect general demographic and invasive species attitude information to bett er understand the factor s influencing choice preferences and the WTP to reduce inva sive species in natural areas. The two surveys differentiated by animal and pl ant species diversity had the same format. The animal species sample (WAS) had 640 respo ndents and the plant species sample (WPS) had 648 respondents. These complete responses were a bout 89% for each survey after leaving out the incomplete responses. About 10% of the respond ents skipped the MAU questionnaire and 12% skipped the demographic questi onnaire with most of them sk ipping the household income question (13.5%). The surveys had high response ra tes for an Internet survey, which was 8.49% for animal species and 8.69% for plant species. Responses mean the number of people in each sample and we have observations which refer to the total number of choice experiments completed. Since we had seven choice pair s for each survey we multiply the number of responses by seven leading to a total of 4480 obs ervations for animal species and 4536 for plant species survey. Respondents Profiles Overall, the surveys had over 55% respondent s from Central Florid a, about 24% from South Florida and only 20% from North Florida. Social economic and demographic profile of respondents differed from Florida state demogr aphics (Table 5-1). A comparison of these characteristics with the 2000 US census revealed some potential non-coverage bias3. Relatively, 3 Non coverage bias is the deviation between the sampling frame and the population.

PAGE 52

52 the preliminary surveys samples represented the states demographic characteristics better than the main survey except for education and the high income groups. Selected characteristics of the sample popul ation compared to th e state profiles are presented in Figure 5-1 to 5-5. In relation to location, one woul d think that because we used Internet survey the samples would be skewed to wards more urban residents. Surprisingly, the samples were skewed in a different direction with more people from suburban and rural areas above the state representation (Figure 5-1). In both samples 55% of respondents were from sub urban population. Respondents were overwhelmingly female, with 66% females in animal species survey and 64% in plant species compared to the state statistics which is 51% female (Figure 5-2). Majority of the respondents were also mature people between age 46 and 65 who accounted for over 50% of respondents compared to about 11% of people under 35 years old (Figure 5-3). About 30% had an annual taxable house hold income between $35,000 and $60, 000. A total 6% had less than $15,000. The survey wa s an exact representation of people with over $150,000 taxable income in the state which is 4% of the population (Figure 5-4). The samples had more highly educated peopl e (Bachelors degree and above about 40% respondents) compared to the state (22%). The p opulation with at least so me college classes was well represented compared to people with high sc hool education and college degrees (Fig 5-5). Invasive Species Knowledge Respondents were asked to indicate how much knowledge they had on invasive species prior to our survey. This knowledge assessment results revealed the results close to the preliminary survey (11% vs. 13%) for people who said they were expert s on invasive species (Figure 5-6). The moderate knowledge group resu lts in this survey were higher and the no knowledge group number was lower compared to th e preliminary survey results on knowledge.

PAGE 53

53 Respondents were asked to give their opinion as to whether inva sive species have effect on the parks and would affect their choice, enjoyment and frequency of visit to parks. This was done on a Likert scale of 1-5, where Strongly agre e=1, Undecided=3, and Strongly disagree =5. Results for the two samples were almost the sa me with less than 30% (cumulative of strongly agree and somewhat agree) of the respondents ag reeing to the effects. On the same cumulative basis, over 15% respondents said that invasive sp ecies could benefit Florida parks (Figure 5-7). Respondents were asked to indicate whether th ey have taken any action in response to invasive species threat. Examples of actions against invasive species were defined to the respondent .They included helping to remove inva sive species from natura l areas, travel farther to visit an alternative location with fewer i nvasive species and donati ng money or supplies to help remove invasive plants in natural areas. Le ss than 16% of residents in the state have taken some action against invasive sp ecies (Figure 5-8). For the re spondents who had taken action, over 50% participated in removi ng invasive plants in natural areas and some respondents said that they have participated in removing i nvasive species in their communities. Some communities have organized volunteer sessions to combat invasive plants. Here in Gainesville, one such event is organized annually and th is year it took place on January 27. Over 1200 volunteers gathered to collect the ex otic invasive plant, Air potato, ( Dioscorea bulbifera ) which is threatening native plants in th e area (The alligator 01/29/2007). Respondents were also asked how frequently they have participated in nature related outdoor activities in the last twelve months (Figure 5-9). Although more than 35% of the respondents did not visit the park at all, over 15% visited the st ate parks at least once a year. Cumulatively, over 65% of the respondents visite d Florida state parks last year. On average, survey respondents visited the park at least once a month (11.11 times a year).

PAGE 54

54 Table 5-1. Demographic profiles for su rveys compared to Florida profiles Preliminary Surveys Main Surveys Survey KnowledgeAttributesWAS WPS Florida Urban 49%47.2%27.2%28.0% 47.0% Suburban 41.7%44.1%55.5%55.0% 44.0% Rural 9.3%8.7%17.3%17.0% 9.0% Male 55.3%49.0%34.5%36.0% 48.8% Female 44.7%51.0%65.5%64.0% 51.2% 18 25 years 15.3%13.2%2.0%1.2% 7.8% 26 35 years 14.3%12.4%9.7%11.2% 16.9% 36 45 years 19.6%18.8%20.0%19.0% 20.1% 46 55 years 29.9%24.1%25.2%27.3% 16.8% 56 65 years 15.6%17.3%25.3%24.8% 12.6% More than 65 years 5.0%14.3%17.8%16.5% 25.9% High School or less 4.0%18.1%32.5%39.1% 48.9% Associate or some college 25.7%17.0%26.1%27.6% 28.8% Bachelor's degree 18.3%22.4%24.2%19.8% 14.3% Advanced degree beyond bachelor's 52.0% 42.7% 17.2%13.5% 8.0% Less than $14,999 15.1%22.8%6.1%5.9% 16.3% $15,000 $34,999 14.1%13.1%18.9%21.5% 28.7% $35,000 $59,999 13.7%11.0%29.2%31.5% 24.8% $60,000 $74,999 12.3%10.1%13.3%14.0% 11.1% $75,000 $99,999 13.0%12.7%17.0%13.8% 8.7% $100,000 $149,999 12.0%10.5%11.4%9.4% 6.3% More than $150,000 19.7%19.8%4.1%3.9% 4.1% US Census 2000 0% 10% 20% 30% 40% 50% 60% UrbanSuburbanRural Location WAS n=640 WPS n=648 Florida Figure 5-1. Location profiles of samples and Florida residents

PAGE 55

55 Figure 5-2. Gender profiles of samples and Florida residents 0% 5% 10% 15% 20% 25% 30% 18 25 26 35 36 4546 55 56 65Over 65 Age WAS WPS Florida Figure 5-3. Age profiles of samples and Florida residents 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% WAS WPS Florida Female Male WAS n=640 WPS n=648 WAS n=640 WPS n=648

PAGE 56

56 0% 5% 10% 15% 20% 25% 30% 35% < $14.9$15-34.9$35-59.9$60-74.9$75-99.9$100149.9 >$150,000 Income WAS WPS Florida Figure 5-4. Income profiles of samples and Florida residents Figure 5-5. Education profiles of samples and Florida residents 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% High School Some College Bachelors Graduate Education Level WAS WPS Florida WAS n=640 WPS n=648 WAS n=640 WPS n=648

PAGE 57

57 Figure 5-6. Comparison of knowledge of invasive species in three surveys Figure 5-7. Perceived impact of invasive species on enjoym ent and other aspects-(main) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% EnjoymentFrequencyChoiceBeneficial Effects of invasive plants WAS Yes WAS No WPS Yes WPS No 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% WPS WAS Preliminary Surveys None Moderate Expert WAS n=640 WPS n=648 WAS n=640 WPS n=648

PAGE 58

58 Figure 5-8. Respondent s taking action against invasive species 0%10%20%30%40% Percentage of respondents Daily Weekly Monthly Once every 2 to 3 months Once every 4 to 6 months Once every 7 to 12 months Not at all Figure 5-9. Park visit frequencies 0% 10% 20% 30% 40% 50% 60% 70% 80% 90%Participation WAS WPSSurveys Yes No WAS n=640 WPS n=648 n=640

PAGE 59

59 CHAPTER 6 MODEL SPECIFICATION AND EMPIRICAL RESULTS Model Variables This chapter presents the empirical results es timated using the conditi onal logit models as described in chapter 3. In the model the dependent variable is the respondents choice which takes a value of Y=1 if park A is chosen and Y=0 ot herwise. The independent variables include the park attributes and fee and th e respondents socioeconomic and invasive species attitude characteristics as presented in Tables 6-1 and 62. Socioeconomic variable s and invasive species attitude variables are together referred to as individual specific variab les. Invasive species attitude variables includ e knowledge of invasive sp ecies, if affected or ha ve taken action against invasive species and if the respondent think th at invasive species co uld be beneficial. Non-Response Errors Testing Since we used internet as the mode of data collection, we believe that problems like non response errors which are common when data is co llected on the web, may exist. We checked for the presence of non-response bias by testing for statistical sign ificance between the distribution of demographic characteristics in sub-samples of the early and late re spondents (Armstrong and Overton, 1997). The early respondents are the fi rst fifty and the late respondents are the last fifty. We conducted a chi-square distribution of age, income, e ducation, gender and location for early and late responses to determine whether th e characteristics of early survey occur in the same proportion as those who answered later. Late comers are thought to resemble-non respondents so a statistically significant diffe rence between the sub samples will give an indication for non response bias, which would n eed to be corrected. Results from these subsamples are presented in Table 6-3. The chi s quare test indicated no statistical difference between the early and late respondents sub-sample s at the 95% confidence level, therefore, no

PAGE 60

60 evidence that non-response bias exists. The ratio of males to females for the early and late subsamples was statistically significant at 99% conf idence level in the plant species model. The early respondent group had 40% male while th e late respondent group had only 26% males. Hypothesized Signs on Parameters The other specification requirements of the m odel are hypothesized relationships between the independent variables and the probability of choosing a certain park alternative. Based on general economic theory, we hypothesize a negative sign to the fee parameter. It is expected that as fee increases the probability of choosing a pa rticular park would d ecrease. While invasive species can provide some positive utility to some individuals, such as in their own yards, this study sought to aid managers in making decisi ons regarding ongoing a nd planned control and mitigation programs. As such, this study focused on the negative impacts of invasive species. For this reason it was assumed that invasive plants have negative impact on utility. We hypothesize that respondents will be more likely to choose parks with le ss invasive plants giving a hypothesized negative sign to the invasive species attribute. Therefore, the negative coefficient indicates a willingness to pay to reduce invasive species. We also hypothesize a positive sign to facilities, plant and animal species diversity. The interaction variables were not assigned specific signs. In summary, the models specification for plant species follows: i jS fe is ps fa Y 4 3 2 1 0 (6-1) The model specification for animal species: i jS fe is as fa Y 4 3 2 1 0 (6-2) where Y is the respondents choice, i and j are preference parameters to be estimated, Si represent the individual specific variable inte racted with alternativ e specific attribute Xi (fa, as, ps, is and fe). The signs on parameters in th e specified models indica te the hypothesized sign.

PAGE 61

61 Parameters are estimated with the maximum likelihood procedure for logit model with the STATA statistical package version 9. Statistically Significant Individual Specific Variables Part of the process of selecting a model is to compile the individual estimated coefficients. A good model will include estimated coefficients that are statistically different from zero (McDonald et al., 2005). For this reason individual specific va riables which play a role in preference variability had to be determined. Socioeconomic and invasive species attitude variables were interacted with park attributes and the parameters were estimated. Table 6-4 shows the results of test of signi ficance for these variables when inte racted with park attributes in the plant species model. Variables that were test ed and found not significant display an ns and if the interaction was not test ed it is indicated by a double da sh. The coefficients for the interaction terms which were f ound significant in this process ar e indicated by asterisks (*) for significance at the 5% confidence level. In order to maintain consistency, the same model was used for each of the sub-samples. The same process was completed for the animal sp ecies survey and interestingly, the interaction terms had relatively similar results as the plant species sample (Table 6-5). A complete list of parameter estimates for the attributes, social economic and invasive species attitude variables is presented in Table 6-7. Generally, for both models, the socioeconomic variables were not significant. The invasive species attribute when interacted with specie s attitude characteristics were significant and therefore, included in the final model. The in teraction with invasive species resulted into Knowledge*Invasive species where knowledge was classified as Expert =1, Moderate = 2 and None =3; Affected*Invasive species where a ffected =1 and not affected = 0. Taken

PAGE 62

62 action*Invasive species where action =1 and no action =0; Bene fit*Invasive species where benefit =1 and no benefit = 0. Specification Returning to the theoretical model described in chapter 3 and the four parks attributes, the utility functions for the two surveys are: UWPS = 0 + 1fa + 2ps + 3is + 4fe + (6-3) UWAS = 0+ 1fa + 2as + 3is + 4fe + (6-4) For each survey the utility difference between choi ces with invasive species attitude included: dUi AB = X + S + AB (6-5) where dUi AB = the utility difference between choice A and choice B; S = Si (xjA xj B); x=( xjA xj B); and AB = (eiAeiB). x can be defined as changes in attribute levels as follows: Facilities = faA-faB (fa ranging from 0-2) Native Plant diversity = psA-psB (ps ranging from 0-2) Animal species diversity = asA-asB (as ranging from 0-2) Invasive species =isA-isB (is ranging from 0-2) Fee =feA-feB ( fe ranging from $0-$20) where A and B represents park choices for exam ple fa =0 if park A has minimal facilities and fa=1 if the facilities were adequate as shown in Table 6-1. An Example of S will be Si(isAisB) for change in invasive species level as it inte racts with one of the i nvasive species attitude variables from the respondent for example invasive species knowledge. The determined utility difference goes into the logit function fo r the estimation of preference parameters.

PAGE 63

63 Results Overall Model Fit The estimated models based on the animal and plant species samples are presented in Table 6-6. The focus is on the probability of choosing an option with the underlying theory that maximum individuals utility will determine wh ich option they choose. The probability of choosing an option is a function of facilities condition, species diversity ( both with positive relationship), invasive sp ecies and fee (both with negative relationship). In models of choice, R2 for good models ranges from 0.2 to 0.4 (Louviere et al., 2000). From the likelihood ratio index (Pseudo R2 or McFaddens R2), both models do not appear to fit the data well as the McFaddens R2 was 0.0346 for the WPS model and 0.0226 for the WAS model. However, for large numbers of observations a model can fail goodness of fit test but still be adequate for practical purposes and it is possible to focus on estimation rather than hypothesis testing (Agresti, 1996). So the analysis was conducted on this basis because we had about 4500 observations for each sample. The plant species model relatively has a higher prediction rate. This might be an indication that there is less preference variability with in the respondents about plant specie diversity in natural areas than it is for animal species. Attribute Variables The parameter estimates for the variables in the two empirical models are included in Table 6-7. The conditional logit models indicate that each of the attribute parameter estimates were significant with the hypothesized sign. Fee and invasive species had negative signs and facilities and species diversity had positive signs. The intercept was not significant so the results interpreted are from the models repeated without a constant. Comparison of the parameter estimates across the models allows us to draw conclusions about the relative importance of these attribut es for the two samples. For both samples the

PAGE 64

64 invasive species variable shows a negative coefficient that is larg er in magnitude than any other attribute. This indicates that in choosing a pa rk, respondents were more sensitive to changes in invasive species in natural areas th an any other factor in the models. Probabilities and Marginal Effects for the Models The sensitivity for choice preference can be fu rther explained by the marginal effects of the attributes to the probabilities. Using the plant species results as an example, we calculated the probabilities of choosing park A for the seven ch oices which were presented to the respondents. The choice with the highest level of invasive plants and fee had the lowest probability while the choice with the lowest level of invasive plants but free has the highest pr obability as shown in Table 6-8. It is recommended that parameter estimates fr om choice models should be transformed to yield estimates of the marginal effect, that is, th e change in predicted probability associated with changes in the explanatory variables (Green, 200 3). The marginal effects are nonlinear functions of the parameter estimates and the levels of the explanatory variables, so they cannot generally be inferred directly from the parameter estimate s. The marginal effect in this study as they compare to the actual parameter esti mates are presented in Table 6-9. The plant species model shows a marginal eff ect of -0.127 for invasive species compared to 0.079 plant species diversity. In the animal sp ecies model, the marginal effect of invasive species is -0.114 compared to 0.094 for animal sp ecies diversity. These attributes were valued more than facilities and fee. Based on estimated coefficients and marginal effects, overall fee was the least important attribute in impacting th e probabilities but it plays an important role in determining the willingness to pay as w ill be reflected in the next section.

PAGE 65

65 Marginal Willingness to Pay Measures Willingness to pay was another goa l of this study and its correct interpretation from the choice experiment could benefit the state in planning invasive species management programs. However, the CE in this study pr esented the respondent with hypothetical c hoice sets without the Neither or status quo option. This forced choice exercise is said to limit the usefulness of the CE methodology as we can only estimate the marg inal price of the attr ibute but not the net willingness to pay for different choices (Longo, 2007). Therefore, the MWTP for attributes in this study are simply the implicit prices and the choices are only ranke d by using the utility scores. The study had four attributes in each survey and the MWTP or the implicit prices were determined for each attribute for the two surveys. The implicit price of the attribute was made on the basis of ceteris paribus4 assumptions. By factoring out the marginal utility of money (Holmes et al., 2003), the coefficients for the attri butes can be translated into the implicit prices for attribute (Table 6-10) as follows: fee attribute Price Implicit (6-6) The marginal utility of money is simply the negative of the fee coefficient. Since we assumed linear relationship in attribute levels, with three levels of invasive species, a change from numerous and dense to few and dispersed in the WPS survey is valued at $6.85 which is the same value for moving from few and dispersed to none. 4 With all the other attributes remaining at the mean levels.

PAGE 66

66 The implicit price was the highest for invasive species in both models followed by species diversity (Figure 6-1). Implicit pr ices were slightly higher for a ll the attributes in the animal species model. Facilities implicit prices ($4. 49 for WAS and $4.13 for WPS) were the lowest. Marginal willingness to pay a nd individual specific variables Although demographic characterist ics per see (gender, educati on, age, income) overall did not cause variations in choice preferences when interacted w ith attributes, some special characteristics of the respondent had effects on es timated parameters and MWTP for the invasive species attribute. These characte ristics included knowledge of inva sive species, if the person has been personally affected or participated in ta king action against invasive plants. People with these characteristics were willing to pay more to reduce invasive species. On the other hand, respondents who claimed that invasi ve plants were beneficial for the natural areas were willing to pay less compared to those who did not thi nk these plants were beneficial (Table 6-11). Marginal willingness to pay by regions The MWTP or implicit price for invasive sp ecies was further analyzed by regions for North, Central and South Florida. We asked each respondent to indicate which County they reside. Using the results from this question we grouped the 68 state counties in their respective regions and generated MWTP estim ates for attributes by regions. South Florida had the highest implicit prices for all the attributes in both th e samples (Table 6-12). For the WAS sample, South Florida was followed by North Florida and Centra l Florida last (Figure 6-2B). For the WPS sample, South Florida had the highest price for invasive plants but Ce ntral and North Florida switched places. In the WPS there are no visibl e variations between regions for the implicit prices for facilities and plant species diversity bu t the price difference for invasive species is very evident (Figure 6-2A). This implies that South Florida region have a relatively higher willingness to pay to reduce invasive species than the other two re gions. In addition, within regions in the

PAGE 67

67 WPS sample respondents appeared to value speci es diversity and facili ties about the same. Respondents in three regions were practically simila r in their preferences. They all had invasive species with the highest implicit price, followed by species diversity and then facilities. Relative Weighting of Attributes The importance of each of the model attributes had to be expressed in relation to all the other attributes in the model and how they im pact recreational utility. The normalized weights for attributes were calculated fo r this purpose (Figure 6-3 A and B)5. Respondents gave negative weights to fee and invasive species and positive we ights for facilities and species diversity. From these weights the most important attribute is th e fee followed by invasive species. By looking at Figure 6-3A, the weights for plant species and f acilities in the WPS mo del look the same. Since native plant species diversity and facilities conditi on have the same mean level, a test statistic was conducted for this sample and we fa iled to reject the null hypothesis that 1 = 2. 1 and 2 represents the attribute coeffi cients for facilities conditi on and native plant diversity, respectively. The same test was applied to the WAS sample for facilities and animal species diversity attributes but this time we rejected the null hypothesis. Therefore, the ranking for the attributes based on the relative weights in bot h models, fee is the most important followed by invasive species. Native plant species are tied with facilities in the WPS model while animal species diversity is ranked number three followed by facilities in the WAS model. Choices Ranking by Utility Scores As previously mentioned ranking in this study could only be done by ut ility score, not the net willingness to pay. The most preferred attribut e levels were excellent facilities, high diversity in plant and animal species, no invasive plants and free. The least preferred combination of 5 The normalized weight were computed by multiplying each attribute coefficient with its mean level and then multiplying the product times ten (Milon et al., 1999)

PAGE 68

68 attribute levels was the minimal facilities, low diversity in plant and animal species, numerous and dense invasive species at a $20 fee. Park choices according to ut ility for all 81 possible alternatives were calculated. The choices with positive derived utility for both the plants and animal species samples are presented in Appendi x B. Utility scores are 1.33 and 1.246 for the most preferred and -2.194 and -2.498 for the le ast preferred for the animal and plant species samples, respectively. The largest contributor to th e utility in the alternativ e is fee as reflected by high utility for the alternatives which were fr ee. This is clearly presented in Table 6-13 by changing just one attribute level in a park choice and calculating th e change in utility as a result of the level change. For example in the WPS mode l, a change in fee from free to $10 will reduce the utility by 0.74 units while a change in inva sive species from none to few and dispersed will reduce the utility by 0.51 units. Results for the Combined Model Since there was not much differences in the estimated coefficients and MWTP for the two samples, at the end of the analysis we decide d to combine the WAS and WPS samples to get the results for one model. The MAU survey for the animal species had no native plants attribute and vice versa for the plant species survey. We made an assumption that for each survey the missing attribute was the same in the two alternatives For example, in WPS when a respondent is presented with a choice set of park A and park B the animal species level is the same in both parks. Therefore, asA-asB =0 for the plant species sample and psA-psB =0 for the animal species sample. We estimated the model only for the attri butes and the results ar e presented in Table 614. For the attributes which were in WAS a nd WPS, their estimated coefficients and MWTP were between the values for th e separated model. For example, the MWTP to reduce invasive species in the combined model was $6.99 which is between $6.85 and $7.15, the values of MWTP in WPS and WAS, respectively. The ranki ng of the attributes based on relative weights

PAGE 69

69 turned out exactly as they were when the models were separated (Figure 6-4). Fee is the highest ranked followed by invasive species, then anim al species diversity. Native plant species and facilities received the same weight. Hypotheses Testing The basic hypothesis in this study is that invasive plants are undesirable and are considered unsightly in parks and recreational places such th at recreation satisfaction in these areas will be reduced by the presence of invasive species. For th is reason natural area users will be willing to pay to reduce invasive species in outdoor recreation areas. We fail to reject this hypothesis. Invasive species had a statistica lly significant negative coefficient indicating a willingness to pay to reduce invasive species. The MWTP to re duce invasive species was $6.85 and $7.15 per person for the plants and animal species samples, respectively. It was anticipated that the willingness to pay for invasive species wi ll not only be because of the effect of invasive plants on recreational utility but also b ecause of the perceived impact of these species on the environment. We fail to reje ct this hypothesis. Our estimates suggest that invasive species attitude char acteristics like the le vel of knowledge of i nvasive species and whether people had taken action against invasive species has an impact on MWTP. For example people with higher knowledge on inva sive plants were willing to pay more to reduce invasive species than people with less knowledge. Some socioeconomic variables such as income age, and education were expected to be significant determinants of the impa ct of invasive plants on the recreational value. We reject this hypothesis. Socio-economic charact eristics had no statisticall y significant impact on park preferences and MWTP when interacted with the invasive species attribute. It was also anticipated that since Florida regi ons are affected at diffe rent levels by invasive plants, willingness to pay will likely be differe nt between regions and perhaps higher for the

PAGE 70

70 most affected region. We fail to reject this hypothesis. MWTP for invasive species differed significantly between regions a nd South Florida which is the most affected region had the highest MWTP to reduce invasi ve species in natural areas.

PAGE 71

71 Table 6-1. Basic models independent variables Attribute Levels Scores Variable Minimal 0 Facilities Condition Adequate 1 Fa Excellent 2 Low 0 Native Plants Diversity Moderate 1 Ps High 2 Low 0 Animal Species Diversity Moderate 1 As High 2 None 0 Presence of Invasive Species Few & dispersed 1 Is Numerous &dense 2 $0 0 Fee $10 10 Fe $20 20 Table 6-2. Socioeconomic and inva sive species attitude variables Variable Type Meaning Response/categories Age Continuous Age of the respondent (years) 1-6 (six categories) Gen Categorical Gender 1,2 (male, female) Edu Categorical Attained level of Education 1-4 (four categories) Emp Categorical Employment st atus 1-5 (five categories) Inco Continuous Annual Gross household income 1-7 (seven categories) Env Categorical Membership in Environmental org. 1,2 (yes, no) Loc Categorical Location where respondent live 1,2,3 (urban, suburban, rural) Reg Categorical Region in Florida 1,2,3 (Central, South, North) Mar Categorical Marital status 1-4 (four categories) Act Categorical Took action against Invasive plants 1,0 (yes, no) Ben Categorical Invasive plants are beneficial 1,0 (yes, no) K Categorical Knowledge of invasive plants 1,2,3 (expert, moderate, none) Aff Categorical Affected by invasive plants 1,0 (yes, no) Table 6-3. 2 Tests statistics of first 50 and la st 50 respondent characteristics Location( df =2) Gender( df =1) Education( df =3)Income( df =4) Age ( df =3) WPS 2 4.964 7.040***1.9740.616 1.896 p-value 0.084 0.0080.5790.961 0.593 WAS 2 0.891 1.6301.8276.163 5.055 p-value 0.641 0.2020.6080.188 0.168 Yates chi-square test used instead of the Pearson chi-square when df = 1. *** Significant at the 99 % level of confidence

PAGE 72

72 Table 6-4. Significance test for attribute interaction with individual specific variables-WPS Variable Fee Plant Sp. Facilities Invasive Sp. Knowledge ns ns Income ns ns Education ns ns ns ns Affected ns ns Benefit ns ns ns Action ns ns ns Location ns ns ns ns Gender ns ns ns ns Age ns ns ns Marital status ---ns Environment -ns ns ns Employment ----* Significant at 5% confidence level Table 6-5. Significance test for attribute inte raction with individual specific variables-WAS Variable Fee Animal Sp. Facilities Invasive Sp. Knowledge ns ns ns Income ns Education ns ns ns ns Affected ns ns Benefit ns ns Action ns N* ns Location ns ns ns ns Gender ns ns ns ns Age ns ns ns Marital status ns ns -ns Environment ns ns ns ns Employment ns ns ns Significant at 5% confidence level Table 6-6. Overall fit for the multi-attribute models WPSWAS No. of Observations 45364480 Likelihood function value -2994-3027 Pseudo R2 0.03460.0226 Sensitivity (actual 1s correctly predicted) 66.2%64.3% Specificity (actual 0s correctly predicted)53.6%51%

PAGE 73

73 Table 6-7. Coefficient estimate for the multi-attribute models Plant Species Model (WPS) An imal Species Model (WAS) Variable Coefficient Std. Error Variable Coefficient Std. Error Facilities 0.306* 0.038 Facilities 0.287* 0.038 Native Plant Species 0.308* 0.038 Animal Species Diversity 0.379* 0.039 Invasive Species -0.497* 0.040 Invasive Species -0.458* 0.040 Fee -0.074* 0.006 Fee -0.064* 0.006 constant 0.031 0.034 constant -0.002 0.034 Facilities 0.307* 0.038 Facilities 0.287* 0.038 Native Plant Species 0.316* 0.037 Animal Species Diversity 0.378* 0.038 Invasive Species -0.509* 0.038 Invasive Species -0.457* 0.038 Fee -0.074* 0.006 Fee -0.064* 0.006 age*fa -0.005 0.015 age*fa 0.028 0.014 age*ps 0.022 0.043 age*as -0.009 0.015 age*Is -0.089 0.016 age*Is -0.087 0.016 age*fe 0.004 0.002 age*fe 0.008 0.002 gen*fa 0.016 0.039 gen*fa 0.039 0.039 gen*ps 0.022 0.043 gen*as -0.022 0.043 gen*Is -0.002 0.042 gen*Is -0.104 0.042 gen*fe 0.002 0.005 gen*fe 0.011 0.005 edu*fa 0.007 0.017 edu*fa -0.006 0.016 edu*ps 0.025 0.019 edu*as 0.009 0.018 edu*Is 0.022 0.019 edu*Is 0.001 0.018 edu*fe 0.000 0.002 edu*fe -0.001 0.002 inco*fa 0.040* 0.012 inco*fa 0.065* 0.012 inco*ps 0.021 0.013 inco*as 0.053* 0.013 inco*Is -0.003 0.013 inco*Is -0.014 0.013 inco*fe 0.005* 0.001 inco*fe 0.009* 0.001 env*fa -0.139 0.081 env*fa 0.172 0.078 env*ps -0.079 0.090 env*as 0.165 0.085 env*Is 0.124 0.091 env*Is 0.153 0.086 env*fe 0.002 0.005 env*fe 0.016 0.009 loc*fa 0.002 0.028 loc*fa 0.010 0.028 loc*ps 0.012 0.031 loc*as -0.013 0.031 loc*Is -0.031 0.031 loc*Is -0.053 0.031 loc*fe 0.001 0.003 loc*fe 0.005 0.003 ac*fa 0.033 0.051 ac*fa -0.090 0.052

PAGE 74

74 Table 6-7. Continued Plant Species Model (WPS) An imal Species Model (WAS) Variable Coefficient Std. Error Variable Coefficient Std. Error ac*ps -0.066 0.056 ac*as -0.217* 0.058 ac*Is -0.169* 0.057 ac*Is -0.424* 0.061 ac*fe 0.010 0.006 ac*fe 0.004 0.006 aff*fa 0.085* 0.040 aff*fa 0.068 0.039 aff*ps -0.023 0.044 aff*as -0.064 0.043 aff*Is -0.237* 0.044 aff*Is -0.359* 0.044 aff*fe 0.023* 0.005 aff*fe 0.029 0.005 ben*fa -0.043 0.052 ben*fa 0.057 0.047 ben*ps 0.090 0.057 ben*as 0.221* 0.052 ben*Is 0.120* 0.056 ben*Is 0.242* 0.052 ben*fe -0.008 0.006 ben*fe 0.003 0.006 k*fa -0.021 0.029 k*fa -0.009 0.004 k*ps 0.019 0.032 k*as -0.008 0.035 k*Is 0.168 0.032 k*Is 0.123* 0.035 k*fe -0.013 k*fe -0.016 0.031 Central Florida mar*fa -0.024 0.025 Facilities 0.289* 0.050 mar*as -0.058 0.028 Native Plant Species 0.306* 0.049 mar*Is -0.056 0.028 Invasive Species -0.509* 0.050 mar*fe 0.000 0.003 Fee -0.072* 0.008 Central Florida South Florida Facilities 0.260* 0.051 Facilities 0.287* 0.076 Animal Species Diversity 0.380* 0.051 Native Plant Species 0.279* 0.076 Invasive Species -0.444* 0.051 Invasive Species -0.454* 0.076 Fee -0.066* 0.008 Fee -0.059* 0.012 North Florida South Florida Facilities 0.397* 0.089 Facilities 0.279* 0.076 Native Plant Species 0.402* 0.087 Animal Species Diversity 0.355* 0.078 Invasive Species -0.592* 0.088 Invasive Species -0.390* 0.077 Fee -0.102* 0.015 Fee -0.048* 0.012 North Florida Facilities 0.374* 0.085 Animal Species Diversity 0.408* 0.085 Invasive Species -0.583* 0.086 Fee -0.079* 0.014 *Indicates Significance at the 0.05 level Bold Indicates Variables in the Final Model

PAGE 75

75 Table 6-8. Probabiliti es for park choices-WPS CHOICE FA PS IS FE P(Yi=A) PARK A Minimal Moderate Few and dispersed $10 0.653 PARK B Adequate High Numerous and dense$20 PARK A Minimal Low None Free0.679 PARK B Excellent High Few and dispersed $20 PARK A Excellent High None $20 0.615 PARK B Adequate Low Numerous and denseFree PARK A Minimal High Few and dispersed $10 0.484 PARK B Excellent Moderate None $20 PARK A Adequate Mode rate None $10 0.414 PARK B Excellent High Numerous and denseFree PARK A Excellent Moderate Few $10 0.516 PARK B Minimal High Numerous and denseFree PARK A Excellent High Numerous and dense $20 0.374 PARK B Minimal Low None $10 Table 6-9. Marginal effects for the models Animal Species (WAS) Plant Species (WPS) Parameter Coefficient Marginal E ffects CoefficientMarginal effects Facilities 0.287 0.0710.3070.076 Species Diversity 0.378 0.094 0.316 0.079 Invasive Species -0.457 -0.114-0.509-0.127 Fee -0.064 -0.016-0.074-0.019 Table 6-10. Comparison of implicit prices estimates for recreational attributes Animal Species(WAS) Plant Species (WPS) Marginal utility of money $0.064 $0.074 Invasive Species $(7.15) $(6.85) Facilities $4.49 $4.13 Diversity $5.92 $4.25

PAGE 76

76 $0.00 $1.00 $2.00 $3.00 $4.00 $5.00 $6.00 $7.00 $8.00 FacilitiesSp. DiversityInvasive Sp Attributes WAS WPS Figure 6-1. Implicit prices for recreational attributes Table 6-11. Implicit prices to reduce invasive species with attitude variables Variable WASWPS Knowledge Expert 9.439.63 Moderate 7.527.39 None 5.605.15 Affected Yes 10.869.00 No 5.355.84 Actions Yes 12.768.76 No 6.226.49 Benefits Yes 4.095.48 No 7.847.09 Table 6-12. Regional marginal will ingness to pay for the attributes Animal Species Model (WAS) Plant Species Model (WPS) North FL Central FL South FL North FL Central FL South FL Facilities $4.76 $3.93 $5.83 $3.88 $3.99 $4.83 Sp.Diversity $5.19 $5.75 $7.40 $3.93 $4.24 $4.69 Invasive Sp. $(7.42) $ (6.72) $ (8.15) $ (5.78) $ (7.05) $ (7.63)

PAGE 77

77 North FL Central FL South FL Facilities Plant Sp Invasive Sp. $0.00 $1.00 $2.00 $3.00 $4.00 $5.00 $6.00 $7.00 $8.00 Facilities Plant Sp Invasive Sp. North FL Central FL South FL Facilities Animal Sp. Invasive Sp. $0.00 $1.00 $2.00 $3.00 $4.00 $5.00 $6.00 $7.00 $8.00 $9.00 Facilities Animal Sp. Invasive Sp. Figure 6-2. Regional marginal willingness to pay A) plant species B) animal species A B

PAGE 78

78 Figure 6-3. Relative weight s of attributes A) plant species B) animal species -8-6-4-20 2 4 Relative Weights Facilities Plant Sp. Invasive Sp. Fee -8 -6 -4 -2024Relative Weights Facilities Animal Sp. Invasive Sp. Fee B A

PAGE 79

79 Table 6-13. Impact of change in attribute level on the total utility Animal Species Choices Facilities Diversity Invasive species Fee Utility Attribute Utility 1 Excellent High None Free 1.33Facilities 0.287 2 Adequate High None Free 1.043 1 Excellent High None Free 1.33Diversity 0.378 2 Excellent Moderate None Free 0.952 1 Excellent High None Free 1.33Invasive Sp. 0.457 2 Excellent High Few and dispersed Free 0.873 1 Excellent High None Free 1.33Fee 0.64 2 Excellent High None $10 0.69 Plant Species 1 Excellent High None Free 1.246Facilities 0.307 2 Adequate High None Free 0.939 1 Excellent High None Free 1.246Diversity 0.316 2 Excellent Moderate None Free 0.93 1 Excellent High None Free 1.246 Invasive Sp. 0.509 2 Excellent High Few and dispersed Free 0.737 1 Excellent High None Free 1.246 Fee 0.74 2 Excellent High None $10 0.506 Table 6-14. Coefficient estimates fo r the attributes (combined model) Variable Coefficient Std.ErrMWTPRank Facilities 0.297 0.027$4.30 4 Animal Species Diversity 0.410 0.030$5.94 3 Native Plant Species 0.284 0.030$4.11 4 Invasive Species -0.483 0.027($-6.99) 2 Fee -0.069 0.004 1

PAGE 80

80 -8.00-6.00-4.00-2.000.002.004.006.00Relative wei g hts facilities animal species plant species invasive species fee Figure 6-4. Relative weights of the attributes for the combined model

PAGE 81

81 CHAPTER 7 INTERPRETATION OF THE RESULTS FOR INVASIVE PLANT MANAGEMENT Invasive Species and Outdoor Recreation Regarding the invasive species attribute, based on respondents interviews, our results are consistent with some studies that while Florid a residents know that invasive plants have a negative impact to natural areas, the plants do not seem to affect peoples enjoyment in outdoor recreation (Wirth and White, 2006; Finn, 2006). Our preliminary surveys while searching for attributes also indicated that people do not car e about invasive species when engaged in outdoor recreation in natural ar eas. However, a study by Adams and Lee (2006) found that the presence of hydrilla in Florida lakes had a negative influen ce on anglers decisions to use certain lakes. But there is no indication of this happ ening with recreati on in wooded parks. Surprisingly, in our study both the animal a nd the plant species models found high invasive species sensitivity in choosing a wooded park for r ecreation. One explanation is that no previous work has measured the effect of invasive plan ts in recreation using a survey based methodology and analysis. The other studies were not desi gned to capture the c hoice behavior of the respondent as in this study. Th ese past studies analyzed the re lationship between park enjoyment and the presence of invasive spec ies based on what people said and not on how they behaved in choosing a park with different levels of invasive species. Estimations of Willingness to Pay to Reduce Invasive Species Based on the determined implicit price for the invasive species attribute, the statewide willingness to pay to control invasive species in state parks was estimated. The attribute levels for invasive plants were N one, Few and Dispersed, and Numerous and Dense. As respondents chose their preferred park, they re vealed their average willingness to pay for each attribute. This included the willingness to pay to reduce invasive species from Numerous and

PAGE 82

82 Dense to Few and Dispersed and from Few and Dispersed to None. Given a known current condition of state parks w ith respect to invasive plants, and with the assumption we made that the average value associated with each leve l changes is the same, we estimated the total willingness to pay to implement an invasive pl ant control policy that achieves reductions along these levels. From an online survey with the park managers to determine the current conditions of invasive plants in state parks, we can presume that the levels of invasive species in the majority of Florida state parks are curren tly Few and Dispersed; there were 4% responses indicating no presence of invasive species 50% indicated few and disper sed, 33% indicated an obvious presence and 13% of the park managers indicate d (in an open ended questi on) that the invasive species problem is being taken care of in thei r respective parks. Florid a has various types of parks; wooded, ocean and beach and river and la kes. From the 159 state parks, 116 parks offer hiking, nature trails, biking and horse trails ac tivities and we consider ed them to be wooded parks. We applied the MWTP estimates using the Few and Dispersed level as the status quo and calculated the range of WTP to prevent i nvasive species from becoming Numerous and Dense for wooded parks for the entire state. We applied the MWTP measures with the annual estimated park attendance for the fiscal year 2005-2006 (FLDEP, 2006); the state parks received about 18.2 million visitors last year with the 116 wooded parks taking 16.1 million visitors. The number of visitors per year ha s been fluctuating ar ound 18 million for Florida state parks with the local visitors estimated at 35% of total park visitors (FLDEP, 2004). For the low estimate, we multiplied the invasive species MWTP by 35 % of annual park attendance; which was 5.6

PAGE 83

83 million visitors for the wooded parks. For the high estimate, we multiply MWTP by the total annual park attendance; which was 16.1 million visitors for the wooded parks. The total amount that Florida residents would be willing to pay to keep invasive plants from becoming Numerous and Dense in Florida wooded parks amounts to $38.7 million per year in the WPS sample. With the applica tion of the MWTP estimates to 100% of 2005/2006 wooded park attendance for high WTP estimates, we arrive at $110.48 million per year (Table 71). Since the MWTP measures for the WAS survey was relatively higher, the estimates ranges from $40.4 to $115.3 per year for this model. We compared the WTP ranges for the plant species model with ranges of WTP for the other attributes in the same model. Using th e same calculation method, the WTP to improve facilities from adequate to the excellent level ranged from $23.3 to $66.6 million per year while the WTP to improve native plants diversity from the moderate to high level ranged from $23.9 to $68.5 million per year. For the animal species mo del, the annual statewide WTP ranged from $25.3 to $72.4 million to improve facilities and $ 33.4 to $95.5 million dollars to improve animal species diversity. Taking care of the invasive species has the highe st range of total WTP than to improve the other two park attributes. Since the money could be collected through an in creased park entrance fee, we assume that non-Florida residents have the same MWTP as Fl orida residents and inte rpret these amounts as the ranges of annual WTP by state park visitors to prevent invasive species from becoming Numerous and Dense and to improve facilities and species diversity in the wooded parks. Also, since Floridas regions are affected at different levels by invasive plants and not all the state parks have invasive pl ants, it will be practical to es timate the willingness to pay based on this reality. From the status quo determinatio n with the park managers, we can assume that

PAGE 84

84 4% of the state parks have no invasive plants 63% (50% + 13%) of the parks have few and dispersed invasive plants and 33% have numerous and dense invasive plants. The MWTP also differed between regions, ranging from $5.78 to $7. 63 for invasive species in the WPS sample. The Florida recreation and park syst em is divided in five districts which can be classified into the three regions of the state as follows: the Northwes t and Northeast districts as North Florida, the Central district as Central Florida and the Sout hwest and Southeast distri cts as South Florida (Appendix C). The MWTP from each level of invasive speci es was calculated by multiplying last years park attendance for the regions with the regi onal invasive species implicit price and the percentage of parks assumed for each level of invasive plants. South Florida, with the second largest number of parks but the highest attendance and highest MW TP has the willingness to pay of $20.8 million per year to impr ove its parks from Numerous and Dense invasive species to the Few and Dispersed. North Florida, with the highest number of parks but low MWTP has only $9.2 million WTP per year to achieve the same improvement. The remaining values by region are in Table 7-2. Marginal Willingness to Pay a nd the Invasive Species Attitude A study using data collected from th e National Survey on Recreation and the Environment reported that 96.7% of the respondents supported user fees or a combination of user fees and taxes to fund the serv ices (Bowker et al.,1 999). Howe ver, the implementation of an additional park fee for the purpose of invasive management should be considered carefully. Although people have indicated some willingness to pay to reduce invasive plants through an increased park usage fee, the amounts they are wi lling to pay are very low even less than one dollar per visit. An example of low willingness to pay to reduce inva sive species in outdoor recreation areas in Florida from Finn (2006) is pr esented in Table 7-3. On average people were

PAGE 85

85 willing to pay less than $4 per visit to reduce Malaleuca in recreationa l areas. Compared to these figures our estimate for invasive pl ants MWTP is relatively high. An important observation is that although the overall MWTP for invasive species was up to over $7 dollars for this study, the ranking of pa rk preferences based on u tility gives a different outcome. In both samples, the top preferences th at gave the respondent utility greater than zero, over 84% of the choices are free and on ly one park was chosen for the $20 fee. This might be an indication that the increase in pa rk fee may not be received positively by Florida state park users. In this study the respondents invasive speci es attitude had an impact on the MWTP to reduce invasive plants. These char acteristics included the level of knowledge of invasive species, if the respondent was affected by invasive sp ecies and people who have taken action against invasive plants. Thus the state could improve the awareness and knowledge of invasive species among Florida residents to increase the MWTP valu es. There have been some indications that Florida residents have interest in receiving in formation on invasive species (Wirth and White, 2006). As we calculated the statewide WTP to control invasive plants, we calculated the statewide WTP changes for a change in knowledge level. We had three levels of knowledge; none, moderate and expert. Since it was determined th at 60% of Florida residents have no knowledge of invasive species, it could be reasonable for the state to implement an educational campaign that could give Florida resident s at least the moderate level of knowledge a nd raise the awareness of the impact of invasive plants. The marginal willingness to pay to redu ce invasive species in WPS was $5.15 and $7.39 for the none and moder ate knowledge, respectively. The different MWTP were multiplied with the 2006 annual park a ttendance of local visitors to determine the

PAGE 86

86 change of WPT with the change in invasive species knowledge le vel. From the WPS model, an educational program that could increase the kno wledge from none to moderate level would increase statewide WTP by $12.6 million per year (Table 7-4). In summary, invasive plant management pr ograms in Florida could achieve the goal of reducing invasive species through education campaigns on invasive species to Florida residents in order to increase park users willingness to spend on park fees, which may in turn be used to control invasive plants in state parks. In addition, the knowledge and awareness of the impact of invasive species among Florida resi dents may lead to increased participation in invasive species management voluntary activities like the Gaines ville community annual ev ent of collecting the invasive Air potato which was mentioned in chapter four. The Importance of Invasive Species Knowledge To assess the importance of knowledge of i nvasive plants among th e state residents on invasive species management programs, we c onducted a small survey on 27 invasive plants experts as park visitors in Flor ida at the end of the study. We used the same MAU questionnaires for this small survey. From this survey, two attr ibutes (facilities and f ees) were not significant while the invasive species attribute was statistica lly significant (Table 7-5). The invasive plant experts did not give importance to the park fee and facilities but they ranked the invasive species attribute the highest among the four attributes affecting recr eational utility (Figure 7-1). In the overall models (WAS and WPS), fee was the highest ranked in impacting recreational utility. The MWTP to reduce invasive species for the experts was almost three times as much as the MWTP for the general population ($19.25 compared to $6.85 in the plant species model). The MWTP to reduce invasive species fo r these proven experts was twice as much as the MWTP for respondents who claimed to be i nvasive species experts ($19.25 vs. $9.63). The WTP to reduce invasive species from Florida re sidents who visit state parks varies between

PAGE 87

87 $29.1 million and $108.7 million per year with i nvasive species knowledge level differences. This is the comparison of WTP between residents who have no invasive sp ecies knowledge at all and residents who are invasive species experts. From this observation, it is evident that improving invasive species knowledge could si gnificantly increase w illingness to pay for invasive species management.

PAGE 88

88 Table 7-1. Annual WTP to control invasive plants in Florida wooded parks (million $) WPS WAS Attribute Low High Low High Facilities $23.3 $66.6 $25.3 $72.4 Species Diversity $23.9 $68.5 $33.4 $95.5 Invasive Species $38.7 $110.5 $40.4 $115.3 Table 7-2. Annual regional WTP to reduce invasive plants levels -WPS Regional WTP to reduce invasive plants Region Number of parks Attendance MWTP Total Few and Dispersed Numerous and Dense North FL 52 4,820,594 $5.78 $27,863,033 $17,553,711 $9,194,801 Central FL 24 3,032,471 $7.05 $21,378,921 $13,468,720 $7,055,044 South FL 40 8,275,978 $7.63 $63,145,712 $39,781,799 $20,838,085 Table 7-3. WTP per visit to reduce Melaleuca in recreational areas by Florida residents Expense Range Number Percent $0 275 44 $0-$1 48 8 $1-$4 171 27 $5-$9 58 9 $10-$15 42 7 $16-$25 23 4 $25 + 9 1 Source: Finn, (2005) Table 7-4. Annual WTP to control invasive spec iesFL residents with knowledge (million $) Level of knowledge WPSWAS None $29.1$31.6 Moderate $41.7$42.5 Experts (self declared) $54.3$53.2 Real Experts $108.7$108.7 Table 7-5. Coefficient estimate for the attributes (invasive plants experts) Variable Coefficient Std. Error P>|z| Relative Importance MWTP$ Facilities 0.170 0.230 0.460 4 3.82 Species Diversity 0.513 0.279 0.066 2 11.52 Invasive Species -0.856 0.280 0.002 1 -19.25 Fee -0.044 0.038 0.244 3

PAGE 89

89 -10-8-6-4-20246 Relative Weights Facilities Sp. Diversity Invasive Sp. Fee Figure 7-1. Relative weights of attr ibutes for invasive plants experts

PAGE 90

90 CHAPTER 8 SUMMARY AND CONCLUSIONS The determination of the impact of invasive plants in outdoor recreat ion utility in Florida parks is important for proper decision-making about invasive plant management. The revenues and satisfaction realized from stat e parks are important to many residents quality of life and are an important element of the states economy. This study provides information about residents preferences for state parks based on park attributes including invasive plants. A Multiattribute model was designed to assess these preferences and was implemented through the Internet to Florida residents. The model consisted of seven different choices sets each with two alternatives for each of the two surveys separated by animal and plant species diversity. The plan t species survey contai ned four attributes; facilities condition, native plant diversity, pres ence of invasive species and the park fee. The animal species survey had the same attributes as the plants survey but with animal species diversity instead of the native plant diversity attribute. Responses to the Multi-attribute surveys were analyzed using the conditional logit model to determine which factors significantly influenced park choices, the implied ranking of the attributes and the alternatives based on the utili ty and the marginal willingness to pay for the attributes. All the attributes were significant but demographic characteristics like age, income and education had no influence on the relationshi p between recreational utility and invasive species. The invasive species attribute had the mo st impact on preference probabilities while the park fee impacted utility the most. Invasive species attitude characteristics (which included invasive species knowledge, if a person have been affected by the species or have taken action against invasive plants and if the person think invasive plants are bene ficial) had a significant

PAGE 91

91 impact on choice preference and, therefore, willingne ss to pay for the invasive species attribute. To the contrary being a member of an environm ental club did not have any impact, most likely because we had a negligible number of e nvironmentally conscious respondents (6%). The MWTP for the various attribute levels ranged from $4.13 to $7.15. The MWTP to reduce invasive species was higher than the MWTP to improve facilities or increase native plants and animals. Estimated annual statewide WTP based on fiscal year 2005-2006 park attendance, ranged from $38.7 to $110.5 to reduce invasive plants, $23.3 to $66.6 million to improve facilities and $23.9 to $68.5 million dollars to improve plant specie s diversity. For the animal species model, the annual WTP ranged from $40.4 to $115.3 to reduce invasive plants, $25.3 to $72.4 million to improve facilities and $33.4 to $95. 5 million dollars to improve animal species diversity. The WTP to reduce invasive species ranged from $29.1-$108.7 million per year with different levels of invasive species knowledge among Florida residents. Since the results from the two surveys were rela tively similar, a quick analysis was done at the end of the study combining the two surveys. Results for the combined model for all the attributes are between the plant species and the animal species model results since overall the animal species model had relatively higher estimates. MWTP were calculated for three regions in or der to capture any differences in WTP and determine if they were related to th e extent of invasive species currently in the region. South Florida, which is relatively highl y affected by invasive plants, ha d the highest MWTP for all the attributes including the inva sive species attribute. The study determined the extent to which the park preferences and MWTP varied based on individual specific factors. These results allow us to identify various aspects within the population, which offers the opportunity for policy makers to capture the aspects to focus on in

PAGE 92

92 dealing with invasive plants. The realized effect of knowledge and perception of invasive species on choice preference and MWTP for the invasive species attribute could give some direction on how to deal with inva sive species in the state more sp ecifically in recreational parks. Since our MWTP values were higher than in pa st studies and free park s were preferred the most, increasing park fee may not be a feasible option. An educational program would be an appropriate approach towards the solution to inva sive species in Florida parks. Residents could be educated on the overall impact of invasive sp ecies both to the state economy and destruction to the environment to improve their MWTP to reduce invasive species. Since demographic characteristics when interacted with the invasive species attribute did not play an important role in choice preference, the question is not who should be targeted to eradicate invasive species but rather how the populations knowledge and awareness should be improved. The next question is how the invasive sp ecies knowledge was acquired for respondents who had the knowledge. The answer to this ques tion will determine how the education program on invasive species could be impl emented. It is therefore recomm ended that future studies find out how residents acquire knowledge about invasive species and use these means to educate the public about invasive species. One limitation of this study is that it was too broad considering the various types of parks in the states park system. It would also have be en best to gather research data through face to face interviews at the parks with the actual park users rather than gathering the information from any Florida resident through Internet. Finally, caution should be taken when referr ing to the statewide willingness to pay estimates because the MWTP for the attributes made inference to out of state park visitors who were not part of the analyzed samples.

PAGE 93

93 APPENDIX A MAIN SURVEY Survey Cover Letter Dear Florida Resident, We are requesting your participation in a University of Florida survey on Recreation and Invasive Plants in Floridas State Parks (the link to the survey webpage is located at the bottom of this letter) You have been selected as a part of a small sample of Florida residents who are being asked to complete this online questionnaire Please take a few minutes to complete the survey. This survey is divided in three parts. In the first part you will be asked to provide different valuations about a specific natural area and a second one of your choice, which is optional. In the second part you will be asked to give your opinion about what effects invasive species have had in your decision of which location to attend and enjoyment when engaging in outdoor recreational activities. Finally, we will ask you to give us some socio-economic information for our analysis. Remember that to participate in this survey you must be 18 years or older. Participation is voluntary. You do not have to answer any questions you do not wish to answer. You are free to stop the questionnaire at any time. There are no anticipated risks, compensation, or other direct benefits to you as a participant in this study. You may be assured of complete confiden tiality. You will not be identified or connected with the questionnaire in any way and participation is totally anonymous. Results will only be reported as summarized data. The information gathered in this study may be published in professional journals or presented at scientific meetings, but will not be accessible as individual data. The survey is funded by the Florida Department of Environmental Protection and is administered by the University of Florida and the Institute of Food and Agricultural Sciences. For questions about this study, please feel free to contact graduate student investigators Santiago Bucaram (santibu@ufl.edu) or Frida Bwenge (fbwenge@ufl.edu). For questions about your ri ghts as a research participant, please contact the University of Florida Institutional Review Boar d (PO Box 112250, Gainesville Fl 32611, telephone 352392-0433). Please remember that your answers to this survey ar e extremely important and may impact your future enjoyment of Floridas state parks. Thank you for your cooperation. WEB SURVEY LINK: http://www.surveymonkey.com/s.asp?u=864193701263

PAGE 94

94 Q UESTIONNAIRE ( WPS )

PAGE 95

95

PAGE 96

96

PAGE 97

97

PAGE 98

98

PAGE 99

99

PAGE 100

100 Sam p le for WAS

PAGE 101

101

PAGE 102

102

PAGE 103

103

PAGE 104

104

PAGE 105

105

PAGE 106

106 APPENDIX B PARK RANKING BY UTILITY Table B-1. Park ranking by utility -WPS FA PS IS FE FA PS IS FE Utility 2 2 0 0 Excellent High None Free 1.246 1 2 0 0 Adequate High None Free 0.939 2 1 0 0 Excellent Moderate None Free 0.930 2 2 1 0 Excellent High Few and dispersed Free 0.737 0 2 0 0 Minimal High None Free 0.632 1 1 0 0 Adequate Moderate None Free 0.623 2 0 0 0 Excellent Low None Free 0.614 2 2 0 10 Excellent High None $10 0.506 1 2 1 0 Adequate High Few and dispersed Free 0.430 2 1 1 0 Excellent Moderate Few and dispersed Free 0.421 0 1 0 0 Minimal Moderate None Free 0.316 1 0 0 0 Adequate Low None Free 0.307 2 2 2 0 Excellent High Numerous and dense Free 0.228 1 2 0 10 Adequate High None $10 0.199 2 1 0 10 Excellent Mode rate None $10 0.190 0 2 1 0 Minimal High Few and dispersed Free 0.123 1 1 1 0 Adequate Moderate Few and dispersed Free 0.114 2 0 1 0 Excellent Low Few and dispersed Free 0.105 0 0 0 0 Minimal Low None Free 0.000

PAGE 107

107 Table B-2. Park ranking by utility-WAS FA AS IS FE FA AS IS FE Utility 2 2 0 0 Excellent High None Free 1.330 1 2 0 0 Adequate High None Free 1.043 2 1 0 0 Excellent Moderate None Free 0.952 2 2 1 0 Excellent High Few and dispersed Free 0.873 0 2 0 0 Minimal High None Free 0.756 2 2 0 10 Excellent High None $10 0.690 1 1 0 0 Adequate Moderate None Free 0.665 1 2 1 0 Adequate High Few and dispersed Free 0.586 2 0 0 0 Excellent Low None Free 0.574 2 1 1 0 Excellent Moderate Few and dispersed Free 0.495 2 2 2 0 Excellent High Numerous and dense Free 0.416 1 2 0 10 Adequate High None $10 0.403 0 1 0 0 Minimal Moderate None Free 0.378 2 1 0 10 Excellent Moderate None $10 0.312 0 2 1 0 Minimal High Few and dispersed Free 0.299 1 0 0 0 Adequate Low None Free 0.287 2 2 1 10 Excellent High Few and dispersed $10 0.233 1 1 1 0 Adequate Moderate Few and dispersed Free 0.208 1 2 2 0 Adequate High Numerous and dense Free 0.129 2 0 1 0 Excellent Low Few and dispersed Free 0.117 0 2 0 10 Minimal High None $10 0.116 2 2 0 20 Excellent High None $20 0.050 2 1 2 0 Excellent Moderate Numerous and dense Free 0.038 1 1 0 10 Adequate Moderate None $10 0.025 0 0 0 0 Minimal Low None Free 0.000

PAGE 108

108 APPENDIX C WOODED PARK CLASSIFI CATION BY REGIONS Table C-1. Park cl assification by regions NORTH FLORIDA NORTH FLORIDA CENTRAL FLORIDA ALFRED B. MACLAY BIG SHOALS ANASTASIA BALD POINT BIG TALBOT ISLAND BLUE SPRING BIG LAGOON CEDAR KEY BULOW CREEK BLACKWATER RIVER CEDAR KEY SCRUB BULOW PLANTATION RUINS CAMP HELEN CRYSTAL RIVE R PRESERVE CATFISH CREEK DEER LAKE DEVIL'S MILLHOPPER DE LEON SPRINGS ECONFINA RIVER DUDLEY FARM DUNN'S CREEK FALLING WATERS FANNING SPRINGS FAVER-DYKES FLORIDA CAVERNS FORT CLINCH FORT MOSE GRAYTON BEACH FORT COOPER GAMBLE ROGERS HENDERSON BEACH FORT GEORGE ISLAND HONTOON ISLAND LAKE JACKSON MOUNDS GOLD HEAD BRANCH KISSIMMEE PRAIRIE LAKE TALQUIN HOMOSASSA SPRINGS LAKE KISSIMMEE LETCHWORTH MOUNDS ICHETUCKNEE SPRINGS LAKE LOUISA OCHLOCKONEE RIVER LITTLE TALBOT ISLAND LOWER WEKIVA RIVER PONCE DE LEON SPRINGS MANATEE SPRINGS NORTH PENINSULA ROCKY BAYOU MARJORIE KINNAN RAWLINGS RAVINE SAN MARCOS DE APALACHE O'LENO ROCK SPRINGS RUN ST. ANDREWS OLUSTEE BATTLEFIELD SEBASTIAN INLET ST. GEORGE ISLAND PAYNES PRAIRIE SILVER RIVER ST. JOSEPH PENINSULA PEACOCK SPRINGS ST.SEBASTIAN RIVER TARKILN BAYOU RAINBO W SPRINGS TOMOKA THREE RIVERS SAN FELASCO HAMMOCK WASHINGTON OAKS TOPSAIL HILL STEPHEN FOSTER WEKIWA SPRINGS TORREYA SUWANNEE RIVER WAKULLA SPRINGS TROY SPRINGS

PAGE 109

109 Table C-1. Continued. SOUTH FLORIDA SOUTH FLORIDA ALAFIA RIVER / CYTEC PAYNES CREEK ANCLOTE KEY WERNER-BOYCE AVALON BAHIA HONDA CALADESI ISLAND CAPE FLORIDA CAYO COSTA CORAL REEF COLLIER-SEMINOLE CURRY HAMMOCK DADE BATTLEFIELD FORT PIERCE INLET DON PEDRO ISLAND FORT ZACHARY TAYLOR EGMONT KEY HUGH TAYLOR BIRCH FAKAHATCHEE STRAND INDIAN KEY GASPARILLA ISLAND JONATHAN DICKINSON HIGHLANDS HAMMOCK KEY LARGO HAMMOCK HILLSBOROUGH RIVER LLOYD BEACH HONEYMOON ISLAND LONG KEY KORESHAN MACARTHUR BEACH LAKE JUNE OLETA RIVER LOVERS KEY SAVANNAS MADIRA BICKEL MOUND SEABRANCH MOUND KEY ST. LUCIE INLET MYAKKA RIVER WINDLEY KEY FOSSIL REEF OSCAR SCHERER

PAGE 110

110 LIST OF REFERENCES Adamowicz, W., J. J. Louvier e, and M. Williams. 1994. Combining Stated and Revealed Preference Methods for Valui ng Environmental Amenities. Journal of Environmental Economics and Management 26:271-292. Adamowicz, W., J. Swait, P. Boxall, J. J. Louviere, and M. Williams. 1997. Perceptions Versus Objective Measures of Environm ental Quality in Co mbined Revealed and Stated Preference Models of Environmental Valuation. Journal of Environmental Economics and Management 32:65-84. Adamowicz, W., P. Boxall, M. Williams, a nd J.J. Louviere. 1998. Stated Preference Approaches for Measuring Passive Use Va lues: Choice Experiments and Contingent Valuation. American Journal of Agricultural Economics 80:64-75. Adams, D.C., and D.J. Lee. 2006. Statewide Bioeconomic Model of the Invasive Aquatic Plants Hydrilla, Water Hyacinth, and Water Lettuce. Draft report to FLDEP-BIPM. Unpublished, University of Florida. Agresti, A. 1996. An Introduction to Categorical Data Analysis. New York John: Wiley & Sons, Inc. Alberini, A., P. Rigati, and A. Longo. 2003. Can People Value the Aesthetic and Use Services of Urban Sites? Evidence from a Survey of Belfast Residents. Journal of Cultural Economics 27(3-4):193-213. Alberini, A., P. Rosato, A. Longo, and V. Zana tta. 2005. Information and Willingness to Pay in Contingent Valuation Study: The Value of S. Erasmo in the Venice Lagoon. Journal of Environmental Planning and Management 48:155-175. Alberini, A., A. Longo, and P. Rigati. 2006. Using Survey to Compare the Publics and Decision makers Preferences for Urban Rege neration: The Venice Arsenale. Working Paper 137.06, FEEM, Milan. Alvarez, R.M, R.P. Sherman, and C. VanBes elaere. 2003. Subject Acqu isition for Web-Based Surveys. Political Analysis 11(1):23-43. Armstrong, J.S., and T.S. Overton. 1977. Es timating Non-response Bias in Mail Surveys. Journal of Marketing Research 14:396-402. Bateman, I.J., R.T. Carson, B. Day, M. Hane mann, N. Hanley, T. Hett, M. Jones-Lee, G. Loomes, S. Maurato, E. Ozdemiroglu, D.W. Pearce, R. Sugden, and J. Swanson. 2002. Economic Valuation with Stated Preference Techniques: A manual. Cheltenham: Edward Edgar, Ltd. Bennett, J., R. Blamey, and M. Morrison. 1997. Valuing Damage to So uth Australian Wetlands Using the Contingent Valuation Method. Occasional Paper 13/97, LWRRDC, Canberra.

PAGE 111

111 Berrens, R.P., A.K. Bohara, H.C. Jenkins-Smith, C.L. Sivla, and D.L. Weimer. 2003. Advent of Internet Surveys for Political Research: Comparison of Telephone and Internet Samples. Political Analysis 11(1):1-22. Berrens, R.P., A.K. Bohara, H.C. Jenkins-Smith, C.L. Sivla, and D.L. Weimer. 2004. Information and Effort in Contingent Valuat ion Surveys: Application to Global Climate Change Using National Internet Samples. Journal of Environmental Economics and Management 47:331. Blamey, R.K., J.C. Rolfe, J.W. Bennett, and M.D. Morrison. 1997. Environmental Choice Modeling: Issues and Qualitative Insights. Choice Modeling Research Report No. 4. University College, The University of New South Wales, Canberra. Bowker, J.M, H.K. Cordell, and C.Y. Johns on. 1999. User Fee for Recreation Services on Public Lands: A national Assessment. Journal of Parks and R ecreation Administration 17(3)1-14. Couper, M. P. 2000. Web surveys: A Review of Issues and Approaches. Public Opinion Quarterly 64:464-494. Couper, M. P., M. Traugott, and M. Lamias. 2001. Web Survey Desi gn and Administration. Public Opinion Quarterly 65:230-253. Darby, K.J., 2006. Consumer Preferences for Lo cally-Grown Berries: A Discrete Choice Model Estimating Willingness-to-Pay. MS Thesis Ohio state university, Columbus. Dillman, D.A. 2000. Mail and Internet Surveys: The Tailored Design Methods, 2nd Ed. New York: John Wiley and Sons, Inc. Doren. R.F., A. Ferriter, and H. Hastings (eds ). 2002. Weeds Wont Wa it! An Assessment of Invasive Exotic Plants in Florida. A Report to the Sout h Florida Ecosystem Restoration Task Force and Working Group 305/348.1665, Sout h Florida Ecosystem Restoration Task Force, Miami, Florida. Eiswerth, M.E., T.D. Darden, W.S. Johnson, J. Agapoff, and T.R. Harris. 20 05. Input Output Modeling, Outdoor Recr eation and the Economic Impacts of Weeds. Weed Science 53:130137. Finn, K. 2006. Social Econo mic Impact of Controlling Melaleuca in South Florida. MS Thesis, University of Flor ida, Gainesville. Florida Department of Community Affairs. 2000. Florida Assessment of Coastal Trends (FACT). Florida Coastal Management Program, Tallahassee, Florida. FLDEP. 2000. Outdoor Recreation in Florida 2000: Florida s Statewide Comprehensive outdoor Recreation Plan (SCORP). Division of Recreation and Parks, Tallahassee, Florida.

PAGE 112

112 FLDEP.___ The Environmental Impact and Regional Differences of In vasive Plants in Florida. Bureau of Invasive Plant Management, Circular 10, downloaded on 04/20/07 from www.dep.state.fl.us/la nds/invaspec/Circular FLDEP. 2004. FY 2003/2004 Florida State Park Sy stem Economic Impact Assessment. Division of Recreation and Parks, Tallahassee, Florida. FLDEP. 2004. Upland invasive exotic plant manageme nt program, Annual Report FY 2002-3. Bureau of Invasive Plant Management, Tallahassee, Florida. FLDEP. 2005. Upland invasive exotic plant manageme nt program, Annual Report FY 2004-5. Bureau of Invasive Plant Management, Tallahassee, Florida. FLDEP. 2006. Final Long Range Program Plan for FY 2007-2008 through 2011-2012. Department of Environmental Protection, Florida. FLEPPC. 2006. Florida Exotic Pest Plant Council. Fort Lauderdale, FL, http://www.fleppc.org visited on 09/16/06. FLEDP. 2006. FY 2005/2006 Florida State Park Sy stem Economic Impact Assessment. Division of Recreation and Parks, Tallahassee, Florida. FLDEP. Division of Recr eation and Parks Website http://www.dep.state.fl.us/parks/ visited between September 06 and May 07. Glisson, M. W. 1994. Invasive Non-indigenous Species in Floridas State Parks. In An Assessment of Invasive Nonindigenous Species in Floridas Public Lands. Technical Report No. TSS-94-100, Florida Department of Environmental Protection, Tallahassee, Florida. Gordon, D.R. 1998. Effects of Invasive, NonIdigenous Specied on Ecosystem Process: Lessons from Florida. Ecological Applications 8(4):975-989. Green P.E., and V. Srinivasan. 1978. Conjoint Analysis in Consumer Research: Issues and Outlook. Journal of Consumer Research 5:103-123. Greene, W.H. 2003. Econometric Analysis, 5th ed. Upper Saddle Rive r, NJ: Prentice-Hall. Grosz, A. 2007. Volunteers Scour the City for Invasive Air Potatoes. The alligator January 29, pp 3, University of Florida. Halbrendit, C.C., F. Yang, L. Thomas, and K. Krisnankumar. 2007. Analysis of Farm Household Preferences in Management of Invasive Species: The case of Miconia in Hawaii. Paper Presented at the 17th Annual World Food and Agribusiness Forum Symposium, Parma Italy, 23-26 June. Hanley, N., R. Wright, and W. Adamowicz. 19 98. Using Choice Experiments to Value the Environment. Environmental and Resource Economics 11:413-428.

PAGE 113

113 Harding, D., and M. Thomas. 2003. The Economics of Selected Fl orida Wildlife Management Areas. Tallahassee, Florida: Florida Fish and wildlife conservation Commission. Rep. 21, September. Hodges, A. 2006. Valuing Environmental Services for Forest Ecosystems: Implications for Policy and Outreach. Paper presented at a forum on conserving ecosystems: A cooperative research and outreach approach Gainesville, Florida, 05/01/2006. Holmes, T.P., and W.L.Adamowicz. 2003. Attri bute-Based Methods. In P.A. Champ, K.J. Boyle, and T.C. Brown, eds. The Economics of NonMark et Goods and Resources: A primer on Nonmarket Valuation. Dordrecht: Kruwer Academ ic Publishers, pp.171-219. Lancaster, K. 1966. A New Approach to Consumer Theory. Journal of Political Economy 74:132-157. Lee, D. J. and C.S. Kim. 2005. Managing Upla nd Invasive Plants on Florida Public Lands. Paper presented at USDA-TSTAR Annual Mee ting, San Juan Puerto Rico, 01/06/05. Letson, D. and J.W. Milon.(eds). 2002. Florida Coastal Environmental Resources: A Guide to Economic Valuation and Impact Analysis. Gainesville: Florida Sea Grant College Program. Longo, A. 2007. The Use of Conjoint Choi ce Experiments in Valuing Cultural Tourism Programs. Working paper, Pro-active mana gement of the Impact of Cultural Tourism Upon Urban Resource and Economies (PICTURE), European Union Sixth Framework Programme. Louviere, J. J., and G. Woodworth. 1983. Desig n and Analysis of Simulated Consumer Choice or Allocation Experiments: An A pproach Based on Aggregate Data. Journal of Marketing Research 20 (4):350-367. Louviere, J. J. 1988. Analyzing Decision making: Metric Conjoint Analysis. Newbury Park, CA: Sage Publications. Louviere, J.J, D. A., Hensher, and J. D. Swait. 2000. Stated Choice Methods Analysis and Application. Cambridge: Cambridge University Press. MacDonald, D.,. and M. Morrison. 2005. The value of Habitat and Agriculture. Client Report, CSIRO Land and Water, Australia. Maddala, G.S. 1983. Limited-Dependent and Qualitative Variables in Econometrics. New York: Cambridge University Press. Makokha, S.N, J.Karugia, S.Staal, and O. Ko sura. 2006. Valuation of Cow Attributes by Conjoin Analysis: A case Study in Western Kenya. Paper Pres ented at the International Association of Agricultural Ec onomists conference, Gold Coast, Australia, 12-18 August.

PAGE 114

114 Mazzotti, F.J., W. Ostrenko and A.T. Sm ith. 1981. Effects of the Exotic Plants Malaleuca quinquenervia and Casuaria equisetifolia on Small mammal Populations in Eastern Florida Everglades. Quarterly Journal of the Fl orida Academy of Sciences Spring 1981, Vol. 44 No. 2. McFadden, D. 1974. Conditional Logi t Analysis of Qualitative Choice Behavior. In P. Zarembka, ed. Frontiers in Econometrics. New York: Academic Press, pp. 105-142. Milon, J.W., A.W. Hodges, A. Rimal, C.F. Kike r, and F. Casey. 1999. Public Preferences and Economic Values for Restorat ion of the Everglades /South Florida Ecosystem. Economics Report 99-1, Food and Resource Economics Department, Univers ity of Florida, Gainesville. National Invasive Species Council. 2001. Meeting the Invasive Species Challenge: National Invasive Species Management Plan. http://www.invasivespecies.gov/council/main.shtml visited 11/20/06. Office of Technology Assessment, U.S. Congre ss. 1993. Two Case Studies: Non-indigenous Species in Hawaii and Florida. In Indigenous Species of the United States. OTA-F-565, U.S. Government Printing Office, Washington, DC., pp. 233-266. Olson B.E. 1999. Impacts of Noxious Weeds on Ecological and Economic Systems. In R.L. Sheley, and J.K. Petroff, eds. Biology and Management of Noxious Rangeland Weeds. Corvallis, OR: Oregon State University Press pp, 4-18. Pearce, D., and D. Moran. 1994. The economic value of Biodiversity. IUCNthe World Conservation Union: Earthcan Publications. Pimentel, D., L. Lach, R. Zuniga, and D. Mo rrison. 2000. Environmental and Economic Costs of Non-indigenous Species in the United States. Bioscience 50:53-65. Pimentel, D. 2002. Biological Invasions. Boca Raton, FL: CRC Press. SAS Institute. 1996. The FACTEX Procedure SAS Procedure Guide. Cary, N.C, pp.209-235 Snowball, J.D., and K.G. Willis. 2006. Estimating the Marginal Utility of Different Sections of Art Festival: The case of Visitors to South African national Arts Festival. Leisure Studies 25(1):43-56. Siikamki, J. 2001. Valuing Benefits of Finnish Forest Biodiversity C onservation-Logit Models for Pooled Contingent Valuation and C ontingent Rating/Rank ing Survey Data. Department of Agricultural and Reso urce Economics Seminar, UC Davis, http://are.berkeley.edu/courses/ envres_seminar/s2001/siikamaki.doc Swanson, C.S., and J. B. Loomis. 1996. Role of Non-market Economic Values in Benefit-Cost Analysis of Public Forest Management. Portland, Oregon, USDA, Forest Service, Gen Tech Report PNW GTR-361, March.

PAGE 115

115 Swait, J., and W. Adamowics. 2001. The influe nce of Task Complexity on Consumer Choice: A Latent Class Model of Decision Strategy Switching. Journal of Consumer Research 28:135-148. Throsby, D. 2003. Determining the Value of Cultural Goods: How much (or How Little) does Contingent Valuation Tell Us? Journal of Cultural Economics 27: 275-285. Tsuge, T., and T. Washida. 2003. Economic Va luation of the Seto Inland Sea by Using an Internet CV Survey. Marine pollution bulletin 47:230-236. U.S. Department of Comme rce, Census Bureau. 2001. U.S. Census 2000 American Fact Finder, U.S.. http://www.census.gov U.S. Department of Comme rce, Census Bureau. 2005. Computer and Internet Use in the United States: 2003, Special studies. Economics and Statistics Ad ministration, Washington DC, October http://www.census.gov U.S. Department of the Interior, Fish and Wild life Service and United States Department of Commerce, Bureau of the Census. 1998. 1996 National Survey of Fi shing, Hunting, and Wildlife-Associated Recreation (Florida). Washington, D.C. U.S. Department of the Interior, Fish and Wild life Service, and United States Department of Commerce, Bureau of the Census. 2002. 2001 National Survey of Fi shing, Hunting, and Wildlife-Associated Recreation (Florida). Washington, D.C. Wirh, F.F., and J.K.White. 2006. Florida Resi dents Perception and Willingness to Pay for Invasive Plants Management in Public Lands. Report No SL49-063 Submitted to FLDEP Tallahassee, Florida. Zapata, H.O., P.R. Sambidi, and E.A. Duf our. 2007. Choice Models in Policy Analysis Selected Paper prepared for presentation at the Southern Agricultural Economics Association Annual Meetings, Mo bile, Alabama, 3-6 February.

PAGE 116

116 BIOGRAPHICAL SKETCH Frida Bwenge is from Tanzania. She rece ived her Bachelor of Science degree in Agriculture from Sokoine University in Tanzan ia in 1989 with specializ ation of rural economy. After graduation, Frida worked with the Tanzan ian Ministry of Agricu lture in the Planning Division as an agricultural economist. While worki ng with the ministry of agriculture, she went to the United Kingdom and receiv ed her Master of Science degr ee in national development and project planning from the Univ ersity of Bradford in 1992. She began another Masters of Science program in food and resource economics in August, 2005. Frida plans to work for an international organization in developm ent planning and poverty eradication in her home continent, Africa, after completing her M.S. degree requirements at the University of Florida.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101118_AAAACM INGEST_TIME 2010-11-18T19:30:23Z PACKAGE UFE0021470_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 12471 DFID F20101118_AABPUC ORIGIN DEPOSITOR PATH bwenge_a_Page_116.QC.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
fa8fb9907072b19f849d5f85b5b61d18
SHA-1
d7118b7270ede4b43b27c9091a104b91531602fa
10673 F20101118_AABPTO bwenge_a_Page_109.QC.jpg
851103c95d9eaeeed5f25a58a1be9a19
b50e0abbf622e5e525d1e4dc2cfc47fff02322a6
75845 F20101118_AABOPX bwenge_a_Page_012.jpg
0f0b8f5d12c09d82a32290e0aec97496
9c79bf1d72bff27e4e49b73949196a7bb7fc265e
5226 F20101118_AABPSZ bwenge_a_Page_100thm.jpg
b70c7d02f0709bd965633e8781de17c0
d81a1d8fa249e713f9f81a09ad790bba70473368
78920 F20101118_AABORA bwenge_a_Page_041.jpg
1009567519ec87ce464e5886feba3c5c
d453c2ebe71ae28ad977bbaeb0d85c4e4c47feb2
77167 F20101118_AABOQM bwenge_a_Page_027.jpg
3d1773d39f8e057480d3b4902701a38b
b885b8d2f4e3594a9122208745c34dcccd53de6c
3626 F20101118_AABPUD bwenge_a_Page_116thm.jpg
348c5f09d649f175a25f2f7167002a14
2637911b068e2bffd2db109e59d18d072b538b4b
3319 F20101118_AABPTP bwenge_a_Page_109thm.jpg
5895681689888101b126157b225b401b
6f2d60e96ef12c78da2d21de423188879395d81f
80865 F20101118_AABOPY bwenge_a_Page_013.jpg
f762f390e18738200accf3ad004cdc13
e17158a4bea1d44e029b59126591522732b0faad
77552 F20101118_AABORB bwenge_a_Page_042.jpg
8b407317c5e0228804eec7f44dcd33f7
d0e436f43241767a81149ac64ca12ef5604f686a
76239 F20101118_AABOQN bwenge_a_Page_028.jpg
130daa8acc619db4a94cd76f52e2d774
7f2938a0eb8dd329d4cb6a8fe65c1db311857bc0
134693 F20101118_AABPUE UFE0021470_00001.mets FULL
ae6be6cce9ccb134f71db46c9bdb7f42
f0ff27dba77d09e5df510d756f13cf92d368d4a0
73998 F20101118_AABOPZ bwenge_a_Page_014.jpg
3d15e64e17f3f2c7abf241ce9db53cc9
b25de8d8abc1dc41a68ebbb738a0fb2838553255
75989 F20101118_AABORC bwenge_a_Page_043.jpg
fc2700a043c9432620fa1602d059234b
523ceaa99ccd83708565b99d270824b58af1d5e7
26163 F20101118_AABPTQ bwenge_a_Page_110.QC.jpg
d5cf306247efd7c67b32e0319a99b12e
507bbc03eb5f8c3618358d0a8f528ec804c44970
69198 F20101118_AABORD bwenge_a_Page_044.jpg
e4e0fafa028dd8f7f0f25c0a5037405b
14e264303a33a11000e441cee3765f3e2739d4e0
72844 F20101118_AABOQO bwenge_a_Page_029.jpg
83da4b6139447f0361783e633fc8a1cb
3bbb82dd48f507eb74817e15baa65a9bc17549dd
7282 F20101118_AABPTR bwenge_a_Page_110thm.jpg
22468a5f0bab4e30957bddde0cf65f4d
386b183fc7b08fa32c4a71589c63832a3cabd8d7
76546 F20101118_AABORE bwenge_a_Page_045.jpg
2f5654040b773308be1d6f4a8c53a796
9025aa59e65e827f7c90102471dac771ecb79fab
72223 F20101118_AABOQP bwenge_a_Page_030.jpg
0c695f4e03f704fc0d3e6f6cfb564bc7
17a11a947d0af0ee0a7e0e5e3a48883e2e170daf
25813 F20101118_AABPTS bwenge_a_Page_111.QC.jpg
ba286a71625faef21f2bbc2259e13de4
435caa45940143789959aad9b59ab82f09ce6d12
83383 F20101118_AABORF bwenge_a_Page_046.jpg
b28cb92fb8562278f3cd92a737a0b67c
3e3a02d768190ab66da722c28089f66e49fbf5c6
74398 F20101118_AABOQQ bwenge_a_Page_031.jpg
5789ca5983cb6e2695b5d80a1d80719c
1f019e76cf300f2ee34a8f20a08ce338b5bfe7e0
7115 F20101118_AABPTT bwenge_a_Page_111thm.jpg
02e415ea2d5bc456a71234b2045810c5
5b75e5cfa19c4a6b3438add686e5f2035c3133bf
33889 F20101118_AABORG bwenge_a_Page_047.jpg
05e7437bc5818ad0e5f4abc2fa3a9d10
84a8443cd2b626cfaae15366a02d108b08a97d37
48722 F20101118_AABOQR bwenge_a_Page_032.jpg
fb2ac5ea604ea6924c00866d48a3224d
64958369ac1f5bafe9e84b33dbb591c23b4ba6c2
28315 F20101118_AABPTU bwenge_a_Page_112.QC.jpg
31dfb6275708cc72655da31944d5b317
36dfd4cfa9278e93741f0c6eaa7085b52f727247
48130 F20101118_AABORH bwenge_a_Page_048.jpg
d30f52478520c55688bb8e90c3d00d12
a868649e5dd7132f0de425f5682296541952f4b8
68561 F20101118_AABOQS bwenge_a_Page_033.jpg
c165160b67d44aef58ffbe05ef787fda
0f7112538b4611e175ef0fe25d6da577c04c0c29
7652 F20101118_AABPTV bwenge_a_Page_112thm.jpg
ec273161cba2f11cdb03321e6de82d04
971e9e678906bc06fa4cf16840b94dc7f05a96a4
31336 F20101118_AABORI bwenge_a_Page_049.jpg
792f61b0e4f5df8bcb8b4621e416b3cf
ee6389f9c49d29594bf8ab2d46404bc0cf699309
70664 F20101118_AABOQT bwenge_a_Page_034.jpg
fb1159f1232b7156f2a30a7cd285e5cd
37e299d196394b0abc10038d246a03571b4b4a99
26388 F20101118_AABPTW bwenge_a_Page_113.QC.jpg
1f61e2b086a553f54e9f6c85556e102e
e83fb434257010fa4fc82f334edd14bbf66e0a55
42685 F20101118_AABORJ bwenge_a_Page_050.jpg
b6df6c20f49f09f4a297d42c2612cfb9
11ecc1ffe5f6556eca37870a6755fa279592513c
60358 F20101118_AABOQU bwenge_a_Page_035.jpg
fb45bce06f688304ecd5372dc13c0a0e
b8f747df4b8fee9bd1c540fc594ea78f6c1799ff
6961 F20101118_AABPTX bwenge_a_Page_113thm.jpg
4b8f51f23a55f3b71650d9bac5bdfb2e
99467155352f92a415bef28e59d11f610787ce9f
70681 F20101118_AABORK bwenge_a_Page_051.jpg
243f47d42ef9c7a8d27be97857dbed99
30a23358be4c2df4f783f8c02872c0a035dcb4ee
33223 F20101118_AABOQV bwenge_a_Page_036.jpg
04a8c5d0b7a97324b2417b96cf38d4d7
e07c46e54011e0c7bff5d701b45121ea0fb30d15
28855 F20101118_AABPTY bwenge_a_Page_114.QC.jpg
266583d3a864df038fffe55c9e30d114
925a6c110097de3b5df4ed191dc0dd0a282320c6
76238 F20101118_AABORL bwenge_a_Page_052.jpg
69a59ad22400d4514c4f22251613e43e
dcb59502800a4b9f336580fdaf84e61cc6a9b902
72763 F20101118_AABOQW bwenge_a_Page_037.jpg
267fe6a61d53f7f4398e8e05bfc095fa
b047ae0bb11dbe5fc5c46af605a08f04e12618b8
7700 F20101118_AABPTZ bwenge_a_Page_114thm.jpg
4258a26011153b380334514ed336772a
500d422226ddf89c393dfde7a7102fdab6d19beb
77085 F20101118_AABOSA bwenge_a_Page_067.jpg
ef7b144f3bf1c69b5838fb995f7020c4
800b460943dd031f030368cff6e88646776eb8ca
78316 F20101118_AABORM bwenge_a_Page_053.jpg
13174503c4ff9858794c1a83ca71fe5e
dc811bc0151a2b111c1eb1b02597a0faed0b72ce
70818 F20101118_AABOQX bwenge_a_Page_038.jpg
d600839b1b5153c1f29f3622591c21f7
110a23f905f09ef99f3b29afc222376de2f21a13
80078 F20101118_AABOSB bwenge_a_Page_068.jpg
7834f1dddaaf3883d1b73e35c0df42c4
3d09c32f247198e36b19cf3c6a8be45dbd5e63be
67493 F20101118_AABORN bwenge_a_Page_054.jpg
bc2d27e2c3a1e4501d241d1ab9f60fc1
b60a7839be251ab796bd3f675b2714fab448275b
80180 F20101118_AABOQY bwenge_a_Page_039.jpg
f44e39f5c9bd26012075c46764cb4b24
9cdf84ecb19af0e0284aee78d451d5b8fbd79a91
76816 F20101118_AABOSC bwenge_a_Page_069.jpg
6ba32de90a4ad13a96f7116d5274b6db
07d56aae62f7c3daee2b0ca6c816619ef47643d7
41005 F20101118_AABORO bwenge_a_Page_055.jpg
3beff7e0127b364bb7c977583c3d77dc
02519b30c14b86272d0b3077e96b468880d27ef9
75373 F20101118_AABOQZ bwenge_a_Page_040.jpg
5f600bc745ee8182963a1fabe6960392
88479a5cd429b03e998902f6f7482622f3fab4e4
16796 F20101118_AABOSD bwenge_a_Page_070.jpg
4b9cda11470b263387d1c8fc8134e30d
ce034cdd78a10035e785bcb031facc3bcc9f3914
66254 F20101118_AABOSE bwenge_a_Page_071.jpg
cff5e4add7096189f43414b8da35d892
6c3747acd7f108d4ff286a8660d61fa84fe8ba9c
39285 F20101118_AABORP bwenge_a_Page_056.jpg
3b749e46f4821b4448059c1bb2fe9207
c6541c62cd71e9081c076aaae90f5c14a90e1a77
47303 F20101118_AABOSF bwenge_a_Page_072.jpg
a899582ee795f9b4755dbe51d8928048
a021e275273a0e2c620dd4a262d7e457f060f897
44161 F20101118_AABORQ bwenge_a_Page_057.jpg
7c0a240bd3edcc4d8d574442adafe494
d22de81b935dfc50e8baee48a1ca20fcc4abdd7a
55536 F20101118_AABOSG bwenge_a_Page_073.jpg
49eacad6b67e9c44b5f47d01c3a24ea2
65f24c7a9382cff3a3843b3f82e21274293cbfac
33879 F20101118_AABORR bwenge_a_Page_058.jpg
3e2ac9150bb80586cc91b83e37bfe353
19a09fcd8db17340680fc2763f6644e36891c228
58271 F20101118_AABOSH bwenge_a_Page_074.jpg
8663b13510a9dac03b0cb08896a9d487
a09d4cbb685b27e91b6941620863a29a5c1e597b
74512 F20101118_AABORS bwenge_a_Page_059.jpg
0d8738e34c90454fddde52ae442366df
e2246f7f352f3de426f749f35ff25cc73e470ebe
54216 F20101118_AABOSI bwenge_a_Page_075.jpg
596cbddfd82086cd0f16376331fdac6e
c2bb5061a6c2d300feb339326ced514ea313479d
73821 F20101118_AABORT bwenge_a_Page_060.jpg
ead0de37369b3843ce3b38163938703a
76cad8b6fbd6b13208f7166d18e4e78ae700a5f3
46356 F20101118_AABOSJ bwenge_a_Page_076.jpg
9f240e1bbaaff4673c540be028a95304
37282ca5c7679e2ee9f89b3cba2a4b6aced717ec
72409 F20101118_AABORU bwenge_a_Page_061.jpg
cb73795271e9c96080816c4e0a4298be
8da7b5adaa4d4dddc705b9d29097333412ee1507
41603 F20101118_AABOSK bwenge_a_Page_077.jpg
fade98ce5148550ebf11758bff11e4a6
eef26dbe3f0779ceb8a0798de3b0324bde3d4537
53122 F20101118_AABORV bwenge_a_Page_062.jpg
abbbedbe4d0c8f5ed1c2e6812e46159c
a172c02ef6a0217d93b1572d833d841e75a2efd0
27307 F20101118_AABOSL bwenge_a_Page_078.jpg
6cfe659482df46fe46f1f74acac1bffc
babb819fd3a7d8f216f73f5fa3e4eb82afd60fba
73631 F20101118_AABORW bwenge_a_Page_063.jpg
a8a1db01ab4a94b00e5fdfdd267cffad
06209589a90150a49589eb4720728cb61a65bdca
46466 F20101118_AABOSM bwenge_a_Page_079.jpg
4bee4f7e803adf5635cd8578795cb560
a0ccaba624056ef9ff13518cec4b796ca7d2d335
70984 F20101118_AABORX bwenge_a_Page_064.jpg
989d41018b6d56665523ddc9a1632768
480b86086aec1484bc697774a29932b9c73c8ad2
90530 F20101118_AABOTA bwenge_a_Page_093.jpg
029f38351af173cd395d34036a93bb71
4feabadd06bea90a2c48b76a3d363a280ca4a798
19813 F20101118_AABOSN bwenge_a_Page_080.jpg
a5acee91e33a83c44fed9beb5e968805
6ebaaca1b2f6a436d7d1ebd8c7080aed4ec33bb2
61355 F20101118_AABORY bwenge_a_Page_065.jpg
2ecc86f00f29ab97d45cf3b0f5a81f60
04ac6875a0964e75fd0642515d49234840ace000
34614 F20101118_AABOTB bwenge_a_Page_094.jpg
49d6740753a10f8a1a38902d7d2473f9
a3c03cf7df76656e6d78c78c2cabd563a62ea2bd
81105 F20101118_AABORZ bwenge_a_Page_066.jpg
82309309c6aa0128aac670c3fb049ac3
b81e0ade026143a2591c15319841ee6370a9cb3b
45265 F20101118_AABOTC bwenge_a_Page_095.jpg
f003e51dbf4d122a5d99fd66902c08ec
1ed4545e2bde279835f4121201781c34eea363db
77722 F20101118_AABOSO bwenge_a_Page_081.jpg
670d24ae610d9a8ff2d112ffb4dbaa48
6626feda6439f71d33ceea48b9f26f5fcaf791dd
46839 F20101118_AABOTD bwenge_a_Page_096.jpg
e8ca0407b22aae87b55a24e683383c45
0c42350eb0f7f163f040b5097eaa4bfac22098e3
73353 F20101118_AABOSP bwenge_a_Page_082.jpg
fd5f48410ddc8a23000f60d45afd58ce
c35e98acab3bd52ef197aa0b98151da59b0871dc
45615 F20101118_AABOTE bwenge_a_Page_097.jpg
6b4c387334caa340b8332ff469e82653
97053f3ac379b4bbed4c9388c4982d9baa838163
F20101118_AABOTF bwenge_a_Page_098.jpg
50199af60390a33ff305e3ee311f6aec
12998d6b0900bcdfa53140ededa241e87951af79
77129 F20101118_AABOSQ bwenge_a_Page_083.jpg
3ec5587cfc334650231a527b1db615e0
4545c718f80f6b90e05b717baadb3f610dce2178
37842 F20101118_AABOTG bwenge_a_Page_099.jpg
68b718aceff8f99caec8c7f0a2055368
ce4f04e2affc73649444de07225a47a487467d7e
77480 F20101118_AABOSR bwenge_a_Page_084.jpg
2ff2a2f0f19111be5db733795fe3bab9
e57752cfbfb6e2621dbbef4cf63bb442fa368064
46449 F20101118_AABOTH bwenge_a_Page_100.jpg
a13f1e90c7f4b9186a04e87c961a1263
1e11bf9a531f5b231c513df769402c3789c3cf14
73901 F20101118_AABOSS bwenge_a_Page_085.jpg
21a7fc77daf76993e3883871756090b8
b4deadd4582a3ce4f3f1d25addda707d5352f3bd
41049 F20101118_AABOTI bwenge_a_Page_101.jpg
fb5d52fb2d69e61f1e2dce6b1ec22ac3
6cdcf9ee797f342ca467632fadcbb65604b83ffe
81227 F20101118_AABOST bwenge_a_Page_086.jpg
0613f17236ce6ac094d8f084c1cd79e7
61e0862e8803a8d40806f6103753000ede198e3c
25103 F20101118_AABOTJ bwenge_a_Page_102.jpg
5553eaa662319c1f0ed2f9beafcc6b9f
dfb43fcf10ecd81b9a169327fd023f9a90396c1f
21866 F20101118_AABOSU bwenge_a_Page_087.jpg
173d61cd3219774a557115c96258d846
fc9ed122efdd2b19ba220bb195e2e498f3ec02ac
26453 F20101118_AABOTK bwenge_a_Page_103.jpg
391b2fc84b7226bc7023a67f724b1810
a2fb2e00a6e73d78c4e4222655651b2ba1916fd2
56748 F20101118_AABOSV bwenge_a_Page_088.jpg
1754a3b382f3789f8c96ea309520eaa8
f03dbc1215e8616fea81c31397d6d9a56515d9ab
42404 F20101118_AABOTL bwenge_a_Page_104.jpg
c7df2df460e333bb11e280f2a0edbd1a
b872efa28e3fe1d44b65f750b6d20668221d712b
18878 F20101118_AABOSW bwenge_a_Page_089.jpg
fab1e6ff655f66c102489ae85b7be39d
b7714771e2a6e3a9dea51619a8781fa79f151e3d
7637 F20101118_AABOUA bwenge_a_Page_003.jp2
0c25d4a7a013c871074ddcbc1aaa1d9e
7ad3499c1eef46051c574ac382fe5b8616f521ac
33327 F20101118_AABOTM bwenge_a_Page_105.jpg
92f360a996501f1f2ca27ce77a5cbe4c
b3bdc67446556367cbd1d01edc59d9cac49db24c
72653 F20101118_AABOSX bwenge_a_Page_090.jpg
889700d2112ae2a4f1eef8fbf4287a4b
22c6b9339c8ec3f288ccb4b1ff98150211b67dac
62752 F20101118_AABOUB bwenge_a_Page_004.jp2
25ce4f6c9d559293c9c4ab1ec0a4d0a3
652989f1117457dd520e3b2b446d0273713fab2e
42392 F20101118_AABOTN bwenge_a_Page_106.jpg
457a1bd0210d7325825fca7439467e9e
0bce7ddd91fd4f1d6337ca4bd3e4328ca97f2c4c
85132 F20101118_AABOSY bwenge_a_Page_091.jpg
57ade8f30c7b9ecaa10eb2dd7f140141
cf007de269a3f9bd499e97d643238995671ca7a5
1051979 F20101118_AABOUC bwenge_a_Page_005.jp2
50325b253d9494f094b1ad522b7b9b90
a5528a3ad15e240cc62cf3423025c3c3ef07963e
49945 F20101118_AABOTO bwenge_a_Page_107.jpg
b91416c361053cc25d6e1d2bf586d4ad
9923acb386b02fffd40f5703207f05d793f4cd91
74538 F20101118_AABOSZ bwenge_a_Page_092.jpg
f6523ed8fdc47dfc4a88ae691e67549e
5d62d1adab4eaa6c0d64755e746ae683f8375241
1051970 F20101118_AABOUD bwenge_a_Page_006.jp2
4d0de254576c3ce65e20075543f203f9
deb4800c025edb82c9f34bffdc91af82a3b07f72
64362 F20101118_AABOTP bwenge_a_Page_108.jpg
4db5b71d73398ed23dbc9f83ad283d52
b2221e449f233ccb07bfa4eb3caa2cf9175dbb49
1051960 F20101118_AABOUE bwenge_a_Page_007.jp2
4a4378550f8fad122e023d0b8afb2886
67162b6686b6d2fd91bd75ddcbaed93973dd7eb4
40650 F20101118_AABOTQ bwenge_a_Page_109.jpg
8ead90c50a1e6f277f0a59595e72fbb2
b87ce4a11a89c54ea0013f0345b9e5b06781d92c
171014 F20101118_AABOUF bwenge_a_Page_008.jp2
384bf66e3c3a3c03a95f979ceab6a005
d65a7d3503b09d480a08a1e56bceb23736287c07
1051975 F20101118_AABOUG bwenge_a_Page_009.jp2
2ce067dcf40af868d3411d62d25f8fae
77f780f8c9eda1cf0114518f3820a55f3c413747
87853 F20101118_AABOTR bwenge_a_Page_110.jpg
7d5ddbaf1af301a27d372df8d74bb20c
ef307e9e2e5dee9ac85b91493f3774083557537c
1053954 F20101118_AABPAA bwenge_a_Page_043.tif
bc4ff2a59e96d0646077cf580aa991dc
135428dc35b914de673424722925710b51f7bc91
53195 F20101118_AABOUH bwenge_a_Page_010.jp2
729669c75056bb3e2a66fca0f655840e
21da88c5e6baf245c05b0d76c50ee5f20603dafe
84965 F20101118_AABOTS bwenge_a_Page_111.jpg
1212fef02f36e580cc2a8dad894b8ae5
ae4a8e16b129bc64f8e924aff5c08e836fa53d21
F20101118_AABPAB bwenge_a_Page_044.tif
bc2d1abfef15ad1bde423669d09317ad
545cd39fc393673da6c8c94a9a5e80e6b4909f00
103355 F20101118_AABOUI bwenge_a_Page_011.jp2
5cfb68fea1cc5baedeae9a2626e677ee
645fa5b2de63ebd21fbd40b05544596a1d547159
95815 F20101118_AABOTT bwenge_a_Page_112.jpg
4d2dae4b399f2ad7de5e92f921ffbe64
31a9e43282cde8e7e3e61d88b02fbe6e4d15a509
F20101118_AABPAC bwenge_a_Page_045.tif
5a46066cae140d228eeb7063d79d93d8
c7dcb81a864cc18d43cbf47783ab41027033d41c
114859 F20101118_AABOUJ bwenge_a_Page_012.jp2
62a1c7ea76b892940e4cb8c5bbf54f52
88dfcebbc83fa3158e124f2c9d5253c2f4789999
91889 F20101118_AABOTU bwenge_a_Page_113.jpg
75b458b8a70cecd46f32355916a419bb
b11f0748bb6ab4fa238d8ac084dfe0622ca95a92
25271604 F20101118_AABPAD bwenge_a_Page_046.tif
45669b3d015f71b9201bfe270de22695
ad4a8bfbc6e8c2cac0a05d90d1323b3ab506f47f
121307 F20101118_AABOUK bwenge_a_Page_013.jp2
1bcdb785c672747c110e3f59a90ed101
c75ff4cad3cbd122c3b389574567c535086ee233
105776 F20101118_AABOTV bwenge_a_Page_114.jpg
c748ee27f78070b1b6d9a6d599cc3f31
1a67db279ffea54f14e290b48dcb785ec782a06c
F20101118_AABPAE bwenge_a_Page_047.tif
a74ae0452fb8b1dda473bd29e3064e4c
68494521d5f9d05944654490aeb8759fe6b5f947
111071 F20101118_AABOUL bwenge_a_Page_014.jp2
2698bbde89ca166cde6695ad1c5abda3
1ba3c538cac94f33344c6a4c0b770de3bd30792d
78961 F20101118_AABOTW bwenge_a_Page_115.jpg
cb304c16768b62879cc0075eaf6edfea
415dc6f126a31b60697f6bc6801a874e69faff65
F20101118_AABPAF bwenge_a_Page_048.tif
539bab71a420b2ffa0f00358fd362031
b86edf3b4c4da69ee8e82d6b845a31d864a660f6
1051900 F20101118_AABOUM bwenge_a_Page_015.jp2
9d2a7e02ff0492ff0d8b9f13c9e43607
b434c976b8c10af0de0b62c67ddc9c8a9e0bc271
38134 F20101118_AABOTX bwenge_a_Page_116.jpg
238174cd6076ae3cfcca7109ad3e8945
f2fd9595297294e98adabe9d63b23cae6ef36e79
F20101118_AABPAG bwenge_a_Page_049.tif
f22626633e0c34c16837d2c250c1efb7
5533012bb09bef52d8ed64c1799f3772bb6a711c
111298 F20101118_AABOVA bwenge_a_Page_029.jp2
688e726e7d6282fdddb9979b45c613a6
bd6c0ae8e494f59bd8c1d0f955288316b0c20919
116595 F20101118_AABOUN bwenge_a_Page_016.jp2
aeec0381cb82011abe151efac08c3a2f
170133c7d0c043c1da6a2e0c9137c0b38d8a4172
26410 F20101118_AABOTY bwenge_a_Page_001.jp2
6b8860ceb2d3419db8b4b8f1958fc798
c35d534e38dfe938309eb0baa7ff2461b6f0a4cb
F20101118_AABPAH bwenge_a_Page_050.tif
e688d5221df66a233b3416c2c4e6f0aa
a395775a449c3c6507b2cfd69b3496c67c6603cd
110239 F20101118_AABOVB bwenge_a_Page_030.jp2
ba120ad6d9ee4d5e4c5fb95e94de8a7a
1a9315f9f9d0191525658ca12c155c387747acfd
1051955 F20101118_AABOUO bwenge_a_Page_017.jp2
6b231cb1d0cde716e08593ecab0f3369
f4d7ac35220524bea9d534a0a1fd30b12dc0e493
5645 F20101118_AABOTZ bwenge_a_Page_002.jp2
efa352b66312edd16f4f5c5ca355fd4d
b1e84f0d4fea3a0fa4ae27aa41749502dad2d093
F20101118_AABPAI bwenge_a_Page_051.tif
9bbd63e9c5a5b5feaf3bd8d34d2b46cc
6741a713428cb53e28683a3d84826224a4bdafc5
113582 F20101118_AABOVC bwenge_a_Page_031.jp2
a0c374f6afc0a18d094b49bf1b7ac897
6b1d70efadd6fbb404c27b6db2ed74e03cda33b8
1051930 F20101118_AABOUP bwenge_a_Page_018.jp2
5720b283a09cc7c6ee7980a4ce3dddd3
2da870db1cee714282df58510712885a52b7ae5d
F20101118_AABPAJ bwenge_a_Page_052.tif
a7ceac0a46b431f286f088ab37d3763c
072a73d8ecfb33837af8f6c0f7085147de45add0
69885 F20101118_AABOVD bwenge_a_Page_032.jp2
d7668c2274f5796e2259a11e8218c8de
178b1c2ec5626fdcaa2633e7e8a4c14b1b942ee3
110524 F20101118_AABOUQ bwenge_a_Page_019.jp2
8f33ff2fa9721597aedf477869c59ed9
b290f7177109277787ae309608363569bd30f6b5
F20101118_AABPAK bwenge_a_Page_053.tif
13afc41fb5f52143d7aa7a719f0d4d3d
9d0bf5ad422cdef003e12bab8f607a6656825add
100525 F20101118_AABOVE bwenge_a_Page_033.jp2
0a76a593017adf1f68363cad73aa4048
1f9625f4f8180627223c4ef5e81579e0617a802a
50017 F20101118_AABOUR bwenge_a_Page_020.jp2
2883d7bc64e7d6e3032fe7dc8b43ec56
0632203f00427880e306d905cd35b814bea7f0cb
F20101118_AABPAL bwenge_a_Page_055.tif
4bbaa1eea0e498458fdd35a6755c597f
84322a014637e0a112fcb9b526e42b00500932dd
105292 F20101118_AABOVF bwenge_a_Page_034.jp2
17ef83e0b95e2696a67ef71d5bf9dab0
f2fbe5083defd7d02e63c81fa80b8c18110378f7
F20101118_AABPBA bwenge_a_Page_071.tif
d4e67ceabc59e499dda078b70c763b17
70ffaa34f10ecd43439d96d8e33443821a97e251
F20101118_AABPAM bwenge_a_Page_056.tif
3395de6be91dab9527f00a5021ac1f44
1077f79ea1f5ade74a5c78d573f33c80d30754c0
89534 F20101118_AABOVG bwenge_a_Page_035.jp2
6f83f5e4907ddce4137b78597670891d
8c048d7e35147fd4fd49cde90c344201bbef7486
414374 F20101118_AABOUS bwenge_a_Page_021.jp2
abab3a98d5db40cd6781c1a47935e3e3
be789bd9c50a6b605ecdf6cdc13eec2f7476c141
F20101118_AABPBB bwenge_a_Page_072.tif
cdbd90c30ef0eb71d9b375059c59ef45
203c3e926193b0a994b1f8d42afd4d7e162c69e9
F20101118_AABPAN bwenge_a_Page_057.tif
6028284c173e7bb2892beae5480d3ef3
06d5180bb846588a13226743512c4d095cc26241
44539 F20101118_AABOVH bwenge_a_Page_036.jp2
9a56d48c3e484b996809c6c58ddd6f9f
6970ce130ce520899c5862f12e4ece1d6cd71de3
38940 F20101118_AABOUT bwenge_a_Page_022.jp2
f808f8307190cd00d34300ad5c757013
70ff83ef5bba46b8a6737104b1658ee25814bffa
F20101118_AABPBC bwenge_a_Page_073.tif
fa06c2c7090f80107c5f0877050595b1
0be9be2f6952b1d4ae76cd204a722d51b7eb86ac
F20101118_AABPAO bwenge_a_Page_058.tif
dfde4a9decbf4ae8d86de77a09b68cb3
9cf35feffebce283df6d06b548dda4ccbe184878
108029 F20101118_AABOVI bwenge_a_Page_037.jp2
e167ed344ebb6ecee005f0187eba3856
8c56166b413623c348d7b7d13f03f8c34570836f
108285 F20101118_AABOUU bwenge_a_Page_023.jp2
f503cb1594f9d2337d2920d9c68cf186
55fd2fe0fe512b94fa97016c205334b035c79ee4
F20101118_AABPBD bwenge_a_Page_074.tif
9200f9b515ecc0839becb175656aefc8
b6a6281dcbdd144b1e37c3e3a407a68ef06b806b
F20101118_AABPAP bwenge_a_Page_059.tif
a24ac9cd61a7f04d3ecc7fd92ab7455a
d8d43b26f6fded7003e8fec44f71844499181fc4
108699 F20101118_AABOVJ bwenge_a_Page_038.jp2
f7a4270926904f1671f83434432aac68
0447c339dafee44f99c1976f89f4435e51720946
116165 F20101118_AABOUV bwenge_a_Page_024.jp2
e56163540fcd7127491beff61ffa59d4
b9a60329494d6243eff45116913b3112eebfbc96
F20101118_AABPBE bwenge_a_Page_075.tif
b7d2321bfda8fbb89d21454c62e329f6
2dc64fd90696f939571310bae872da3234b95d58
F20101118_AABPAQ bwenge_a_Page_060.tif
c4e54636e24ac6bb6c7ba0f63658ed43
f53816ee20047be71a4c960df87ebd4bcf8eb3f9
124154 F20101118_AABOVK bwenge_a_Page_039.jp2
947a142ec7a0fa711e002ab18e2be264
ef295eff9161dbb9e8da1ff81ef37318a4b4d40b
115604 F20101118_AABOUW bwenge_a_Page_025.jp2
fdce255ef4651c2adc71a554c894a09b
31793caf62078b8965fa44a393b0bed28767e038
F20101118_AABPBF bwenge_a_Page_076.tif
b35c8dd5f3500411b4b5b02513a565ff
3200e7dcafef983f99285adb772fb41e2b25ab28
F20101118_AABPAR bwenge_a_Page_061.tif
6f8c38146ad22c114fbc2b282e613181
5d34432335db98ee86bb394e99a0ad51dc6cd583
117043 F20101118_AABOVL bwenge_a_Page_040.jp2
295ef19508d3f55cb318ec20e2257474
16967eb7ec77458f7ae39684cab467d2ac8b7a94
114162 F20101118_AABOUX bwenge_a_Page_026.jp2
e5b2c5c425b8cb4c8c36b77ece5dbd43
8c06fca580dc71d99ff16a673ac9696cda47fcdd
F20101118_AABPBG bwenge_a_Page_077.tif
5028fbfccf522c20faade5cf89c071f0
963234dff5427e9fed1ce73f190bc8207c3539c0
449610 F20101118_AABOWA bwenge_a_Page_055.jp2
23b9dc3b81a9f5cb880cd7d5313fe5d1
184772f22f8745afee117bb3e0728faa013beb1b
F20101118_AABPAS bwenge_a_Page_062.tif
cff6187c0f0aa60873cf704290c0c8df
2b9500591369dff7baad7a4f0dd0d77e4c726171
118731 F20101118_AABOVM bwenge_a_Page_041.jp2
7c2b659c74ebae1d6002c21a697bba72
359b81c034e33db529eb259e0a3b8f6a3e2b34a2
115540 F20101118_AABOUY bwenge_a_Page_027.jp2
525fc1b4cdf2f427ad0e828fbf2e3021
215febf2810ffef9632e4c8a9dd192657d1739cc
F20101118_AABPBH bwenge_a_Page_078.tif
a983526ca7e242651f100afbdf23202b
ca22b317729f5c771607c5133edfe3db3ee337f7
447443 F20101118_AABOWB bwenge_a_Page_056.jp2
9c1c8a59b7f2fd19ed7ed136dae2c182
4ddccd7d6c411501a5296946cb34c45ef44d706c
F20101118_AABPAT bwenge_a_Page_063.tif
7c9920acb952c70e5b1c33e6394c1a1b
fe9f843c055e78bfb6f3326e69031e920018b340
116500 F20101118_AABOVN bwenge_a_Page_042.jp2
45fc4427d152c9ecd20af095053136b7
9081a554c070343e9c4a4907fbb77849efe001ea
116487 F20101118_AABOUZ bwenge_a_Page_028.jp2
30bdf0a7d6eae8eb905700a2ccb8fc48
3bc24bbff80c25a497c18f9e57d79ad146ccb032
F20101118_AABPBI bwenge_a_Page_079.tif
04bbf1e88c2c5b6b89fe89470158371f
f17c198a09157e95cb9a3f5f2d032194870e995d
517547 F20101118_AABOWC bwenge_a_Page_057.jp2
1d5d703ad4a459b9186ef99262eec480
3061d18f085f76e987dfcb05df74cd6603be6ec1
F20101118_AABPAU bwenge_a_Page_064.tif
446267faf439f5028c09bcee9f85ad58
421e6880dc9c346cc12bd310b4aa257ad01dfb2f
115846 F20101118_AABOVO bwenge_a_Page_043.jp2
3a8dc6fc993956d95d5e2816a7136910
2f3fe8021be5f670026fba42e781cf9fdf9e4d4b
F20101118_AABPBJ bwenge_a_Page_080.tif
64ba392103b88ac5f65ee9a89961d784
6027e48ad5518437587581f7e61efb2923daa539
352076 F20101118_AABOWD bwenge_a_Page_058.jp2
75fdedc132487c0cc5da3510228c77b4
4f4c26598fdc249350517f5fdb641c4d429a2014
F20101118_AABPAV bwenge_a_Page_065.tif
7218c0c3b6899913d508785e53077275
5a1c04044b3f1840d5b5105a656faa7db70f09e4
104377 F20101118_AABOVP bwenge_a_Page_044.jp2
9a48e1ff083c64cdd1d6bc6d494633e8
613412668e085d31b95a367ef78a1f06eacb1c0d
F20101118_AABPBK bwenge_a_Page_081.tif
28d09e20a8da8ef899cb2a0eb36e3945
c30b6d8e3169a61e3d341299b35804b79c487a0e
113818 F20101118_AABOWE bwenge_a_Page_059.jp2
2ba86c32df245d5bc6d8c786cea552a3
3b5eac551538dddaa318a713e0249f9896242cba
F20101118_AABPAW bwenge_a_Page_066.tif
0283b3853fa7814cacec668e51b71b1d
6e7431815ded4eabb19a3c49ff8e5d532de653be
116809 F20101118_AABOVQ bwenge_a_Page_045.jp2
5859955fd5bff9fea672efeb48806006
e7ae252841479ecc5a11e1cfb7be3f4ea72c77ce
F20101118_AABPBL bwenge_a_Page_082.tif
a56de45118c8a7a458788af194ba1738
0de7e14cfdb2c7bbfde222e1f4ed5a0a954c4830
114251 F20101118_AABOWF bwenge_a_Page_060.jp2
aecc6ecdac6c75638176b7984f043a73
f41b446aa1d671302d3b8bd2049aba6b0397d93d
F20101118_AABPAX bwenge_a_Page_067.tif
59dc56f77a22c39590c4f11327965836
cff093c8b1aae97d4ac971b081081403e37b4e89
1051974 F20101118_AABOVR bwenge_a_Page_046.jp2
02846beb6189b74317dd97723ecb9b3b
eba702f8df65325fd455ed51974a889fb18ebb27
F20101118_AABPBM bwenge_a_Page_083.tif
1738af269213cb37e7b3ea9acebdbb11
26410dc4184869bce9e75dda317343c7093bab6c
110526 F20101118_AABOWG bwenge_a_Page_061.jp2
233e9a1118e8a8ee7385ce7647499941
7ce466ad9d2732e7e81bdbd59f034ccbee72de5a
F20101118_AABPAY bwenge_a_Page_068.tif
1255f5e6e2599e76d9fecb30582b1f2f
4db0fc97b8158b0bc7cdad782acc825a3361efb9
346954 F20101118_AABOVS bwenge_a_Page_047.jp2
a4d110a50fb0debcc89f6d50eb31daa0
d73bea1cbd2aee1482e76d22332a80ace01ceabf
F20101118_AABPCA bwenge_a_Page_097.tif
682524c8dbbcee3ecac421a157602fe8
74d39c10025d8104f67ae762b0cafbee0214a634
F20101118_AABPBN bwenge_a_Page_084.tif
ee96cfe5910abef08eb25be0d98b5f6e
a4c31dfd05af8fd7739b6e6b9b14043709ce4418
81504 F20101118_AABOWH bwenge_a_Page_062.jp2
cc9c68d742e49baf0fc575d476530a8d
94d89166bbb30bb424da2b6ce7cc4ed1536e308d
F20101118_AABPAZ bwenge_a_Page_069.tif
1d8c122fe4952f6b4dc54265447db664
dc34da429a50444f27ff9d0a9720e4ff588da72f
F20101118_AABPCB bwenge_a_Page_098.tif
e78400df7c17baa934bc04d77c744f5e
69abc957271fb17e4b5082916a8981010045c662
F20101118_AABPBO bwenge_a_Page_085.tif
31b835b4ed1819c1e8b360b401bd05fb
bc6183b03aecb2f72b8ee2255725f0e3f7efed34
110537 F20101118_AABOWI bwenge_a_Page_063.jp2
8dfd995602d42321b9481805f04c2f7a
61bb6fbc5dcafbb0abf516b3040b0d94e750465e
514807 F20101118_AABOVT bwenge_a_Page_048.jp2
ea6b174f30761cb364091dcb57803552
6fc3addca6e082da240120cb2eb805225818ea27
F20101118_AABPCC bwenge_a_Page_099.tif
1370f31defa5fcba7584ffcf2b43fabc
d82e84f42828ef0346b4ce8bf60222a8e3963e5f
F20101118_AABPBP bwenge_a_Page_086.tif
89378e3b02c6135ca223b55f11d173c2
78f5cb63899d14a57d810672013a6049d1594004
108738 F20101118_AABOWJ bwenge_a_Page_064.jp2
1c03530ade99b4f5aac82bc2ae54a1c8
1950d9d057a36fcb004329ad0e3357cc108b26b6
331679 F20101118_AABOVU bwenge_a_Page_049.jp2
26b581587818db3a34ea0a17a25f5f9a
44c1cdd939b404e8f0c2e4801bc657620b093204
F20101118_AABPCD bwenge_a_Page_100.tif
77f136df251bc543dbbb16d814f45835
1827725d04752c56f088a1fcb9927889495d5d51
F20101118_AABPBQ bwenge_a_Page_087.tif
4fdedc79604af35ad5766f3ac84f1e76
4fabacd2d1b8e09b778e1e81126742da1c6dac86
88796 F20101118_AABOWK bwenge_a_Page_065.jp2
df8775996563f80f5d49317808c1b5bc
33cde3c90bd408603f56786984bb390fb379a6ef
840905 F20101118_AABOVV bwenge_a_Page_050.jp2
2d5b5640637265b6cff85c960cc9646b
417ec5ed4b2f05ecd7e19c9651c5952518a28c10
F20101118_AABPCE bwenge_a_Page_101.tif
f5b8a2deb02766653fb692f63325e4e0
569683f80903b7a812b412ffd77f1d5f9f5bb75e
F20101118_AABPBR bwenge_a_Page_088.tif
4d059e4a9d8fe7631ce13974016212a7
2287aa87bd747b45f427e671d882beec909ece50
124119 F20101118_AABOWL bwenge_a_Page_066.jp2
450d77b6ee24e6869c5342ddb646377e
46e56de49ab6702be001dc50c37adef0b3498d15
105198 F20101118_AABOVW bwenge_a_Page_051.jp2
065b71afadedb125512dadc899f4e1d5
2a0870b6e93bb195d5992217072e79d158d8caf0
F20101118_AABPCF bwenge_a_Page_102.tif
57ab2dc8cbea70338bd5b7ff54b544f6
bfd0d7ed8e0ff7f5d24915587a507ef1b7f12f82
115553 F20101118_AABOXA bwenge_a_Page_081.jp2
cb2216c3cf46d779d4f8ee03f0790316
ac552fc246a8662d7012852d840b8a0762cd722a
F20101118_AABPBS bwenge_a_Page_089.tif
5ab384e63215de3be90de4f11c3bdcdf
f364a8fafb2a6283242cb849362ac62727674b38
117540 F20101118_AABOWM bwenge_a_Page_067.jp2
9d7a72b7ec623f941edac4eeec44a482
1fde49e254fce0648da576de759e2789b58d8be6
118438 F20101118_AABOVX bwenge_a_Page_052.jp2
f7cb1c2f82c991f391efe6debf89db40
a7acb5dfd078fd1ddbaaba02641cbb2b6b743e62
F20101118_AABPCG bwenge_a_Page_103.tif
37e5e94da1dc6f2e3a026db445e9ed19
1f675c74063b08b19f12003e1d3776a28670e423
109127 F20101118_AABOXB bwenge_a_Page_082.jp2
1eb9a17a7b25b4dde32fdae71a34f370
16adc4778a92dbb9e9d5a3ae92f351b6ab297c4f
F20101118_AABPBT bwenge_a_Page_090.tif
c91d795718f03d96e2dd70662343de9e
550d0c2324d25f4304bd288f845e833b9ab56d39
120717 F20101118_AABOWN bwenge_a_Page_068.jp2
82de38dc7031274651e6e93330fda077
9dbf266df00c04bdb49a00e60570a392b5324523
120199 F20101118_AABOVY bwenge_a_Page_053.jp2
99faa959036574de2ab253c31191f067
f9741b67b6b267e3f77b34b44d7d8d06c9b7dc71
F20101118_AABPCH bwenge_a_Page_104.tif
b885c216825cc54f7eccde96b7e7d260
0e54ab78ea2bff7315a5543dc22b7329a50def38
115831 F20101118_AABOXC bwenge_a_Page_083.jp2
436da8a6ddad12ca35b4a6104b72d7b2
64a67350af1222c801bfba7e2a249d48298b501e
F20101118_AABPBU bwenge_a_Page_091.tif
359900c08b011daa2b2983ee4703ce54
06db5c0d5ec36ed93c271727a1367333e07171e7
117671 F20101118_AABOWO bwenge_a_Page_069.jp2
780e1776566294fd30d4b57b5ea577d1
4144a80842ca6e459df9c0bd1ed053b1fc8e732c
867759 F20101118_AABOVZ bwenge_a_Page_054.jp2
1f840c4577cfd23e9d58b8efc98a15cc
372c124c87780d4f9b15d55c557a617e4352dd90
F20101118_AABPCI bwenge_a_Page_105.tif
a9498befe62b4f1f866f65d3a65dd173
54627af8cb669e564c140f4a02e279cb47ba798a
118054 F20101118_AABOXD bwenge_a_Page_084.jp2
cad2134d3235746035b14489dff3f5ff
e18ba1d11332554a78fdd29ff77f770be8522943
F20101118_AABPBV bwenge_a_Page_092.tif
f64129f8606cf016bc1932561a6cadd6
1cdbdf7d72ad467b256e3abeba48d16661af6563
17934 F20101118_AABOWP bwenge_a_Page_070.jp2
c7b279a8b88dade66c86b19f2ade6bab
64ac4725c79b34e2bc11ff8b8eec73a3c980d5d7
F20101118_AABPCJ bwenge_a_Page_106.tif
6d79ab6b5d79c3e3ebe8a9b1c1bf0323
87a39b9c2ac73f0919b6f3b0b13613dd6546299c
114264 F20101118_AABOXE bwenge_a_Page_085.jp2
fb1493935c68ee0dd264b93e32a5a3ea
ac5a1a31ac12616cb4874452915ce0aa3c170a71
F20101118_AABPBW bwenge_a_Page_093.tif
a51361e2eedd9d32754792072409013d
d97434298903980c15e4702f09c6201928c75f3e
96048 F20101118_AABOWQ bwenge_a_Page_071.jp2
666ed1bf2af8e6ebb42732c4bdee0d43
08064ee3892162b2d10e5a0ee35266565261ee63
F20101118_AABPCK bwenge_a_Page_107.tif
053674fb3146ee3fa3147c1b649cc2e1
97c716466ce90f261a6d6015c54ca5a46c7511fa
121712 F20101118_AABOXF bwenge_a_Page_086.jp2
dd041c01a7083d6623350a0d46f205d0
e00a7faabe90fd20739bf152c9edf28ada7ff7da
F20101118_AABPBX bwenge_a_Page_094.tif
35847f327c6f01d1babb067c4408b561
e21df8b4d755446e2872a8e118852a49d7f3b79d
63961 F20101118_AABOWR bwenge_a_Page_072.jp2
c1810276f9f9566507a07f98e37dc60a
fc1f6720fdc737f0f3b4a1166c759c381373094f
F20101118_AABPCL bwenge_a_Page_108.tif
8c743d049722e34403749b3a9d9fee1d
0f878fd35f44a728e410ee1646fb3cd90028ba6a
26803 F20101118_AABOXG bwenge_a_Page_087.jp2
3dd3a39a083b161448c6bb54a504d69d
1d0c0801ad8a61e871367df052373ca93f7b3f74
F20101118_AABPBY bwenge_a_Page_095.tif
08e7da7aee5ab461c1194074863d5a2e
ceb0a61853a64958e87d4caed46f06cd5ff935df
88565 F20101118_AABOWS bwenge_a_Page_073.jp2
c671ff083cd64ef08291ab167f4fcb67
463819e4a0126b03197c0b5eaef318a0726c1154
66819 F20101118_AABPDA bwenge_a_Page_009.pro
32b7783a95dd02e12b7af4de8df37f17
bc5581f2ac536029853612ee98807b8062306a34
F20101118_AABPCM bwenge_a_Page_110.tif
00cfc9906c4d1ad6967fb315c55b4488
4174a73a59135319dfc3980951e96fcb12cc7692
80016 F20101118_AABOXH bwenge_a_Page_088.jp2
f3483e49f21469229d45a2702c98a8c6
293a90ee7f143b705befc3b5ad805c8c729c68a5
F20101118_AABPBZ bwenge_a_Page_096.tif
14d6baa4c163c9fd06070e9de9d40a59
236b807a96f57dd2343a7df7f8aa4b0feabf2054
86980 F20101118_AABOWT bwenge_a_Page_074.jp2
adee29aba369581598d9121a0acb0acf
d0cad3feddc7a25eeabc13fc2090fabd1e327f50
21035 F20101118_AABPDB bwenge_a_Page_010.pro
6fba206a38e9f59a1797f05a83f0545d
c284846773063336497d718f81bae790f7eeae47
F20101118_AABPCN bwenge_a_Page_111.tif
5eb7e1411f271990b4a1c8eb8bfb0671
265c0a50ee595dbbf3a5d63a34ce1123d928b520
158297 F20101118_AABOXI bwenge_a_Page_089.jp2
1eec192095c61daee4efd1ef0c9c5443
f4efab79aae78710e136b68977f73af3f51998c0
47208 F20101118_AABPDC bwenge_a_Page_011.pro
c756b259f930fbe79b614c299349f277
ca16e1c1dc59685c9999d4db502a3222a89915fb
F20101118_AABPCO bwenge_a_Page_112.tif
aca6860e49ecb9b958b82354fa378668
8408c16d4a16d6a85bdbb54a665802bd601792f3
109995 F20101118_AABOXJ bwenge_a_Page_090.jp2
93653ad868f1ef3ca7d6440a31a18209
96e778667ecf181bbe614fd50a124b279ac22a8d
74159 F20101118_AABOWU bwenge_a_Page_075.jp2
5f9421ae3f1d8bcaa96855942e01a20d
e74a533e414a89840b8ae1e1ba021dd6f3aafd19
52700 F20101118_AABPDD bwenge_a_Page_012.pro
9f7eb31f9e5f4bd4b5cf4c8da36099af
d4b0d28e749179fc0e749292bcd1c36c4116a5e5
F20101118_AABPCP bwenge_a_Page_113.tif
0ff2511c6d581a8474d14d3a517eb5d8
8904080e3892c3edfde97fd5e4b4186b7ee864fd
1051973 F20101118_AABOXK bwenge_a_Page_091.jp2
5f0a348b0beb197fb54e217ab3fbec45
fdc34a13afb2910e8039119bdd248518618d0a81
559915 F20101118_AABOWV bwenge_a_Page_076.jp2
77ee2264ec8404be0d32e78880bb2090
0d2f25b67579f6e95d4b0366b5e4f37ee8d2039d
57064 F20101118_AABPDE bwenge_a_Page_013.pro
cc326c4974de4175335d8df7e810b3cd
fe046e9ca70af8efce60a605098dd385328e0383
F20101118_AABPCQ bwenge_a_Page_114.tif
0733df213831c18495e2f96043cc819b
d9e62f554814b9816ceff753205d9f0f2bebcecb
506010 F20101118_AABOWW bwenge_a_Page_077.jp2
286a8708cf6ebda8a4d07bd40e325d2c
f8b23a2674bb3262c37ab932b1698d7f84e9e032
51598 F20101118_AABPDF bwenge_a_Page_014.pro
8d1b0fb33dc4e63a89e3909b5dd83f95
aaf1169f7f96a52dc78dcfedb8e7e09dc8d9f158
F20101118_AABPCR bwenge_a_Page_115.tif
4372fa9296e15ab2a6fc46f85d4c129e
1187e28878dbcd0c5a6b4b54f4b9cf579172e0bc
112673 F20101118_AABOXL bwenge_a_Page_092.jp2
8e45e0222b731ffff29e2f08d8fdef9e
d7266f36377469efbbe04e2f22da21573094965c
262439 F20101118_AABOWX bwenge_a_Page_078.jp2
731b7322312c2fba478d9134c1804b69
f955f6161808f58d2d1215171a73f6d3a42da52c
55648 F20101118_AABPDG bwenge_a_Page_015.pro
bcb7a3ef0b3818900f7be0c30305844d
a6cad2aaf965fb82161b06bda7d5cc0b0d5afdd6
68489 F20101118_AABOYA bwenge_a_Page_107.jp2
3b5964172913399bb349c8a9188d9b3f
d4cda436e73c1a83cea49f1530f1d821cc39b29b
F20101118_AABPCS bwenge_a_Page_116.tif
46657631605b420e3de54ecdf9320e30
44aa2accddac83db705a6fe241cf0ea8dab302df
1051985 F20101118_AABOXM bwenge_a_Page_093.jp2
dffa7099dd658b0bff033505d1f1b27b
a47d04ce4ac8259ed32748f3c078907af21ee76e
64820 F20101118_AABOWY bwenge_a_Page_079.jp2
bee251e7bd28b438a45c6d5af8ec9368
7c97c69180bf451646ae37125cbb87496169726b
53890 F20101118_AABPDH bwenge_a_Page_016.pro
322df336b364db3626643c0462bf6a1f
a259a9f21d24543acdd61f009897cc4b83bcfbc9
85267 F20101118_AABOYB bwenge_a_Page_108.jp2
dd59384d4b349a7b5a3dc53b45825df0
8836140d450df93d0e3df2b76a67557706c49198
8919 F20101118_AABPCT bwenge_a_Page_001.pro
4447679ddb3ee68b413b522cf5bc5761
e0739a85c369131b76c55b333968a1cdb30ba956
372462 F20101118_AABOXN bwenge_a_Page_094.jp2
2187f855ce1c10bd8d57b044ea82b374
530cfba17616fbcca849e3a7478d95710370c800
177408 F20101118_AABOWZ bwenge_a_Page_080.jp2
be68f7f22431b80ca987930939079afb
642d203b9fea4388ebf9c20a948f2b6cd78d5139
50348 F20101118_AABPDI bwenge_a_Page_017.pro
54f000389313a31412638c96591cc97c
619f717115644f7b0111a77ff57531cad2c68f0d
43994 F20101118_AABOYC bwenge_a_Page_109.jp2
3811aaff83a8aaaefb4329ae9bab0a77
6fc8951b1216212020bdde1c5d87468c2f03df9e
939 F20101118_AABPCU bwenge_a_Page_002.pro
105033efd150c158ff2d0afa79162f26
72dde78038c5dbe43fc8dafd5e4b334234b93488
773955 F20101118_AABOXO bwenge_a_Page_095.jp2
ade7a1e4054ea0665d51fd73cf16b8e6
98328edf89821bad0e20f108be7175292d824ba0
51757 F20101118_AABPDJ bwenge_a_Page_018.pro
e65a8dc7a11f7b2168e5bd241e4aeaeb
ecf9abb67ab09e1a0fbf9323c357a37fd0a91bf9
136656 F20101118_AABOYD bwenge_a_Page_110.jp2
21f005ba8493b990836923efc666921d
c1f05f87dc212aaf36feddbc16dc26ab7c9c53e5
1899 F20101118_AABPCV bwenge_a_Page_003.pro
c118e74fa042664180e0df7a49011be9
a4f020ef97556dff418c0da1c58b9664b24394f9
851407 F20101118_AABOXP bwenge_a_Page_096.jp2
d1b8c05055364b4cc4ad53be640bf85f
91bdd866f96a5263bd984ac4568653b190234b34
52128 F20101118_AABPDK bwenge_a_Page_019.pro
ad3e358e7704c2ca72bcfdb16ecb38dd
b3e70480398694508b9b3b656154f50fa4139903
134729 F20101118_AABOYE bwenge_a_Page_111.jp2
5b70503b95604e75e3c2c43dd08700a9
1b1e8cf1c1317199279feb724045ae7913c72ff9
27560 F20101118_AABPCW bwenge_a_Page_004.pro
ed7102106f2fe84f95278c395dc6f466
a05ee646fb9ec51c120a9edd8c8b2689bf56098c
844770 F20101118_AABOXQ bwenge_a_Page_097.jp2
d89c46f956359c9352b40671e67873cf
95f6cb3c76334478a50805282afdf763853dc688
43531 F20101118_AABPEA bwenge_a_Page_035.pro
2ef65c9ea2eefc06ef9bccdef7006134
44283c5af026d251746b88eaf65b42132bfe850c
23036 F20101118_AABPDL bwenge_a_Page_020.pro
3b25e7b8721cf044c36f5f384481de6f
140d1753c803777e3fa83b5f24a618d68df54e7c
1051986 F20101118_AABOYF bwenge_a_Page_112.jp2
db779b6a1f5bce49a59339a0a1360b3a
72fc2e40d05ec3051ebde21e27ce0b19f857ac29
87126 F20101118_AABPCX bwenge_a_Page_005.pro
d0018ca05d3f3efe3d717c55e8d4733c
c9f5f550c19673bc39030e2af17e745bb3f7cd37
872405 F20101118_AABOXR bwenge_a_Page_098.jp2
6386e08f9dfcf2201ab72cde261f9c37
ee002d54289e157a56faaaac06950fe1364b1f2e
14066 F20101118_AABPDM bwenge_a_Page_021.pro
32ade20d6f91c2131cc1344f8d7c3619
82b76d64fa3db28a9a689dcae27aa7bf04395cf0
136633 F20101118_AABOYG bwenge_a_Page_113.jp2
ca35ec4c737c6d8ff9462d9ac4e54e3d
6534e7b87caf4dec607cca2a86fe1424c94020bd
69671 F20101118_AABPCY bwenge_a_Page_007.pro
87b53ef2df46a7f5bf560cb0d9532704
31dc5714676c46a2f9566cfb276b56c38200cdfc
584519 F20101118_AABOXS bwenge_a_Page_099.jp2
b3c3bf0f12645d88b0448c1906f0d739
1097cc3453722c67d078a43e00ddb75402b6d11e
21705 F20101118_AABPEB bwenge_a_Page_036.pro
ac18985205045e33a9a98f4f43378e63
9b8f038d69b04997789be0901fb72b09eeaf0be9
21786 F20101118_AABPDN bwenge_a_Page_022.pro
512bf45b274d34608be016851a8c69ba
809707d97c301486977eddd9b231ec5a996becf5
1051968 F20101118_AABOYH bwenge_a_Page_114.jp2
2f004370acd6073aa80418c86c5551b1
f409e4846f8bd7f78f53b7677bc82ff6a77f3cc0
7074 F20101118_AABPCZ bwenge_a_Page_008.pro
2add80d5b9ee1c9daf31b587420089e0
1c660ed7efae7299f89c2a864153bf097d555807
681698 F20101118_AABOXT bwenge_a_Page_100.jp2
84619d8f5bfd4e6031f7ddec3ac84261
33a05057cabc3250666ba88df04f3dfd892f7565
48959 F20101118_AABPEC bwenge_a_Page_037.pro
a715a4b5313ef5f223decb2b8437cf9c
7058a57f62487b534e4353b95e4736f5d8af9394
49978 F20101118_AABPDO bwenge_a_Page_023.pro
4081c9eb64e9d093b114fb556b4715db
bf04242486730906ffb751317f8ec18dd8647f97
F20101118_AABOYI bwenge_a_Page_115.jp2
6e0a123c6133520c2bf478474fe73941
cfc774b60c514d1d7763cd6850c051a371394cf4
626895 F20101118_AABOXU bwenge_a_Page_101.jp2
2823937d062b9c75209cb0e4c078312e
49dac8945b98856ff19fb3e5fd7763264d0e400b
50231 F20101118_AABPED bwenge_a_Page_038.pro
9f2ff4d1e1d293df6b91be10690bf655
f4aa69d5f23ca8701a520b701cb0d8ad218be2d4
53637 F20101118_AABPDP bwenge_a_Page_024.pro
22fecdb8965bc28c7fd2d47e9adf3973
e873076043b6d5b77c583f2da8c29e7f536a43b6
51559 F20101118_AABOYJ bwenge_a_Page_116.jp2
18b24710fd5a0c9b2b6e027a7cff941e
3639ed5d517efd1c083c51caaaa46aa21f649e10
56592 F20101118_AABPEE bwenge_a_Page_039.pro
2f11438d67ae26f089cb50b672cebbef
afbff3b5476f89c979a1c8ee4117894dd281a036
54179 F20101118_AABPDQ bwenge_a_Page_025.pro
9ef53cc7f431d4c745bce6cc57f671d8
3258f6ab17f332d8a4a19ef3766fc52c1a7d0e5f
F20101118_AABOYK bwenge_a_Page_001.tif
73e3b306b7b90ba4945644da4b8618a0
c2532452eb92cbea9b3d81f920c6b789b13ec994
208170 F20101118_AABOXV bwenge_a_Page_102.jp2
1291f2ab27916eb94056be3bf90c9790
d57837cdb2853d8ad1a1e97466137942454915c2
52956 F20101118_AABPEF bwenge_a_Page_040.pro
f542d1e2696f3ec0241dfd1a284bd9fd
85154a7bc5f7da3f1ffab736d4251eb28804deec
52996 F20101118_AABPDR bwenge_a_Page_026.pro
7408620786b9a73b582e3fdd0389e8ff
09028644d10a7edf9ba9b6895a6bc758ab4a496c
F20101118_AABOYL bwenge_a_Page_002.tif
e3c0390b6be28378fcf67627b6fbcbc2
ffa3a5c8fe7dfa6865ed18a92e567a99cc92bd3d
217440 F20101118_AABOXW bwenge_a_Page_103.jp2
4d8f36b44574eeca096e456fdae860d9
1a509595aaf6a402bd54793e7aec9d60ff15686b
55659 F20101118_AABPEG bwenge_a_Page_041.pro
33dd8abeab9f2a8ac5d7a318c984a720
119e95f53849d677d39be8b2986d3b1d49599945
F20101118_AABOZA bwenge_a_Page_017.tif
ef4cace3178c11c49d1627db05da57d7
18da2e7e1d095ecbf878d6fbc1e37e55c22d550a
53751 F20101118_AABPDS bwenge_a_Page_027.pro
4807adbb4fadccd4add2751d4bc43ec2
bedd964cdf57ba92c62532b8ffdca5007ec17c2a
F20101118_AABOYM bwenge_a_Page_003.tif
6821617898bc0ac4baa89c226ca470de
77f8aa711c112d5e57175599857eaab36d6ff332
483845 F20101118_AABOXX bwenge_a_Page_104.jp2
0771213f3ccf41e316b7ddaac46aeede
6ce9741a19acfb9fc35a73b1f4268e14f89a9b3c
54455 F20101118_AABPEH bwenge_a_Page_043.pro
5b0705f9e971f26937b03e4b9ac157ba
b11d8e1164a138dffe56e973790337f77991f536
F20101118_AABOZB bwenge_a_Page_018.tif
b7861d83ebe3391285e467ac0fbc3ed0
0944b2d670957084d7c6c83523dd4e4f5622376c
54459 F20101118_AABPDT bwenge_a_Page_028.pro
f675e511715850508b92e7bc936bbf9c
531bdf5616333f5aa1d30a512daec6c035f85018
F20101118_AABOYN bwenge_a_Page_004.tif
7d728d75443288265ed000ca7b74d1e8
d1ee9c5cb0cf96439d2df52a5fe44549fc0ee316
433211 F20101118_AABOXY bwenge_a_Page_105.jp2
f8f5757f86887afa54bb3804773b0278
25c8fd7172338c45f723c16dfec01f012cf39809
F20101118_AABOZC bwenge_a_Page_019.tif
383cf90d1b6a751b27c7e08a03585e9f
4a9ce590f10024f8b79370c160373c1835a5d74e
50605 F20101118_AABPDU bwenge_a_Page_029.pro
c8b9b3c36f24be85f02c0459350bc41d
f5c38dd9d6dc95acc28fdc8a289342cb5c681f95
F20101118_AABOYO bwenge_a_Page_005.tif
fe3506e8ddaff307de6613b50658ba99
1778655e372a965beb41cebcbc95201bcdc1884f
55264 F20101118_AABOXZ bwenge_a_Page_106.jp2
81273e1158ea6fd24f3f0ac89c2b5bc3
f25f177f07a771334afd990846c2ed4013d8cd45
48731 F20101118_AABPEI bwenge_a_Page_044.pro
17802eb4387f10cee8260ea8e7448fa5
4f9ef25aca4ad9407f38d5a79fe1054ed1890508
F20101118_AABOZD bwenge_a_Page_020.tif
53aa1ac96ff6bda13d059663020a8222
3df1f4514493364686142539e11cdd820b6d9ca7
50046 F20101118_AABPDV bwenge_a_Page_030.pro
87fd6820256d65a74b6b88785265813b
097e3b5b721897da7ead00460d33cb8ef4ad0065
F20101118_AABOYP bwenge_a_Page_006.tif
78f44b7b795585b5d6bd38b83cb15d2f
e61f8fb4a751d020a18d37f564c0d9feae4fdeb0
53842 F20101118_AABPEJ bwenge_a_Page_045.pro
6cf96a6c5014b459f0b82ffaec570715
7710fa39ad5171315ef06c7e3d48dec1e1fbcd7b
F20101118_AABOZE bwenge_a_Page_021.tif
875e2f3e6c0be335e3128e8453a14c11
d184d7c560ce8cccab5b515167cafe20d520ca29
53460 F20101118_AABPDW bwenge_a_Page_031.pro
3f8d7c7740b1da8e3d81792c85400442
ce394c2f054fb221abf666b3bf17b3b38cb4dbb2
F20101118_AABOYQ bwenge_a_Page_007.tif
51fcbe83c011785a9129dbb97b9b980f
ceac51759d41b13e95be713009fa4eccf1ad97e1
53561 F20101118_AABPEK bwenge_a_Page_046.pro
48aa480f7ce9cbff79b00b274fc5e695
ed80007b5655f6c3660d42cfb4f07d8c4e01dfb5
F20101118_AABOZF bwenge_a_Page_022.tif
ff0694f1c00614635bd83d082b66d3c6
d6c6ec3d1cc6c0d6fc767a11189cca5757a44c0e
31252 F20101118_AABPDX bwenge_a_Page_032.pro
12391acb8ff3e73c416ccb8834734ca1
40aac2d1aeeca8ea9a50ce2c14c21115f2c378c4
F20101118_AABOYR bwenge_a_Page_008.tif
094cced15104a672d13509167f24efb1
a1c721d5e1ec7534608abcf527e30be2e5389c1f
38188 F20101118_AABPFA bwenge_a_Page_062.pro
119c9c1b0ce330be2e1dd2cdbc7b42e2
fc32b2fdabcd17c8d7bf4ef5d5776d4d3697e758
12504 F20101118_AABPEL bwenge_a_Page_047.pro
1b9f8732f4b9b86ee661002b7cdde9fd
f630a9b5607bc0983b2c29807f0e7b2fc5f90d26
F20101118_AABOZG bwenge_a_Page_023.tif
16614d932107ca04949344eb3fb283e3
abf7ac2f2cbcececefa073c7fecbd67b35e2b4e4
49049 F20101118_AABPDY bwenge_a_Page_033.pro
dc2bbc586394f488071b2f946a5ca271
f50416dc52cb62613ba18f3866e6729a59ccc1f1
F20101118_AABOYS bwenge_a_Page_009.tif
1193973809aab5452d5d61c2fbf3ef64
e2de021ce43c171e3d620511b196c563ce46a730
51707 F20101118_AABPFB bwenge_a_Page_063.pro
ab6cfd26d19d512b0bd0a06d75716329
4b1fe6de8d059b1cae21d431497d0bac6f34b91e
26155 F20101118_AABPEM bwenge_a_Page_048.pro
91b9eea9de2ea300a2e12e8435906a86
6b329fab11136018a5863f131b776d8fca8a1ad4
F20101118_AABOZH bwenge_a_Page_024.tif
e23f07cda16793f56c00c8ca5f2f6478
f6d7669e87910e13c12115110831ad336f8d38fa
50901 F20101118_AABPDZ bwenge_a_Page_034.pro
d704ce17edaa7eaa3de22fe5008e6866
fff866c4b9f70bc3f3f416657f5f0230a0a81e57
F20101118_AABOYT bwenge_a_Page_010.tif
0a6cdc538939d25909de3a6219751fb3
4b2e472e4a8ede7ed8a245b74d45cc5c94164085
10339 F20101118_AABPEN bwenge_a_Page_049.pro
a4cd7499b72bbdbb17a1a1d409e20003
996e47a210c2beba552b45bd191a5dc420d32e75
F20101118_AABOZI bwenge_a_Page_025.tif
8e63dbee92bdea36dad0c36bfcf7fabc
013b64df0d0336f2023599c8408578b4e7eb86d9
F20101118_AABOYU bwenge_a_Page_011.tif
b9cae41100a2019f49856cf032c21603
d6952db1509ef72af7d4228e91644842810b69a9
F20101118_AABPFC bwenge_a_Page_064.pro
751a6fecdcedf5041dec25ec16f6d76a
84566ea29c0eac417f195f64c5eff6bf5c4bcaff
10757 F20101118_AABPEO bwenge_a_Page_050.pro
6b96c66ebd0b54e128fe93cfe4c2e2d5
c8e1b8404e87a97003d6d77580dc4a94d6a67b66
F20101118_AABOZJ bwenge_a_Page_026.tif
d68a33654a9bd71874bacb5673617e4f
b6a3ce989b9b7cc5035cbba8f51befe766cbed17
F20101118_AABOYV bwenge_a_Page_012.tif
d86266be02a5f6ec000ace95299b4928
05110accdfbf06e74449167b36bca7005f7b8eef
41664 F20101118_AABPFD bwenge_a_Page_065.pro
9615533af18d99198fe04f6286efb2f0
1a0af834ff3f0499ea265fa7e160e017083ea60c
47161 F20101118_AABPEP bwenge_a_Page_051.pro
283c42a8260a24be6bb605ba90ee00bc
cf212acd44d7192b67e90f51af0910efc5f1e02f
F20101118_AABOZK bwenge_a_Page_027.tif
21b193f2eae7924f96fdb5297bee8f92
c5f9c62803a6c18118d1c41b72f0b533026c7059
58015 F20101118_AABPFE bwenge_a_Page_066.pro
116b535920b73c1e582dcc42273b3036
68d9632336690cf1a34c52100b04d86841d2de90
53077 F20101118_AABPEQ bwenge_a_Page_052.pro
da661fa644dde1d839df477975cc1d9d
8031c8dfb47c093ef936b7282e039030a18052e5
F20101118_AABOZL bwenge_a_Page_028.tif
8ef9cd6ca083fef7dcc81d91be6d7ec6
fb31e3b39fa01a9b378f176c801e817f20368ec1
F20101118_AABOYW bwenge_a_Page_013.tif
91be575acffdaed935f5a4c09d11d906
fafd91fc512361fcd4d9b75a26b5643505b64617
56672 F20101118_AABPFF bwenge_a_Page_067.pro
ead39fe4ac5bb3afead521635dacddb3
e8c56cc91f2e82ccb16e19d8a366a6123a60f639
55574 F20101118_AABPER bwenge_a_Page_053.pro
d4889e415dd17767b0550b1c94d3ce13
248c723b1f04887855b34abebdd71a2bdaaea010
F20101118_AABOZM bwenge_a_Page_029.tif
ffe851589d4ec86a6274f6d7710c6e9e
2e2d918916db4e01be2767a61b98aed88c9847bd
F20101118_AABOYX bwenge_a_Page_014.tif
6615eb9bcba74a4090d74d65e730ae64
e3fd739dc7d1755b14a9c53756e9139ca91b3fe4
57085 F20101118_AABPFG bwenge_a_Page_068.pro
9081738b944ef752b033d01ab5cdad61
976ff7b6f8780c44fc4a65ab6f5f5b244c345c81
39151 F20101118_AABPES bwenge_a_Page_054.pro
cae61e17e39440ea782cc8e75047af47
08a9a9cfad621398d337ac74598a8c85f719afaf
F20101118_AABOZN bwenge_a_Page_030.tif
3fc77fcf9124c5e7da900f237df91cc8
271aa48f035b2c1ba221cf0c46e0994b7c31149a
F20101118_AABOYY bwenge_a_Page_015.tif
5528f4b1d905d09e7efc094b39fe7b99
2cef875bfb53a0a87f1fddfdf334ec32c20acb35
54419 F20101118_AABPFH bwenge_a_Page_069.pro
69f90fec491c3c4f696601ccbd47695a
5b5e1ace73c73908b2adcf8bdd0737b9eb7910c7
10675 F20101118_AABPET bwenge_a_Page_055.pro
fcd2b62d9552d8bdfdf3bb58e3b8f161
abbdc95c4ae4b3f562758f1b7dfa5b72e5f534ab
F20101118_AABOZO bwenge_a_Page_031.tif
d000d5d95c3c01643ae81ab0da10ae75
ff31185e3068e6f579fa4d3ccdedf32f85eec084
F20101118_AABOYZ bwenge_a_Page_016.tif
897b9cfb229f70b178af70eb4eb202d6
0cf910d76e577af8ba3659e484a1e2207e8f0e00
6531 F20101118_AABPFI bwenge_a_Page_070.pro
e72712880a89b8c0fb9a2bbbdda4cb68
fa62f0c0894446579b2f03ed960080c3d6dce269
12614 F20101118_AABPEU bwenge_a_Page_056.pro
de57ce686f0423ca0e804dd95e444512
e119af01b53838c57e3a6a49fbd87fb7c1d0e43f
F20101118_AABOZP bwenge_a_Page_032.tif
2a65b62bb04a78dec79c43164ab95e77
bea9ec5d9442d12564a03dcf9db87de29e14f2d9
47517 F20101118_AABPFJ bwenge_a_Page_071.pro
3ea618bd2f283e3dd143aa0227ee6d06
255c4d840664e5518d6d5665add2c49d44c1d17d
16483 F20101118_AABPEV bwenge_a_Page_057.pro
32513799a9e30322985376e250056834
b7d0b96bb26516b6caff9df6f44a706bc98919b4
F20101118_AABOZQ bwenge_a_Page_033.tif
9a2d193a121c26c8349ba99cb1744602
fa7a7c86ea5b13428590c0bde27a9874457ca84e
53549 F20101118_AABPFK bwenge_a_Page_072.pro
2a6f18d485244735da92dadb079f173d
ed69c0f5353c564dee8f45f19a80172eae73c405
13714 F20101118_AABPEW bwenge_a_Page_058.pro
c06eaf4932e0050e06602c9fb2567792
cbe5ee4f55b20810b2a23c5e7d01f4da6eb886df
F20101118_AABOZR bwenge_a_Page_034.tif
43de10c422cc962b1d24d98dbe2bf6d0
58262fc691c9c79d3335ace4ccf3a6643fa7644d
47956 F20101118_AABPGA bwenge_a_Page_088.pro
a19c71dc7c509f5d7a1a807616dc5abd
1641db6fd24e295259be5d5823db0333e91a8201
48704 F20101118_AABPFL bwenge_a_Page_073.pro
be6098b237c02f648a8c01e13181663d
0a3095b7b2125a0b1ba4950b06e3b818000ed675
52436 F20101118_AABPEX bwenge_a_Page_059.pro
a333f7c5c95b7797346f5f6b34d9586e
c368a38bb775e3f6db1013e2f59b46bfc26c9770
F20101118_AABOZS bwenge_a_Page_035.tif
182cfdc190b24b4c2581a34962b14592
d4fb7997943b111631da8cb36be9a7ec3fd9ca48
5007 F20101118_AABPGB bwenge_a_Page_089.pro
392c0df41db2c97714b9617cd01a2874
9a57def68d92777a07b827ccba87d6bd060e734c
46792 F20101118_AABPFM bwenge_a_Page_074.pro
5ff404f9a896bbff89897aff631fa67f
8bbf5fb7babe7bce345ec4df5690ca4aceaf4a35
52661 F20101118_AABPEY bwenge_a_Page_060.pro
bfb6202cf65a84adda044c0a3dda186c
e5e183129a50b13080075dabc8f700826acd8bec
F20101118_AABOZT bwenge_a_Page_036.tif
35e843713a3d6152a2ca8930b439a48b
6da6a8e752e8b7ea6d5edef14f2d185407868531
51517 F20101118_AABPGC bwenge_a_Page_090.pro
ece49866ce2bd8318c10a720801f143f
a40f681520fd6757c5077226bed46cffe1608488
38282 F20101118_AABPFN bwenge_a_Page_075.pro
9d7e076ef3dbaa6c400675384b078bf5
f9d6df30a641fa1ebc2935cc28208332873f2916
51175 F20101118_AABPEZ bwenge_a_Page_061.pro
c617c759468b055e117dca93e3ffdba8
1fbb38c8268df6ad8a54fb9f05c291f685daf398
F20101118_AABOZU bwenge_a_Page_037.tif
c80908dc2b1a949ce890f843ea23e592
66a207258199473a52073e119856fffa453e13ba
23941 F20101118_AABPFO bwenge_a_Page_076.pro
18d50a88d6249fee707e7185fecfbfcb
c677f3ae0978f5c557371f745e446911dcfb447b
F20101118_AABOZV bwenge_a_Page_038.tif
a81fe90f6091c7e15ca23cef2fa776e4
5d9210a25267cc59872a2563e3bbeaf3516a9cc5
53991 F20101118_AABPGD bwenge_a_Page_091.pro
f2efe2b18d6ca08a6b78d5741d1fb16a
b4462e4840ffba9e5ec7ee75048a83ea103d7d8e
12342 F20101118_AABPFP bwenge_a_Page_077.pro
990785ca764c8401f6deaab706775bd8
caf9671c93b1e1fdd57b28f156568025205d445f
F20101118_AABOZW bwenge_a_Page_039.tif
1f4957f9ebe7fd3cc706d7baa9769d61
50b62ec8b1a0bf584e29e7c72106f554d7eb2a4e
51849 F20101118_AABPGE bwenge_a_Page_092.pro
c880e170b0cfc67ccc856387583839f7
bf45d3445fef28a85ae2fb6334f5891a35d4ab95
8681 F20101118_AABPFQ bwenge_a_Page_078.pro
9108da88424d56d8f572d393b8f04094
cfb3160ddfa3f8a549d7de5329d94701519bb5f7
63579 F20101118_AABPGF bwenge_a_Page_093.pro
dba5dcb15745173cb5da9990b6312702
98e0803e2b4c71f8718bf11c92547dd9180fc8fb
36239 F20101118_AABPFR bwenge_a_Page_079.pro
5ee9f3e3d84f4d656efc4e191b06311c
24ab24073c51830abfcd507c2f6df7795de69391
F20101118_AABOZX bwenge_a_Page_040.tif
a687b2c8c0554d85814285cf0ef29064
c8e2ce10d06d95fbc3245e2ebd945f4fff1cd5a2
15224 F20101118_AABPGG bwenge_a_Page_094.pro
8785066155e8b4f71a45c239d398c838
a3264f4bc4802ab568dc256b2c72cff08afdfd6d
6772 F20101118_AABPFS bwenge_a_Page_080.pro
38dfa16fd07a8bef4edef095a6028763
64a90b2fe08b60d10c6f76b543594d1542e021d6
F20101118_AABOZY bwenge_a_Page_041.tif
a029deb1a0b70e863c4cdcedec1c1665
5991623b5c2af21d20062f8315d6306a33fbe9b3
27868 F20101118_AABPGH bwenge_a_Page_095.pro
e8f2df1daaaf06f9f1360c70d2517688
e08e01212bea4fae40465b8b002ea48f5eeaa226
53168 F20101118_AABPFT bwenge_a_Page_081.pro
cffe8ec016360ae0d0234b493133ac75
ea6cdfb26852f0b7e98bf87bacfc4a2b3a15fe96
F20101118_AABOZZ bwenge_a_Page_042.tif
2b9cf0879611eb754caf5275e6b64e4b
ab5e0f78f732fbd8c31c46f7cdc9b97ef1d5ec78
8615 F20101118_AABPGI bwenge_a_Page_096.pro
061a8d03d990df9582ad699015e6062d
5a7f8bc14dc1efd2f69905d7d6234f618d589979
49943 F20101118_AABPFU bwenge_a_Page_082.pro
19b4898a6eb7513358b6441a7cf4ecac
4ba87b45fab4185afee501be467dfdc4abb4e4fa
8252 F20101118_AABPGJ bwenge_a_Page_097.pro
5064d479bc61cd13a79e5c5690c0a8ac
6767081964929a9b71c6eeb589d591f38c0c942f
53608 F20101118_AABPFV bwenge_a_Page_083.pro
fa9bc3a45a2259e43713a9048c1093b0
c89c2c478c116cb00617d2b97d78d9a2d7ee9e39
9054 F20101118_AABPGK bwenge_a_Page_098.pro
842179d3422366920501a01ba7d771e6
28d6a8ce6a19495dbfb6f066043f35102221831e
54547 F20101118_AABPFW bwenge_a_Page_084.pro
115f0cb33af480a11f7a0a4833b99c40
5fe7197543450465d43c0f4ba78974d07d473912
10569 F20101118_AABPGL bwenge_a_Page_099.pro
73bc390a48f4faa540d450a1d31b0318
99f43e94f8d39c0f64347b3d9feb54bc6cda72ea
52341 F20101118_AABPFX bwenge_a_Page_085.pro
8aa43da0ed8d263cdbf44de69d9fb659
160707f838f5692f40b9d35bc8d35d919c6eb976
66875 F20101118_AABPHA bwenge_a_Page_114.pro
2a42d55d2427c9506f6d21ca90de97bd
9b5b6251b3ab761b90caab6d5a56fa121064c03d
23305 F20101118_AABPGM bwenge_a_Page_100.pro
20486147f2d7ca26426c7917784b33ee
41b8cf5c27af822b10dd35566b8d16ff9de5a71e
56467 F20101118_AABPFY bwenge_a_Page_086.pro
ff72fcd18644789336887997bdb8741f
b4453aa661b50f06d20475887346cdace83fd90a
44989 F20101118_AABPHB bwenge_a_Page_115.pro
b759a76315246e240139a68d0bc24e88
82a3635ec5248ead6617a2bf5881ed8cacc5f089
21374 F20101118_AABPGN bwenge_a_Page_101.pro
cbd5fdb37ba92c0957457ed2e39ccc11
ac7e8c00599583519cad391ebd66882e2b0d7d6b
10591 F20101118_AABPFZ bwenge_a_Page_087.pro
00fd62f1d6d5b277459a2906f1d79157
d50a861fee5e9dfd5098013f697caeb1a58016a7
22044 F20101118_AABPHC bwenge_a_Page_116.pro
449c1b903c2f59bff03ab87e1f7256c7
0985508e3d6eed5b1205df832e70e90a88e5f720
9659 F20101118_AABPGO bwenge_a_Page_102.pro
6d3ee30cb58400a3a997b81f1e61dc00
09c2df7868e9e997a9251138d46293d90e42fa55
92 F20101118_AABPHD bwenge_a_Page_002.txt
f6b72dd685f608b9fc3adf0a41ae9316
b3a31c0f1f75a23022c891caab000776b702b23d
8143 F20101118_AABPGP bwenge_a_Page_103.pro
eaf2fcc8cedab54df6911d8181fcd71c
220a059a8dd01e839837375ab82cfe6c5dbe835f
25002 F20101118_AABPGQ bwenge_a_Page_104.pro
0a7a9fbf10b360f59097a1a0fcea4683
d74703f31b8d4fd4cc28dddf860bd0db70d9af18
127 F20101118_AABPHE bwenge_a_Page_003.txt
4ac4b51b7e7d2a6c2aba404fee7deb0b
1d9da6d66b64f3987dca6e442416097ea7bb5bb3
24135 F20101118_AABPGR bwenge_a_Page_105.pro
37805ed8703a59ea48515087ff3a1611
a0206825267dbc49c772fae0b3cc4e8817fec2fd
1137 F20101118_AABPHF bwenge_a_Page_004.txt
26c1a17083bca56748649913ea0eeb0c
9e43a9d4be56e70b2790e6e6aa418f57fbe0c899
35254 F20101118_AABPGS bwenge_a_Page_106.pro
356096025870bdd9c4440287cf44d4e8
541ee71a179c3dfc21a583c2d00327bfe28d3966
3565 F20101118_AABPHG bwenge_a_Page_005.txt
473117c7c732caac8d57dce4edbe4d3a
ed190fe86e2951b5e8c7a1de006ec433101cb371
43584 F20101118_AABPGT bwenge_a_Page_107.pro
533a67888112a9b85bbc9658cf30ab94
5686dfc24a4d5db69f411128281e0915c841d9f0
3446 F20101118_AABPHH bwenge_a_Page_006.txt
ef8ace7bd53e8e3f0d5653bc0870ca61
e59b63f0762f572a6dd8f36c86782e44d385b054
35260 F20101118_AABPGU bwenge_a_Page_108.pro
f30263a8594648a0107b67aa313a8110
ef11372989af56b7aa071b8da72058cb8b374a71
2762 F20101118_AABPHI bwenge_a_Page_007.txt
179f16c8d29be8dff43e26cb2c18b000
8ea1bc7e3d773b4e976570785e583c4cf35ae1c7
18341 F20101118_AABPGV bwenge_a_Page_109.pro
0d05924b73df1a5ab815e9d0ad96352f
d804cc8c7a3a107c772cd43f368cc522ddfde84a
430 F20101118_AABPHJ bwenge_a_Page_008.txt
cc6f42bfcf806f772e36655f9f367d06
28aec96a1ccd8edc8e3d408680e9ba4ce100e799
62966 F20101118_AABPGW bwenge_a_Page_110.pro
25c6e1dea94f2eb4a18f3ba332fb45a9
9e598bcf3bb2a4f78b4bb6cd42f508cebe5ff187
2664 F20101118_AABPHK bwenge_a_Page_009.txt
6b78d0430ba367a02e035ad7f59f9453
a7df35d72360e8bba1885a7b2ad7eb297747a74c
62438 F20101118_AABPGX bwenge_a_Page_111.pro
f26c1e01fc2ad3cb954bb9b26398ac09
68da6bcfb89941a15e801f4141c2a46d79df161a
2162 F20101118_AABPIA bwenge_a_Page_025.txt
a23b2420864f35e1dd68760f365bc452
c8845a4970b1c10ff637c1456b653faf46807a0b
928 F20101118_AABPHL bwenge_a_Page_010.txt
32b94f6288470a8650aea24aa0c57610
78d2f5d66cb2311affefc5d70f3b7e25731852a3
62230 F20101118_AABPGY bwenge_a_Page_112.pro
a5d7a0c988791e4f5ddd2c60efe63645
8453bede3b75cda937182d84568a16579970bb8b
2088 F20101118_AABPIB bwenge_a_Page_026.txt
2f947a3486cdecf8b5b7a057e66aaf7f
081ad553d9ec79b2d1bfbf5f0f206038782c5def
2034 F20101118_AABPHM bwenge_a_Page_011.txt
8f8eb3ba18f5ee548b6b4dadca35c195
80cba906557d2dbbb231711d98d9c129d9cddc36
62533 F20101118_AABPGZ bwenge_a_Page_113.pro
daa7569e24de10e296e46560812e6c29
e2a05e5800229085da2afcc7e37524d7458fd34b
2118 F20101118_AABPIC bwenge_a_Page_027.txt
5f0b7a92d838926f5289e7dd6c7312f8
096be04493d288609b2451b464be668594e25d21
2190 F20101118_AABPHN bwenge_a_Page_012.txt
6ade94bee4df7e213fd08f5b4ce0bc78
95c212b25d20c5cdeb3c15f18e6035da2e41b903
2177 F20101118_AABPID bwenge_a_Page_028.txt
fa1926a2614a92718a577227dbd06c54
d7840d9f53d8d2ba8032651aaeaae043e7186070
2267 F20101118_AABPHO bwenge_a_Page_013.txt
c4ba9cd5ff8fd1847bf6a2c8f6e9bce4
96ca87c5ea627509cf5354f24eb8c748fd9b9929
2002 F20101118_AABPIE bwenge_a_Page_029.txt
31f4ca1cd9241e82cbb1b2366a569140
66a6e94b9f0db107241a315bdb3ec788bcc8934d
2039 F20101118_AABPHP bwenge_a_Page_014.txt
ada0f5efa9f5ae25490dbf2078c47796
d7d6c69e673a982514deedf500cbc99a7acbf673
2212 F20101118_AABPHQ bwenge_a_Page_015.txt
1fa8d518c1a7a7a7f6eddc6a1036ee0f
6e651b91f6f194269ba2ad4b115998d3391b726f
1983 F20101118_AABPIF bwenge_a_Page_030.txt
33bb15caddb290bfaeb6eb0e07b3543e
42f7ade71b800fa228576e4fc8cd01103270d3a5
2122 F20101118_AABPHR bwenge_a_Page_016.txt
4a024545dbfce61fff103deddbf85de7
45467c6df8f0d35b3f13a6e28f8362275a4247b0
2115 F20101118_AABPIG bwenge_a_Page_031.txt
168fc6ae44c44ef0b987951ac9985b6f
49326d976a7ec90be648b6bb546a770e4795f327
1991 F20101118_AABPHS bwenge_a_Page_017.txt
881675cafc8bfc750dbd094efe8ceaf7
4d99e9df6508b123221dd24a218c04054e276c47
1243 F20101118_AABPIH bwenge_a_Page_032.txt
823c8b4c176f6204cadf5b7812482708
3a4cf1c0f1ba2a70688f199b461031f44c64187f
2080 F20101118_AABPHT bwenge_a_Page_018.txt
e85b15d69edbef3cf25745185a4d3816
9373a4b543d35bd5bdb8442bfbf51334da86032f
2125 F20101118_AABPII bwenge_a_Page_033.txt
9c7f3ee2f9ed08439d3941f7a661db19
c5d6734a9a79e844437c360790cea684fc2faf71
2163 F20101118_AABPHU bwenge_a_Page_019.txt
bf2c5d638f8e39dec03904193956d1a0
0c4b67d034e076d19c25a6c6e239ddddede147ba
2045 F20101118_AABPIJ bwenge_a_Page_034.txt
1cced853d6318c9c8c001ad37921c180
928c42010b1091ee2e380d08520e35a5ab858cf1
967 F20101118_AABPHV bwenge_a_Page_020.txt
6462b635795f63170bc3805210e865e9
7c16c063db64da4f3e9231f72b8606e3c5c36b97
1875 F20101118_AABPIK bwenge_a_Page_035.txt
8a3e674b718260797b022b4a7f36ebba
0d21826b0c4392bc22a6c9e134c73a738c6b2210
751 F20101118_AABPHW bwenge_a_Page_021.txt
70274a9dee16a66ac717a80ea363aabd
16a66f7c90403323940f90d38028b59beb0f53be
2011 F20101118_AABPJA bwenge_a_Page_051.txt
d0041004291f40f182f1bccd042c0195
7ec59f3c8529fe1e6ea3b7c48e0be0bf64dc3ca8
1056 F20101118_AABPIL bwenge_a_Page_036.txt
4593894fd8350c2169254d8d457f408f
bd2d04f8bf73472bda4bcea98241e952d37fe1b7
1100 F20101118_AABPHX bwenge_a_Page_022.txt
56d00cfe7e4df05c983dcb86214775e2
748baf9d93b7f1abbd4b0d5e0080480c1a7b29a6
2129 F20101118_AABPJB bwenge_a_Page_052.txt
aec06c0b29513b1d6f6dd82358b04e33
119f521fab8b33f1b3dbc60ac085f4704cec3f92
2035 F20101118_AABPIM bwenge_a_Page_037.txt
afda6c664907369b1e048a813aa12bcc
d00ce580e3c91daa9e7b3afd1c1326b31f6ef85a
2070 F20101118_AABPHY bwenge_a_Page_023.txt
8b793eba2d8d1d304c4977c92115197c
49a56edd6edd9d8421a49396e1a600f835994682
2188 F20101118_AABPJC bwenge_a_Page_053.txt
430434a984acc5468b6d05352059ea3f
cfa94e2bbee1458b598655cc7a9d02436626251a
2006 F20101118_AABPIN bwenge_a_Page_038.txt
aa721d0a2671923293f6c16aabbe2e15
53660109eb74b628a10b646695f89d5d860e66cc
2114 F20101118_AABPHZ bwenge_a_Page_024.txt
2895e4833c949edcb75910fd5299672d
952aff57b99f2081a5329a4cef57d05815ede3ee
1996 F20101118_AABPJD bwenge_a_Page_054.txt
0d0d3cc3e13a23ec32fb7811e5eaf046
cafb08bb88e2e329b34a8f4b03bce4a029b7ad48
2229 F20101118_AABPIO bwenge_a_Page_039.txt
90398e12fd96eb07ad0ad07d4f24e7a4
a784d591cb4c2155e33ed51fe2b88897f7299fed
504 F20101118_AABPJE bwenge_a_Page_055.txt
556cebc4ae2b0a383fbb2c5bacc8112c
0299fc0fce821a06dbf685e254f737eb7eecb639
2086 F20101118_AABPIP bwenge_a_Page_040.txt
414285a4e4ce2164aed0794cf254f2c5
bcf784b987ec6369786d7d5660dfa72b097448ca
672 F20101118_AABPJF bwenge_a_Page_056.txt
4e89abb2430b912257dbe14305cb5882
08c9519d8a4845cd74890111c6bedae8b4a55a2c
2192 F20101118_AABPIQ bwenge_a_Page_041.txt
4406a5f3937a3a7deaa359512861e051
9d65b5eff3088d8dd71ee08b4acf73e3d2ca228c
2191 F20101118_AABPIR bwenge_a_Page_042.txt
16393d5f70c3d295ee840dd66043ddc6
8a465561a49424f21e565ed2a92df74fa9861fa4
2137 F20101118_AABPIS bwenge_a_Page_043.txt
b4b48f2db4a7cd398bf413f53d0fa942
ebc46109390a043a59808d6f7f90b86cb084ae9d
860 F20101118_AABPJG bwenge_a_Page_057.txt
61f410b9c8e9e183df314349f69427ff
63c69ec0d2316690058277461f852e96002bc9d4
1925 F20101118_AABPIT bwenge_a_Page_044.txt
1bb9dfcde0254886ff8e1c8578c28038
4b0ccb53bec8eb5a7664590179c809b06c26ad07
949 F20101118_AABPJH bwenge_a_Page_058.txt
9ad8b2022774877a90ac8ece90b7cb47
7ad841789fcb25164f41950e2fdffd458bde5dad
2136 F20101118_AABPIU bwenge_a_Page_045.txt
39e559e473de5ee5d7508b0029f5503b
e03d342fb6750bbb5cfad5a78f0f81e6d52e34cb
2185 F20101118_AABPJI bwenge_a_Page_059.txt
a0f2d40389b7a9e344be11c35a99815b
019f2f81a2b3ce5412bf02ddde85bfb4f31069aa
2110 F20101118_AABPIV bwenge_a_Page_046.txt
422254e38275f0e9bf9594056e9083b1
a604e511206e7c358582e98c51f0aa1668cb0efb
2146 F20101118_AABPJJ bwenge_a_Page_060.txt
778517603885f04b993a4dfbb599f158
98b367013fa52bdd4b694b8f06409770ab502c89
601 F20101118_AABPIW bwenge_a_Page_047.txt
440134e7bf3165a234feda08833beec9
f70bcec7d181936a3f0c0a9f5d7ad92829350f28
2040 F20101118_AABPJK bwenge_a_Page_061.txt
90ca05fec10e6ded3d5ebf8e9e2fdf7a
e1ac298e3f23dfe22978a36b5f056b3eb34e1d75
1581 F20101118_AABPIX bwenge_a_Page_048.txt
ab1e61e19f165dbf4107b41094f7f797
18f1ad68c9eaa7b771112775b11f3f180a4533d4
463 F20101118_AABPKA bwenge_a_Page_078.txt
aa1c58999c42bb23c8ed4b914c14d36e
82e99f094739d8dd2f9a27fbebc9033e654b28f2
1681 F20101118_AABPJL bwenge_a_Page_062.txt
7a1408544cba748cb7f6b935a35e1304
b3777f03578912f81db34919367808deac05e054
503 F20101118_AABPIY bwenge_a_Page_049.txt
f07ba56f26e8187203825f7890f69722
79dbdc215d03fc0f6fe2bb123354d903cda6205f
1741 F20101118_AABPKB bwenge_a_Page_079.txt
03504e73f65ad2ee35f0df2e16575217
1079ff2411798e595a9d1f1792b243544aeb901e
2044 F20101118_AABPJM bwenge_a_Page_063.txt
c54c172428d48419e60e5c37c8afe6a6
cd5eb33d7d099280f90aa1d2b8448b014aee1f07
853 F20101118_AABPIZ bwenge_a_Page_050.txt
080d3c5a8d5907f8b9483123765eb2ff
23b8a4cfc537421412929290c619fbc675cdd848
363 F20101118_AABPKC bwenge_a_Page_080.txt
a14cbf48109685ae49385fd421b73ddf
8bebc1a469b6bf26d5c5f10afde15eb6a70b66c3
F20101118_AABPJN bwenge_a_Page_064.txt
603343aea795213b4f52ec86e3c5d74c
a03c71920fa4dbf7b2e273755bda9a20cd2c7c63
2174 F20101118_AABPKD bwenge_a_Page_081.txt
ed0b6531697374b6be0b18dc5df538cd
dc06c591bd0db719ca3a4b7eeed809ecf79eeb72
2277 F20101118_AABPJO bwenge_a_Page_066.txt
bdb5e3d4e70b7bce2831bb04b44e6687
a97175777b85c6a7278d44af46532c082c73c655
1969 F20101118_AABPKE bwenge_a_Page_082.txt
92cda5a2f6438eb6517f904b3aa85486
4b9fb5a61117bee7f72ea62bbc762b6ac8025db7
2220 F20101118_AABPJP bwenge_a_Page_067.txt
86942b006f46975fdfcf8378110de14e
b71278c02654e360313bf1e4a6cf11e5d1c6e43a
2116 F20101118_AABPKF bwenge_a_Page_083.txt
03905da019e976d4226fcdbc16f915dd
2b94b9f99cf20294a59d5d04f9f61f19ddc70113
2230 F20101118_AABPJQ bwenge_a_Page_068.txt
5cfcfa9274ca0b10dc70be89a704c459
cf3df0f1a5f14077c496c975cfe3132f9dc0aed2
2152 F20101118_AABPKG bwenge_a_Page_084.txt
1697c633a3170e98cc08944eef86dfbc
388ce3c4f553642265b7b9074e0db11c844809e5
2181 F20101118_AABPJR bwenge_a_Page_069.txt
dbf6404f104038c7c6248e569032f274
01992d7e0700f7a116ebbeaf03ec78b45fa7fb0c
262 F20101118_AABPJS bwenge_a_Page_070.txt
191d107ad7ad24ad7331cc0dcad41c86
2a0f1c48273adae5d295a73cabfad7ea9f15587e
2062 F20101118_AABPKH bwenge_a_Page_085.txt
77524758ac06030313a806dc9bd27e3f
a492ed9963efcb00403861c89bd3b575a61752f3
2049 F20101118_AABPJT bwenge_a_Page_071.txt
03d6e3f75dddf5b9b42b8d0f2c2b989f
7af8afb446567ca2a90e31ed922a87a49c5360d2
2238 F20101118_AABPKI bwenge_a_Page_086.txt
354c784cc17e613682a6c0cc1ae41fb1
f55c39135b25f2dbc97ba8ce6e52a0c2b112700f
2851 F20101118_AABPJU bwenge_a_Page_072.txt
8b1c50860909bee9bbf845262ecb1445
9b061ff2b855b33adc80ebc14d4ec3fca6829933
2408 F20101118_AABPKJ bwenge_a_Page_088.txt
84c2764691b096d68006ab7c090943d5
ce66114e30a0e2880b910816af74fb402be75ff1
2286 F20101118_AABPJV bwenge_a_Page_073.txt
348dbd6ec8095e09604c8496733cca2d
cc1bedaba18001bc4aa286b4a82a38099047242d
324 F20101118_AABPKK bwenge_a_Page_089.txt
9608004048e7dc419de5728520a9d691
97fe7198487c74455d73fcd9c0e217f0ec55ee65
2311 F20101118_AABPJW bwenge_a_Page_074.txt
78213e111e4ebb6544651dcc74c4976d
9147257762304853eb560a5a15931af36ad9a37a
953 F20101118_AABPLA bwenge_a_Page_105.txt
28add946662fdb8eed3f8e7a0054ef54
7bed682d82733ab6e5515956ddaa226f206a732a
2100 F20101118_AABPKL bwenge_a_Page_090.txt
1c68e58402a0644371519b0f50c2923f
f6de1692003f9ca1ec6d4efb9f38ac618671be6a
1827 F20101118_AABPJX bwenge_a_Page_075.txt
072fb8425d3a80fa32b828f7f3871cc2
c917476d1823ef2dc423846666706aebe8efe5f7
1829 F20101118_AABPLB bwenge_a_Page_106.txt
cab1a0bd18dfcfca9f1660f0b183cf53
4b3a49a4849def5b2dd30e34200d090f195f8d56
2135 F20101118_AABPKM bwenge_a_Page_091.txt
c95ccea373d6cf8fc8a82456a64099d2
4da1ba60efcc7ef131e2d84f132993274edf889b
1169 F20101118_AABPJY bwenge_a_Page_076.txt
03220e8aedebaad0a022c2c950175053
4da6ca7e4916740c61e8c70ed34b9e40551c5cc3
2205 F20101118_AABPLC bwenge_a_Page_107.txt
90b90d76e530dcc288eadef4917a629a
db00011ab56bb52576db0f37b70134d4def34daa
2058 F20101118_AABPKN bwenge_a_Page_092.txt
deac63bc4ae610a9255c718c29978386
f8362e39e9b33400ce7629cad7c0b8df7c673993
667 F20101118_AABPJZ bwenge_a_Page_077.txt
9c93dd0c35a467adf5aae3b65a8cda47
3ccc69b5449bd6764a1e092f9c884858a6a37400
1452 F20101118_AABPLD bwenge_a_Page_108.txt
7f2b67e70c7578f4fbc98aa9bb47160e
09a5069549ef1dffd54356392dda7d5f4351448d
2532 F20101118_AABPKO bwenge_a_Page_093.txt
0811d383df6209aa420c1707fe35ec53
7fef2ef5fe92316ba554c778014367193ff10a67
741 F20101118_AABPLE bwenge_a_Page_109.txt
be473beffbe024861209b00e74b92829
ee85eb3c9576bad8fb86472f860d99c08c6af6a1
811 F20101118_AABPKP bwenge_a_Page_094.txt
bae6f1dffc3e8c75d187188ee752f20b
511d93d9aefab8372cd69c0b2fa5fddd857de619
2542 F20101118_AABPLF bwenge_a_Page_110.txt
6b11c774a6473f7c3885c9204d604055
63324bc57c3fff2419506daa970a6f78b695eb38
1063 F20101118_AABPKQ bwenge_a_Page_095.txt
08caf768e4ef87ff291487c415631d5d
5c5e66f9a81893f156fb635bd48d65ae39533178
2524 F20101118_AABPLG bwenge_a_Page_111.txt
8a599eee9497dfab58876f6a8f50ecec
a5e1011ed70c20b4c9d9ee13f0aaf025166a5a63
485 F20101118_AABPKR bwenge_a_Page_096.txt
04603afe31f7177423939b0cded4d4c4
8f5d2566f1f77e91c84fb04bc83fc3d961995161
2505 F20101118_AABPLH bwenge_a_Page_112.txt
9332333438f7248470be2602e0776a6a
d09234b70593d2d42e888e986540ac7573311f54
496 F20101118_AABPKS bwenge_a_Page_097.txt
241f0d565c5872d9e2d82a36bdde399f
78c651d6198263fd91e6f51c68256a46db000301
537 F20101118_AABPKT bwenge_a_Page_098.txt
b99c97bce2e31e516e01ad58b639aac2
8ab1ecffd525a81039a60e10064e81ead0ac5f2f
2522 F20101118_AABPLI bwenge_a_Page_113.txt
7dbd97755c856ca3877cade202fc72b7
2d046c80904bad0dac4df09bcce2eea8f47216b6
F20101118_AABPKU bwenge_a_Page_099.txt
f1b3cee91875404617ab5255a2e6dd72
66709007974ea61a44f5b80e047ba8ac5d516ed3
2687 F20101118_AABPLJ bwenge_a_Page_114.txt
555721ef36594dcb3f1b7294b025264c
b3bf7ebc6e1552123e7f20b1b050e7385728ab42
1332 F20101118_AABPKV bwenge_a_Page_100.txt
9746266b855629bea5e1435b9125d9c2
2d0e6e3b86d50b98f2466ddfba77c261c1568d3b
1837 F20101118_AABPLK bwenge_a_Page_115.txt
07249ddfa55d9590da96b0b41dec0cf9
af3f30d1f0d458598913bc26336fe021fac1875d
814 F20101118_AABPKW bwenge_a_Page_101.txt
226805722f4130d5fd7a1d2fceae9c38
4adbf2e417b3383f7d2a893ae7207e39c8c826a7
921 F20101118_AABPLL bwenge_a_Page_116.txt
7f69d77ec8bc24c6db5314bee951056a
52ae10e77c65bc22f3ce452de0632972cd2d30b6
387 F20101118_AABPKX bwenge_a_Page_102.txt
09b1773fda6417c42bf142ea16e7ae8e
a68d3a7a7bbce8efdfec40d0b73471baf48ad4c4
6064 F20101118_AABPMA bwenge_a_Page_007thm.jpg
bbcd19fdbce426d826e7fa7f35b359aa
44a10958b07ece2aac23c8edd9fcf847bf51d12e
F20101118_AABPLM bwenge_a_Page_001thm.jpg
967c29711eb4b9a9c8c103a6bc379c7f
508d592d015f6de84debf06825f216c2e1e8bde5
323 F20101118_AABPKY bwenge_a_Page_103.txt
c5b56907b97e1da91a6fcaa3b79d6d65
f6cdda60b21470440e69464cce18d5db8f006bae
4091 F20101118_AABPMB bwenge_a_Page_008.QC.jpg
3654f89edaf25e90d12b8f8f7b150d2e
48966e9b72713d74717b867c37ca4fdecac296b9
1534841 F20101118_AABPLN bwenge_a.pdf
77824b5fbb3df1e18ef54294387bd6a0
a0642cdaf886c31b8bc0fbdbbee60dab93f4eda8
964 F20101118_AABPKZ bwenge_a_Page_104.txt
d750a7579f4c1ac01798c5e5d8f53a74
ca8f91325d935eaac224c53ff9419401b8f69bbd
1642 F20101118_AABPMC bwenge_a_Page_008thm.jpg
93af254650ea10efcc3b01c87b15aa33
e0239a80a9a96648f6a986e601c09f128692a8df
7635 F20101118_AABPLO bwenge_a_Page_001.QC.jpg
80ce8dfd7639ecea711449ad12f3860d
1f88cbc97a6d04c4a15f7be02365112e320cef96
22735 F20101118_AABPMD bwenge_a_Page_009.QC.jpg
441837d555a6c9c823dda0504cbad2d5
c457c5b162b6c69adbbf7154a3b43713b4a2cb62
3105 F20101118_AABPLP bwenge_a_Page_002.QC.jpg
23a7cd4ab68b7b62b635b0399224dc0f
3f947f52781dd1d1f3e557b047b8b0ff61122ecf
5827 F20101118_AABPME bwenge_a_Page_009thm.jpg
95f11c21fc09859070f81e1d49ec8869
6935da0cf9e65035eb94598d550071c62878d003
1346 F20101118_AABPLQ bwenge_a_Page_002thm.jpg
ef841ffe9fdb9d4e352f54ab7d2ba027
bfc34c7043da6c0f668f92b95d66f0a9be9b2692
13128 F20101118_AABPMF bwenge_a_Page_010.QC.jpg
eb7074662d14d2704a448017fd9f2966
654fc5d994afc7e86e51ee90ca3e2772216bacee
3435 F20101118_AABPLR bwenge_a_Page_003.QC.jpg
1d9bc2a905c60e8f4bd154a46f96c860
612c3f79a042d63a9605117d5a357d9e56a70dcc
4071 F20101118_AABPMG bwenge_a_Page_010thm.jpg
3b7c1fb859798ba5c5bc568ac514161c
73c2831f061ac2b368fa31a8087086dca869061e
1607 F20101118_AABPLS bwenge_a_Page_003thm.jpg
927687f6fe51200e1dacfdd5c0546b47
3e1e4ec31d574e6e5ddb3d75e6bf211af218d158
21993 F20101118_AABPMH bwenge_a_Page_011.QC.jpg
4d3d4c75d9a74bb15ffbf843dbf81981
97b9d7d294746740bdf61512be44766c3059bb73
14774 F20101118_AABPLT bwenge_a_Page_004.QC.jpg
295b227e9b4d3a5444275e50561d0068
13ff54c03b3b157c058daf2a6f071803a36c2ed4
6151 F20101118_AABPMI bwenge_a_Page_011thm.jpg
a344919e7edd15fb0088f7e9cf1f7d3a
a539b4e794568271d301534fa38ad827819922d5
4164 F20101118_AABPLU bwenge_a_Page_004thm.jpg
52be1cfe58b781d1e69d7fc76a2c05bd
c19b2c672de9edbb041c1eae483e57dd2146713e
19470 F20101118_AABPLV bwenge_a_Page_005.QC.jpg
b074e99de5208dbffe0c2c945f1212a1
4ab771dea40a4f41f06d8e1b23b8f80f090514d1
24680 F20101118_AABPMJ bwenge_a_Page_012.QC.jpg
91fe033944ed47399974b72bcb101bb2
e35d781868b924e2d48cb9cdfecd8cec28d3032a
5193 F20101118_AABPLW bwenge_a_Page_005thm.jpg
f1d6694628f03f48427fe58630616dfa
59bfa903f49d3c4bc9ffa41939bcbcecb0b229bd
6734 F20101118_AABPMK bwenge_a_Page_012thm.jpg
c4bebabc33fb055b4608068ec917c598
58fa67fbdd26d47b30fd7c7cf634b276e69115a3
20862 F20101118_AABPLX bwenge_a_Page_006.QC.jpg
bdd97e6449cfc25c354168e3061dd32e
48050a028695c187699cf481fff589e4e8d267cc
3466 F20101118_AABPNA bwenge_a_Page_020thm.jpg
304780d7b10f95b299f2c30aecaa3c9f
c33a6572eaee70d620b893f1c0eba10ebf8e1b59
26537 F20101118_AABPML bwenge_a_Page_013.QC.jpg
1a4373c3d452cb281139f79673d0e3f9
758460702b910e3da52337460cf57ba7af49b101
5495 F20101118_AABPLY bwenge_a_Page_006thm.jpg
8527c6a522dd80b41a071e63b0fca3cd
43cf7be358985b66f377f1037448017276267e65
10704 F20101118_AABPNB bwenge_a_Page_021.QC.jpg
8013636f5b4e6e779296f65e72ff31c7
b6f9bb383269b20ae2514862bb825ecbb12bba7c
7222 F20101118_AABPMM bwenge_a_Page_013thm.jpg
48a02797ec0098396d6fab0843dc4c8e
39c1907fcf2e2038e21ed85106e3511886204705
24234 F20101118_AABPLZ bwenge_a_Page_007.QC.jpg
5ba2d4ccbfc5a06196ee062b852f99e7
134cde141999e495f1d506ae7f4568ca7d18016f
3449 F20101118_AABPNC bwenge_a_Page_021thm.jpg
ef9ca84baf6014f42770f8f481cfcb23
80bab7d89b0fb4a3cabeab488034310343dffcbf
23760 F20101118_AABPMN bwenge_a_Page_014.QC.jpg
d097066213ee5f499ca678c96bd8fd34
0dd282a20760f505747edbb7c68d64279cc72bb2
8847 F20101118_AABPND bwenge_a_Page_022.QC.jpg
07ea8c898e126d32db2aabf94914ba85
b0e31ac90572b46f7700b5200ba73aa903b58c55
6508 F20101118_AABPMO bwenge_a_Page_014thm.jpg
6118110985a30c524fe604657f253c48
63149da7c19a8084264f9f5e7e20579131b40266
2853 F20101118_AABPNE bwenge_a_Page_022thm.jpg
c4cbefaf3b7303be6a92c17625046fed
0c01d81ac89807279f59e30ceec337cf6b8e64f7
27358 F20101118_AABPMP bwenge_a_Page_015.QC.jpg
1e64722acf180472d2f46d88e1b3ea02
baa34fdeb8eb3b86c5f452116d354d0b9380f684
23407 F20101118_AABPNF bwenge_a_Page_023.QC.jpg
b59d923af443f833f26ecc777da81613
28db9c78b8afbf79357340a86429eb0aa209579a
7300 F20101118_AABPMQ bwenge_a_Page_015thm.jpg
c43c2b21a4043b4d7600900d09ebe1fd
a5189fc1078253513748bdd86f962571ae12668a
6378 F20101118_AABPNG bwenge_a_Page_023thm.jpg
00d7614c454b1130f472f9d07b96b142
bf6d72fb7e2b09a7948c2252f7bbbeedff4306bd
25346 F20101118_AABPMR bwenge_a_Page_016.QC.jpg
9fc17a546e72ac0ea4dd04f4853f9606
c2dffaa21087f18faecec2183fd7ad35cd86a85f
25230 F20101118_AABPNH bwenge_a_Page_024.QC.jpg
4f2bba901826a87a4874575503bd5167
c179895693d37891343879d86cea3f8a7eddb4e7
6835 F20101118_AABPMS bwenge_a_Page_016thm.jpg
4edcec43700690a9a77b289b4368413f
5d467fe5f178fab7c0c6c04a78cfebc5d5413390
6805 F20101118_AABPNI bwenge_a_Page_024thm.jpg
82343bb289589639e945bf856bac7965
4390e66f5217a8d10477776160c7ab4660b127c8
25094 F20101118_AABPMT bwenge_a_Page_017.QC.jpg
4ed04729e4f0964f0b542ca96e91a710
4c556a00580e76067e12aba441e3ba35052b11dd
25420 F20101118_AABPNJ bwenge_a_Page_025.QC.jpg
2977071ddd46710b7fff3ab5ffd08f44
e4be670eab5703f41328f0384673c9fcc584dd78
6868 F20101118_AABPMU bwenge_a_Page_017thm.jpg
7051537108175c9091eeb325d824ec53
c52f0c7e9389cfb0139074eb8b003ca61c134e7d
25959 F20101118_AABPMV bwenge_a_Page_018.QC.jpg
6d51a03d72eca55918712002dd121864
cd916f23e73f9e6e78375bf152adcdbaf7fdd41a
7017 F20101118_AABPNK bwenge_a_Page_025thm.jpg
d053df3cc6a24e20e2ef82073f37dbab
26627d3275c757c271f5b864a0c1656d730b42be
7220 F20101118_AABPMW bwenge_a_Page_018thm.jpg
04eb3bc3cfaa970922fd417be9bc2bf5
043cac8a6b075bf6863e62fc3f384524cec0ec67
23035 F20101118_AABPOA bwenge_a_Page_034.QC.jpg
582da4c01dd8079bc4a0e59f8fba8b76
587283dd91e658dc6e58fd2ff05ac3457aeaba4c
24503 F20101118_AABPNL bwenge_a_Page_026.QC.jpg
2420660c58255b0d72b002f27d8e2bcc
dc56653aa624039885be07f762b55d8df0369d56
23006 F20101118_AABPMX bwenge_a_Page_019.QC.jpg
b397d2ee3a26faef67ffafd7ad473862
8d21eddd9859588156ee769f68629e80840ca1a4
6545 F20101118_AABPOB bwenge_a_Page_034thm.jpg
348ff25adc6441f69371309c9d3f5150
f9491205d5fe927e15aba38bb1dfd9d90973d2a3
6864 F20101118_AABPNM bwenge_a_Page_026thm.jpg
75d8009aab8b0c0d550b32ffd37cd89b
412876fa6393019c121f7cf81002c2fc10047a7d
6457 F20101118_AABPMY bwenge_a_Page_019thm.jpg
1d391596b03f08132a7475450dd89ae3
9f9896083512ecb84ee58c78b0b22ebe4b3229c4
19864 F20101118_AABPOC bwenge_a_Page_035.QC.jpg
2bb8b30508f96444b23f3f726664c4c9
b4e52aa99b01997bfdcd83aa7db0da773abd801f
25130 F20101118_AABPNN bwenge_a_Page_027.QC.jpg
f20108d194d23cfadae9c864dd8b60a1
891a546caaf0b0787bdc8fe357445d95e1c06512
11610 F20101118_AABPMZ bwenge_a_Page_020.QC.jpg
fbe63b0123f92108da2a6dfc2d027eb1
fb91d85014d3b96430a19e3d165fa7c9f186f9bf
25574 F20101118_AABPNO bwenge_a_Page_028.QC.jpg
913958ba78d603e69936d5cbe771ea46
2abaa349bfc5859bec8f297116301db72dba790f
5815 F20101118_AABPOD bwenge_a_Page_035thm.jpg
f1872ac89160fa63ea2144ec6743ab2c
a50477e7904445f9c1000e79126fb8537baf918f
6927 F20101118_AABPNP bwenge_a_Page_028thm.jpg
b71e7c5b7cd7556c9a19b6c6880c1ed3
96baa4e39b884d610f04e9ef6fb3beecb8f9452a
F20101118_AABPOE bwenge_a_Page_036.QC.jpg
850e5a911111510989849d5351a3dcdd
a25962110abcc0e7c5dc7fe4855292bcb7fbceb5
24500 F20101118_AABPNQ bwenge_a_Page_029.QC.jpg
aede7fe5678015f488725fd3a6243c51
71fb4ec706fd85e2f8c8ed87edfd0f2cb84bcc96
3418 F20101118_AABPOF bwenge_a_Page_036thm.jpg
fc8cb269f37529aa4931576422f4fef9
550859bdbeaaa7c3768b18a21bd1c417a5ff2116
6852 F20101118_AABPNR bwenge_a_Page_029thm.jpg
9fad24396418c50e7c4dd1ca9a9c0e1e
07c5bebade41aa2cd950a19679de222c4f9a1248
23845 F20101118_AABPOG bwenge_a_Page_037.QC.jpg
b7870d737e8cee88c534d9291bb83b62
e134f63dbe64e2812bd74440d6d3d6f3e4c0d079
23824 F20101118_AABPNS bwenge_a_Page_030.QC.jpg
9e04b2ddf1de0964fe0ad325d9760c21
c053c0a24a8043ffd5008d149011fa992562ebc4
6498 F20101118_AABPOH bwenge_a_Page_037thm.jpg
9991ce33623d8655d7bc5b3f8f51956b
48d24585dd6c84213d4288fa5cc5f6c73c8efae9
6578 F20101118_AABPNT bwenge_a_Page_030thm.jpg
fa8c3e50c38b8fcbe8fcc0731cf74b19
4c91e31fdb05e3c785ae3284e9abbc767f1ccf4b
22980 F20101118_AABPOI bwenge_a_Page_038.QC.jpg
42dd40cc2e58d19919c10dd2b297e41d
4cf7f745dea1d5d40ef8a755cbec78afd67cfacc
24651 F20101118_AABPNU bwenge_a_Page_031.QC.jpg
7329baaeefeddfcf4b495c70dde83245
e611a74d8dc4084e27086c7427cca2ce2fe08577
6574 F20101118_AABPOJ bwenge_a_Page_038thm.jpg
01f19d6d8aa2cb307f1378a1ba1190e2
acffffa59bbf02535e140271712c82efbd840358
6802 F20101118_AABPNV bwenge_a_Page_031thm.jpg
f704b08f67c9bf8a0f53c417b8ad34f8
d6bde716c324c0754446bcd9c16150768c9b5f3f
26967 F20101118_AABPOK bwenge_a_Page_039.QC.jpg
84cc61bf9d88aa9ac6dac06cdb02607c
0ad552c6cc9a89712135e35e7c5f60a0d122ce4f
16011 F20101118_AABPNW bwenge_a_Page_032.QC.jpg
35a32c3923b686e25d1fd834a8962c32
cf66458957f00165cc4dea567567e31e76433348
4483 F20101118_AABPNX bwenge_a_Page_032thm.jpg
c00a6273d7b06ef3ccde8a1094e51b67
7c5a603ea4ae6eade7a175559a5201f05b5e944e
4050 F20101118_AABPPA bwenge_a_Page_047thm.jpg
a643ff659e1334637e042efd4ff70fbe
7c6ecd39fc9688a16826dbc8431f4a897206dd96
7341 F20101118_AABPOL bwenge_a_Page_039thm.jpg
b4c8214f25e24f716020faa7a5dd8c49
a282b03ff219c449134396bb7d749dad8f86b596
22473 F20101118_AABPNY bwenge_a_Page_033.QC.jpg
68c64f099f38a07aa56fe5f833841d46
4f9510b7cc5d1de8d6bd801d4dee6e657416e637
4249 F20101118_AABPPB bwenge_a_Page_048thm.jpg
edf1489d46340ed05827b7549e0e9eeb
8073cea21a09e34b4d82325342ef903c2034f246
24966 F20101118_AABPOM bwenge_a_Page_040.QC.jpg
f8c51d2528aa20428191b8bb30bc0298
da05da04eb1a9a434274fe5b8202be4684c92887
6303 F20101118_AABPNZ bwenge_a_Page_033thm.jpg
aa9b2fe84a38265ea4c56d55e95e762c
ee0ad73ad9365e83b60e0b949198fcc54eca13c0
10705 F20101118_AABPPC bwenge_a_Page_049.QC.jpg
dfade6e09ebe3b6211ab9df1b73e55c7
3e7a6528e98c8246b9fc172d8014839f1ec14fa5
6782 F20101118_AABPON bwenge_a_Page_040thm.jpg
41fd3c0561c8a79944ffe58a70f36809
1b669055192f1e47bd41ebacfffd7dfee105797a
3805 F20101118_AABPPD bwenge_a_Page_049thm.jpg
bba5cbdbd19dc1f4d5404c0cb533d84d
69c4d2cb5484d84f4ea647098ed8176f9ad5d05d
25458 F20101118_AABPOO bwenge_a_Page_041.QC.jpg
6797ce50213650965118527f83597c8d
4fb19319259c8d5558b5bedfb01cb97b8b6f45a9
14903 F20101118_AABPPE bwenge_a_Page_050.QC.jpg
6c3d91444f2752098cef41dce6f4b3d3
587ad56ea591caf1e9992afbfce2705a581dbf2a
6899 F20101118_AABPOP bwenge_a_Page_041thm.jpg
5182eedd94a92eb3d4ce41392f015b72
333ddacf370863a4db368bfec534f674228084db
4356 F20101118_AABPPF bwenge_a_Page_050thm.jpg
295d2c69d27ad2e71f2833ebb039266d
ebbbc512b83b0bd35cc976c83afbac93bdcc945b
25710 F20101118_AABPOQ bwenge_a_Page_042.QC.jpg
f7c4c4ce670c145156d92e4ff04ab48b
03497a7c6b10bff281a68f7d1fe16cb93073b820
22741 F20101118_AABPPG bwenge_a_Page_051.QC.jpg
a8a91450317ccc6ef2bb2ad9544f0ada
b5be1c635acc588b9b5ab16a41c30a340e521c53
25347 F20101118_AABPOR bwenge_a_Page_043.QC.jpg
b6f74b660bcff61188dcf14b13d263ef
f0a528848e26426d3229af6056d683f7d3792fa4
6464 F20101118_AABPPH bwenge_a_Page_051thm.jpg
5c3c810238fbaa4e24bf1b56dae892dc
0f71227c8ff10fded3408b315a57bd5223ce0782
6860 F20101118_AABPOS bwenge_a_Page_043thm.jpg
a59b751a3f1f863b73db2eb2392183bd
b12dfa0bcfbad1ae9a0150ce3dd2ef5be332120c
25278 F20101118_AABPPI bwenge_a_Page_052.QC.jpg
f30769b83a23385446ce3224ff376654
0a079948c75af42e95b9e87312ad5a58335be9a6
23199 F20101118_AABPOT bwenge_a_Page_044.QC.jpg
eb6417e35e206694d3d183936d5f6506
edff3d281008d5ddffeae77d5b8df7b59c6b85ed
6784 F20101118_AABPPJ bwenge_a_Page_052thm.jpg
6ed4212c45da4d647112935ecc629f24
a2b705c30e8e9b7b0ccd652783821e844b6e7604
6250 F20101118_AABPOU bwenge_a_Page_044thm.jpg
4e5c5be36dc82d42c1eb515026dbca08
affe1ba3fd1e5df04d8be5be2d4c5b09335891d7
25423 F20101118_AABPPK bwenge_a_Page_053.QC.jpg
9eb6daa434beddd0dbf2d73fc3cb86a0
c54410011f13997e10e792238f2d8cf54c993a43
25591 F20101118_AABPOV bwenge_a_Page_045.QC.jpg
4637bca148d8c5f328a264354fbfe163
1d2c2507f524e04c580b0b64cf7a9826185d7471
F20101118_AABPPL bwenge_a_Page_053thm.jpg
ab55f6018845135e9e132269fb67ac67
1886dea24b29add8293dcff0b32ab28d4eb46847
6827 F20101118_AABPOW bwenge_a_Page_045thm.jpg
9715a97cb9920f6aea3e3dd431562a93
4ba6f320e9b18c90c2a332dcb01b74dc2ca7d670
24027 F20101118_AABPQA bwenge_a_Page_061.QC.jpg
d7bf439f028f1a53e3a80af31c897545
0f1e64f609ea9c017e5f1f0cc3d012eb225625a2
26715 F20101118_AABPOX bwenge_a_Page_046.QC.jpg
6164139adbad9f5d4a02511807ce58e5
01a83609899a72232b43b5f0400c9efcd326e108
17597 F20101118_AABPQB bwenge_a_Page_062.QC.jpg
523270b0d845fb562f0a82e6feb59253
cdf633a620b45fe490344976045cd6ce1346f8e0
18877 F20101118_AABPPM bwenge_a_Page_054.QC.jpg
282382f6573505ab3ee83ec3f88db463
8a0eced10ed8829e9c344435f9d55b1bd458dc48
7256 F20101118_AABPOY bwenge_a_Page_046thm.jpg
afc200e47bf521c694b38977dca50f61
04c100c9376be882e5fc5aee97e4efdca54db27c
5323 F20101118_AABPQC bwenge_a_Page_062thm.jpg
e54120edb68b91001b712f0db92d004b
8c981f15dd420623babde1778d1ba98bfa0e7941
5161 F20101118_AABPPN bwenge_a_Page_054thm.jpg
205171ca6dd8455ae2ee504bca7cd7cc
a777d66266068ef1944dfa7179f3fc23866265c2
11763 F20101118_AABPOZ bwenge_a_Page_047.QC.jpg
52b7419485ab6f69670b96463d68dc80
2a67a48124d5a581891d3d43cbb1f75ac060dbf6
23980 F20101118_AABPQD bwenge_a_Page_063.QC.jpg
04fa626bae6b55ec199c23844c7f390e
edce137b55f5431eb50e3c64d6860ddd751e24a6
13871 F20101118_AABPPO bwenge_a_Page_055.QC.jpg
64a87757419059a053323515f8453e74
b7a1c8a57c1f5f0fef565f592df33be4b029e5d7
6713 F20101118_AABPQE bwenge_a_Page_063thm.jpg
b6946cd004c50ceeddf12ca94e38f06e
1de705e8540c8d8bcfbd1aecccc071f359fcd119
4566 F20101118_AABPPP bwenge_a_Page_055thm.jpg
a30c2a4d9c9f9fd208cd6db0361144bf
0091a60d7e29b3fc085c90a6938183d6c861f3aa
23155 F20101118_AABPQF bwenge_a_Page_064.QC.jpg
8f976d5ac15ca3d8877ae98034865efd
e8e32c1f6df011502ba9c662db44a5af4352d60f
12608 F20101118_AABPPQ bwenge_a_Page_056.QC.jpg
95e82e087820f5ff3b397f2e0f527af6
860522c6a67d804e055d61adc9174d3d3147a871
6700 F20101118_AABPQG bwenge_a_Page_064thm.jpg
45ff5cffa60e9adf0269189035cfbfc1
cf51d010659016f3bbe1999626e948ef64c6f1f7
4117 F20101118_AABPPR bwenge_a_Page_056thm.jpg
b8ec743993059ea1d0781e94e841c7e6
a6b3144413de020badfbd0795d94947490c36125
19401 F20101118_AABPQH bwenge_a_Page_065.QC.jpg
7c701a61adbb39c46f10976189b2cecb
ebee60841a4c0d16258a5b3c89b0af939cfcd4a9
14234 F20101118_AABPPS bwenge_a_Page_057.QC.jpg
1b5bcf9ddfa8a51aec33222d0d8bd9b7
740e119333e948b918d774f63687e6b2422d3413
5920 F20101118_AABPQI bwenge_a_Page_065thm.jpg
bc9d25b4bbaeba9126bfcb666c8e48a8
46495303803f059bd3410bc2b93da8932455f4f0
4551 F20101118_AABPPT bwenge_a_Page_057thm.jpg
2bfddc92f601bdb3992e31e08cf891d6
5adae0a6eab0c649a1ea7f8e70a23f824ceaf2d3
26926 F20101118_AABPQJ bwenge_a_Page_066.QC.jpg
bc0207dad7050610307ae4d3464c3f1c
25bdc40c633a21d6b06d2f66cf4b0290f187c81e
10972 F20101118_AABPPU bwenge_a_Page_058.QC.jpg
83f4a6cc52b19a6f075ea030cdaa7cf7
c08ccda02ede9de837881335daaf07d18b740ab1
7244 F20101118_AABPQK bwenge_a_Page_066thm.jpg
10d9242471c18bcbe26007176c1fdc75
5975247669a30c6fb073c5779bd974705f842c35
3729 F20101118_AABPPV bwenge_a_Page_058thm.jpg
b96811affedf4218510b35786b0b42d0
a7a11315b3a4ed9ccc9a31157f9e9fc3521bc11a
25438 F20101118_AABPQL bwenge_a_Page_067.QC.jpg
e63afc1416436e3fc340e8798fc65f07
1a628a6619205d50df4a8396982e6c7494f80b7d
25048 F20101118_AABPPW bwenge_a_Page_059.QC.jpg
cee908e047c55e2b74ae0d01b83d68ff
a7776cb634aec06eb60aa3efcb30d4b48f691b6b
7042 F20101118_AABPQM bwenge_a_Page_067thm.jpg
e0054cc23c664bc62fe7a73d9397b417
37c719aa39d0263fc2f39230c86e0bb9f67ffb07
6770 F20101118_AABPPX bwenge_a_Page_059thm.jpg
cf6e589cc39e700a5498bbaf8681166a
395e401cdfc1e4a2ab887177a2e211c4e12e52b1
5286 F20101118_AABPRA bwenge_a_Page_074thm.jpg
8a5bf730f8ef3de7b95a71d643c50c3c
052d03dfca2b4f41a6cf6318f953a74cac3e858e
24717 F20101118_AABPPY bwenge_a_Page_060.QC.jpg
6349709f80b572b4c66e8fb560322f13
6f05f908c339953dcda7837fddf9f1118b3b5e3f
17758 F20101118_AABPRB bwenge_a_Page_075.QC.jpg
de0f30421aa21cea2824da14737e4570
26a71c3bdf46768a36a5bddd7ac56cec3493bffd
26778 F20101118_AABPQN bwenge_a_Page_068.QC.jpg
f013e7fd41551f6ff68622f8794b2288
1f0c987978446d2ba8e75bd83b9425219798405f
6619 F20101118_AABPPZ bwenge_a_Page_060thm.jpg
d98b3244328530df505d27fdede75b00
28fc3136d2c7f0a0d66b29f463d6b61b195c02a4
5011 F20101118_AABPRC bwenge_a_Page_075thm.jpg
f88cd98d23a014d9d4a00856b9887611
0831a2765c96ea92a5313843199fe3e5dc0b1082
7293 F20101118_AABPQO bwenge_a_Page_068thm.jpg
cb8c9dd1bb53dfba8057432de0dd51ec
83e3ce0f785792c8d83d81377fbea0af3dc136a0
14669 F20101118_AABPRD bwenge_a_Page_076.QC.jpg
050106ca36138f5de412567654ed73a1
beedc091127385dbaa104a477772de02435050df
25470 F20101118_AABPQP bwenge_a_Page_069.QC.jpg
f83cb1db6c04614cfae43bc958d76e71
8bf5a5409f31c3af205b339062ed51f1021289fd
4474 F20101118_AABPRE bwenge_a_Page_076thm.jpg
70462fad6fa7592d852e51c3ef0a6beb
386b24a6b81bf06cc310a84fbbebecc43d87354c
7035 F20101118_AABPQQ bwenge_a_Page_069thm.jpg
e44c66fb9d85e64f6f944fc651617459
be230cd9fb96bfd98e05a0e74dd0fef1720d068d
12964 F20101118_AABPRF bwenge_a_Page_077.QC.jpg
fce9a6d4ad3d2654b2a5577d14088c32
dd48b26b9e24328ee14450b97b432090b93d63d4
5775 F20101118_AABPQR bwenge_a_Page_070.QC.jpg
577f7409a49465dd15218a8cb8901b6f
22882e0b56caa7bd9ff0fe94750a73ca5ebfc5a4
3937 F20101118_AABPRG bwenge_a_Page_077thm.jpg
d7b12729fc25161fd52853bdaf7f8b8a
5e330c0d8d3d19d3123523c3393289300ca14b8f
1999 F20101118_AABPQS bwenge_a_Page_070thm.jpg
cddad1dc215536abd7013a9b99f6b1df
732d7f59c75dab2db580ab59c2ce55f1f96642a0
8909 F20101118_AABPRH bwenge_a_Page_078.QC.jpg
cabc29763d5fa9a2f3b2ccb5fbfb3fb4
a0216709fc1578e755b76d28ec749a437bdcdb2f
19554 F20101118_AABPQT bwenge_a_Page_071.QC.jpg
2c025b946a572a9b85a4386482575088
a72a73622b82e3099017ab2386324884e906f4ce
3382 F20101118_AABPRI bwenge_a_Page_078thm.jpg
21a29252fd6d55e0ca70f5432214ca7b
666548e7340cd4d759322dd414474da54da2207a
5612 F20101118_AABPQU bwenge_a_Page_071thm.jpg
ef53e6108c6970bd1b5e93eac0925e69
65591816ffa1a05e36676624949dfec8e2d8ecf0
14994 F20101118_AABPRJ bwenge_a_Page_079.QC.jpg
d2514cf334f73781617d3d032e3b86be
e1dd3620c399f9660bd6d020beb9b089a434980e
14664 F20101118_AABPQV bwenge_a_Page_072.QC.jpg
bba5758e1fbc6032ae05f350911229ef
d84a00f82d7841d1619028b2d600190af8e74f04
4352 F20101118_AABPRK bwenge_a_Page_079thm.jpg
4be02841ef7857925edcd6e54b9c55ca
d339ea4417815f72c952ce2837f04be1aaae8387
4419 F20101118_AABPQW bwenge_a_Page_072thm.jpg
059aa41157cf19db1d947558aec10230
f86197579fb319050db3c97d359370228597a9e7
6416 F20101118_AABPRL bwenge_a_Page_080.QC.jpg
a30805167bc106c2e06767356a8a0e41
0cc1453141aa16e89481f001996e639da789d966
17228 F20101118_AABPQX bwenge_a_Page_073.QC.jpg
c91c4e0763f7e9e51d2953956293edde
d3a71f26b31417af3937459865982c292a2ad7dd
17067 F20101118_AABPSA bwenge_a_Page_088.QC.jpg
1871e899aa02a26398e23de43e5fb825
aece6b70b1d553ca00c251b4218b78ca21d75753
2483 F20101118_AABPRM bwenge_a_Page_080thm.jpg
862afc6d90de33b526e445ed81358f00
5633a2eade4897475fdfef3d38a495ee42d70a7b
5178 F20101118_AABPQY bwenge_a_Page_073thm.jpg
f3cd95023c7d38449ee42c2c1115fc05
4689e385c1e9b394f8b8990537031a1435b415a7
5103 F20101118_AABPSB bwenge_a_Page_088thm.jpg
40a3656f71029cb6d2721b1aa57333f0
afc96ed3bcfe12b642a46847af36d3e552f63ea5
25635 F20101118_AABPRN bwenge_a_Page_081.QC.jpg
29512fcd2c1e290f2fec5f96fc81afc2
b95db17482de793098ccf5c315d233839a4a4d3c
17759 F20101118_AABPQZ bwenge_a_Page_074.QC.jpg
b67ffb40121e32cc7fc3b2d46fd16a21
217f979f3b19d3c92bff811a1fbd3e194c937632
54916 F20101118_AABOPA bwenge_a_Page_042.pro
8e5e39fc5eb75a74e104eaf50968f697
4db460143e1fc01e1ab698496e15d716e5e25be5
6451 F20101118_AABPSC bwenge_a_Page_089.QC.jpg
825d755f46c619d435d81ae870d35c26
f9c6528255f49673afedfbfae1afbd781af06aed
6941 F20101118_AABOPB bwenge_a_Page_084thm.jpg
0940ee691e4a48c5b708be1766fa3841
554fcc3d85521371b70f885e68bb23e4e3e61452
2402 F20101118_AABPSD bwenge_a_Page_089thm.jpg
fdf9ba97bb03b31a2598748a65449332
e621030b3fc5ab948655913ea2a3052c6cc01053
7007 F20101118_AABPRO bwenge_a_Page_081thm.jpg
b7b5f97f83c38f9b67054fde0d366153
01ba9fcb7999268f2136b3da7705c24dfd47f1e0
F20101118_AABOPC bwenge_a_Page_054.tif
8553ce78e34c8791c6e6919c08bea746
376ade2dd0cf0423c5468fdc1e6d9cf349554860
23881 F20101118_AABPSE bwenge_a_Page_090.QC.jpg
5bcf4b3d5d3fcb6a672148cb70792c7a
8c72376d3a706cf2e30ec268630d58c966b549e0
23696 F20101118_AABPRP bwenge_a_Page_082.QC.jpg
836e1be8c620a9b130b096707f8a9666
6b32675e823916e521de79c9740b1e40d41b575c
F20101118_AABOPD bwenge_a_Page_065.txt
d7401e8e499eef6677d6b39549d7a7a7
a89848e5dcd3e0f9004c8e39bffaf6c4d04ddc43
6661 F20101118_AABPSF bwenge_a_Page_090thm.jpg
306cb2d41c639595ba3e6a337a66ae68
86b7ad20d96d310100b1615448275538ceacff96
6696 F20101118_AABPRQ bwenge_a_Page_082thm.jpg
a0b947dae98ba26f4aa17adb0084e106
10a702fa01b6ba71814b3f1a67a2ea9c96f6363d
85434 F20101118_AABOPE bwenge_a_Page_006.pro
481b24113c30a3fdf7b519795c7551f3
1f2480caf377db15370017e42d7b6719e977da78
27435 F20101118_AABPSG bwenge_a_Page_091.QC.jpg
546d7659f82ba195acdcdd04f25ed330
27a81a16463f2a8d797ec1ae84c48ad4f718e675
25828 F20101118_AABPRR bwenge_a_Page_083.QC.jpg
92c40ae4865d65f6dd1b8affe0b155ca
7cf13b64d2d76da8984cfedbfc7da8edced84d4a
14535 F20101118_AABOPF bwenge_a_Page_048.QC.jpg
316c2ea9ea16652bacc6a70caafd0027
927a4f7e473f29ea5e9592d3d3e5eac7e2e1f0aa
7370 F20101118_AABPSH bwenge_a_Page_091thm.jpg
652d3cde433e12dfceb7d27b38cd71f2
79c864daf5548856dd275c8d082222964c64b163
6991 F20101118_AABPRS bwenge_a_Page_083thm.jpg
d1b5828badc2e2ca42a68883777d8c95
24eaefbe32f0bf49210c9c53919bfcbb569ed3b5
6356 F20101118_AABOPG bwenge_a_Page_061thm.jpg
31d5c3bd26872906bbbb5523667e80d4
c3fb5146571284093386669d52b006bde1c660b1
24440 F20101118_AABPSI bwenge_a_Page_092.QC.jpg
c65442649c1484ab55938dfe3e14d9bb
f02259d0687fdbba3b9f5338278d4d7767729cfd
26079 F20101118_AABPRT bwenge_a_Page_084.QC.jpg
f6428d111ae08860fbac5f059c4cd292
5193956e2e2dd09cb1695261486bb21c795a6eff
F20101118_AABOPH bwenge_a_Page_109.tif
45641ab75407b3e04c3c927834da0ed1
238462c339bf2eb5a3a49db3fef1d28cbb2ae820
6796 F20101118_AABPSJ bwenge_a_Page_092thm.jpg
a3d92981ee41ca53eee1bc83aa125356
f3fcd238e01e8684e51f6b9ed691fa86f954a197
24324 F20101118_AABPRU bwenge_a_Page_085.QC.jpg
b1732f9dbb17067432fe5d003e18f786
57b6a92e8878ed9a71896f7e2322acffc1f6f77b
6652 F20101118_AABOPI bwenge_a_Page_027thm.jpg
6b40744ab448993ee396440e3d4e76ea
9122872362541c6398492acbf02a9909653868d0
24166 F20101118_AABPSK bwenge_a_Page_093.QC.jpg
484a2eb7ea4b376cc8362fc3b4846e65
6bf7f198447d68ab29d370c65a2237713ab0b07b
6706 F20101118_AABPRV bwenge_a_Page_085thm.jpg
72ccc8b489256e645dfeb5cf34de8c97
09999121d770e55c79d8544f1f118dd9eb18bf92
174306 F20101118_AABOPJ UFE0021470_00001.xml
b3b32bd818a3314bae545c2081539fc8
b65485ac62d3dbc6352464703dbccdbead58fc81
6198 F20101118_AABPSL bwenge_a_Page_093thm.jpg
6ab786593f1a0f58656130ca222e16e4
73eba55aa0f7476af0cfca94cfde60d88949413c
507 F20101118_AABOOU bwenge_a_Page_001.txt
cd99e8242e95bed01d50e682fbdb2b25
b72328a93b19297fa7f32e56c80e5a8f1f3ddebf
26387 F20101118_AABPRW bwenge_a_Page_086.QC.jpg
1a1a22f0b82e83f6b3f2a8aa57a4caeb
c9969faf7ec529842a6a587c25e09efe7e591ce6
13444 F20101118_AABPTA bwenge_a_Page_101.QC.jpg
2f067c6fbee9523af710d00ca8583752
3b572e16b0d7faeb2c63659bd2af59b75a32f3a8
12171 F20101118_AABPSM bwenge_a_Page_094.QC.jpg
a96e04ebbc16d28fb2d23667cc7ba1bd
63f3787117f4a1ab4f7017f7a691dcca93244ecd
F20101118_AABOOV bwenge_a_Page_070.tif
9592d67c55f41e94ab37b47cf34590ad
891e14ed1c4c61b6f1a2adbf6d6443da8bff89bd
7162 F20101118_AABPRX bwenge_a_Page_086thm.jpg
216ab0e9c82cb20f2fed5969e61a2b98
8af643f52457f0943dd439a4802a8315a95a89a1
4235 F20101118_AABPTB bwenge_a_Page_101thm.jpg
0f4932adbe569cac7eac55aa4d210e0b
8cf22932aaac1ba6670174283af4305418eb6767
3615 F20101118_AABPSN bwenge_a_Page_094thm.jpg
e130522b61cf28a8d12818283131a2bb
60bb9671d1fbb6ff37b112a81ad07f1daa1f48eb
13213 F20101118_AABOOW bwenge_a_Page_106.QC.jpg
7079e8ff57b39e340e3477d8f3e4c8c0
a4e6fb9f09c60dff476d3d4ce354b91b1abe2d1d
7183 F20101118_AABPRY bwenge_a_Page_087.QC.jpg
21ab622bb5d5868d2e21dd8fbd47d976
368597ae86f2f79e3279e2ce3b00315b0c948660
8463 F20101118_AABPTC bwenge_a_Page_102.QC.jpg
0a7c18dbfe7dc951ccaa84bd3c20a6f0
f6124121dea6930ba1e9ac006d90cf1da938626d
24443 F20101118_AABOPM bwenge_a_Page_001.jpg
168547c7a6d254c70cd0ebac1516a3f4
c0f4006200d35fda9c0a9fb59db242acd0733f05
14156 F20101118_AABPSO bwenge_a_Page_095.QC.jpg
a74bfd8372158bf72c348ddcc31fe7bd
0e1da7b764a681cf2caaee773dfe79c4e5f4a750
7024 F20101118_AABOOX bwenge_a_Page_042thm.jpg
b9e3223b077d555b40235218a53bd11f
47c069d518d12d7ce146e86118690d32911172a2
2381 F20101118_AABPRZ bwenge_a_Page_087thm.jpg
931ce5715c0a966fba3d65a99591ad3a
2b50eb5617762501bfd97a3054856942b3fae157
86317 F20101118_AABOQA bwenge_a_Page_015.jpg
4de495d7843465dccccef07103e58533
6e52130ccab929cedf6b1384df6a46715796479e
2854 F20101118_AABPTD bwenge_a_Page_102thm.jpg
ec5ecaca48ec0f5ff8aef61d108aab9f
e7cc8e81df8717620a00cdc055c827c15071e6db
423 F20101118_AABOOY bwenge_a_Page_087.txt
4cfa2c8103a1dcab620d1faeeec28d87
b5316bd8e9705da802a7ae5e40f89714cc6ba134
77361 F20101118_AABOQB bwenge_a_Page_016.jpg
5066977714ea2f3778ef7706ae1c44ad
97777c290d7a7f8dbf8b18ceeb5b184852b78134
9402 F20101118_AABPTE bwenge_a_Page_103.QC.jpg
b17c819ae8a37371f50129c15b65302b
ad99eca2bb6eabaddfe5be68db05c4ff7a33b585
10055 F20101118_AABOPN bwenge_a_Page_002.jpg
f5b2ae2afc91702be520e42a98f7a4ab
c9d5af3b09af4a3aa118da15d42017d2e42604ca
4132 F20101118_AABPSP bwenge_a_Page_095thm.jpg
9680de2a86607772e849a234f534e754
bffaf8be40f7c94e9fafc8f3ccb38d752ecf9c60
4282 F20101118_AABOOZ bwenge_a_Page_104thm.jpg
dadfb46ec3cdfa1c350156a61cb8a011
374d7097fa518551789451174609a49874530215
80445 F20101118_AABOQC bwenge_a_Page_017.jpg
85d87d622914ba9f75b20e68b89b5119
b83ee17abfcf6ed365b84c5fd4f789da159a35d3
3011 F20101118_AABPTF bwenge_a_Page_103thm.jpg
dc4baa6d515261b95e4309caac057210
46b4915cdcdacf62715552de3c6d08cc807598a0
11336 F20101118_AABOPO bwenge_a_Page_003.jpg
0955804c2f6793132781ab9478b609e3
2c7842762a8987800a00467b51f241d99960c1f6
16617 F20101118_AABPSQ bwenge_a_Page_096.QC.jpg
b082683f1f5ccf0c0d93a0aacad070fe
a21549cd81b7279ce9b5d10c3d78a3cd17f48e66
80294 F20101118_AABOQD bwenge_a_Page_018.jpg
b24e61f5c8d3dbe357915c01313bdffc
205ed4c56c10678a2cc89cfbfd4894055b11c3c8
14475 F20101118_AABPTG bwenge_a_Page_104.QC.jpg
57c862896336cbbf0a1bc2e3315d3ba1
13d4e48c7c9b8ec146b0d9f2056c56ddcc014f98
45335 F20101118_AABOPP bwenge_a_Page_004.jpg
9b7390b61f9d8573be17f15172cdabe2
5e804fe7fdccafb1e06826b6052988951ad46d3d
5415 F20101118_AABPSR bwenge_a_Page_096thm.jpg
2e3d38f1b406550c31ea74eeceb551ad
c724b4ddc3ffa89983cc66515bed437a9382831c
71009 F20101118_AABOQE bwenge_a_Page_019.jpg
44fdb71a47aca8ebb98f3b91582d29e9
a08c9262e67c667dba475ab281311f58c7eb59aa
10869 F20101118_AABPTH bwenge_a_Page_105.QC.jpg
de9baa32e62589cc1ccff504fda41cde
1fb36fb247b29b5c12cae0d61580e723cce777f2
76176 F20101118_AABOPQ bwenge_a_Page_005.jpg
4f1ed35444f8962bcd12387dd38fbb18
7b6dbd5c7d2f42ecb6e45df5240f0f41125fe742
16099 F20101118_AABPSS bwenge_a_Page_097.QC.jpg
c5dc0bd2c53b9760ee6909133fec12a0
f8b9411275ad2bdee54e3b5052a4631722affc09
35006 F20101118_AABOQF bwenge_a_Page_020.jpg
e61b449e72e79c5ba9ec4d01558ae0b0
5f725d029b3e3464e42ecb835b3d6175d7b7629a
3392 F20101118_AABPTI bwenge_a_Page_105thm.jpg
05a4b089544a42693a5a926f69ee55b3
4935e392b4c73b61e21b2aa764aa88f92b17f38f
82090 F20101118_AABOPR bwenge_a_Page_006.jpg
dbab64c87390d259a48d5c854b806b11
bfe5b5d9c390313e1b34ca749805ef548ec4d144
5493 F20101118_AABPST bwenge_a_Page_097thm.jpg
6d28739d518d295fed251144e01e04e9
3f340f03bd4b3fd7a45f0f271ee885c977da2741
32027 F20101118_AABOQG bwenge_a_Page_021.jpg
db14f89b81259803faf9debaa55d1e1d
5abd531f4d61e37d8d0b7383674dccf82e20101b
4038 F20101118_AABPTJ bwenge_a_Page_106thm.jpg
e8c9809d4f4ddedc8b37988d83066f46
33fc8284ac3efd5cdd8cce14eada699a43dc0735
83390 F20101118_AABOPS bwenge_a_Page_007.jpg
1a3d76f4997f28320f34e90e4076b916
16141db3ad4a90dc5b5b68a83a73ae5025f55599
16236 F20101118_AABPSU bwenge_a_Page_098.QC.jpg
f740a01ee4edb768f66be5f3a8ac6d72
9a69c1ac611f7235923288080d03155e0875f5b5
28843 F20101118_AABOQH bwenge_a_Page_022.jpg
b32176318317eb9d18b4bb10266d6a0d
6b14e9a77402e92c36527cf749bcee3eda767597
15707 F20101118_AABPTK bwenge_a_Page_107.QC.jpg
c0fd666e688c73603c751e6e5053e508
c07bf8cce646d1a65fee031856a8f445866cd538
13506 F20101118_AABOPT bwenge_a_Page_008.jpg
dc1aec30930b0879eaf0fb0f7c379667
81fec02b693533f5b68088e6412cad9bcb096c1e
5510 F20101118_AABPSV bwenge_a_Page_098thm.jpg
0c1b8043c202dea6a23a18f5505bc65d
dcd4b4ec33cced91cf2bc8881752a096dc31c603
71433 F20101118_AABOQI bwenge_a_Page_023.jpg
ae7928a09cfd1429e1303a3a939261b7
10dec0e7934cbdcd364c0e6a63a58d1d65dea5e5
4773 F20101118_AABPTL bwenge_a_Page_107thm.jpg
6f43aca94ddaae657d52b17a2f12e106
a0b92818f38ee31208993e28699c45d13dd84a07
78847 F20101118_AABOPU bwenge_a_Page_009.jpg
3d612749ba4ce8b6d4205105bdf3661e
843b4397c714a38fe703706abe6cab6d8b51ae0c
13083 F20101118_AABPSW bwenge_a_Page_099.QC.jpg
92d4d1e9f24caae956f3e08e19d3fd4f
3b076cbc2dbe438ebe87a5ed10d9a35c875d7514
76753 F20101118_AABOQJ bwenge_a_Page_024.jpg
e3bb39b67611dfdc6f92fad7390e7484
bd1f335745189b41a1b6c4986f25151901d58ef4
21338 F20101118_AABPUA bwenge_a_Page_115.QC.jpg
5fbf23f0e96e58bd39c8ccc9dfed08e9
e8918e6b6268971e66208d99c1d5e28232b38014
19601 F20101118_AABPTM bwenge_a_Page_108.QC.jpg
36fa409f820c94bdec654ecdce931956
d9130b79c079c0e30a1c944df28714d748039bae
39062 F20101118_AABOPV bwenge_a_Page_010.jpg
6203eec60c7406127a316ef8731ec8a6
9c2ab0d1f0a62c2fc8e633eff4e463026cfde479
4496 F20101118_AABPSX bwenge_a_Page_099thm.jpg
326fbaad053489812142c3f61c298be2
612128af6b0d3a6ca5d3038aa039efd3f0996dfd
75947 F20101118_AABOQK bwenge_a_Page_025.jpg
51cbcfa4260eecf4b4b8a766e7a4b2ae
3333b1a6a43bb361a8c0fa993cce35de0d0e294b
5866 F20101118_AABPUB bwenge_a_Page_115thm.jpg
a1ac0bbbc876518174f63dfb032b91c1
25065ce82da0fe0061d01d5cfd7d8d76a11dfc7e
5192 F20101118_AABPTN bwenge_a_Page_108thm.jpg
3593d6425f4d28350a6ab2818a1244bf
37a77b61dbdb4603332aefd4eba58eef5e02889b
71626 F20101118_AABOPW bwenge_a_Page_011.jpg
aaa87c697c1b802202575da5f1cac43e
760a0eff73e49a932844d29cd3937ad128541f82
16566 F20101118_AABPSY bwenge_a_Page_100.QC.jpg
7f02669c2552d103d15175009c57babe
c8ca1e233e4c1b7a94f325b04adcebea930f48d7
75032 F20101118_AABOQL bwenge_a_Page_026.jpg
d330cc0f18145e748500fa4a78950a29
1bb3f53c459291cc4d9fb636f220ae4becfece77