Citation
Evaluation of Mosquito Trapping Efficiency and Determination of Seasonality for Mosquitoes at the University of Florida Horse Teaching Unit

Material Information

Title:
Evaluation of Mosquito Trapping Efficiency and Determination of Seasonality for Mosquitoes at the University of Florida Horse Teaching Unit
Creator:
DILLING, SARAH COURTNEY
Copyright Date:
2008

Subjects

Subjects / Keywords:
Carbon dioxide ( jstor )
Diseases ( jstor )
Horses ( jstor )
Magnetism ( jstor )
Mosquito control ( jstor )
Odors ( jstor )
Rain ( jstor )
Species ( jstor )
Stall ( jstor )
Surveillance ( jstor )
City of Gainesville ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Sarah Courtney Dilling. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
12/18/2004
Resource Identifier:
57722322 ( OCLC )

Downloads

This item has the following downloads:

dilling_s ( .pdf )

dilling_s_Page_013.txt

dilling_s_Page_088.txt

dilling_s_Page_083.txt

dilling_s_Page_047.txt

dilling_s_Page_103.txt

dilling_s_Page_089.txt

dilling_s_Page_068.txt

dilling_s_Page_085.txt

dilling_s_Page_001.txt

dilling_s_Page_044.txt

dilling_s_Page_009.txt

dilling_s_Page_104.txt

dilling_s_Page_022.txt

dilling_s_Page_134.txt

dilling_s_Page_031.txt

dilling_s_Page_055.txt

dilling_s_Page_114.txt

dilling_s_Page_024.txt

dilling_s_Page_036.txt

dilling_s_Page_058.txt

dilling_s_Page_081.txt

dilling_s_Page_023.txt

dilling_s_Page_115.txt

dilling_s_Page_092.txt

dilling_s_Page_125.txt

dilling_s_Page_039.txt

dilling_s_Page_098.txt

dilling_s_Page_132.txt

dilling_s_Page_102.txt

dilling_s_Page_120.txt

dilling_s_Page_101.txt

dilling_s_Page_062.txt

dilling_s_Page_090.txt

dilling_s_Page_099.txt

dilling_s_Page_043.txt

dilling_s_Page_106.txt

dilling_s_Page_111.txt

dilling_s_Page_005.txt

dilling_s_Page_054.txt

dilling_s_Page_095.txt

dilling_s_Page_117.txt

dilling_s_Page_006.txt

dilling_s_Page_131.txt

dilling_s_Page_060.txt

dilling_s_Page_108.txt

dilling_s_Page_050.txt

dilling_s_Page_057.txt

dilling_s_Page_107.txt

dilling_s_Page_091.txt

dilling_s_Page_100.txt

dilling_s_Page_015.txt

dilling_s_Page_061.txt

dilling_s_Page_059.txt

dilling_s_Page_113.txt

dilling_s_Page_011.txt

dilling_s_Page_026.txt

dilling_s_Page_129.txt

dilling_s_Page_070.txt

dilling_s_Page_051.txt

dilling_s_Page_045.txt

dilling_s_Page_053.txt

dilling_s_Page_016.txt

dilling_s_Page_082.txt

dilling_s_Page_019.txt

dilling_s_Page_048.txt

dilling_s_Page_027.txt

dilling_s_Page_010.txt

dilling_s_Page_056.txt

dilling_s_Page_074.txt

dilling_s_Page_018.txt

dilling_s_Page_041.txt

dilling_s_Page_087.txt

dilling_s_Page_035.txt

dilling_s_Page_037.txt

dilling_s_Page_112.txt

dilling_s_Page_025.txt

dilling_s_Page_109.txt

dilling_s_Page_002.txt

dilling_s_Page_079.txt

dilling_s_Page_073.txt

dilling_s_Page_071.txt

dilling_s_Page_075.txt

dilling_s_Page_066.txt

dilling_s_Page_076.txt

dilling_s_Page_052.txt

dilling_s_Page_007.txt

dilling_s_Page_118.txt

dilling_s_Page_008.txt

dilling_s_Page_032.txt

dilling_s_Page_127.txt

dilling_s_Page_065.txt

dilling_s_Page_123.txt

dilling_s_Page_105.txt

dilling_s_Page_097.txt

dilling_s_Page_034.txt

dilling_s_Page_033.txt

dilling_s_Page_094.txt

dilling_s_Page_128.txt

dilling_s_Page_042.txt

dilling_s_Page_084.txt

dilling_s_Page_077.txt

dilling_s_Page_096.txt

dilling_s_Page_003.txt

dilling_s_Page_017.txt

dilling_s_Page_014.txt

dilling_s_Page_029.txt

dilling_s_Page_133.txt

dilling_s_Page_040.txt

dilling_s_Page_072.txt

dilling_s_Page_030.txt

dilling_s_Page_078.txt

dilling_s_Page_080.txt

dilling_s_Page_046.txt

dilling_s_Page_067.txt

dilling_s_Page_063.txt

dilling_s_Page_122.txt

dilling_s_Page_119.txt

dilling_s_Page_049.txt

dilling_s_Page_124.txt

dilling_s_Page_093.txt

dilling_s_Page_121.txt

dilling_s_Page_012.txt

dilling_s_Page_020.txt

dilling_s_Page_038.txt

dilling_s_Page_069.txt

dilling_s_Page_116.txt

dilling_s_pdf.txt

dilling_s_Page_086.txt

dilling_s_Page_110.txt

dilling_s_Page_028.txt

dilling_s_Page_126.txt

dilling_s_Page_064.txt

dilling_s_Page_021.txt

dilling_s_Page_004.txt

dilling_s_Page_130.txt


Full Text












EVALUATION OF MOSQUITO TRAPPING EFFICIENCY AND DETERMINATION
OF SEASONALITY FOR MOSQUITOES AT THE UNIVERSITY OF FLORIDA
HORSE TEACHING UNIT















By

SARAH COURTNEY DRILLING


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


2004

































Copyright 2004

by

Sarah Courtney Dilling

































This document is dedicated to Belle the Appaloosa mare who unexpectedly became ill
and died November 2004.















ACKNOWLEDGMENTS

I would like to thank Saundra TenBroeck for having confidence in me and showing

supportive guidance throughout my program. I thank Jerry Hogsette for constantly and

willingly being there for support and assistance while educating me in a subject area

where I was just a novice. I would also like to express my gratitude to Dan Kline for his

encouragement and suggestions. I thank Alyce Nalli for her constant assistance and

genius ideas; I could not have completed this research without her. I would also like to

thank the staff at the University of Florida Horse Teaching Unit, especially Joel

McQuagge and Tonya Stephens. I thank Aaron Lloyd and Joyce Urban for helping me

identify mosquitoes and always making time to help. I would like to thank Kelly

Spearman, Talia Bianco, Tonya Stephens and Kylee Johnson for all of their assistance

teaching me proper horsemanship; I have become a better rider because of them. The

utmost appreciation goes to my loving and supportive husband, Bradley Dilling. I thank

Brad for helping me with fieldwork, getting me to finally write this thesis and keeping me

focused when times got tough.

















TABLE OF CONTENTS

page

ACKNOW LEDGM ENTS ........................................ iv

LIST OF TABLES ............... ............. ...................... vii

LIST OF FIGURES ..................................... ix

ABSTRACT.................. .................. xi

CHAPTER

1 LITERATURE REVIEW .................. ......... ........................1

Taxonomy .................................................... .........2
M orphology and Life Cycle...................................................... 3
Behavior and Ecology............. ... ..............
Flight C categories ........................................6
Meteorological Conditions Affecting Flight ..........................................7
Feeding B ehavior................................
H ost L location B ehavior................................................9
Host Preference .............. ............. ........ ...............1
Reproduction Behavior.............................................. 12
Seasonality ......................... ..................... ........12
H health and Econom ic Im pact...................................... ............... 13
Control ............. ........... .... ......._. ........................18
Chemical and Biological Mosquito Control............... .................18
Nonchemical M osquito Control and Surveillance ............................................20
Sum m ary ................................. .................. ............... ........ 25

2 SEASONALITY OF MOSQUITOES AT AN EQUINE FACILITY IN NORTH
CENTRAL FLORIDA..... ........... ........ ..................32

Introduction.............................. ..................32
M materials and M methods ............................................................33
Experimental Design .............................................. .... ....34
Data Analysis................................ ........34
Results ................................... .................34
Discussion ................................. ...... ...........35
Conclusions............................................. .........36


v











3 MOSQUITO TRAPPING STUDIES AT AN EQUINE FACILITY IN NORTH
C E N TR A L FL O R ID A .......... ..... ........... ........................................................ 54

Introduction ...................... ..... ..... ............ .54
M materials and M methods ............................................................55
Experimental Design .............................................. .... ....56
CDC 1012 Trapping Study .............................. ........56
M M Pro Trapping Study .............................................. ............... 57
H orse O dor Study .................. ......... .. ........... ............ 57
Location Profile Study.............................................................. ..............58
Horse Vacuuming Study............................... ...............59
Separate Entity Study .............................. ............................. 59
Data Analysis............................... .... ......... 60
Results ................. ...................................... ..................... ........ 60
M M Pro Trapping Study .............................................. ............... 61
H orse O dor Study .................. ......... .. ........... ............ 62
Location Profile Study.............................................................. ..............63
Horse Vacuuming Study............................... ...............63
Separate Entity Study .............................. ............................. 64
Discussion ......................... .. ... .... ....... .........65
CDC 1012 and MMPro Trapping Studies...................................................65
H orse O dor Study .................. ................................... ............ 67
Location Profile Study.............................................................. ..............67
Horse Vacuuming Study............................... ...............68
Separate Entity Study .............................. ............................. 70
Conclusions............................ ........... .......... 71

4 SUM M ARY ...................................... ................................. ........ 113

LIST OF REFERENCES ..................................... ........ ........... ......... 115

B IO G R A PH IC A L SK E T C H ...................................................................................... 12 1
















LIST OF TABLES


Table page

1-1 Classification of family Culicidae ................................................. .........30

1-2 Systemic list of mosquitoes found in Florida............ ........................31

2-1 Total count, and percent of total count of mosquito species trapped during
seasonality study .....................................................52

2-2 Mean numbers ( standard deviation) of mosquitoes trapped each month .........53

3-1 Schedule of CDC 1012 trapping study .................................. ........101

3-2 Schedule of M M Pro trapping study................................................................. 102

3-3 Interval schedule of horse odor study. ................................................................ 103

3-4 Schedule of horse odor study................................................. 104

3-5 Schedule of separate entity study stall assignments .......................................... 105

3-6 Mean numbers ( standard deviation) of mosquitoes captured during CDC 1012
trapping study. .............. ................... ............ .106

3-7 Mean numbers ( standard deviation) of mosquitoes captured during MMPro
trapping study. .............. ................... ............ .107

3-8 Mean numbers ( standard deviation) of mosquitoes captured during horse odor
study. ............................. ................. ......... 108

3-9 Mean numbers ( standard deviation) of mosquitoes captured during location profile
study. .............................................. ......... 109

3-10 Mean numbers of mosquito species trapped per position during location profile
study. ....................... ................. .. .................... 110

3-11 Mean numbers ( standard deviation of difference of mosquitoes vacuumed from
horse 1 and horse 2 during vacuuming study. ..................................................1111









3-12 Mean numbers ( standard deviation) of total mosquitoes vacuumed from horse 1
and horse 2 during vacuuming study.............................................112
















LIST OF FIGURES


Figure page

1-1 CDC 1012 (John W. Hock Company, Gainesville, FL) mosquito trap....................27

1-2 Mosquito Magnet X (American Biophysics Corporation, East Greenwich, RI)
mosquito trap ............... ..... ....................... .28

1-3 Mosquito Magnet Pro (American Biophysics Corporation, East Greenwich, RI)
mosquito trap .................................................29

2-1 Weather station at UF Horse Teaching Unit. .......................................39

2-2 Aerial photograph of the UF Horse Teaching Unit showing the four MMPro trap
sites for the seasonality study. ........................................ ................. 40

2-3 Total mosquito counts as related to months during the seasonality study...............41

2-4 Total mosquito counts as related to rainfall during the seasonality study...............42

2-5 Maximum and Minimum temperatures recorded at the meteorological station during
the seasonality study..................... ...................43

2-6 Total rainfall (centimeters) recorded during the seasonality study. ........................44

2-7 Total numbers of Culex nigripalpus trapped by MMPro during seasonality study. ..45

2-8 Total numbers of Culex erraticus trapped by MMPro during seasonality study. .....46

2-9 Total numbers of Culex salinarius trapped by MMPro during seasonality study......47

2-10 Total numbers of Mansonia spp. trapped by MMPro during seasonality study.......48

2-11 Total numbers of Anopheles spp. trapped by MMPro during seasonality study......49

2-12 Total numbers of Coquillettidiaperturbans. trapped by MMPro during seasonality
study. ............................. .................... ........ 50

2-13 Total numbers of Psorophora spp. trapped by MMPro during seasonality study. ..51

3-1 Portable vacuum aspirator (DC Insect Vac. BioQuip, Rancho Dominguez, CA) used
for collection of mosquitoes during vacuuming studies.............. ..............73









3-2 Paint gelding that was sampled for mosquitoes during trapping studies..................74

3-3 Appaloosa mare that was sampled for mosquitoes during trapping studies...............75

3-4 CDC 1012 trap placement during CDC 1012 trapping study...............................76

3-5 CDC 1012 trapping study placement used for mosquito trapping. .........................77

3-6 MMPro trap placement during MMPro trapping study............. ..............78

3-7 MMPro trap placement used during mosquito trapping study. ...............................79

3-8 Modified CDC 1012 mosquito trap used during horse odor study.............................80

3-9 CDC 1012 trap layout used for mosquito location profile study.............................81

3-10 Technique of aspirating mosquitoes off of horse during horse vacuuming study and
separate entity study. ....................... ........... ........82

3-11 Feeding slips at UF Horse Teaching Unit where the 2nd trial of horse vacuuming
study and separate entity study was conducted. .............................................83

3-12 Distribution and mean numbers of mosquitoes trapped during CDC 1012 trapping
study. ............................. .................... ........ 84

3-13 Total numbers of mosquitoes captured during CDC 1012 trapping study............85

3-14 Mean numbers of mosquitoes captured during CDC 1012 trapping study. ..........86

3-15 Distribution and mean numbers of mosquitoes trapped during MMPro trapping
study. ............................. .................... ........ 87

3-16 Total numbers of mosquitoes trapped during MMPro trapping study. ....................88

3-17 Mean numbers of mosquitoes trapped during MMPro trapping study...................89

3-18 Distribution and mean numbers of mosquitoes trapped during horse odor study....90

3-19 Mean numbers of mosquitoes trapped during horse odor study..............................91

3-20 Mean numbers of mosquitoes trapped during the different treatments of horse odor
study. ............................. .................... ........ 92

3-21 Percent of mosquito species trapped during location profile study..........................93

3-22 Distribution and mean numbers of mosquitoes trapped per position during location
profile study..................... ................. ......... 94

3-23 Mean numbers of mosquitoes trapped per position during location profile study...95









3-24 Percent of mosquito species captured during vacuuming study, trial 1. ...............96

3-25 Percent of mosquito species captured during vacuuming study, trial 2. ...............97

3-26 Percent of mosquito species captured from Appaloosa mare during vacuuming
study. ............................. .................... ........ 98

3-27 Percent of mosquito species captured from Paint gelding during vacuuming study.99

3-28 Distribution and mean numbers of total mosquitoes captured during separate entitiy
study. ............................. .................. ......... 100
















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

EVALUATION OF MOSQUITO TRAPPING EFFICIENCY AND DETERMINATION
OF SEASONALITY FOR MOSQUITOES AT THE UNIVERSITY OF FLORIDA
HORSE TEACHING UNIT

By

Sarah Courtney Dilling

December 2004

Chair: Saundra H. TenBroeck
Major Department: Animal Sciences

Traps are effective surveillance tools for monitoring seasonal prevalence and the

species composition of mosquitoes, along with reducing mosquito numbers nearby. There

is little research to study the efficiency of mosquito traps in a competitive environment

with a natural host. Such research is warranted because these pests can cause substantial

economic losses to the equine industry through nuisance biting and disease transmission.

A series of studies were conducted to achieve this objective, including competitive

studies with traps and a horse acting as a natural host, determining prevalent species

feeding on a horse, location profiles of mosquitoes through trapping, adding horse odors

to the trap airstream, and determining the distance required between two horses to

achieve separate entities. Seasonal population trends were evaluated, along with

temperature and rainfall. Trends of mosquito populations were monitored using the

Mosquito Magnet Pro (MMPro) (American Biophysics, Corp., North Kingston, RI),









trapping systems from September 2003 through September 2004. Peaks in mosquito

populations correlated with changes in temperature and rainfall, the highest occurring

from August to October. Mosquito location profile through trapping was evaluated using

three CDC 1012 traps baited with CO2. There were significantly (P < 0.05) higher

mosquitoes captured in the trap closest to a body of water, when compared to traps in

open pasture.

Studies evaluated competitive trials with traps and a horse acting as a natural host.

The CDC 1012 trapping study was conducted using a CO2 baited CDC 1012 trap (John

W. Hock Company, Gainesville, FL). The MMPro trapping study was conducted using a

C02, heat and moisture baited MMPro trap. The horse odor study was conducted using a

CO2 and equine odor baited modified CDC 1012 trap. A trap was set up next to a stall

for five 20-minute intervals starting approximately 30 minutes after sunset. There were

two treatment groups, horse present and no horse present for studies CDC and MMPro;

and one additional treatment group horse being vacuumed for odor study. When a horse

was placed in the stall, the mosquitoes trapped significantly (P < 0.05) decreased. The

quantity of mosquitoes feeding on two horses at a given time period of time was

examined using a portable vacuum aspirator (DC Insect Vac. BioQuip, Rancho

Dominguez, CA). There was a significant difference (P < 0.05) in mosquitoes captured

between horses and seasons. The distance required for mosquitoes to distinguish two

horses as separate entities was evaluated using a vacuum aspirator. The horses were

separated by predetermined distances and vacuumed for 20-minute intervals. There was

no significant difference of the mean total mosquitoes vacuumed between horses per

distance.














CHAPTER 1
LITERATURE REVIEW

Mosquitoes have been around since the beginning of time and have survived years

of changing conditions that humans could never withstand. Mosquito control is a major

entomological concern worldwide, so why is it necessary to control mosquitoes? The

two main reasons are, to prevent or reduce nuisance biting and preclude the spread of

mosquito-borne diseases (Dwinell et al., 1998). The number of people worldwide

affected by these diseases is staggering. Each year 300 to 500 million cases of malaria are

reported, resulting in 1.5 to 2.7 million deaths (Centers for Disease Control and

Prevention, 2004). Malaria is not common in the United States, but recently the

introduction and spread of West Nile Virus (WNV) has become a major concern. In 2003,

9,862 human cases and 5,181 equine cases were reported (Stark and Kazanis,

2003). The most serious consequence of WNV infection is fatal encephalitis in humans

and horses, and mortality in certain domestic and wild birds. The state of Florida has an

active equine industry, where some of the top thoroughbreds in the nation are bred and

trained. Since the mortality rate of horses infected with WNV is around 30%, the threat

of this arbovirus is tremendous (Porter et al., 2003). Mosquitoes also pose a nuisance

factor to equine and livestock industries, causing reduced feed conversion efficiency,

weight gain reductions, and decreased milk yield (Steelman, 1979, and Byford et al.,

1992). Florida has been working since the early 1900s to control the threat of mosquito-

borne diseases through surveillance and chemical spraying. These methods are still used

today, but chemical control may be greatly restricted in the future due to environmental









effects, insecticide resistance and health concerns (Kline and Mann, 1998). New trapping

innovations have given mosquito control agencies and backyard enthusiasts a way to

safely and effectively control the nuisance mosquitoes and possibly provide an accurate

picture of local populations. Mosquito traps are becoming a popular merchandizing

commodity, but their effectiveness could still be greatly enhanced in the future. Knowing

the biology and natural behavior of the mosquito is essential to effectively controlling

this pest. Extensive research involving humans and livestock animals concerning

mosquito attraction has been conducted, but little has been done with horses. In Florida,

the equine industry is a multi-billion dollar business, and mosquito control is essential.

Evaluating and understanding natural behavior of mosquitoes as they interact with horses

could provide valuable information about trapping and control. This chapter is a review

of selected literature on the taxonomy, morphology, behavior, ecology, seasonality,

public health importance, and potential control of mosquitoes found in Gainesville,

Florida.

Taxonomy

"Mosquito" is a Spanish word meaning little fly, and its use dates back to about

1583; in England they are known as gnats (Spielman and D'Antonio, 2001). Mosquitoes

are insects that belong to the order Diptera and family Culicidae. Culicidae consists of

about 3200 recognized species. Currently Culicidae are classified into three subfamilies:

Anophelinae, Culicinae, and Toxorhynchitinae (Table 1-1). There are slight taxonomic

differences between the three subfamilies, mostly during their larval stages. Anophelinae

is the most distinct group compared to the other two subfamilies. Their eggs have floats,

larvae lack air tubules, and adults have characteristic palps that are the same length as the

proboscis. Members belonging to the subfamilies Culicinae and Toxorhynchitinae,









during their larval stages have, air tubules and the palps of all adult females are

significantly shorter than their proboscis. Toxorhynchitinaes larvae are all predaceous

and adults are quite large in size. They also have a characteristic proboscis, which is

curved and has been adapted for feeding only on nectar (Woodbridge and Walker, 2002).

There are 38 genera of mosquitoes worldwide, 13 of these encompassing 77 species are

found in Florida (Table 1-2): Anopheles (Meigen), Aedes (Meigen), Ochlerotatus (Lynch

Arribalzaga), Psorophora (Robineau-Desvoidy), Culex (Linnaeus), Deinocerites

(Theobald), Culiseta (Felt), Coquillettidia (Dyar), Mansonia (Blanchard),

Orthopodomyia (Theobald), Wyeomyia (Theobald), Uranotaenia (Lynch Arribalzaga)

and Toxorhynchites (Theobald) (Public Health Entomology Research and Education

Center, 2002).

Morphology and Life Cycle

The mosquito goes through four separate and distinct stages in its life cycle: egg,

larva, pupa, and adult. The eggs of most mosquitoes are found in various shapes

including elongate, ovoid, spindle, spherical and rhomboid. Eggs of Anopheles,

Toxorhynchites, Wyeomyia, Aedes, Ochlerotatus, Psorophora, and Haemagogus species

are laid individually, whereas in Culex, Culiseta, Coquillettidia, and Mansonia species,

they are attached together in a single clump, forming a floating egg raft or a submerged

cluster (Woodbridge and Walker, 2002). Approximately 2 to 3 days after a female has

taken a blood meal an average of 75 eggs per ovary develop. Culex, Culiseta, and

Anopheles lay their eggs on the water surface while many Aedes and Ochlerotatus lay

their eggs on damp soil that will be flooded by water. Most eggs hatch into larvae within

48 hours; others might withstand subzero winters before hatching (Harwood and James,

1979).









Mosquito larvae have three distinct body regions: head, thorax and abdomen. The

head is usually broad and flattened with lateral antennae. Mouthparts usually consist of

brushes and grinding structures that filter bacteria and microscopic plants, however some

larvae are predaceous and will grasp their prey. The thorax is broader than the head and

somewhat flattened. The structure and number of hairs on both the head and thorax aid

in identification of species. The abdomen is elongated and cylindrical, consisting of nine

well defined segments. The first seven segments are similar to each other, but the last

two are modified with specific structures. The eighth segment contains the respiratory

opening, and in most species this is a siphon. The ninth segment is the anal segment.

During larval development mosquitoes pass through four instars, and at the end of each

one they molt and increase in size. The average size of a fourth-instar larva is 6.35 to

12.7 millimeters in length. Depending on temperature and other environmental factors,

mosquito species require about 7 days to complete larval development. At the end of the

fourth instar, larvae molt again and become pupae (Ogg, 2002).

The pupal stage of development prepares the juvenile mosquito to become an

adult. The pupa is shaped like a comma and has hard scales made of chitin that protect

the body. Although the pupae are non-feeding, they are mobile and often called

"tumblers." When pupae are disturbed, they will move in ajerking, tumbling motion

toward protection and then float back to the surface. The pupae are less dense than

water, so they float on the surface and receive oxygen through two breathing tubes called

trumpets (Woodbridge and Walker, 2002). If the ninth segment of the pupa's abdomen is

examined along with the overall size of the pupa, the sex can be determined. The ninth

segment on male mosquitoes is more prominent during this stage, while the female pupa









is usually larger in size than that of the male. In Florida, larvae can pupate in water

temperatures of 170C to 350C for a total period of 1 to 4 days; however, temperatures

above or below these cause increased mortality in the population (Nayar, 1968). The

metamorphosis of the mosquito into an adult is completed within the pupal case. The

adult mosquito splits the pupal case and emerges to the surface of the water where it rests

until its body dries and hardens (Floore, 2003).

Bodies of adult mosquitoes are slender, with thin narrow legs, and elongated

wings. The body surface is covered with scales and setae providing characteristic

markings and colors for identification. The long and filamentous antennae arise between

the eyes, and are usually sexually dimorphic. The prominent proboscis of the adult

female projects anteriorly at least two-thirds the length of the abdomen (Woodbridge and

Walker, 2002). Only female mosquitoes feed on blood, which is essential for egg

production. Females feed on animals-warm or cold blooded-and birds. Stimuli that

influence biting (blood feeding) include a combination of carbon dioxide, temperature,

moisture, smell, color and movement (Floore, 2003). Male mosquitoes do not bite, but

feed on the nectar of flowers or other suitable sugar sources. Females also feed on nectar

for flight energy. During the summer, adult mosquitoes have a life span of a few weeks.

However, it has been found that some species can spend the winter as adults, and can

therefore have a life span of several months (Nasci et al., 2001).

Behavior and Ecology

Once the mosquito has emerged from its pupal case, it will seek shelter where it

will rest and await activity periods. Every mosquito species has a characteristic pattern

of daily activity, which is intrinsically known through the natural circadian rhythm of the

daily light-dark cycle. Generally a mosquito will take flight during one or two periods









each day, depending on whether the specific species of mosquito is characterized as being

diurnal, nocturnal or crepuscular. During these periods, both male and female

mosquitoes will take flight without external cues (Woodbridge and Walker, 2002).

Generally mosquitoes do not actively fly over ranges greater than 2 kilometers. Yet some

species of mosquitoes, such as the salt-marsh mosquito, Ochlerotatus taeniorhynchus

(Wiedemann), will travel long distances by wind and will be carried hundreds of

kilometers from their origins. Ochlerotatus taeniorhynchus emerge in remote salt-marsh

locations where hosts are sparse, so they will make extended round trip migrations to

complete their life cycle while ovipositing at their original breeding sites (Woodbridge

and Walker, 2002).

Flight Categories

Mosquito flights can be classified into three categories: migratory, appetential or

consumatory (Bidlingmayer, 1985). A migratory flight lacks an objective, does not meet

any individual need (Provost, 1952), and is only a one-way flight with no return.

Generally only newly emerged mosquitoes venture out with a migratory flight. The

direction of migration is dependant on wind conditions at the time of departure and

duration is limited by the mosquito's energy bank reserves and the meteorological

conditions during flight; the destination is accidental (Bidlingmayer, 1985). An adult

mosquito will respond to a physiological stimulus by taking an appetential flight. When

a resting mosquito is in need of a blood meal, an oviposition site, or a better resting place,

it will begin a searching flight for this need. During an appetential flight, the appropriate

sense organs (olfactory, visual, thermal, auditory, or humidity receptors) will be alert for

cues that indicate the presence of the objective, and this flight will end when the objective

is located, or continue until energy reserves are depleted (Bidlingmayer, 1985). Once the









objective is located the next flight category, the consumatory flight, begins. This flight is

direct and brief, since visual and biochemical cues operate only over distances short

(Bidlingmayer, 1985). If the cue encountered was olfactory, a direct upwind flight is

conducted until other cues, visual perception, movement or thermal, enable the female to

locate her goal more precisely (Gillies and Wilkes, 1972).

Meteorological Conditions Affecting Flight

Meteorological conditions greatly influence mosquito flight; the most influential

factors are light, temperature, humidity and wind (Day and Curtis, 1989). Dry windy

conditions can completely inhibit mosquito flight, especially during the winter months in

Florida (Day and Curtis, 1989). Nightly variations in wind, rainfall, and relative

humidity influence mosquito flight patterns and possibly feeding success. During late

summer and fall, daily rainfall patterns can potentially influence whether the mosquito

population continues to build, remains constant, or declines by affecting feeding and

oviposition behaviors (Day and Curtis, 1989). In Florida, research conducted by the

Middlesex County Mosquito Extermination Commission indicated that most mosquito

species possessed a bimodal flight activity pattern during the night, with the larger peak

occurring soon after sunset and the smaller peak just prior to dawn (Schmidt, 2003).

These are times of rapidly changing light levels. Temperature and relative humidity

greatly influence the flight behavior and activity of mosquitoes, but optimal conditions

vary between species. Many researchers disagree considerably as to what are optimal

conditions for flight (Rowley and Graham, 1968). In Florida, Bradley and McNeel

(1935) and Bidlingmayer (1974) found that temperatures below 210C and 190C

respectively, reduced trap catches. After temperatures have risen above the minimum









flight threshold for individual mosquito species, higher temperatures do not affect flight

(Taylor, 1963). Optimal relative humidity for flight is also greatly disagreed upon.

Rowley and Graham (1968) found that relative humidity between 30 and 90% had no

demonstrable influence on flight performance. Mosquito cruising speeds are generally

less than 1 meter per second but flight activity is greatly reduced if winds speeds exceed

flight speeds (Grimstad and DeFoliart, 1975).

Mosquito activity can be forecasted using the four meteorological factors listed

above. American Biophysics Corporation, manufacturer of the Mosquito Magnet, has

teamed with The Weather Channel, which displays on its website, www.weather.com, the

first-ever "Mosquito Activity Forecast." The "Mosquito Activity Forecast," developed

and managed by a team of meteorologists from The Weather Channel, provides hourly

predictions of mosquito activity nationwide. This information is very useful to people

who want to take part in outdoor activities in areas prone to high incidence of mosquito

vector-borne disease.

Feeding Behavior

Adult mosquitoes of both sexes of most species feed regularly on plant sugar

(nectar) throughout their lives, but only females feed on hosts for a blood meal. Females

of some mosquito species feed on sugar infrequently or never [e.g. Ae. aegypti (Linnaeus)

and An. gambiae (Meigen)], and utilize blood for both energy and reproduction. Females

of some mosquito species feed entirely on plant sugar, such as Toxorhynchites, and do

not require a blood meal for egg development. Generally, however, a blood meal is

required by female mosquitoes to obtain protein from the blood to develop eggs

(Woodbridge and Walker, 2002).









Host Location Behavior

Generally 1-3 days after the emergence of the female mosquito, she will look for a

host from which to feed. Research has long been conducted to determine why

mosquitoes are attracted to certain hosts and what attractants are responsible for the

mosquito's odor mediated behavior. In 1942, researchers showed that unwashed naked

children were more attractive to species of the genus Anopheles than naked children who

had washed. The same group also showed that dirty clothes in a hut attracted more

mosquitoes than an empty hut (Haddow, 1942). Individual variation in attractiveness to

mosquitoes was shown in 1965, when Khan and his associates were able to isolate one

person who was very attractive and 3 people who were very unattractive to Ae. aegypti

(Khan et al., 1965). Mosquitoes seeking a host are exposed to a wide variety of visual,

olfactory, and physical stimuli. Any one or combination of these stimuli could act as

cues for host identification and location. Host selection is mainly determined by host

preference and availability, but stimuli that the mosquito detect help locate the host.

Some of the best-documented olfactory attractants are carbon dioxide (C02), lactic

acid, and octenol. Generally, carbon dioxide is universally attractive to mosquitoes, and

is probably the best understood of the volatile host cues (Gibson and Torr, 1999). Some

researchers believe that carbon dioxide acts as an attractant, which mediates orientation

towards a host, and can exhibit a synergistic response with other host odors (Gillies,

1980). Gillies (1980) also found that the carbon dioxide and whole-body odors have an

orientating effect of variable extent when presented singly and a greater enhanced effect

when presented together. It has been shown that mosquito light traps baited with CO2

capture 8-30 times more mosquitoes than traps without CO2 (Kline and Mann, 1998).

Although it has been concluded that CO2 increases mosquito catches in traps, it was









found that CO2 appears to be of little importance in host discrimination by mosquitoes

(Mboera and Takken, 1997). It is believed that one of the volatile compounds

mosquitoes use to discriminate hosts is lactic acid. Lactic acid is a by-product of

anaerobic metabolism common to all animals and humans. Skin emanations are

important, because odors from live hosts are always more attractive than any combination

of these chemicals in a warm, humid airstream (Woodbridge and Walker, 2002).

1-octen-3-ol (octenol) is another olfactory attractant documented as an effective mosquito

attractant. In 1984 Hall et al., through studying the attractiveness of oxen to Tsetse flies,

discovered octenol and isolated it the breath of oxen in Africa. Field tests have

demonstrated that octenol serves as a powerful attractant for certain species of

mosquitoes and flies (Kline, 1994). Currently manufacturers of commercial mosquito

traps, such as American Biophysics Corporation, makers of the Mosquito Magnet Pro,

suggest the use of octenol as a supplementary additional bait to trap mosquitoes.

Visual attraction of hosts to mosquitoes has been thoroughly documented. Adult

mosquitoes possess two compound eyes and two ocelli. Compound eyes are used for

navigation and sensing movement, patterns, contrast, and color, while ocelli are believed

to sense light levels and possibly polarized light (Allan et al., 1987). The compound eyes

of adult mosquitoes have relatively poor resolution but overall high light sensitivity (Muir

et al., 1992). It has also been reported that diurnal species respond to visual

characteristics of hosts, such as color, brightness, pattern, and movement (Allan et al.,

1987). Field trials conducted to investigate color affinity ofMansonia mosquitoes

showed that Mansonia have marked attractiveness towards blue and red followed by

white, yellow and green, with the least numbers attracted by black (Bhuyan and Das,









1985). Movement may also play a role in host location by mosquitoes, and a consistently

small but positive attraction has been affirmed (Wood and Wright, 1968). Once the

female is within 1 meter of a host, convective heat and humidity become the main

attractant opposed to chemical or visual stimuli (Woodbridge and Walker, 2002).

Mosquitoes are also attracted by physical stimuli like temperature and humidity. A

source of heat has shown positive attractiveness from some species of mosquitoes

(Howlett, 1910). Brown (1951) showed that mosquitoes landed three times as often on a

clothed robot when the robot's "skin" temperature was 980F than when the "skin"

temperatures were 50-650F (Brown, 1951). He also noted that moisture coming off the

robots' clothing increased the number of landings by 2 to 4 times, but only at

temperatures above 600F. It has not been determined what single stimulus causes

mosquitoes to locate and feed on a host, but it has been determined that a combination of

visual, olfactory and physical stimuli are effective as attractive factors.

Host Preference

Host preference varies widely between different genera of mosquitoes, changes

within a genus depending on geographic location. Some species feed almost entirely on

members of one genus of host animal; others opportunistically attack members of two or

three vertebrate classes. Mosquitoes belonging to the genus Culex are primarily avian

feeders, but if the population of birds is insufficient or unavailable, Culex will

contentedly feed on mammals (Braverman et al., 1991). Some Florida mosquitoes prefer

to feed on mammals, including Aedes, Anopheles, Coquillettidia, Mansonia and

Psorophora. However, the genus of animal differs between mosquito genera, and if given

the opportunity, they will also take a blood meal from a bird (Edman, 1971). Livestock









are exposed to mosquitoes in very high numbers during most of the year in Florida;

because some of these species are known disease vectors, this causes concern. When

trying to understand and control the transmission of vector-borne diseases, knowledge of

the feeding behavior of mosquitoes is of prime importance (Defoliart et al., 1987).

Reproduction Behavior

Mating usually occurs a few days after adult emergence. The males generally form

flight swarms around the female's preferred host. When a female enters a swarm, males

detect the characteristic frequency of her wing beat and position with their plumose

antennae and Johnston's organs. The male locates the female, pursues her, and mates

with her. If the female is of another species, males either do not respond to her flight

tone or release her upon detection that she lacks the appropriate species-specific contact

pheromones. There are some exceptions to this mating ritual. Male Deinocerites guard

pupae at the water surface and mate with the females as they emerge (Woodbridge and

Walker, 2002).

Seasonality

In Gainesville, Florida, mosquitoes are present twelve months a year (Crowley,

2003), yet species are seasonally dependant. Identification of population trends of

mosquitoes is essential for developing appropriate control methods and disease

prevention. In Florida, horse owners can base mosquito-borne disease vaccinations,

such as the vaccine for West Nile Virus, depending on seasonality trends of Culex

nigripalpus. For example, Fort Dodge Animal Health suggests vaccinating horses

against West Nile Virus in March and August. Mosquito population variations are

closely linked to rainfall and temperature. In Gainesville, the highest numbers of

mosquitoes are generally trapped in August, and the lowest are generally in January









(Crowley, 2003). Population peaks usually occur 2-3 weeks following a heavy rainfall

(Crowley, 2003). Mosquito species seasonality even varies within the same genus, for

instance, Culex nigripalpus (Theobald) is generally abundant during the late summer and

early winter seasons while Culex salinarius (Coquillett) is abundant during the late winter

until early summer seasons (Zyzak et al., 2002). Culex nigripalpus is a species that

thrives under the hot, humid conditions commonly reported in Florida, while Culex

salinarius is a species that is most abundant in Florida during cool, dry months. Two

species ofMansonia are commonly trapped in Gainesville, including Ma. dyari (Belkin,

Heinemann and Page) and Ma. titillans (Walker), but their seasonality trends differ

slightly. Ma. dyari are generally abundant mid-summer until late-fall, while Ma. titillans

are abundant late summer until early spring. Coquillettidia perturbans (Walker) is found

most of the year between early spring until late fall (Slaff and Haefner, 1985).

Health and Economic Impact

Mosquitoes are vectors for many diseases that cause millions of human and animal

death every year. Mosquitoes are the sole vectors of pathogenic organisms causing

human malaria, yellow fever, and dengue fever. They are also of prime importance in the

transmission of diseases like filariasis and viral encephalitides of man. Every year there

are between 300-500 million cases of humans infected with malaria, resulting in 1.5 to

2.7 million deaths. Patients with malaria typically are very sick with high fevers, shaking

chills, and flu-like illness. Even though malaria was eradicated from the United States in

the 1950s there are approximately 1,200 cases of malaria diagnosed in the United States

each year. Most of these cases are from people that traveled and returned from malaria-

risk areas, such as Africa. The mosquito that transmits malaria, Anopheles, is found

throughout much of the United States. This causes concern because it is possible that a









person who entered the United States from Africa, carrying the plasmodia responsible for

Malaria, could infect native Anopheles quadrimaculatus and this mosquito could transmit

the disease in the United States (Centers for Disease Control and Prevention, 2004).

Dengue fever and filariasis are also very dangerous diseases that occur in the

tropics. Most dengue infections result in relatively mild illness, but some can progress to

dengue hemorrhagic fever. With dengue hemorrhagic fever, the blood vessels start to leak

and cause bleeding from the nose, mouth, and gums. Bruising can be a sign of bleeding

inside the body. Without prompt treatment, the blood vessels can collapse, causing shock

(dengue shock syndrome). Dengue hemorrhagic fever is fatal in about 5% of cases,

mostly among children and young adults. Primary vectors of dengue virus are Aedes

albopictus (Skuse) and Aedes aegypti. Yearly cases are found as close to the United

States as Mexico, therefore this virus is a threat to the US, because we have a high

population of competent Aedes mosquitoes (Centers for Disease Control and Prevention,

2003a). Lymphatic filariasis, the second leading cause of permanent and long-term

disability worldwide is caused by parasitic nematodes, which affects over 120 million

people worldwide (Centers for Disease Control and Prevention, 2003B). Culex and

Mansonia mosquitoes are the primary vectors of filariasis, which transmit the filariasis

parasitic nematode, Wuchereria bancrofti and Brugia malayi. Lymphatic filariasis is

rarely fatal, but it can cause recurring infections, fevers, severe inflammation of the

lymph system, and a lung condition called tropical pulmonary eosinophilia. In about 5%

of infected persons, a condition called elephantiasis causes the legs to become grossly

swollen. This can lead to severe disfigurement, decreased mobility, and long-term

disability (Centers for Disease Control and Prevention, 2003b).









Even though malaria, dengue fever and filariasis are not common in the United

States, all of the primary mosquito species are present. The mosquitoes that are

competent to transmit those diseases are Aedes, Anopheles, Mansonia, and Culex. The

composition of mosquito species that live in the United States are potential vectors of

many diseases, therefore we are not completely safe from the diseases that are

concentrated in the tropics. The United States is not completely free from mosquito-

borne diseases. Arboviral encephalitides cause deaths not only in humans, but also birds,

livestock and horses. Arboviral encephalitides have a global distribution, but there are

five main virus agents of encephalitis in the United States: eastern equine encephalitis

(EEE), western equine encephalitis (WEE), St. Louis encephalitis (SLE), La Crosse

encephalitis (LAC), and the newly introduced West Nile virus (WNV), all of which are

transmitted by mosquitoes (Centers for Disease Control, 2001).

Arboviral viruses, short for arthropod-bome viruses, are maintained in nature

through biological transmission between susceptible vertebrate hosts by mosquitoes.

Arboviruses that cause human encephalitis are members of three virus families: the

Togaviridae, Flaviviridae, and Bunyaviridae. The majority of human infections are

asymptomatic or may result in flu-like symptoms. In some cases infection leads to

encephalitis, swelling of the brain, with a fatal outcome or permanent brain damage.

Eastern equine encephalitis (EEE) has been affecting North American horses since

1830s (Mitchell et al., 1985). At that time viruses were unknown until the mid-20th

century, the cause of this disease was unknown. The disease was not officially named

until a major outbreak occurred in horses in coastal areas of Delaware, Maryland, New

Jersey, and Virginia in 1933 (TenBroeck et al. 1935). Outbreaks of EEE are infrequent,









animals are naive to the disease and are susceptible, which causes a significant economic

and social impact once an outbreak has occurred. The first time EEE occurs in an area,

there are illness and death of horses, poultry, and humans (Centers for Disease Control

and Prevention, 2001). Since 1934 there have been 153 confirmed cases of EEE in

humans in the United States (Centers for Disease Control and Prevention, 2001.

Approximately one-third of all people with clinical encephalitis caused by EEE will die

from the disease, and those that recover may suffer from permanent brain damage. There

are many known mosquito vectors of EEE in the United States, including Culiseta

melanura (Coqullett), Aedes vexans (Meigen), Ochlerotatus sollicitans (Walkeri),

Coquilletidia perturbans; there are also potential vectors of EEE, such as Aedes

albopicus, Aedes canadensis (Meigen), Culex nigripalpus, and Culex salinarius

(Woodbridge and Walker, 2002).

Western Equine Encephalitis (WEE) was first isolated in California in 1930 from

the brain of an infected horse. Today WEE is still an important cause of encephalitis in

horses and humans, mainly in the western United States. The symptoms are similar to

EEE, and the mortality rate of WEE is approximately 3% (Centers for Disease Control

and Prevention, 2001). The major WEE vector in the United States, Culex tarsalis

(Coquillett), but Aedes melanimon and Culiseta inornata (Williston) have been identified

as potential vectors (Woodbridge and Walker, 2002).

St. Louis Encephalitis (SLE) is the most common mosquito-transmitted human

pathogen in the United States. It is widely distributed in the lower 48 states. Since 1964,

there have been 4,437 confirmed cases of SLE with an average of 193 cases per year.

The symptoms are similar to EEE, and the mortality rate of SLE is approximately 5-15%









(Centers for Disease Control and Prevention, 2001). Culex is the main vector for SLE in

the United States, however some other confirmed vectors are Cx. tarsalis, Cx. pipiens,

Cx. quinquefasciatus (Say), and C.x nigripalpus. The remainder of Culex species, Cx.

restuans, Cx. salinarius are potential vectors (Woodbridge and Walker, 2002).

La Crosse Encephalitis (LAC) was discovered in Wisconsin in the 1960s. Since

then LAC has been identified in 11 states, mostly in the Midwestern region, but also in

Virginia, North Carolina, Alabama, and Mississippi. An average of 75 cases are reported

in the United States every year. The symptoms are similar to EEE, and the mortality rate

of SLE is approximately less than 1% (Centers for Disease Control and Prevention,

2001).

Ochlerotatus triseriatus is the primary vector of LAC in the United States, but

several species of mosquitoes have been identified as potential vectors: Ae. canadensis,

Ae. vexans, Oc. sollicitans (Walkeri), Oc. taeniorhynchus, and Ae. albopictus

(Woodbridge and Walker, 2002).

Since the early 1950s, West Nile Virus (WNV) has affected people in African,

Middle Eastern and some Mediterranean countries, usually causing epidemics every

10 years (Malkinson et al., 2002). This arbovirus had never been reported in North America

until 1999, when it was isolated from birds found in New York City (NYC). There was a

high degree of similarity among the various strains circulating throughout NYC and

surrounding counties, which indicated that a single WNV strain (WN-NY99) had been

introduced (Lanciotti, 1999). When WN-NY99 was isolated and compared with various

other West Nile Virus strains, a significant similarity was found between it and the WNV

strain WN-Israel 1998 isolated in Israel in 1998 (Lanciotti, 1999). The virus rapidly









spread throughout the United States in less than 2 years, producing a mortality rate of

30% in infected horses. This is a significant threat to the bustling equine industry in the

state of Florida, where many of the top horses are bred and trained. Mosquitoes that

belong to the genus Culex are the main vectors of WNV, the species include Cx. pipiens,

Cx. salinarius, and Cx. tarsalis. Several genera have been identified as potential vectors

of WNV: Aedes, Anopheles, Coquillettidia, Culiseta, Culex, Deinocerites, Ochlerotatus,

Orthopodomyia, Psorophora, and Uranotaenia (Woodbridge and Walker, 2002).

Control

To control mosquitoes and the public health hazards they present, many states and

localities have established mosquito control programs. These programs consist of three

main categories, chemical, biological, and physical control along with surveillance

methods to monitor mosquito populations. With more than 1,197 miles of coastline, a

warm subtropical climate, and heavy rainfall, Florida produces an unusually rich fauna,

including 77 species of mosquitoes. The Florida Anti-Mosquito Association (FAMA)

now known as the Florida Mosquito Control Association (FMCA), was formed in 1922.

Shortly thereafter the legislature created the first mosquito control district (Indian River,

1925). Since then Florida has had an active mosquito control program focusing on

surveillance and population control.

Chemical and Biological Mosquito Control

Treatment of adult mosquitoes is the most visible practice exercised by mosquito

control operations. Every mosquito control district in Florida uses adulticides, applied

either aerially or with ground equipment. Adulticiding is usually the least efficient

mosquito control technique. Some of the approved adulticides include, malathion,

permethrin, resmethrin, fenthion, naled, and chlorpyrifos (Dwinell et al., 1998). Some









Florida mosquitoes are resistant or more tolerant to some adulticides, thus affecting the

adulticide selection. All insecticide applications must be made during periods of adult

mosquito activity, this also decreases the chance of affecting non-target insects such as

bees and butterflies. Larvicides kill mosquitoes by applying natural agents or

commercial products designed to control larvae and pupae present in aquatic habitats

(Dwinell et al., 1998). Larviciding is generally more effective and target specific than

adulticiding. The use of adulticides and larvicides may be restricted in the future

because of growing concern about potential negative effects on human health and

environmental hazards.

Biological control should only be used to augment other mosquito control measures

as part of an Integrated Pest Management program (IPM). In most cases, mosquito

biological control targets the larval stage. Some examples of biological control affecting

mosquito larvae are fish that prefer to feed on the larvae. One such fish is Gambusia

affinis, also known as the mosquito fish. Studies were conducted in New Jersey

Mosquito Control Agency (NJMCA) looking at the effectiveness of mosquito fish as

predators towards mosquito larvae. The NJMCA Research and Development Committee

concentrated on investigating the effectiveness of Gambusia in the following areas:

woodland pools, mine pits, stormwater management facilities, ornamental pools,

abandoned swimming pools, ditches, brackish marshes, and freshwater swamps. Results

showed very effective mosquito control, as long as adequate fish reproduction occurred

and accessibility to larvae by fish was assured (Duryea et al., 1996). There are a couple

of other biological control agents that are occasionally used, such as the predaceous

mosquito Toxorhynchites, which feeds on larvae in its larval stage, and does not feed on









blood as an adult. Also the parasitic nematodes, such as Romanomermis, the fungus

Lagenidium giganteum, and the bacteria Bacillus sphaesicus have been used for

biological control. Biological control certainly holds the possibility of becoming a more

important tool and playing a larger role in mosquito control in the future (Dwinell et al.,

1998).

Personal protection and education are very important aspects of mosquito control.

It may not be mosquito control in a population sense, but it is personal mosquito control

and helps protect from nuisance bites and potential disease infection. The most popular

insect repellent is called DEET (N,N-diethyl-metatoluamide). DEET is one of the few

repellents that can be applied to human skin or clothing. DEET does not kill insects, but

repels them from treated areas. Scientists are not completely sure how it repels insects,

yet it is believed that DEET affects the insect's ability to locate animals to feed on by

disturbing the function of receptors in the mosquito antennae that sense host location

chemicals (Rose, 2001). Education is key to protection against mosquitoes. Such tips as

avoiding outside activities during peak mosquito times, or if outside activities are

necessary; wear proper clothing and insect repellent. Reducing areas of mosquito

breeding, such as tires and water filled containers will decrease mosquito populations.

Proper education of horse-owners will help protect horses from the nuisance biting and

potential disease infection. Keep horses inside under fans during mosquito peak times,

vaccinate horses against mosquito borne diseases prevalent in the area, and use mosquito

spray regularly.

Nonchemical Mosquito Control and Surveillance

Mosquito traps are very effective as a surveillance tool to monitor seasonal

prevalence and the species composition in a specific area. Traps are also effective in









reducing numbers of mosquitoes in the location of the trap. Over the years there have

been lots of different kinds of mosquito traps developed, varying greatly in effectiveness

and usefulness. The most effective traps use a combination of factors that attract

mosquitoes such as light, heat, moisture, carbon dioxide and synthetic chemicals for host

attraction, such as octenol.

In 1962, the Center for Disease Control miniature light trap was introduced

specifically for arbovirus surveillance and other short-term mosquito investigations

(Stamm et al., 1962). The CDC trap model 1012 (CDC 1012)(John W. Hock Company,

Gainesville, FL) (Figure 1-1) attracts mosquitoes with a white light and carbon dioxide

and captures them with the down draft produced by a fan. The CDC 1012 trap is a

lightweight trap, weighing less than 1 kilogram, it is powered by a 6-volt battery

(McNelly, 1989). This trap is useful for surveillance studies because it is versatile and

catches the mosquitoes alive and unharmed. The main disadvantage to the CDC 1012

trap is that the white light attracts other insects, such as beetles, moths and other

nontarget species, which are drawn into the capture net (Kline, 1999).

Recently new traps have been developed that use the innovation of carbon dioxide

plumes and counterflow geometry together. These traps have been found to increase

mosquito catches, and reduce capture of nontarget species. The counterflow geometry is

achieved with two fans operating simultaneously, the first fan provides an outflow

attractant such as carbon dioxide and the second fan has an inflow current to capture

mosquitoes. Around the attractant plume any insect with a flight speed less than

approximately 3.5 m/sec will be captured (Kline, 1999). The first mosquito trap to utilize

this technology is the Mosquito Magnet X (MM-X) (American Biophysics Corporation,









North Kingston, RI). The MM-X trap (Figure 1-2) is constructed from a modified plastic

pretzel jar and is operated by a 12-volt battery. The attractant is carbon dioxide, which is

supplied by compressed gas cylinder at a flow rate of 500 ml/minute. Octenol, which can

also be used as additional attractant, is supplied by American Biophysics Corporation

OCTI slow release packets, which have a release rate of 0.5 mg/hour (Kline, 1999). This

trap is very effective in attracting a wide variety of mosquito species (Kline, 1999 and

Burkett et al., 2001), but its major disadvantage is the external source of power and

mosquito attractant requirements that must be provided (Kline, 2002).

American Biophysics Corporation has considered the drawbacks of the MM-X trap

and has improved this technology by developing a new category of traps that utilize

propane as an energy source. These traps also use counterflow geometry technology, but

have added catalytic combustion of propane to produce carbon dioxide, heat and water

vapor. A thermoelectric generator uses excess heat from the combustion process to

generate electricity to run the traps' fans (Kline, 2002). Another trap recently developed

is the Mosquito Magnet Pro (MMPro) (American Biophysics Corporation, North

Kingston, RI), which is commercially available to the general public (Figure 1-3). The

use of propane, which is supplied in the same cylinders as used for gas grills, makes it

easier for the general public who are concerned about local mosquito populations. Now

American Biophysics Corporation has six traps currently available for commercial sale,

the Mosquito Magnet Pro, Pro-Plus, Liberty, Liberty-Plus, MM-X and Defender. The

technology behind the traps is similar; the traps differ in acre coverage and power

sources. The Liberty and Defender both use a 12-volt power source and only use propane

for production of carbon dioxide. These traps claim to cover 1 acre and 1/2 acre









respectively. The effectiveness of the traps mentioned above has been researched by

Burkett et al. (2001) and Kline (2002) who compared MMPro, MMX, and the CDC 1012.

Burkett et al. (2001) and Kline (2002) both found that the MMPro was more effective

than CDC 1012 but was outperformed by MMX.

Mosquito larva surveillance is also critical to effective control; it is used to

determine the location, species and population densities of pest and vector mosquitoes. It

is also vital for predicting adult emergence and establishing optimal times for application

of larval control measures (Belkin, 1954). There are two main ways to survey larvae in

the field: by dipping and with emergence traps. The kind of mosquito larvae one is

looking for will determine the sampling technique to be used. Dipping involves

collecting larvae present in water with a dipper. It is an effective tool to monitor larvae

(Waters and Slaff, 1987). Emergence traps are placed on the surface of a body of water

to trap mosquitoes as they emerge as adults and take flight. These traps are usually

pyramidal in shape with netting or screen trimming the walls of the pyramid. There is a

collection cup at the top of the trap, so when the mosquitoes emerge into adults they fly

upward and into the collection cup (Appleton and Sharp, 1985). Emergence traps have

not been found to be effective in mosquito control, but they are very helpful in

surveillance.

Host-baited mosquito traps are a useful tool to evaluate mosquito host preference

and possible host location cues. Entomologists have used host-baited traps since the

early 1900s for collecting Anopheline mosquitoes during malaria investigations (Mitchell

et al., 1985). In Florida, mosquito-borne diseases that affect horses have a significant

economic impact on the equine industry, yet there is limited published information on the









collection of mosquitoes from horses (Fletcher et al., 1988). In 1944 Kumm and Zuniga

used an animal-baited trap to record seasonal variations of Anopheles. Jones et al. (1977)

collected mosquitoes directly from ponies and donkeys as part of an equine encephalitis

survey, the trap used was modified from Magoon's trap used in 1935. Most large animal-

baited traps enclose the animal in a stable or stall with small openings for the mosquitoes

to gain entrance. Recently portable animal-baited traps have been used which utilize

screen walls and roofing to allow host odors to flow away from the trap as an attractant

(Mitchell et al., 1985). In 1984 Hall et al. made one of the most significant discoveries

concerning host odor and location cues of flies, using an animal-baited trap. The study

was conducted in Africa, where they placed an ox into a pit. The odors emanating from

the ox increased the catch of tsetse flies compared to a trap with no odors. Eventually

Hall et al. (1984) discovered that an alcohol compound in the breath of the ox, octenol,

was a potent attractant for the flies. Since then researchers started looking at specific

compounds that elicit a response from mosquitoes, indicating host location behaviors.

Fletcher et al. (1988) used a different approach by constructing an open animal-baited

trap, which left a horse naturally exposed to insects. The sides of the trap would close

around the horse and entrap the insects which could then be identified. Since mosquitoes

have a preference for the type of host they feed on, there must be specific factors that

elicit a host location behavior. Researchers are now looking at how effective artificial

attractants are when put in competition with an animal in a natural state. In 2003

Campbell stated that a large variance was found between mosquito species and numbers

aspirated directly from a horse compared to various mosquito traps. There are

significantly more mosquitoes attracted to the horse than all of the mosquito traps that









were tested. Campbell's (2003) research has lead to more investigations looking at

animal-baited traps and traditional baited mosquito traps.

Summary

The utilization of host odors as attractants for trapping mosquitoes has received

increased interest in the past few years for population management and surveillance

(Kline, 1994). Historically mosquito control has been performed with chemical

insecticides and topical repellents, however, these methods will be greatly restricted in

the future because of potential health and environmental hazards. It has been stated by

Kline (1994) that there is a need for new, safe and effective ways to kill pest/vector

species of mosquitoes, and to deter blood seeking mosquitoes from biting humans and

other hosts. During a research investigation conducted in 2002, an interesting skew was

observed in the mosquito species collected directly from a horse compared to those being

captured by various mosquito traps. The primary species found on the horse was

Mansonia titillans, which comprised 40% of the total catch; while Culex nigripalpus

comprised 85-91% trapped in the Mosquito Magnet Pro (Campbell 2003). In Florida

there are two common species of Mansonia, M. dyari andM. titillans. Until recently

Mansonia spp. has been thought of as only a nuisance mosquito with a minor impact on

livestock production or humans. In the United States it has been found that Mansonia

spp. primarily prefer to feed on mammals (Cupp and Stokes, 1973), and mosquito

infestations can cause a significant weight gain reduction in cattle (Steelman, 1979). In

1999, the Center for Disease Control reported that Mansonia spp. has been found a

positive carrier of West Nile Virus in Ethiopia (Hubalek and Halouzka, 1999). It has also

been reported that Mansonia spp. can be positive carriers of Venezuelan Equine

Encephalitis Virus (Mendez et al., 2001) and Japanese Encephalitis Virus (Arunachalam






26


et al., 2002). There could be a disastrous effect on the United States equine industry if

Mansonia spp. are able to carry and transmit these encephalitis viruses because efforts to

control them through trapping have been largely unsuccessful (Campbell 2003).

The objectives of this study is to determine if traditional mosquito traps using

artificial attractants such as carbon dioxide, heat and moisture are effective when put into

competition with a horse. Is the horse eliciting a response from the mosquitoes that is

much stronger than what the trap is doing?









































Figure 1-1. CDC 1012 (John


ock Company, Gainesville, FL) mosquito trap.









































Figure 1-2. Mosquito Magnet X (American Biophysics Corporation, East Greenwich, RI)
mosquito trap









































Figure 1-3. Mosquito Magnet Pro
RI) mosquito trap.


East Greenwich,












Table 1-1. Classification of family Culicidae
Subfamily Tribe Genera
Anophelinae Anopheles, Bironella, C/ igonia
Culicinae Aedeomyiini Aedeomyia
Aedini Aedes, Ochlerotatus, Verrallina, Ayurakitia,
Armigeres, Eretmapodites, Haemagogus,
Heizmannia, Opifex, Psorophora, Udaya,
Zeugnomyia
Culicini Culex, Deinocerites, Galindomyia
Culisetini Culiseta
Ficalbiini Ficalbia, Mimomyia
Hodgesiini Hodgesia
Mansoniini Coquillettidia, Mansonia
Orthopodomyiini Orthopodomyia
Sabethini Sabethes, Wyeomyia, Phoniomyia, Limatus,
Trichoprosopon, Shannoniana, Runchomyia,
Johnbelkinia, Isostomyia, Tripteroides, Malaya,
Topomyia, Maorigoeldia
Uranotaeniini Uranotaenia
Toxorhynchitinae Toxorhynchites
The classification of all mosquitoes into 3 subfamilies, 10 tribes of Culicinae, and 38
genera is based on Knight and Stone (1977).










Table 1-2. Systematic list of mosquitoes found in Florida
Genus Species


Anopheles


Aedes
Ochlerotatus


Psorophora

Culex


Deinocerites
Culiseta
Coquillettidia
Mansonia
Orthopodomyia
Wyeomyia
Uranotaenia
Toxorhynchites
The classification


atropos, barber, bradleyi, crucians, diluvialis, georgianus, grabhamii,
inundatus, maverlius, perplexens, punctipennis, quadrimaculatus,
smaragdinus, walker, nyssorhynchus, albimanus
cinereus, vexans, cim *,,i. albopictus
atlanticus, Canadensis, mathesoni, dupreei, fulvus pallens, infirmatus,
mitchellae, scapularis, sollicitans, sticticus, taeniorhynchus, thelcter,
thibaulti, tormentor, tortilis, hendersoni, triseriatus, bahamensis
columbiae, discolor, pygmaea, cyanescens, ferox, horrida, johnstonii,
mathesoni, ciliata, howardii
bahamensis, nigripalpus, quinquefasciatus, restuans, salinarius, tarsalis,
atratus, cedecei, erraticus, iolambdis, mulrennani, peccator, pilosus,
biscaynensis, territans
cancer
melanura, inornata
perturbans
dyari, titillans
alba, 'i.'i JO
mitchellii, smithii, vanduzeei
lowii, sapphirina
rutilus rutilus, rutilus septentrionalis
of Florida mosquitoes by Genus species based on Public Health


Entomology Research and Education Center 2002














CHAPTER 2
SEASONALITY OF MOSQUITOES AT AN EQUINE FACILITY IN NORTH
CENTRAL FLORIDA

Introduction

Mosquitoes cause two significant issues in Florida's equine industry including a

nuisance factor and potential for disease transmission. The economic loss of biting

mosquitoes on horses has not been studied, but cattle have been found to have significant

weight gain reduction, milk production reduction and a compromised immune system due

to stress (Steelman, 1979 and Byford et al., 1992). In 1999, West Nile Virus started

spreading throughout the United States and has caused the state of Florida's equine

industry millions of dollars every year in disease prevention, health care and overall

morbidity of infected horses (Porter et al., 2003).

Extensive research has been conducted to develop and refine trapping systems

that increase mosquito collection and enable efficient surveillance of local mosquito

populations. The ability of monitoring seasonality population trends enables efficient

mosquito control and knowledge of when to take protective measures for horses against

mosquitoes. For example, Fort Dodge Animal Health suggests, in Florida, vaccinating

horses against West Nile Virus in March and August. In Gainesville Florida, mosquitoes

are generally present twelve months a year, but the abundance and species composition of

mosquitoes varies depending on the season. Mosquito population variations are closely

linked to rainfall and temperature. The highest numbers of mosquitoes are generally









trapped in August, and the lowest are generally in January. Population peaks usually

occur 2-3 weeks following a heavy rainfall.

The following experiment is an evaluation study of the seasonal population trends

of mosquitoes at an equine facility in North Central Florida, using the Mosquito Magnet

Pro (American Biophysics Corporation, North Kingston, RI). Meteorological conditions,

such as maximum and minimum temperature and rainfall, were collected for comparison

of mosquito population trends.

Materials and Methods

Adult mosquito seasonality studies were conducted in Gainesville, Florida at the

University of Florida Horse Teaching Unit, a 60-acre equine facility, housing

approximately 40 head of quarter horses varying in sex and age. The study began

September 26, 2003 and concluded September 2, 2004. The trapping devices used were

four Mosquito Magnet Pro traps, which were placed in predetermined locations at the

equine facility.

The Mosquito Magnet Pro (MM-Pro) (Figure 1-3) uses patented technology that

catalytically converts propane into carbon dioxide, heat and moisture, which attracts

mosquitoes to the counterflow of a suction fan and released attractants. The mosquitoes

are vacuumed into a collection net where they dehydrate and die. The trap is self-

powering when provided with propane through a thermoelectric module, which is

supplied by a standard 20-pound commercial propane tank. MM-Pro stands 2 meters

high and is built of stainless steel with a PVC shell.

A preexisting weather station was utilized to measure maximum and minimum

temperature (C) and rainfall (cm) during the seasonality study (Figure 2-1). The station









is positioned on a wooden post measuring 2.4 meters tall, located in a predetermined

position. A rain gauge is mounted at the top of the station and a waterproof thermometer

is located below the rain gauge.

Experimental Design

The mosquito seasonality study was conducted from September 26, 2003 until

September 2, 2004. The four MM-Pros were placed in the same location for the duration

of the study (Figure 2-2). The MM-Pro is powered by propane and the 9-kilogram tank

was replaced approximately every three weeks. The nylon collection nets were collected

twice a week on Tuesday and Friday mornings around 10am. Mosquitoes were stored at

00C until identification and counting was conducted. When the nylon mosquito

collection nets were retrieved on Tuesdays and Fridays, maximum and minimum

temperature and total rainfall were recorded from the weather station.

Data Analysis


Count data from the studies described above were analyzed with the General

Linear Model (GLM) and Proc Means Programs; Tukey's Studentized Range Test was

used for separation of means (SAS Institute, Version 8.2, Copyright 1999-2001). The

significance level was set at P < 0.05.

Results

Total mosquitoes captured using the Mosquito Magnet Pro plotted against months

and as related to rain measured in centimeters can be seen in Figures 2-3 and 2-4

respectively. Maximum and Minimum temperatures were recorded at the University of

Florida Horse Teaching Unit and are plotted against time during the seasonality study

(Figure 2-5). Temperature ranges were similar to previous years; there were no notable









differences in maximum or minimum temperatures. Rain measured at the UF Horse

Teaching Unit is shown for years 2001-2004 in Figure 2-6. A total of 42,077 mosquitoes

were trapped during this study (Table 2-1). The highest populations of mosquitoes were

trapped during the months of August, September, and October (Table 2-2). The lowest

populations of mosquitoes were trapped during the months of December, January,

February, and March (Table 2-2). An increase in mosquito populations was seen

approximately 2-3 weeks following a high amount of rainfall. A significant difference (P

< 0.05) was found between MMPro traps 2 and 3 when compared to traps 1 and 4. Traps

2 and 3 were placed in similar environments during this study, as were traps 1 and 4.

Seasonality trends of specific mosquito species were analyzed and plotted in Figures 2-7

through 2-13. Only species with trapped numbers higher than 1000 over the course of the

year were plotted individually for seasonality trends.

Discussion

Mosquito populations fluctuated throughout the study period, but populations were

at their highest during the warm, wet summer months of July through September. Lower

counts were seen during the cool, dry months of January through April. With the first

major rainfall of the year in late March, a peak of mosquitoes resulted approximately 2-3

weeks later during early April. This population quickly decreased because temperatures

were still fairly cool. Data suggest that temperature and rainfall appear to be major

factors affecting mosquito seasonality trends. Gentry's (2002) research supports this

observation that was conducted at the UF Horse Teaching Unit. Temperature in

Gainesville, Florida fluctuates greatly during the cool seasons and become quite regular

during the summer months. Even though nighttime temperatures may increase enough









during cool season months for mosquitoes to develop, it quickly drops and disables these

mosquitoes from greatly increasing population numbers. Mosquitoes are found in north

central Florida 12 months a year, but they are present in much more significant numbers

during the warm, wet seasons of summer and fall (Campbell, 2003). Data from this

seasonality study suggest that Anopheles spp., Culex erraticus and Culex salinarius

mosquitoes were found all 12 months of the year, with high peaks in populations during

cool, dry months. The majority of mosquitoes trapped significantly decreased in

November and did not increase until April. There was one species, Coquillettidia

perturbans, which peaked suddenly in April and never again achieved such high numbers

throughout the rest of the year. Other species such as Mansonia titillans, Psorophora

spp. and Culex nigripalpus population peaks did not occur until late summer.

Mosquito surveillance is a requirement to achieve an effective and environmentally

friendly mosquito control program. Surveillance is most effective when combined with

monitoring meteorological and environmental factors that may influence mosquito

population change: For example, rainfall, temperature, relative humidity, wind direction,

and extreme weather such as hurricanes. This year, 2004, Florida had a record number of

hurricanes make landfall, and caused severe flooding. The standing water created mass

numbers of mosquito populations.

Conclusions

Identifying local mosquito seasonality trends is a valuable tool for mosquito control

agencies and the general public. During this study in Gainesville, Florida a total of

42,077 mosquitoes were trapped encompassing 17 different species. The most prominent

genus trapped was Culex including Cx. nigripalpus, Cx. salinarius, and Cx. erraticus.

Culex is one of the main vectors of West Nile Virus in Florida, and was trapped all 12









months of the year at the UF Horse Teaching Unit; therefore a threat of this virus is high.

Mosquito populations peaked during the warm, wet months of July through September.

Mosquitoes were present during the cold months of winter, but in very low numbers.

Data suggest that low nighttime temperatures greatly affect mosquito populations, by

inhibiting larval development. Generally 2-3 weeks following a heavy rainfall, mosquito

populations increased. Therefore, temperature and rainfall are major factors that affect

mosquito populations.

Surveillance of mosquitoes emerging from the floodwater could aid in predicting

dangerous disease outbreaks. Ideally, through surveillance officials potentially could

detect a period of risk in advance of the appearance of a disease. This measure could

provide control agencies the opportunity to spray for population control. The medical

community made aware of potential human cases, and the public warned so that

protective measures can be made. Knowing the times of the year that specific disease

vector mosquitoes are prevalent, can aid in developing a suitable pest management

program for a facility. In 1999 West Nile Virus greatly affected the horse industry in the

United States. By 2002 a vaccine was approved for horses to protect against this disease.

The recommendations made by veterinarians for time of year when this vaccination

should be given correspond with population peaks of mosquito vectors capable of

transmitting WNV. Demographics where WNV vectors are only prevalent for short

periods of time only need one vaccine and one booster shot per year. Places like Florida

where WNV vectors are present 12 months a year require 3 total shots. This example

demonstrates how important monitoring mosquito populations are and how this

information can greatly benefit the general public.









The general public can greatly influence the population peaks and prevalence of vectors

for a number of diseases such as WNV, EEE, SLE, LAC and Yellow Fever. Personal

protection and education are very important aspects of mosquito control. It may not be

mosquito control in a population sense, but it is personal mosquito control and helps

protect from nuisance bites and potential disease infection. The most popular insect

repellent is called DEET (N, N-diethyl-metatoluamide). DEET is one of the few

repellents that can be applied to human skin or clothing. DEET does not kill insects, but

repels them from treated areas. Scientists are not completely sure how it repels insects,

yet it is believed that DEET affects the insect's ability to locate animals to feed on by

disturbing the function of receptors in the mosquito antennae that sense host location

chemicals (Rose, 2001). Education is key to protection against mosquitoes, avoid outside

activities during peak mosquito times, if outside activities are necessary, and wear proper

clothing and insect repellent. Reducing areas of mosquito breeding, such as tires and

water filled containers will decrease mosquito populations. Proper education of horse-

owners will help protect horses from the nuisance biting and potential disease infection.

Keep horses inside under fans during mosquito peak times, vaccinate horses against

mosquito borne diseases prevalent in the area, and use mosquito spray regularly.














































Figure 2-1. Weather station at south side of UF Horse Teaching Unit, used to collect
rainfall, minimum and maximum temperature readings for the mosquito
seasonality study conducted September 26, 2003 through September 2, 2004.
















































Figure 2-2. Aerial photograph of the UF Horse Teaching Unit show the four MMPro trap
sites for the mosquito seasonality study conducted September 26, 2003
through September 2, 2004.


























3500


3000


2500
I-


5 2000


1500

I-
1000


500


0
9 10 10 10 11 11 12 12 12 1 1 1 2 2 2 3 3 3 4 4 5 5 6 6 6 7 7 7 8 8 8
Month


Figure 2-3. Total mosquito counts as related to months during the seasonality study conducted at the UF Horse Teaching Unit from
September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e. = January).





















350( 14

300( 12

| 250( 10
o
o 200( 8-

S150( 6

100 4



50W 0
9 10 10 10 11 11 12 12 12 1 1 1 2 2 2 3 3 3 4 4 5 5 6 6 6 7 7 7 8 8 8
Month

SRain (inches) -*--Mosquito Count

Figure 2-4. Total mosquito counts from the MMPro, related to rainfall during the seasonality study conducted at the UF Horse
Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar
date (i.e. 1= January).



















40.0



30.0



20.0



E 10.0
I-


0.0



-10.0



-20.0
Months

Max Temperatures (C) Min Temperatures (C)



Figure 2-5. Maximum and Minimum temperatures recorded at the meteorological station during the seasonality study conducted at
the UF Horse Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals
corresponding calendar date (i.e. l=January).





















40.00


35.00


30.00


S25.00


E 20.00


^ 15.00


10.00


5.00


0.00
O N D J F M A M J J
Month

2001-2002 02002-2003 02003-2004

Figure 2-6. Total rainfall (centimeters) at the UF Horse Teaching Unit from October 2001 through August 2004.



















3500


3000


2500


2000


5 1500


1000


500 -




b) N N N 1 V 'b 'Ib b I I <0 <0 '\ '\ '\ )b )b Ib
Month


Figure 2-7. Total numbers of Culex nigripalpus trapped by MMPro during the seasonality study conducted at the UF Horse Teaching
Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e.
1= January).























100



80 80



60
40
I-^








20




Cb) N N N 1 b Ib Ib I I <0 <0 Month


Figure 2-8. Total numbers of Culex erraticus trapped by MMPro during the seasonality study conducted at the UF Horse Teaching
Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e.
1= January).
























400


350


*O 300
C.
C.

250


S200


150


100


50


0


Month


Figure 2-9. Total numbers of Culex salinarius trapped by MMPro during the seasonality study conducted at the UF Horse Teaching
Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e.
1 January).




















1200


1000



800
C.
I-

| 600

o 00

I 400 -



200-





Q) p ^O hp \ ^ ^ ^ 'K 'K 'K K) ) ) ^ b b b b b Month


Figure 2-10. Total numbers ofMansonia spp. trapped by MMPro during the seasonality study conducted at the UF Horse Teaching
Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e.
1= January).
























140


120


S100
C-

80


60


40


20




Cb N N N I ^ ^ ^ \ \ ^ 'L 'L ^ 'b Ib b I I Month

An. crucians An. quadrimaculatus

Figure 2-11. Total numbers of Anopheles spp. trapped by MMPro during the seasonality study conducted at the UF Horse Teaching
Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e.
1= January).
























500



2 400
I-



| 300



I 200



100





9) C N N> \ N N 'V 'L 'I I 0b 0b b Rb Rb A A A b \b 'b
Month


Figure 2-12. Total numbers of Coquillettidiaperturbans. trapped by MMPro during the seasonality study conducted at the UF Horse
Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar
date (i.e. 1= January).























300


250


200


150


100


50


0 A
9) N N N 1 N 'b b b I I <0 <0 o b \ \ \ Qb b (b
Month

Ps. columbiae Ps. ferox Ps. cyanescens

Figure 2-13. Total numbers of Psorophora spp. trapped by MMPro during the seasonality study conducted at the UF Horse Teaching
Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e.
1= January).









Table 2-1. Total count, and percent of total count of mosquito species trapped
by MMPro traps during seasonality study at the UF Horse Teaching Unit
conducted September 26, 2003 until September 2, 2004.
Mosquito Species Total Count Percent of Total
Culex nigripalpus 18213 43.3
Mansonia spp. 7398 17.6
Culex salininarius 5374 12.7
Coquillettidia perturbans 3854 9.2
Psorophora columbiae 2216 5.3
Culex erraticus 1566 3.7
Anopheles crucians 1218 2.9
Anopheles quadrimaculatus 1110 2.6
Aedes vexans 372 0.88
Psorophora ferox 365 0.87
Uranataenia sapphirina 144 0.27
Ochlerotatus infirmatus 103 0.24
Psorophora cyanesens 84 0.20
Uranataenia lowii 44 0.11
Ochlerotatus trisariatus 6 0.01


Culiseta melanura
Ochlerotatus mitchellae


5
3
42077


0.01
0.01









Table 2-2. Mean standard deviation of mosquitoes trapped each month during
seasonality study at the UF Horse Teaching Unit conducted September 26,
2003 until September 2, 2004.
Month Mean Catch ( Std) n Total
September 2003 666 ( 221) 4 3566
October 2003 357.9 ( 354.1) 32 10552
November 2003 139.1 ( 173.3) 28 3895
December 2003 19.59 ( 15.90) 29 568
January 2004 13.03 ( 11.46) 35 456
February 2004 11.97 ( 11.06) 31 371
March 2004 29.31 ( 20.03) 26 762
April 2004 114.0 ( 78.9) 23 2623
May 2004 37.53( 23.03) 32 1201
June 2004 96.9 ( 166.3) 30 2908
July 2004 192.2 (267.2) 32 6150
August 2004 178.8 ( 154.0) 31 5542
September 2004 1742 ( 1538.0) 2 3483
42077
Note: n=number of observations














CHAPTER 3
MOSQUITO TRAPPING STUDIES AT AN EQUINE FACILITY IN NORTH
CENTRAL FLORIDA

Introduction

Mosquito traps are very effective as a surveillance tool to monitor seasonal

prevalence and the species composition in a specific area. Traps are also effective in

reducing numbers of mosquitoes in the location of the trap. Over the years there have

been lots of different kinds of mosquito traps developed, varying greatly in effectiveness

and usefulness (Kline, 1999). Many commercial traps have been advertised to control

mosquitoes up to 1 acre and are commonly found around livestock facilities. Most of

these traps use stimuli that mimic behavior such as heat, carbon dioxide, kairomones, and

moisture (Kline and Mann, 1998). But how effective are mosquito traps when they are

competing with a natural host? Campbell (2003) conducted a series of experiments

evaluating mosquito traps and a horse, and found a significant difference between species

and total quantity of mosquitoes collected from mosquito traps compared to mosquitoes

aspirated directly off a horse. If adult mosquito trapping is the main technique for

surveillance, is this the true representation of the actual mosquito species population?

Trap catches could potentially be improved by adding odors directly from the horse to the

specified trap. Little research has been conducted which directly compares the efficiency

of mosquito traps with a natural host in the immediate proximity.

Therefore, the main objectives of the following studies are: (1) to conduct trapping

competitive studies with 2 traps and a horse; (2) to evaluate mosquito location profiles









around the University of Florida Horse Teaching Unit, (3) an evaluation was conducted

looking mosquito species feeding on a horse at a given time by the technique of

aspiration; (4) to determine the distance required between 2 horses to achieve separate

entities.

Materials and Methods

Six studies were conducted in Gainesville, Florida at the University of Florida

Horse Teaching Unit; a 60-acre equine facility housing approximately 40 head of quarter

horses varying in sex and age. Three CDC 1012 traps, a portable vacuum aspirator, a

Mosquito Magnet Pro, a gelding paint Quarter horse, and an Appaloosa mare were placed

throughout the property.

Traps

Centers for Disease Control (CDC) trap. The CDC trap model 1012 (John W.

Hock Company, Gainesville, FL) (Figure 1-1), was used during this experiments with

slight modifications. Generally the trap uses a 6.3-volt light as an attractant, but this was

eliminated to decrease attractive variables of the trap. The CDC 1012 uses a 6-volt

battery to provide energy to a fan, which creates a downward airflow to capture

mosquitoes attracted to the top of the trap. Carbon dioxide was supplied from a

compressed gas cylinder at a rate of 500 ml/min. The carbon dioxide line was attached

with rubber bands to the side of the trap, with the gas being released directly above the

fan. A polypropylene container with screen on the bottom was used as the collection

device and connected to the bottom of the trap.

Mosquito Magnet Pro. The Mosquito Magnet Pro (MM-Pro) (American

Biophysics Corporation, North Kingston, RI) (Figure 1-3) uses patented technology that

catalytically converts propane into carbon dioxide, heat and moisture, which attracts









mosquitoes to the counterflow of a suction fan and released attractants. The mosquitoes

are vacuumed into a collection net where they dehydrate and die. The trap is powered

through propane by catalytically converting the gas into carbon dioxide, heat, and

moisture. The MM-Pro stands 2 meters high and is built of stainless steel with a PVC

shell.

Portable Vacuum Aspirator. Mosquitoes that landed on the horse were collected

using a portable vacuum mosquito aspirator (DC Insect Vac. BioQuip, Rancho

Dominguez, CA) (Figure 1-1). A 12-volt AC supplied by a modified plug, which fits into

a car cigarette lighter socket, powered the vacuum. A plastic collection cup fits into the

vacuum and contains the mosquitoes while the vacuum in turned on; when the vacuum

was turned off a cap was placed on the container.

Horses. The horses used during this experiment were a gelding Paint (Figure 3-2)

and an Appaloosa mare (Figure 3-3), both approximately 7 years of age. The horses were

housed at the University of Florida Horse Teaching Unit and remained there after the

experiment concluded.

Experimental Design

CDC 1012 Trapping Study

The CDC trapping efficiency study was conducted from July 19, 2004 until August

21, 2004. Starting 30 minutes after sunset, a CDC 1012 mosquito trap was placed

directly next to the gate on a stall of a horse barn (Figure 3-4). The position of the barn

and the stall where the study was conducted in shown in Figure 3-5. The trap was baited

with only CO2 and the net was changed every 20 minutes to include five 20-minute

trapping periods. There were two treatment groups: no horse present and horse present,

each chosen at random. During nights when the horse is present, a Paint gelding was









placed in the stall during the second 20-minute trapping period, then removed.

Mosquitoes were stored at 00C until identification and counting was conducted. Each

treatment, horse and no horse, was repeated six times (see Table 3-1 for schedule).

MMPro Trapping Study

The MMPro trapping efficiency study was conducted from October 1, 2004 until

October 21, 2004. Starting 30 minutes after sunset a Mosquito Magnet Pro trap was

placed directly next to side of a horse feeding slip stall (Figure 3-6). The position of the

barn and the stall where the study was conducted in shown in (Figure 3-7). The trap was

baited with C02, moisture and heat; the net was changed every 20 minutes to include 5

20-minute trapping periods. There are two treatment groups: no horse present and horse

present. The treatment "horse present" was conducted coordinating with when the

facilities were available. During nights when the horse is present, a Paint gelding was

placed in the stall during the second 20-minute trapping period, then removed.

Mosquitoes were stored at 00C until identification and counting was conducted. Each

treatment, horse and no horse, was repeated six times (see Table 3-2 for schedule).

Horse Odor Study

Field data for this study were collected beginning August 27, 2004 and concluded

September 24, 2004, examining mosquitoes captured by a modified CDC 1012 trap with

the addition of equine odors to the attractive air stream. A modified hand held vacuum

was constructed for this portion of the experiment. A 2-inch shop vacuum hose was

connected to a 2-inch pvc pipe, with the open end placed near the CDC 1012 trap

airstream. There was an electric powered in-line fan in the pvc pipe which created a

suction and air through the hose and out the open end. Starting 30 minutes after sunset, a









CDC 1012 mosquito trap was placed directly next to the gate on a stall of a horse barn.

The net was changed every 20-minutes to include 4 20-minute trapping periods. The

gelding Paint horse was placed in a stall of a horse barn (Figure 3-5) for 1 hour during

intervals 1, 2 and 3. The mosquito trap was only baited with CO2 during interval 1, and

then again during interval 3. During interval 2, the modified hand held vacuum was

attached to the CDC 1012 trap, so the trap was baited with horse odor and carbon

dioxide. The horse was vacuumed all over the body (Figure 3-8) and the odors traveled

through the PVC pipe and were released directly above the fan close to the carbon

dioxide source. The horse was removed from the stall during interval 4 and collection

continued until for 20 more minutes. This protocol was conducted six times. All

collection intervals were 20-minute intervals. See Tables 3-3 and 3-4 showing time

schedule and treatment date schedule respectively. Mosquitoes were stored at 00C until

identification and counting was conducted.

A similar protocol was conducted to monitor mosquito activity from August 27,

2004 through September 24, when no horse was present. During intervals 1-4, a CDC

1012 trap was set up in the same position as mentioned above, baited only with carbon

dioxide. All collection intervals were 20-minute intervals. Mosquitoes were stored at

00C until identification and counting was conducted.

Location Profile Study

The mosquito location profile through trapping study using CDC 1012 traps was

conducted from July 13, 2004 until August 26, 2004. Three CDC 1012 mosquito traps

were placed in three positions throughout the University of Florida Horse Teaching Unit

(Figure 3-9). The first position was close to a swamp area and the other two positions









were on each side of a horse barn. All three CDC 1012 traps were baited only with

carbon dioxide and allowed to run from 8pm until 10:30pm. The traps were set up 30

minutes before sunset and collected after mosquito peak activity had concluded.

Mosquitoes were stored at 00C until identification and counting was conducted.

Horse Vacuuming Study

The quantity of mosquitoes feeding on a horse at a given time period of time was

examined from June 15, 2004 until July 12, 2004 and then repeated October 4, 2004 and

concluded October 22, 2004. During the first trial, mosquitoes were aspirated from the

skin of a gelding Paint horse (Figure 3-10) located in a stall of a barn using 2 portable

vacuum mosquito aspirators (Figure 1-1) for a 1 hour time period starting 30-minutes

after sunset. During the second portion of this experiment, a mare Appaloosa and a

gelding Paint horse were vacuumed simultaneously starting at for 1 hour and 40 minutes

starting 30-minutes after sunset. Mosquitoes were aspirated from all over the body of

the horse. Mosquitoes were stored at 00C until identification and counting was

conducted.

Separate Entity Study

The distance required for the mosquito to distinguish two horses as separate entities

was examined from October 4, 2004 until October 22, 2004. Two horses, a gelding Paint

and an Appaloosa mare, were placed in adjacent feeding slip stalls starting 30-minutes

after sunset (Figure 3-11). The horses were vacuumed simultaneously using 2 portable

vacuum mosquito aspirators for 20 minutes. The collection cups on the vacuum

aspirators were emptied and new collection cups were placed in the aspirators. The

Appaloosa mare remained in the original stall, and the Paint gelding was moved to









another stall depending on the randomly chosen assignments for the night. The horses

were then vacuumed again for 20 minutes. This procedure was repeated until the horses

were vacuumed in all five stalls. Five distance intervals were set up, D1=3.05 meters,

D2=6.09 meters, D3=9.14 meters, D4=12.19 meters, D5=20.42 meters. See the schedule

of stall assignments for each date of the trial in Table 3-5. The mosquitoes were stored at

00C until identification and counting was conducted

Data Analysis

Count data from the studies described above were analyzed with the General Linear

Model (GLM) and Proc Means Programs; Tukey's Studentized Range Test was used for

separation of means (SAS Institute, Version 8.2, Copyright 1999-2001). The significance

level was set at P < 0.05.

Results

CDC 1012 Trapping Study

The total number of mosquitoes collected by CDC 1012 mosquito trap during this

trial was 2,249 with the most prominent species being Mansonia titillans comprising 76%

of the total catch, followed in descending order by Coquillettidia perturbans (11%) and

Culex salinarius (7%). The distribution and means of mosquitoes trapped during the two

treatment groups are shown in Figure 3-12. Total values of mosquitoes that were trapped

during each interval by treatment are shown in Figure 3-13. The mean number of

mosquitoes captured during each interval by treatment group is shown in Figure 3-14.

There was a significant difference (P < 0.05) in the mean numbers of mosquitoes between

the treatment groups horse and no horse (Table 3-6).









During the treatment "horse present," there was a significant difference (P <0.05)

in the mean numbers of mosquitoes caught between intervals (Table 3-6). There was a

significant reduction (P < 0.05) in the mean mosquito catch in interval 2 when compared

to intervals 1 and 3. There was no significant difference between intervals 1 and 3. The

greatest mean number of mosquitoes was captured during interval 1. The fewest mean of

mosquitoes was captured during interval 2, which coincided with when the horse was

located in the stall next to the CDC 1012 mosquito trap.

During the treatment "no horse present," there was a significant difference (P <

0.05) in the mean numbers of mosquitoes caught between intervals (Table 3-6). Interval

2 was significantly greater (P < 0.05) when compared to all of the other intervals. There

was no significant difference between intervals 1 and 3 or between intervals 4 and 5. The

greatest mean number of mosquitoes was captured during interval 2. The fewest mean of

mosquitoes was captured during interval 5.

MMPro Trapping Study

The total number of mosquitoes collected by MM-Pro mosquito trap during this

trial was 5,533 with the most prominent species being Culex nigripalpus comprising 36%

of the total catch, followed in descending order by Mansonia titillans (31%) and

Psorophora columbiae (23%). The mean number of mosquitoes captured during each

interval by treatment group is shown in Table 3-7. The distribution and mean numbers of

mosquitoes trapped during the two treatment groups are shown in Figure 3-15. Total

values of mosquitoes that were trapped during each interval by treatment are shown in

Figure 3-16. The mean number of mosquitoes captured during each interval by treatment

group is shown in Figure 3-17. There was a significant difference (P <0.05) in the mean









numbers of mosquitoes captured between the treatment groups horse and no horse (Table

3-7).

During the treatment "horse present," there was a significant difference (P < 0.05)

in the mean numbers of mosquitoes caught between intervals (Table 3-7). There was a

significant difference (P < 0.05) in the mean mosquito catch in interval 2 when compared

to 1 and 3. There was no significant difference between interval 1 and 3. The greatest

mean of mosquitoes was captured during interval 1. The fewest mean of mosquitoes was

captured during interval 2, which coincided with when the horse was located in the stall

next to the MMPro mosquito trap.

During the treatment "no horse present," there was a significant difference (P <

0.05) in the mean numbers of mosquitoes caught between intervals (Table 3-7). Intervals

1 and 2 were significant greater than intervals 4 and 5. There was no significant

difference between intervals 1, 2 and 3, or between 3, 4, and 5. The greatest mean

number of mosquitoes was captured during interval 2. The fewest mean number of

mosquitoes was captured during interval 5.

Horse Odor Study

The total number of mosquitoes collected by the CDC 1012 mosquito trap during

this trial was 1,109 with the most prominent species being Mansonia titillans comprising

76% of the total catch, following by Culex nigripalpus (28%). The distribution and

mean numbers of mosquitoes trapped during the two treatment groups are shown in

Figure 3-18. Total values of mosquitoes that were trapped during each interval by

treatment are shown in Figure 3-19. The mean number of mosquitoes captured during

each interval by treatment group is shown in Figure 3-20. There was no significant









difference between the mean numbers of mosquitoes trapped during treatments Horse

Present and Horse Odor Vacuumed (Table 3-8). There was a significant difference (P <

0.05) between the mean numbers of mosquitoes trapped of No Horse Present when

compared to all other treatments.

Location Profile Study

The total number of mosquitoes collected by the CDC 1012 mosquito traps during

this trial was 7,184 with the most prominent species being Mansonia titillans comprising

66% of the total catch, followed by Culex nigripalpus (16%)(Figure 3-21). The

distribution of mean numbers of mosquitoes captured from each trap position is shown in

Figure 3-22. There was a significant difference (P < 0.05) between the mean numbers of

mosquitoes trapped in position 2 when compared to positions 1 and 3 (Table 3-9). There

was no significant difference between the mean numbers of mosquitoes trapped in

positions 1 and 3 (Table 3-9). There was no significant difference between the mean

numbers of Cx. nigripalpus and Cx. erraticus between trap positions (Table 3-10). There

was a significant difference (P < 0.05) in mean numbers of Cx. salinarius trapped in trap

position 2 when compared to positions 1 and 3 (Table 3-10). There was a significant

difference (P <0.05) in mean numbers ofMa. titillans and Cq. perturbans trapped in

trap position 1 when compared to positions 2 and 3 (Table 3-10).

Horse Vacuuming Study

The total numbers of mosquitoes aspirated from the Paint horse from June 15,

2004 until July 12, 2004 was 1,946. The total number of mosquitoes aspirated off of the

Paint and Appaloosa horses from October 4, 2004 until October 22, 2004 was 7,197.

During the first trial, the most prominent species trapped was Cq. perturbans comprising









40% of the total catch, followed in descending order by Cx. salinarius (37%) and Ma.

titillans (12%) (Figure 3-24). During the second trial, the most prominent species

trapped was Cx. nigripalpus comprising 41% of the total catch, followed in descending

order by Ma. titillans (27%) and Ps. columbiae (17%) (Figure 3-25). There was a

significant difference (P <0.05) between the mean numbers of mosquitoes and the

species composition aspirated from the Paint and Appaloosa horses. Almost 50% of the

total mosquitoes vacuumed from the Appaloosa mare were Cx. nigripalpus, followed by

Ma. titillans (21%) and Ps. columbiae (15%) (Figure 3-26). Approximately 40% of the

mosquitoes vacuumed from the Paint were Ma. titillans, followed by Cx. nigripalpus

(25%) and Ps. columbiae (21%) (Figure 3-27).

Separate Entity Study

The total numbers of mosquitoes aspirated from the Appaloosa and Paint horses

from October 4, 2004 until October 22, 2004 was 4,810 and 2,387 respectively. The

most abundant species was Culex nigripalpus comprising 41% of the total catch,

followed in descending order by Mansonia titillans (27%) and Psorophora columbiae

(17%). There was a significant difference (P < 0.05) between the mean numbers of

mosquitoes and species composition aspirated from the Paint and Appaloosa horses.

Almost 50% of the total mosquitoes vacuumed from the Appaloosa mare were Cx.

nigripalpus, followed by Ma. titillans (21%) and Ps. columbiae (15%) (Figure 3-26).

Approximately 40% of the mosquitoes vacuumed from the Paint were Ma. titillans,

followed by Cx. nigripalpus (25%) and Ps. columbiae (21%) (Figure 3-27). There was no

significant difference between the difference of mean numbers of mosquitoes vacuumed

off horse 1 and horse 2 per distance interval (Table 3-10). There was no significant









difference between the mean of total mosquitoes vacuumed from horse 1 and horse 2 per

distance (Table 3-11), the means and standard deviations (shown in Figure 3-28).

Discussion

CDC 1012 and MMPro Trapping Studies

Over the years there have been many different kinds of mosquito traps developed,

varying greatly in effectiveness and usefulness. The most effective traps use a

combination of factors that attract mosquitoes such as light, heat, moisture, carbon

dioxide and synthetic chemicals for host attraction, such as octenol. Trap efficiency and

species composition have been researched thoroughly, yet little research has been

conducted testing trap efficiency when placed in a competitive situation with an actual

host. Before this study was conducted, it was suggested that a host located near a

mosquito trap would cause an increase in the catch of the trap. However, the data in

studies 1 and 2 suggest that the traps (CDC 1012 and MM-Pro) were both significantly

out competed when a natural host was in the vicinity. During the 100-minute trapping

sessions, when there was no horse present the mosquitoes trapped followed a natural

increase the hour after sunset and steadily decreased for the remaining trapping period.

This mosquito activity pattern has been well documented, and it has been stated that most

mosquito species possess a bimodal flight activity pattern during the night, with the larger

peak occurring soon after sunset and the smaller peak just prior to dawn (Schmidt, 2003).

When a horse was added to the system for a 20-minute interval out of the total 100-

minute session, the natural increase did not occur and virtually no mosquitoes were

trapped, but only when the horse was present. As soon as the horse was removed, the

mosquitoes captured in the traps increased significantly.









This study was conducted twice with two different trap types during different

months of the year. The CDC 1012 trap was tested along with the MM-Pro during the

trapping studies. Previous data suggest that the MM-Pro is a superior mosquito trap

when compared to the CDC 1012 (Kline, 1999; Burkett et al., 2001), yet Campbell

(2003) suggested the CDC 1012 trap performed as well or better than the MM-Pro. Since

it has been suggested that both traps perform similarly, their mosquito yield capabilities

were not studied during this experiment. Only their performance when placed in a

competitive situation with a host was examined. The CDC 1012 trapping study was

conducted during the months of July and August, and the MMPro trapping study was

conducted in the month of October. The mosquito species composition changed between

study 1 and study 2 most likely because of change in season. During CDC 1012 trapping

study the most prominent mosquito species trapped were Mansonia titillans,

Coquillettidia perturbans and Culex salinarius, respectively. During the MMPro

trapping study the most prominent mosquito species trapped were, Culex nigripalpus,

Mansonia titillans, and Psorophora columbiae, respectively. The change in seasons

between the studies provided a wide array of mosquito species present, which enabled the

examination of preferences of host or trap.

The traps were placed approximately 1 meter from the horse, but the horse could

move around in the 3 x 3 meter stall. So at times the trap may have been as far away as 3

meters from the horse. This distance between horse and trap has been shown to reduce

the catch of the trap by almost 95%. This is a substantial reduction of mosquitoes, and

could suggest that the traps are inefficient when a host is within 3 meters of the trap.









Further studies should be conducted examining the exact distance between horse and trap

when the mosquito catch begins to decrease.

Horse Odor Study

The next study took the information learned from the previous studies and

elaborated by adding equine odors to a CDC 1012 trap. It was believed that by adding an

odor attractant to the carbon dioxide airstream of the trap, the mosquito catch would

increase. During this study when the horse was being vacuumed to pull off skin odors,

the mosquito catch did not significantly increase. This data may suggest that host

location of mosquitoes may not be from equine skin odors, or that the added attractant to

the trap was still not sufficient enough to out compete the natural host in the vicinity.

The widely used mosquito attractant octenol was isolated from compounds in oxen

breath, so further studies could be conducted examining equine breath for attractiveness.

Only one horse (Paint) was used during this study, and even though it was found that the

Paint was an attractive host, another horse could have potentially caused an increase in

mosquitoes trapped with odors. Another experiment that could be conducted would

remove the visibility of the horse from the mosquitoes, and therefore only have odors as

the attractant.

Location Profile Study

Another study was conducted examining trap location differences at the University

of Florida horse teaching unit. This facility is 60-acres and CDC 1012 mosquito traps

were distributed around the facility in various types of environments. One trap (position

2) was located near a swamp, and the other two traps were positioned on each side of a

10-stall barn. Previous data collected suggested a difference in mosquito quantities found

on the west side (position 1) of the barn when compared to the east side (position 3). The









most prominent mosquito species trapped during this study was Mansonia titillans and

Culex nigripalpus. Position 2 trapped the highest number of mosquitoes, which is

feasible since this trap is located close to a standing body of water. Position 3 trapped a

higher number of mosquitoes than position 1, but this difference was not found to be

significant. Both position 1 and 3 were located the same distance from the nearest body

of water, but their surroundings varied slightly. There was a stall enclosed by white

material adjacent to the CDC 1012 trap at position 1, earlier it was believed that this

enclosure was deterring mosquitoes from the area. Position 3 was adjacent to an open

stall and had no visible structures surrounding it that suggested being any kind of

deterrence. Even though the difference in mean mosquitoes was not found to be

significantly different during this study, the total numbers were different. Position 1

trapped a total of 1134 mosquitoes while position 3 trapped 1801 mosquitoes. These

positions are only 30 meters apart, so what caused this slight difference in total

mosquitoes? It has been documented that mosquitoes are attracted or repelled by

different colors and light intensities (Muir et al., 1992), the white material covered stall

adjacent to the trap at position 1 could have affected the trap totals during this study.

Horse Vacuuming Study

High numbers of mosquitoes feeding on horses in Florida can cause significant

health issues ranging from a direct impact like West Nile Virus to an indirect nuisance

effect causing reduced feed conversion efficiency, weight gain reductions, and decreased

milk yield (Steelman, 1979 and Byford et al., 1992). During this trial high quantities of

mosquitoes were aspirated off of the skin of a Paint gelding and an Appaloosa mare.

Trapping occurred over 11 nights and a total of 9,143 mosquitoes were vacuumed from

the horses. These numbers should cause concern about the sheer abundance of









mosquitoes routinely feeding on horses in Gainesville, Florida. Horses are too busy

fighting off biting mosquitoes that they may neglect grazing or drinking adequate water.

There were common localities where the mosquitoes were most likely found during this

vacuuming study including the neck, legs, heels, barrel and stomach. An interesting note

was made that the Appaloosa mare always had mosquitoes congregating around her

cornet band; this was not seen with the Paint horse. Both horses were very adequate at

shaking off the mosquitoes by twitching their skin or knocking them off with their tail.

Some times the horse's response to the biting mosquito was faster than I was at

vacuuming the mosquito.

The species composition of mosquitoes vacuumed during this study varied by

season and by horse. During the first trial conducted in June and July 2004, the most

abundant species collected was Coquillettidia perturbans (39.5%) and Culex salinarius

(36.9%). Cq. perturbans are primarily mammal feeders, but are opportunistic feeders and

will feed on a bird if available (Sardelis et al., 2001). Cx. salinarius are primarily avian

feeders, but if the population of birds is insufficient or unavailable, they will contentedly

feed on mammals (Braverman et al., 1991). During the second trial the three highest

abundant mosquito species were Culex nigripalpus, Mansonia titillans and Psorophora

columbiae. There was a difference in the percentage of these species found on each

horse. The Appaloosa mare had almost 50% of the total mosquitoes vacuumed being Cx.

nigripalpus, followed in descending order by Ma. titillans (21%) and Ps. columbiae

(15%) The Paint gelding had 40% of the mosquitoes vacuumed being Ma. titillans,

followed in descending order by Cx. nigripalpus (25%) and Ps. columbiae (21%). The

difference in species composition between horses is interesting. As stated earlier, a









congregation of mosquitoes around the comet band was found on the Appaloosa mare

that was not present on the Paint gelding, data suggests that these mosquitoes were Cx.

nigripalpus. Why was Cx. nigripalpus more attracted to the Appaloosa than the Paint is

unknown, it could be the difference in sex or color of hair. Ma. titillans and Ps.

columbiae are primarily mammal feeders, but are opportunistic feeders and will feed on a

bird if available (Edman, 1971). Cx. nigripalpus is an opportunistic feeder and shifts host

selection based on the season, feeding on avian hosts in the winter and spring and on

mammalian hosts in the summer and fall (Sardelis et al., 2001). It has been stated that

male hosts are more attractive to mosquitoes than female hosts (Khan et al., 1965), but

the opposite was seen during this study. More mosquitoes were vacuumed from the mare

Appaloosa than the gelding Paint horse. Since the male horse was a gelding (castrated),

the lack of testosterone could have affected the normal behavior of mosquito preference.

During trial 2 the species composition was much more diverse when compared to trial 1.

There were 11 different mosquito species collected during the MMPro trapping study and

6 species during the CDC 1012 trapping study. The difference in species composition is

probably due to seasonal differences. When the CDC 1012 trapping study was conducted

there was little rainfall and the mosquito populations were still low, the MMPro trapping

study was conducted during optimal meteorological conditions, warm and wet.

Separate Entity Study

The distance needed between two hosts for a mosquito to view them as separate

entities is valuable information. This information could be utilized by mosquito trap

manufactures when determining the range of effectiveness with natural hosts, such as

livestock, situated around the trap. The closest distance examined during this study was

3.05 meters and the farthest was 20.42 meters between horses. There was no significant









difference found between mosquitoes vacuumed at any of the distance intervals, yet there

was a significant difference between total mosquitoes vacuumed off of each individual

horse. This data could suggest a couple of conclusions, either the mosquitoes recognized

the two horses as separate entities immediately at a distance of 3.05 meters apart, or the

goal of separate entities was not accomplished at a distance of 20.42 meters apart. Since

there was a significant difference in mosquitoes vacuumed off of each individual horse, I

believe that the mosquitoes immediately recognized the two horses as separate entities at

3.05 meters apart. Another study could be conducted by placing two horses side by side

and measuring mosquitoes landing through vacuuming, which would clarify at what

distance separate entity status is achieved.

Conclusions

Mosquito traps are thought to be effective tools for surveillance and reducing

numbers in a specific area. Data from this study suggests that the CDC 1012 and

Mosquito Magnet Pro are effective at trapping mosquitoes as long as they are not placed

in a competitive situation with a natural host. If adult mosquito trapping is the main

technique for surveillance, this may not be a true representation of the actual mosquito

species population. There seems to be a different composition of mosquito species being

caught in mechanical traps when compared to mosquitoes vacuumed directly off of a

horse. Therefore, it is suggested that these mechanical traps are only representing a

fraction of the total mosquito population in an area. Mechanical traps could potentially

be improved by adding odors directly from the host to the trap. Even though this

technique was not successful during this trial, more field research should be conducted on

the subject. Hall et al. accomplished isolating a host odor (octenol) that attracted Tsetse

flies in 1984. Octenol was isolated from oxen breath and is currently used as a mosquito









attractant (Kline, 1994). Since the mosquitoes are attracted to horse instead of the traps

tested, this suggests that there is some kind of attractant overriding the standardized

attractants used in traps. This stimulant has not been identified, it may be a combination

of visual and chemical odor attractants. Further studies should evaluate these stimulants

to alleviate such high numbers of biting mosquitoes on horses. If the traps are being out

competed by a natural host, then trap placement is important to effectively reduce adult

mosquito populations around horses. The distance between a natural host and trap where

the competitive state decreases needs to be investigated. This information could provide

mosquito trap manufactures beneficial data to make suggestions about trap placement.











































figure 3-1. Fortawle vacuum aspirator (uc insect vac. tBioyuip, Kancto uominguez,
CA) used for collection of mosquitoes during vacuuming studies.









































figure 3-2. Paint gelding that was sampled tor mosquitoes during trapping studies.









































figure 3-3. Appaloosa mare that was sampled tor mosquitoes dunng trapping studies.








































figure 3-4. CUC 1012 trap placement during CUC 1012 trapping study at the Ut Horse
Teaching Unit from July 19, 2004 through August 21, 2004




...............







.A.-- TURE-'


~I ~I














PAST....RE





















:ure 3-5. C~~~~~~~~~~~DC 1012 trappigsuypaeetue o oqiotapn.N -
. . . . . .P. . .


.. .. ..re. .. 3. .. 5.D.01.rp in.tdypa e en.sd.o.oqut.ra pn


................................
.................................
........................................




................


.........
. . .


Fig








































Figure 3-6. MMPro trap placement during MMPro trapping study at the UF Horse
Teaching Unit from October 1, 2004 through October 21, 2004









































































































5.7 acres 5.0 acres





































































































MMPro trap


....................................................................................................
..................................................
....................................................................................................
..................................................
....................................................................................................
..................................................
....................................................................................................
..................................................
....................................................................................................
..................................................


. . . . . . . . . . . . . . . . .
..................................................
....................................................................................................
..................................................
....................................................................................................
..................................................
....................................................................................................
..................................................
....................................................................................................
..................................................
....................................................................................................
..................................................
....................................................................................................
..................................................
....................................................................................................

LI.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.,.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.,.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.I.1






Figure 3-7. Mosquito Magnet Pro trap placement used during mosquito trapping study.







































Figure 3-8. Modified CDC 1012 mosquito trap used during trapping horse odor trapping
study at the UF Horse Teaching Unit from August 27, 2004 through
September 24, 2004.




























C",

.. .... 72.51 m eters .......lllllll
72STURE
Fu39 C0taauu froulcipfet
. . . .:::
:':::::~::.. .............iiii


6. meters..








Figure 3 -9. CDC 10 12 trap layout used for mosquito location profile study.








































Figure 3-10. Technique of aspirating mosquitoes off of horse during horse vacuuming
study and separate entity study.






























Figure 3-11. Feeding slips at UF Horse Teaching Unit where the 2nd trial of horse
vacuuming study and separate entity study was conducted.







84








120-


100-


o 80 A2.
A-d
o 60-


S40





I I \fflI I



Interval S 1 ? a <

Figure 3-12. Distribution and mean numbers of mosquitoes trapped with CDC 1012 for
treatments, horse and no horse present, and intervals of testing from July 19,
2004 through August 21, 2004, at an equine facility in Gainesville, Florida.

















bUU


500


o 400
















SHorse No Horse

Figure 3-13. Total numbers of mosquitoes captured with CDC 1012 trap when a horse is
present during interval 2 and when no horse is present in interval 2; testing
from July 19, 2004 through August 21, 2004, at an equine facility in
Gainesville, Florida.
Gainesville, Florida.







86








90.0

80.0 -

70.0

1 60.0

|50.0 1 0 3

S40.0

c 30.0 30.5

20.0 -

10.0- \115

0.0
1 2 3 4 5
Time intervals

-- Horse -M-No Horse

Figure 3-14. Mean numbers of mosquitoes captured with CDC 1012 trap when a horse is
present during interval 2 and when no horse is present in interval 2; testing
from July 19, 2004 through August 21, 2004, at an equine facility in
Gainesville, Florida.







87









400-



8A 300-
0
.0

Z 200-




T -


0-



Interval S

Figure 3-15. Distribution and mean numbers of mosquitoes trapped with MMPro for
treatments, horse and no horse present, and intervals of testing from October
1, 2004 through October 21, 2004, at UF Horse Teaching Unit in Gainesville,
Florida.




Full Text

PAGE 1

EVALUATION OF MOSQUITO TRAPPING EFFICIENCY AND DETERMINATION OF SEASONALITY FOR MOSQUITOES AT THE UNIVERSITY OF FLORIDA HORSE TEACHING UNIT By SARAH COURTNEY DILLING 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 2004

PAGE 2

Copyright 2004 by Sarah Courtney Dilling

PAGE 3

This document is dedicated to Belle the Appaloosa mare who unexpectedly became ill and died November 2004.

PAGE 4

ACKNOWLEDGMENTS I would like to thank Saundra TenBroeck for having confidence in me and showing supportive guidance throughout my program. I thank Jerry Hogsette for constantly and willingly being there for support and assistance while educating me in a subject area where I was just a novice. I would also like to express my gratitude to Dan Kline for his encouragement and suggestions. I thank Alyce Nalli for her constant assistance and genius ideas; I could not have completed this research without her. I would also like to thank the staff at the University of Florida Horse Teaching Unit, especially Joel McQuagge and Tonya Stephens. I thank Aaron Lloyd and Joyce Urban for helping me identify mosquitoes and always making time to help. I would like to thank Kelly Spearman, Talia Bianco, Tonya Stephens and Kylee Johnson for all of their assistance teaching me proper horsemanship; I have become a better rider because of them. The utmost appreciation goes to my loving and supportive husband, Bradley Dilling. I thank Brad for helping me with fieldwork, getting me to finally write this thesis and keeping me focused when times got tough. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii LIST OF FIGURES...........................................................................................................ix ABSTRACT .......................................................................................................................xi CHAPTER 1 LITERATURE REVIEW.............................................................................................1 Taxonomy.....................................................................................................................2 Morphology and Life Cycle..........................................................................................3 Behavior and Ecology...................................................................................................5 Flight Categories...................................................................................................6 Meteorological Conditions Affecting Flight.........................................................7 Feeding Behavior...................................................................................................8 Host Location Behavior.........................................................................................9 Host Preference...................................................................................................11 Reproduction Behavior........................................................................................12 Seasonality..................................................................................................................12 Health and Economic Impact......................................................................................13 Control........................................................................................................................18 Chemical and Biological Mosquito Control........................................................18 Nonchemical Mosquito Control and Surveillance..............................................20 Summary.....................................................................................................................25 2 SEASONALITY OF MOSQUITOES AT AN EQUINE FACILITY IN NORTH CENTRAL FLORIDA................................................................................................32 Introduction.................................................................................................................32 Materials and Methods...............................................................................................33 Experimental Design..................................................................................................34 Data Analysis..............................................................................................................34 Results.........................................................................................................................34 Discussion...................................................................................................................35 Conclusions.................................................................................................................36 v

PAGE 6

3 MOSQUITO TRAPPING STUDIES AT AN EQUINE FACILITY IN NORTH CENTRAL FLORIDA................................................................................................54 Introduction.................................................................................................................54 Materials and Methods...............................................................................................55 Experimental Design..................................................................................................56 CDC 1012 Trapping Study..................................................................................56 MMPro Trapping Study......................................................................................57 Horse Odor Study................................................................................................57 Location Profile Study.........................................................................................58 Horse Vacuuming Study......................................................................................59 Separate Entity Study..........................................................................................59 Data Analysis.......................................................................................................60 Results.........................................................................................................................60 MMPro Trapping Study......................................................................................61 Horse Odor Study................................................................................................62 Location Profile Study.........................................................................................63 Horse Vacuuming Study......................................................................................63 Separate Entity Study..........................................................................................64 Discussion...................................................................................................................65 CDC 1012 and MMPro Trapping Studies...........................................................65 Horse Odor Study................................................................................................67 Location Profile Study.........................................................................................67 Horse Vacuuming Study......................................................................................68 Separate Entity Study..........................................................................................70 Conclusions.................................................................................................................71 4 SUMMARY..............................................................................................................113 LIST OF REFERENCES.................................................................................................115 BIOGRAPHICAL SKETCH...........................................................................................121 vi

PAGE 7

LIST OF TABLES Table page 1-1 Classification of family Culicidae..............................................................................30 1-2 Systemic list of mosquitoes found in Florida.............................................................31 2-1 Total count, and percent of total count of mosquito species trapped during seasonality study......................................................................................................52 2-2 Mean numbers ( standard deviation) of mosquitoes trapped each month ...............53 3-1 Schedule of CDC 1012 trapping study ....................................................................101 3-2 Schedule of MMPro trapping study..........................................................................102 3-3 Interval schedule of horse odor study.......................................................................103 3-4 Schedule of horse odor study....................................................................................104 3-5 Schedule of separate entity study stall assignments ................................................105 3-6 Mean numbers ( standard deviation) of mosquitoes captured during CDC 1012 trapping study.........................................................................................................106 3-7 Mean numbers ( standard deviation) of mosquitoes captured during MMPro trapping study.........................................................................................................107 3-8 Mean numbers ( standard deviation) of mosquitoes captured during horse odor study.......................................................................................................................108 3-9 Mean numbers ( standard deviation) of mosquitoes captured during location profile study.......................................................................................................................109 3-10 Mean numbers of mosquito species trapped per position during location profile study.......................................................................................................................1 10 3-11 Mean numbers ( standard deviation of difference of mosquitoes vacuumed from horse 1 and horse 2 during vacuuming study.........................................................111 vii

PAGE 8

3-12 Mean numbers ( standard deviation) of total mosquitoes vacuumed from horse 1 and horse 2 during vacuuming study......................................................................112 viii

PAGE 9

LIST OF FIGURES Figure page 1-1 CDC 1012 (John W. Hock Company, Gainesville, FL) mosquito trap......................27 1-2 Mosquito Magnet X (American Biophysics Corporation, East Greenwich, RI) mosquito trap............................................................................................................28 1-3 Mosquito Magnet Pro (American Biophysics Corporation, East Greenwich, RI) mosquito trap............................................................................................................29 2-1 Weather station at UF Horse Teaching Unit..............................................................39 2-2 Aerial photograph of the UF Horse Teaching Unit showing the four MMPro trap sites for the seasonality study...................................................................................40 2-3 Total mosquito counts as related to months during the seasonality study..................41 2-4 Total mosquito counts as related to rainfall during the seasonality study..................42 2-5 Maximum and Minimum temperatures recorded at the meteorological station during the seasonality study.................................................................................................43 2-6 Total rainfall (centimeters) recorded during the seasonality study............................44 2-7 Total numbers of Culex nigripalpus trapped by MMPro during seasonality study...45 2-8 Total numbers of Culex erraticus trapped by MMPro during seasonality study.......46 2-9 Total numbers of Culex salinarius trapped by MMPro during seasonality study......47 2-10 Total numbers of Mansonia spp. trapped by MMPro during seasonality study.......48 2-11 Total numbers of Anopheles spp. trapped by MMPro during seasonality study......49 2-12 Total numbers of Coquillettidia perturbans. trapped by MMPro during seasonality study.........................................................................................................................50 2-13 Total numbers of Psorophora spp. trapped by MMPro during seasonality study...51 3-1 Portable vacuum aspirator (DC Insect Vac. BioQuip, Rancho Dominguez, CA) used for collection of mosquitoes during vacuuming studies...........................................73 ix

PAGE 10

3-2 Paint gelding that was sampled for mosquitoes during trapping studies....................74 3-3 Appaloosa mare that was sampled for mosquitoes during trapping studies...............75 3-4 CDC 1012 trap placement during CDC 1012 trapping study.....................................76 3-5 CDC 1012 trapping study placement used for mosquito trapping.............................77 3-6 MMPro trap placement during MMPro trapping study..............................................78 3-7 MMPro trap placement used during mosquito trapping study...................................79 3-8 Modified CDC 1012 mosquito trap used during horse odor study.............................80 3-9 CDC 1012 trap layout used for mosquito location profile study................................81 3-10 Technique of aspirating mosquitoes off of horse during horse vacuuming study and separate entity study.................................................................................................82 3-11 Feeding slips at UF Horse Teaching Unit where the 2nd trial of horse vacuuming study and separate entity study was conducted........................................................83 3-12 Distribution and mean numbers of mosquitoes trapped during CDC 1012 trapping study.........................................................................................................................84 3-13 Total numbers of mosquitoes captured during CDC 1012 trapping study...............85 3-14 Mean numbers of mosquitoes captured during CDC 1012 trapping study..............86 3-15 Distribution and mean numbers of mosquitoes trapped during MMPro trapping study.........................................................................................................................87 3-16 Total numbers of mosquitoes trapped during MMPro trapping study.....................88 3-17 Mean numbers of mosquitoes trapped during MMPro trapping study.....................89 3-18 Distribution and mean numbers of mosquitoes trapped during horse odor study....90 3-19 Mean numbers of mosquitoes trapped during horse odor study...............................91 3-20 Mean numbers of mosquitoes trapped during the different treatments of horse odor study.........................................................................................................................92 3-21 Percent of mosquito species trapped during location profile study..........................93 3-22 Distribution and mean numbers of mosquitoes trapped per position during location profile study..............................................................................................................94 3-23 Mean numbers of mosquitoes trapped per position during location profile study...95 x

PAGE 11

3-24 Percent of mosquito species captured during vacuuming study, trial 1...................96 3-25 Percent of mosquito species captured during vacuuming study, trial 2...................97 3-26 Percent of mosquito species captured from Appaloosa mare during vacuuming study.........................................................................................................................98 3-27 Percent of mosquito species captured from Paint gelding during vacuuming study.99 3-28 Distribution and mean numbers of total mosquitoes captured during separate entitiy study.......................................................................................................................100 xi

PAGE 12

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 EVALUATION OF MOSQUITO TRAPPING EFFICIENCY AND DETERMINATION OF SEASONALITY FOR MOSQUITOES AT THE UNIVERSITY OF FLORIDA HORSE TEACHING UNIT By Sarah Courtney Dilling December 2004 Chair: Saundra H. TenBroeck Major Department: Animal Sciences Traps are effective surveillance tools for monitoring seasonal prevalence and the species composition of mosquitoes, along with reducing mosquito numbers nearby. There is little research to study the efficiency of mosquito traps in a competitive environment with a natural host. Such research is warranted because these pests can cause substantial economic losses to the equine industry through nuisance biting and disease transmission. A series of studies were conducted to achieve this objective, including competitive studies with traps and a horse acting as a natural host, determining prevalent species feeding on a horse, location profiles of mosquitoes through trapping, adding horse odors to the trap airstream, and determining the distance required between two horses to achieve separate entities. Seasonal population trends were evaluated, along with temperature and rainfall. Trends of mosquito populations were monitored using the Mosquito Magnet Pro (MMPro) (American Biophysics, Corp., North Kingston, RI), xii

PAGE 13

trapping systems from September 2003 through September 2004. Peaks in mosquito populations correlated with changes in temperature and rainfall, the highest occurring from August to October. Mosquito location profile through trapping was evaluated using three CDC 1012 traps baited with CO2. There were significantly (P 0.05) higher mosquitoes captured in the trap closest to a body of water, when compared to traps in open pasture. Studies evaluated competitive trials with traps and a horse acting as a natural host. The CDC 1012 trapping study was conducted using a CO2 baited CDC 1012 trap (John W. Hock Company, Gainesville, FL). The MMPro trapping study was conducted using a CO2, heat and moisture baited MMPro trap. The horse odor study was conducted using a CO2 and equine odor baited modified CDC 1012 trap. A trap was set up next to a stall for five 20-minute intervals starting approximately 30 minutes after sunset. There were two treatment groups, horse present and no horse present for studies CDC and MMPro; and one additional treatment group horse being vacuumed for odor study. When a horse was placed in the stall, the mosquitoes trapped significantly (P 0.05) decreased. The quantity of mosquitoes feeding on two horses at a given time period of time was examined using a portable vacuum aspirator (DC Insect Vac. BioQuip, Rancho Dominguez, CA). There was a significant difference (P 0.05) in mosquitoes captured between horses and seasons. The distance required for mosquitoes to distinguish two horses as separate entities was evaluated using a vacuum aspirator. The horses were separated by predetermined distances and vacuumed for 20-minute intervals. There was no significant difference of the mean total mosquitoes vacuumed between horses per distance. xiii

PAGE 14

CHAPTER 1 LITERATURE REVIEW Mosquitoes have been around since the beginning of time and have survived years of changing conditions that humans could never withstand. Mosquito control is a major entomological concern worldwide, so why is it necessary to control mosquitoes? The two main reasons are, to prevent or reduce nuisance biting and preclude the spread of mosquito-borne diseases (Dwinell et al., 1998). The number of people worldwide affected by these diseases is staggering. Each year 300 to 500 million cases of malaria are reported, resulting in 1.5 to 2.7 million deaths (Centers for Disease Control and Prevention, 2004). Malaria is not common in the United States, but recently the introduction and spread of West Nile Virus (WNV) has become a major concern. In 2003, 9,862 human cases and 5,181 equine cases were reported (Stark and Kazanis, 2003). The most serious consequence of WNV infection is fatal encephalitis in humans and horses, and mortality in certain domestic and wild birds. The state of Florida has an active equine industry, where some of the top thoroughbreds in the nation are bred and trained. Since the mortality rate of horses infected with WNV is around 30%, the threat of this arbovirus is tremendous (Porter et al., 2003). Mosquitoes also pose a nuisance factor to equine and livestock industries, causing reduced feed conversion efficiency, weight gain reductions, and decreased milk yield (Steelman, 1979, and Byford et al., 1992). Florida has been working since the early 1900s to control the threat of mosquito-borne diseases through surveillance and chemical spraying. These methods are still used today, but chemical control may be greatly restricted in the future due to environmental 1

PAGE 15

2 effects, insecticide resistance and health concerns (Kline and Mann, 1998). New trapping innovations have given mosquito control agencies and backyard enthusiasts a way to safely and effectively control the nuisance mosquitoes and possibly provide an accurate picture of local populations. Mosquito traps are becoming a popular merchandizing commodity, but their effectiveness could still be greatly enhanced in the future. Knowing the biology and natural behavior of the mosquito is essential to effectively controlling this pest. Extensive research involving humans and livestock animals concerning mosquito attraction has been conducted, but little has been done with horses. In Florida, the equine industry is a multi-billion dollar business, and mosquito control is essential. Evaluating and understanding natural behavior of mosquitoes as they interact with horses could provide valuable information about trapping and control. This chapter is a review of selected literature on the taxonomy, morphology, behavior, ecology, seasonality, public health importance, and potential control of mosquitoes found in Gainesville, Florida. Taxonomy Mosquito is a Spanish word meaning little fly, and its use dates back to about 1583; in England they are known as gnats (Spielman and D'Antonio, 2001). Mosquitoes are insects that belong to the order Diptera and family Culicidae. Culicidae consists of about 3200 recognized species. Currently Culicidae are classified into three subfamilies: Anophelinae, Culicinae, and Toxorhynchitinae (Table 1-1). There are slight taxonomic differences between the three subfamilies, mostly during their larval stages. Anophelinae is the most distinct group compared to the other two subfamilies. Their eggs have floats, larvae lack air tubules, and adults have characteristic palps that are the same length as the proboscis. Members belonging to the subfamilies Culicinae and Toxorhynchitinae,

PAGE 16

3 during their larval stages have, air tubules and the palps of all adult females are significantly shorter than their proboscis. Toxorhynchitinaes larvae are all predaceous and adults are quite large in size. They also have a characteristic proboscis, which is curved and has been adapted for feeding only on nectar (Woodbridge and Walker, 2002). There are 38 genera of mosquitoes worldwide, 13 of these encompassing 77 species are found in Florida (Table 1-2): Anopheles (Meigen), Aedes (Meigen), Ochlerotatus (Lynch Arribalzaga), Psorophora (Robineau-Desvoidy), Culex (Linnaeus), Deinocerites (Theobald), Culiseta (Felt), Coquillettidia (Dyar), Mansonia (Blanchard), Orthopodomyia (Theobald), Wyeomyia (Theobald), Uranotaenia (Lynch Arribalzaga) and Toxorhynchites (Theobald) (Public Health Entomology Research and Education Center, 2002). Morphology and Life Cycle The mosquito goes through four separate and distinct stages in its life cycle: egg, larva, pupa, and adult. The eggs of most mosquitoes are found in various shapes including elongate, ovoid, spindle, spherical and rhomboid. Eggs of Anopheles, Toxorhynchites, Wyeomyia, Aedes, Ochlerotatus, Psorophora, and Haemagogus species are laid individually, whereas in Culex, Culiseta, Coquillettidia, and Mansonia species, they are attached together in a single clump, forming a floating egg raft or a submerged cluster (Woodbridge and Walker, 2002). Approximately 2 to 3 days after a female has taken a blood meal an average of 75 eggs per ovary develop. Culex, Culiseta, and Anopheles lay their eggs on the water surface while many Aedes and Ochlerotatus lay their eggs on damp soil that will be flooded by water. Most eggs hatch into larvae within 48 hours; others might withstand subzero winters before hatching (Harwood and James, 1979).

PAGE 17

4 Mosquito larvae have three distinct body regions: head, thorax and abdomen. The head is usually broad and flattened with lateral antennae. Mouthparts usually consist of brushes and grinding structures that filter bacteria and microscopic plants, however some larvae are predaceous and will grasp their prey. The thorax is broader than the head and somewhat flattened. The structure and number of hairs on both the head and thorax aid in identification of species. The abdomen is elongated and cylindrical, consisting of nine well defined segments. The first seven segments are similar to each other, but the last two are modified with specific structures. The eighth segment contains the respiratory opening, and in most species this is a siphon. The ninth segment is the anal segment. During larval development mosquitoes pass through four instars, and at the end of each one they molt and increase in size. The average size of a fourth-instar larva is 6.35 to 12.7 millimeters in length. Depending on temperature and other environmental factors, mosquito species require about 7 days to complete larval development. At the end of the fourth instar, larvae molt again and become pupae (Ogg, 2002). The pupal stage of development prepares the juvenile mosquito to become an adult. The pupa is shaped like a comma and has hard scales made of chitin that protect the body. Although the pupae are non-feeding, they are mobile and often called tumblers. When pupae are disturbed, they will move in a jerking, tumbling motion toward protection and then float back to the surface. The pupae are less dense than water, so they float on the surface and receive oxygen through two breathing tubes called trumpets (Woodbridge and Walker, 2002). If the ninth segment of the pupas abdomen is examined along with the overall size of the pupa, the sex can be determined. The ninth segment on male mosquitoes is more prominent during this stage, while the female pupa

PAGE 18

5 is usually larger in size than that of the male. In Florida, larvae can pupate in water temperatures of 17C to 35C for a total period of 1 to 4 days; however, temperatures above or below these cause increased mortality in the population (Nayar, 1968). The metamorphosis of the mosquito into an adult is completed within the pupal case. The adult mosquito splits the pupal case and emerges to the surface of the water where it rests until its body dries and hardens (Floore, 2003). Bodies of adult mosquitoes are slender, with thin narrow legs, and elongated wings. The body surface is covered with scales and setae providing characteristic markings and colors for identification. The long and filamentous antennae arise between the eyes, and are usually sexually dimorphic. The prominent proboscis of the adult female projects anteriorly at least two-thirds the length of the abdomen (Woodbridge and Walker, 2002). Only female mosquitoes feed on blood, which is essential for egg production. Females feed on animals-warm or cold blooded-and birds. Stimuli that influence biting (blood feeding) include a combination of carbon dioxide, temperature, moisture, smell, color and movement (Floore, 2003). Male mosquitoes do not bite, but feed on the nectar of flowers or other suitable sugar sources. Females also feed on nectar for flight energy. During the summer, adult mosquitoes have a life span of a few weeks. However, it has been found that some species can spend the winter as adults, and can therefore have a life span of several months (Nasci et al., 2001). Behavior and Ecology Once the mosquito has emerged from its pupal case, it will seek shelter where it will rest and await activity periods. Every mosquito species has a characteristic pattern of daily activity, which is intrinsically known through the natural circadian rhythm of the daily light-dark cycle. Generally a mosquito will take flight during one or two periods

PAGE 19

6 each day, depending on whether the specific species of mosquito is characterized as being diurnal, nocturnal or crepuscular. During these periods, both male and female mosquitoes will take flight without external cues (Woodbridge and Walker, 2002). Generally mosquitoes do not actively fly over ranges greater than 2 kilometers. Yet some species of mosquitoes, such as the salt-marsh mosquito, Ochlerotatus taeniorhynchus (Wiedemann), will travel long distances by wind and will be carried hundreds of kilometers from their origins. Ochlerotatus taeniorhynchus emerge in remote salt-marsh locations where hosts are sparse, so they will make extended round trip migrations to complete their life cycle while ovipositing at their original breeding sites (Woodbridge and Walker, 2002). Flight Categories Mosquito flights can be classified into three categories: migratory, appetential or consumatory (Bidlingmayer, 1985). A migratory flight lacks an objective, does not meet any individual need (Provost, 1952), and is only a one-way flight with no return. Generally only newly emerged mosquitoes venture out with a migratory flight. The direction of migration is dependant on wind conditions at the time of departure and duration is limited by the mosquitos energy bank reserves and the meteorological conditions during flight; the destination is accidental (Bidlingmayer, 1985). An adult mosquito will respond to a physiological stimulus by taking an appetential flight. When a resting mosquito is in need of a blood meal, an oviposition site, or a better resting place, it will begin a searching flight for this need. During an appetential flight, the appropriate sense organs (olfactory, visual, thermal, auditory, or humidity receptors) will be alert for cues that indicate the presence of the objective, and this flight will end when the objective is located, or continue until energy reserves are depleted (Bidlingmayer, 1985). Once the

PAGE 20

7 objective is located the next flight category, the consumatory flight, begins. This flight is direct and brief, since visual and biochemical cues operate only over distances short (Bidlingmayer, 1985). If the cue encountered was olfactory, a direct upwind flight is conducted until other cues, visual perception, movement or thermal, enable the female to locate her goal more precisely (Gillies and Wilkes, 1972). Meteorological Conditions Affecting Flight Meteorological conditions greatly influence mosquito flight; the most influential factors are light, temperature, humidity and wind (Day and Curtis, 1989). Dry windy conditions can completely inhibit mosquito flight, especially during the winter months in Florida (Day and Curtis, 1989). Nightly variations in wind, rainfall, and relative humidity influence mosquito flight patterns and possibly feeding success. During late summer and fall, daily rainfall patterns can potentially influence whether the mosquito population continues to build, remains constant, or declines by affecting feeding and oviposition behaviors (Day and Curtis, 1989). In Florida, research conducted by the Middlesex County Mosquito Extermination Commission indicated that most mosquito species possessed a bimodal flight activity pattern during the night, with the larger peak occurring soon after sunset and the smaller peak just prior to dawn (Schmidt, 2003). These are times of rapidly changing light levels. Temperature and relative humidity greatly influence the flight behavior and activity of mosquitoes, but optimal conditions vary between species. Many researchers disagree considerably as to what are optimal conditions for flight (Rowley and Graham, 1968). In Florida, Bradley and McNeel (1935) and Bidlingmayer (1974) found that temperatures below 21C and 19C respectively, reduced trap catches. After temperatures have risen above the minimum

PAGE 21

8 flight threshold for individual mosquito species, higher temperatures do not affect flight (Taylor, 1963). Optimal relative humidity for flight is also greatly disagreed upon. Rowley and Graham (1968) found that relative humidity between 30 and 90% had no demonstrable influence on flight performance. Mosquito cruising speeds are generally less than 1 meter per second but flight activity is greatly reduced if winds speeds exceed flight speeds (Grimstad and DeFoliart, 1975). Mosquito activity can be forecasted using the four meteorological factors listed above. American Biophysics Corporation, manufacturer of the Mosquito Magnet, has teamed with The Weather Channel, which displays on its website, www.weather.com the first-ever Mosquito Activity Forecast. The Mosquito Activity Forecast, developed and managed by a team of meteorologists from The Weather Channel, provides hourly predictions of mosquito activity nationwide. This information is very useful to people who want to take part in outdoor activities in areas prone to high incidence of mosquito vector-borne disease. Feeding Behavior Adult mosquitoes of both sexes of most species feed regularly on plant sugar (nectar) throughout their lives, but only females feed on hosts for a blood meal. Females of some mosquito species feed on sugar infrequently or never [e.g. Ae. aegypti (Linnaeus) and An. gambiae (Meigen)], and utilize blood for both energy and reproduction. Females of some mosquito species feed entirely on plant sugar, such as Toxorhynchites, and do not require a blood meal for egg development. Generally, however, a blood meal is required by female mosquitoes to obtain protein from the blood to develop eggs (Woodbridge and Walker, 2002).

PAGE 22

9 Host Location Behavior Generally 1-3 days after the emergence of the female mosquito, she will look for a host from which to feed. Research has long been conducted to determine why mosquitoes are attracted to certain hosts and what attractants are responsible for the mosquitos odor mediated behavior. In 1942, researchers showed that unwashed naked children were more attractive to species of the genus Anopheles than naked children who had washed. The same group also showed that dirty clothes in a hut attracted more mosquitoes than an empty hut (Haddow, 1942). Individual variation in attractiveness to mosquitoes was shown in 1965, when Khan and his associates were able to isolate one person who was very attractive and 3 people who were very unattractive to Ae. aegypti (Khan et al., 1965). Mosquitoes seeking a host are exposed to a wide variety of visual, olfactory, and physical stimuli. Any one or combination of these stimuli could act as cues for host identification and location. Host selection is mainly determined by host preference and availability, but stimuli that the mosquito detect help locate the host. Some of the best-documented olfactory attractants are carbon dioxide (CO2), lactic acid, and octenol. Generally, carbon dioxide is universally attractive to mosquitoes, and is probably the best understood of the volatile host cues (Gibson and Torr, 1999). Some researchers believe that carbon dioxide acts as an attractant, which mediates orientation towards a host, and can exhibit a synergistic response with other host odors (Gillies, 1980). Gillies (1980) also found that the carbon dioxide and whole-body odors have an orientating effect of variable extent when presented singly and a greater enhanced effect when presented together. It has been shown that mosquito light traps baited with CO2 capture 8-30 times more mosquitoes than traps without CO2 (Kline and Mann, 1998). Although it has been concluded that CO2 increases mosquito catches in traps, it was

PAGE 23

10 found that CO2 appears to be of little importance in host discrimination by mosquitoes (Mboera and Takken, 1997). It is believed that one of the volatile compounds mosquitoes use to discriminate hosts is lactic acid. Lactic acid is a by-product of anaerobic metabolism common to all animals and humans. Skin emanations are important, because odors from live hosts are always more attractive than any combination of these chemicals in a warm, humid airstream (Woodbridge and Walker, 2002). 1-octen-3-ol (octenol) is another olfactory attractant documented as an effective mosquito attractant. In 1984 Hall et al., through studying the attractiveness of oxen to Tsetse flies, discovered octenol and isolated it the breath of oxen in Africa. Field tests have demonstrated that octenol serves as a powerful attractant for certain species of mosquitoes and flies (Kline, 1994). Currently manufacturers of commercial mosquito traps, such as American Biophysics Corporation, makers of the Mosquito Magnet Pro, suggest the use of octenol as a supplementary additional bait to trap mosquitoes. Visual attraction of hosts to mosquitoes has been thoroughly documented. Adult mosquitoes possess two compound eyes and two ocelli. Compound eyes are used for navigation and sensing movement, patterns, contrast, and color, while ocelli are believed to sense light levels and possibly polarized light (Allan et al., 1987). The compound eyes of adult mosquitoes have relatively poor resolution but overall high light sensitivity (Muir et al., 1992). It has also been reported that diurnal species respond to visual characteristics of hosts, such as color, brightness, pattern, and movement (Allan et al., 1987). Field trials conducted to investigate color affinity of Mansonia mosquitoes showed that Mansonia have marked attractiveness towards blue and red followed by white, yellow and green, with the least numbers attracted by black (Bhuyan and Das,

PAGE 24

11 1985). Movement may also play a role in host location by mosquitoes, and a consistently small but positive attraction has been affirmed (Wood and Wright, 1968). Once the female is within 1 meter of a host, convective heat and humidity become the main attractant opposed to chemical or visual stimuli (Woodbridge and Walker, 2002). Mosquitoes are also attracted by physical stimuli like temperature and humidity. A source of heat has shown positive attractiveness from some species of mosquitoes (Howlett, 1910). Brown (1951) showed that mosquitoes landed three times as often on a clothed robot when the robots skin temperature was 98F than when the skin temperatures were 50-65F (Brown, 1951). He also noted that moisture coming off the robots clothing increased the number of landings by 2 to 4 times, but only at temperatures above 60F. It has not been determined what single stimulus causes mosquitoes to locate and feed on a host, but it has been determined that a combination of visual, olfactory and physical stimuli are effective as attractive factors. Host Preference Host preference varies widely between different genera of mosquitoes, changes within a genus depending on geographic location. Some species feed almost entirely on members of one genus of host animal; others opportunistically attack members of two or three vertebrate classes. Mosquitoes belonging to the genus Culex are primarily avian feeders, but if the population of birds is insufficient or unavailable, Culex will contentedly feed on mammals (Braverman et al., 1991). Some Florida mosquitoes prefer to feed on mammals, including Aedes, Anopheles, Coquillettidia, Mansonia and Psorophora. However, the genus of animal differs between mosquito genera, and if given the opportunity, they will also take a blood meal from a bird (Edman, 1971). Livestock

PAGE 25

12 are exposed to mosquitoes in very high numbers during most of the year in Florida; because some of these species are known disease vectors, this causes concern. When trying to understand and control the transmission of vector-borne diseases, knowledge of the feeding behavior of mosquitoes is of prime importance (Defoliart et al., 1987). Reproduction Behavior Mating usually occurs a few days after adult emergence. The males generally form flight swarms around the females preferred host. When a female enters a swarm, males detect the characteristic frequency of her wing beat and position with their plumose antennae and Johnstons organs. The male locates the female, pursues her, and mates with her. If the female is of another species, males either do not respond to her flight tone or release her upon detection that she lacks the appropriate species-specific contact pheromones. There are some exceptions to this mating ritual. Male Deinocerites guard pupae at the water surface and mate with the females as they emerge (Woodbridge and Walker, 2002). Seasonality In Gainesville, Florida, mosquitoes are present twelve months a year (Crowley, 2003), yet species are seasonally dependant. Identification of population trends of mosquitoes is essential for developing appropriate control methods and disease prevention. In Florida, horse owners can base mosquito-borne disease vaccinations, such as the vaccine for West Nile Virus, depending on seasonality trends of Culex nigripalpus. For example, Fort Dodge Animal Health suggests vaccinating horses against West Nile Virus in March and August. Mosquito population variations are closely linked to rainfall and temperature. In Gainesville, the highest numbers of mosquitoes are generally trapped in August, and the lowest are generally in January

PAGE 26

13 (Crowley, 2003). Population peaks usually occur 2-3 weeks following a heavy rainfall (Crowley, 2003). Mosquito species seasonality even varies within the same genus, for instance, Culex nigripalpus (Theobald) is generally abundant during the late summer and early winter seasons while Culex salinarius (Coquillett) is abundant during the late winter until early summer seasons (Zyzak et al., 2002). Culex nigripalpus is a species that thrives under the hot, humid conditions commonly reported in Florida, while Culex salinarius is a species that is most abundant in Florida during cool, dry months. Two species of Mansonia are commonly trapped in Gainesville, including Ma. dyari (Belkin, Heinemann and Page) and Ma. titillans (Walker), but their seasonality trends differ slightly. Ma. dyari are generally abundant mid-summer until late-fall, while Ma. titillans are abundant late summer until early spring. Coquillettidia perturbans (Walker) is found most of the year between early spring until late fall (Slaff and Haefner, 1985). Health and Economic Impact Mosquitoes are vectors for many diseases that cause millions of human and animal death every year. Mosquitoes are the sole vectors of pathogenic organisms causing human malaria, yellow fever, and dengue fever. They are also of prime importance in the transmission of diseases like filariasis and viral encephalitides of man. Every year there are between 300-500 million cases of humans infected with malaria, resulting in 1.5 to 2.7 million deaths. Patients with malaria typically are very sick with high fevers, shaking chills, and flu-like illness. Even though malaria was eradicated from the United States in the 1950s there are approximately 1,200 cases of malaria diagnosed in the United States each year. Most of these cases are from people that traveled and returned from malaria-risk areas, such as Africa. The mosquito that transmits malaria, Anopheles, is found throughout much of the United States. This causes concern because it is possible that a

PAGE 27

14 person who entered the United States from Africa, carrying the plasmodia responsible for Malaria, could infect native Anopheles quadrimaculatus and this mosquito could transmit the disease in the United States (Centers for Disease Control and Prevention, 2004). Dengue fever and filariasis are also very dangerous diseases that occur in the tropics. Most dengue infections result in relatively mild illness, but some can progress to dengue hemorrhagic fever. With dengue hemorrhagic fever, the blood vessels start to leak and cause bleeding from the nose, mouth, and gums. Bruising can be a sign of bleeding inside the body. Without prompt treatment, the blood vessels can collapse, causing shock (dengue shock syndrome). Dengue hemorrhagic fever is fatal in about 5% of cases, mostly among children and young adults. Primary vectors of dengue virus are Aedes albopictus (Skuse) and Aedes aegypti. Yearly cases are found as close to the United States as Mexico, therefore this virus is a threat to the US, because we have a high population of competent Aedes mosquitoes (Centers for Disease Control and Prevention, 2003a). Lymphatic filariasis, the second leading cause of permanent and long-term disability worldwide is caused by parasitic nematodes, which affects over 120 million people worldwide (Centers for Disease Control and Prevention, 2003B). Culex and Mansonia mosquitoes are the primary vectors of filariasis, which transmit the filariasis parasitic nematode, Wuchereria bancrofti and Brugia malayi. Lymphatic filariasis is rarely fatal, but it can cause recurring infections, fevers, severe inflammation of the lymph system, and a lung condition called tropical pulmonary eosinophilia. In about 5% of infected persons, a condition called elephantiasis causes the legs to become grossly swollen. This can lead to severe disfigurement, decreased mobility, and long-term disability (Centers for Disease Control and Prevention, 2003b).

PAGE 28

15 Even though malaria, dengue fever and filariasis are not common in the United States, all of the primary mosquito species are present. The mosquitoes that are competent to transmit those diseases are Aedes, Anopheles, Mansonia, and Culex. The composition of mosquito species that live in the United States are potential vectors of many diseases, therefore we are not completely safe from the diseases that are concentrated in the tropics. The United States is not completely free from mosquito-borne diseases. Arboviral encephalitides cause deaths not only in humans, but also birds, livestock and horses. Arboviral encephalitides have a global distribution, but there are five main virus agents of encephalitis in the United States: eastern equine encephalitis (EEE), western equine encephalitis (WEE), St. Louis encephalitis (SLE), La Crosse encephalitis (LAC), and the newly introduced West Nile virus (WNV), all of which are transmitted by mosquitoes (Centers for Disease Control, 2001). Arboviral viruses, short for arthropod-borne viruses, are maintained in nature through biological transmission between susceptible vertebrate hosts by mosquitoes. Arboviruses that cause human encephalitis are members of three virus families: the Togaviridae, Flaviviridae, and Bunyaviridae. The majority of human infections are asymptomatic or may result in flu-like symptoms. In some cases infection leads to encephalitis, swelling of the brain, with a fatal outcome or permanent brain damage. Eastern equine encephalitis (EEE) has been affecting North American horses since 1830s (Mitchell et al., 1985). At that time viruses were unknown until the mid-20th century, the cause of this disease was unknown. The disease was not officially named until a major outbreak occurred in horses in coastal areas of Delaware, Maryland, New Jersey, and Virginia in 1933 (TenBroeck et al. 1935). Outbreaks of EEE are infrequent,

PAGE 29

16 animals are naive to the disease and are susceptible, which causes a significant economic and social impact once an outbreak has occurred. The first time EEE occurs in an area, there are illness and death of horses, poultry, and humans (Centers for Disease Control and Prevention, 2001). Since 1934 there have been 153 confirmed cases of EEE in humans in the United States (Centers for Disease Control and Prevention, 2001. Approximately one-third of all people with clinical encephalitis caused by EEE will die from the disease, and those that recover may suffer from permanent brain damage. There are many known mosquito vectors of EEE in the United States, including Culiseta melanura (Coqullett), Aedes vexans (Meigen), Ochlerotatus sollicitans (Walkeri), Coquilletidia perturbans; there are also potential vectors of EEE, such as Aedes albopicus, Aedes canadensis (Meigen), Culex nigripalpus, and Culex salinarius (Woodbridge and Walker, 2002). Western Equine Encephalitis (WEE) was first isolated in California in 1930 from the brain of an infected horse. Today WEE is still an important cause of encephalitis in horses and humans, mainly in the western United States. The symptoms are similar to EEE, and the mortality rate of WEE is approximately 3% (Centers for Disease Control and Prevention, 2001). The major WEE vector in the United States, Culex tarsalis (Coquillett), but Aedes melanimon and Culiseta inornata (Williston) have been identified as potential vectors (Woodbridge and Walker, 2002). St. Louis Encephalitis (SLE) is the most common mosquito-transmitted human pathogen in the United States. It is widely distributed in the lower 48 states. Since 1964, there have been 4,437 confirmed cases of SLE with an average of 193 cases per year. The symptoms are similar to EEE, and the mortality rate of SLE is approximately 5-15%

PAGE 30

17 (Centers for Disease Control and Prevention, 2001). Culex is the main vector for SLE in the United States, however some other confirmed vectors are Cx. tarsalis, Cx. pipiens, Cx. quinquefasciatus (Say), and C.x nigripalpus. The remainder of Culex species, Cx. restuans, Cx. salinarius are potential vectors (Woodbridge and Walker, 2002). La Crosse Encephalitis (LAC) was discovered in Wisconsin in the 1960s. Since then LAC has been identified in 11 states, mostly in the Midwestern region, but also in Virginia, North Carolina, Alabama, and Mississippi. An average of 75 cases are reported in the United States every year. The symptoms are similar to EEE, and the mortality rate of SLE is approximately less than 1% (Centers for Disease Control and Prevention, 2001). Ochlerotatus triseriatus is the primary vector of LAC in the United States, but several species of mosquitoes have been identified as potential vectors: Ae. canadensis, Ae. vexans, Oc. sollicitans (Walkeri), Oc. taeniorhynchus, and Ae. albopictus (Woodbridge and Walker, 2002). Since the early 1950s, West Nile Virus (WNV) has affected people in African, Middle Eastern and some Mediterranean countries, usually causing epidemics every 10 years (Malkinson et al., 2002). This arbovirus had never been reported in North America until 1999, when it was isolated from birds found in New York City (NYC). There was a high degree of similarity among the various strains circulating throughout NYC and surrounding counties, which indicated that a single WNV strain (WN-NY99) had been introduced (Lanciotti, 1999). When WN-NY99 was isolated and compared with various other West Nile Virus strains, a significant similarity was found between it and the WNV strain WN-Israel 1998 isolated in Israel in 1998 (Lanciotti, 1999). The virus rapidly

PAGE 31

18 spread throughout the United States in less than 2 years, producing a mortality rate of 30% in infected horses. This is a significant threat to the bustling equine industry in the state of Florida, where many of the top horses are bred and trained. Mosquitoes that belong to the genus Culex are the main vectors of WNV, the species include Cx. pipiens, Cx. salinarius, and Cx. tarsalis. Several genera have been identified as potential vectors of WNV: Aedes, Anopheles, Coquillettidia, Culiseta, Culex, Deinocerites, Ochlerotatus, Orthopodomyia, Psorophora, and Uranotaenia (Woodbridge and Walker, 2002). Control To control mosquitoes and the public health hazards they present, many states and localities have established mosquito control programs. These programs consist of three main categories, chemical, biological, and physical control along with surveillance methods to monitor mosquito populations. With more than 1,197 miles of coastline, a warm subtropical climate, and heavy rainfall, Florida produces an unusually rich fauna, including 77 species of mosquitoes. The Florida Anti-Mosquito Association (FAMA) now known as the Florida Mosquito Control Association (FMCA), was formed in 1922. Shortly thereafter the legislature created the first mosquito control district (Indian River,1925). Since then Florida has had an active mosquito control program focusing on surveillance and population control. Chemical and Biological Mosquito Control Treatment of adult mosquitoes is the most visible practice exercised by mosquito control operations. Every mosquito control district in Florida uses adulticides, applied either aerially or with ground equipment. Adulticiding is usually the least efficient mosquito control technique. Some of the approved adulticides include, malathion, permethrin, resmethrin, fenthion, naled, and chlorpyrifos (Dwinell et al., 1998). Some

PAGE 32

19 Florida mosquitoes are resistant or more tolerant to some adulticides, thus affecting the adulticide selection. All insecticide applications must be made during periods of adult mosquito activity, this also decreases the chance of affecting non-target insects such as bees and butterflies. Larvicides kill mosquitoes by applying natural agents or commercial products designed to control larvae and pupae present in aquatic habitats (Dwinell et al., 1998). Larviciding is generally more effective and target specific than adulticiding. The use of adulticides and larvicides may be restricted in the future because of growing concern about potential negative effects on human health and environmental hazards. Biological control should only be used to augment other mosquito control measures as part of an Integrated Pest Management program (IPM). In most cases, mosquito biological control targets the larval stage. Some examples of biological control affecting mosquito larvae are fish that prefer to feed on the larvae. One such fish is Gambusia affinis, also known as the mosquito fish. Studies were conducted in New Jersey Mosquito Control Agency (NJMCA) looking at the effectiveness of mosquito fish as predators towards mosquito larvae. The NJMCA Research and Development Committee concentrated on investigating the effectiveness of Gambusia in the following areas: woodland pools, mine pits, stormwater management facilities, ornamental pools, abandoned swimming pools, ditches, brackish marshes, and freshwater swamps. Results showed very effective mosquito control, as long as adequate fish reproduction occurred and accessibility to larvae by fish was assured (Duryea et al., 1996). There are a couple of other biological control agents that are occasionally used, such as the predaceous mosquito Toxorhynchites, which feeds on larvae in its larval stage, and does not feed on

PAGE 33

20 blood as an adult. Also the parasitic nematodes, such as Romanomermis, the fungus Lagenidium giganteum, and the bacteria Bacillus sphaesicus have been used for biological control. Biological control certainly holds the possibility of becoming a more important tool and playing a larger role in mosquito control in the future (Dwinell et al., 1998). Personal protection and education are very important aspects of mosquito control. It may not be mosquito control in a population sense, but it is personal mosquito control and helps protect from nuisance bites and potential disease infection. The most popular insect repellent is called DEET (N,N-diethyl-metatoluamide). DEET is one of the few repellents that can be applied to human skin or clothing. DEET does not kill insects, but repels them from treated areas. Scientists are not completely sure how it repels insects, yet it is believed that DEET affects the insects ability to locate animals to feed on by disturbing the function of receptors in the mosquito antennae that sense host location chemicals (Rose, 2001). Education is key to protection against mosquitoes. Such tips as avoiding outside activities during peak mosquito times, or if outside activities are necessary; wear proper clothing and insect repellent. Reducing areas of mosquito breeding, such as tires and water filled containers will decrease mosquito populations. Proper education of horse-owners will help protect horses from the nuisance biting and potential disease infection. Keep horses inside under fans during mosquito peak times, vaccinate horses against mosquito borne diseases prevalent in the area, and use mosquito spray regularly. Nonchemical Mosquito Control and Surveillance Mosquito traps are very effective as a surveillance tool to monitor seasonal prevalence and the species composition in a specific area. Traps are also effective in

PAGE 34

21 reducing numbers of mosquitoes in the location of the trap. Over the years there have been lots of different kinds of mosquito traps developed, varying greatly in effectiveness and usefulness. The most effective traps use a combination of factors that attract mosquitoes such as light, heat, moisture, carbon dioxide and synthetic chemicals for host attraction, such as octenol. In 1962, the Center for Disease Control miniature light trap was introduced specifically for arbovirus surveillance and other short-term mosquito investigations (Stamm et al., 1962). The CDC trap model 1012 (CDC 1012)(John W. Hock Company, Gainesville, FL) (Figure 1-1) attracts mosquitoes with a white light and carbon dioxide and captures them with the down draft produced by a fan. The CDC 1012 trap is a lightweight trap, weighing less than 1 kilogram, it is powered by a 6-volt battery (McNelly, 1989). This trap is useful for surveillance studies because it is versatile and catches the mosquitoes alive and unharmed. The main disadvantage to the CDC 1012 trap is that the white light attracts other insects, such as beetles, moths and other nontarget species, which are drawn into the capture net (Kline, 1999). Recently new traps have been developed that use the innovation of carbon dioxide plumes and counterflow geometry together. These traps have been found to increase mosquito catches, and reduce capture of nontarget species. The counterflow geometry is achieved with two fans operating simultaneously, the first fan provides an outflow attractant such as carbon dioxide and the second fan has an inflow current to capture mosquitoes. Around the attractant plume any insect with a flight speed less than approximately 3.5 m/sec will be captured (Kline, 1999). The first mosquito trap to utilize this technology is the Mosquito Magnet X (MM-X) (American Biophysics Corporation,

PAGE 35

22 North Kingston, RI). The MM-X trap (Figure 1-2) is constructed from a modified plastic pretzel jar and is operated by a 12-volt battery. The attractant is carbon dioxide, which is supplied by compressed gas cylinder at a flow rate of 500 ml/minute. Octenol, which can also be used as additional attractant, is supplied by American Biophysics Corporation OCT1 slow release packets, which have a release rate of 0.5 mg/hour (Kline, 1999). This trap is very effective in attracting a wide variety of mosquito species (Kline, 1999 and Burkett et al., 2001), but its major disadvantage is the external source of power and mosquito attractant requirements that must be provided (Kline, 2002). American Biophysics Corporation has considered the drawbacks of the MM-X trap and has improved this technology by developing a new category of traps that utilize propane as an energy source. These traps also use counterflow geometry technology, but have added catalytic combustion of propane to produce carbon dioxide, heat and water vapor. A thermoelectric generator uses excess heat from the combustion process to generate electricity to run the traps fans (Kline, 2002). Another trap recently developed is the Mosquito Magnet Pro (MMPro) (American Biophysics Corporation, North Kingston, RI), which is commercially available to the general public (Figure 1-3). The use of propane, which is supplied in the same cylinders as used for gas grills, makes it easier for the general public who are concerned about local mosquito populations. Now American Biophysics Corporation has six traps currently available for commercial sale, the Mosquito Magnet Pro, Pro-Plus, Liberty, Liberty-Plus, MM-X and Defender. The technology behind the traps is similar; the traps differ in acre coverage and power sources. The Liberty and Defender both use a 12-volt power source and only use propane for production of carbon dioxide. These traps claim to cover 1 acre and acre

PAGE 36

23 respectively. The effectiveness of the traps mentioned above has been researched by Burkett et al. (2001) and Kline (2002) who compared MMPro, MMX, and the CDC 1012. Burkett et al. (2001) and Kline (2002) both found that the MMPro was more effective than CDC 1012 but was outperformed by MMX. Mosquito larva surveillance is also critical to effective control; it is used to determine the location, species and population densities of pest and vector mosquitoes. It is also vital for predicting adult emergence and establishing optimal times for application of larval control measures (Belkin, 1954). There are two main ways to survey larvae in the field: by dipping and with emergence traps. The kind of mosquito larvae one is looking for will determine the sampling technique to be used. Dipping involves collecting larvae present in water with a dipper. It is an effective tool to monitor larvae (Waters and Slaff, 1987). Emergence traps are placed on the surface of a body of water to trap mosquitoes as they emerge as adults and take flight. These traps are usually pyramidal in shape with netting or screen trimming the walls of the pyramid. There is a collection cup at the top of the trap, so when the mosquitoes emerge into adults they fly upward and into the collection cup (Appleton and Sharp, 1985). Emergence traps have not been found to be effective in mosquito control, but they are very helpful in surveillance. Host-baited mosquito traps are a useful tool to evaluate mosquito host preference and possible host location cues. Entomologists have used host-baited traps since the early 1900s for collecting Anopheline mosquitoes during malaria investigations (Mitchell et al., 1985). In Florida, mosquito-borne diseases that affect horses have a significant economic impact on the equine industry, yet there is limited published information on the

PAGE 37

24 collection of mosquitoes from horses (Fletcher et al., 1988). In 1944 Kumm and Zuniga used an animal-baited trap to record seasonal variations of Anopheles. Jones et al. (1977) collected mosquitoes directly from ponies and donkeys as part of an equine encephalitis survey, the trap used was modified from Magoons trap used in 1935. Most large animal-baited traps enclose the animal in a stable or stall with small openings for the mosquitoes to gain entrance. Recently portable animal-baited traps have been used which utilize screen walls and roofing to allow host odors to flow away from the trap as an attractant (Mitchell et al., 1985). In 1984 Hall et al. made one of the most significant discoveries concerning host odor and location cues of flies, using an animal-baited trap. The study was conducted in Africa, where they placed an ox into a pit. The odors emanating from the ox increased the catch of tsetse flies compared to a trap with no odors. Eventually Hall et al. (1984) discovered that an alcohol compound in the breath of the ox, octenol, was a potent attractant for the flies. Since then researchers started looking at specific compounds that elicit a response from mosquitoes, indicating host location behaviors. Fletcher et al. (1988) used a different approach by constructing an open animal-baited trap, which left a horse naturally exposed to insects. The sides of the trap would close around the horse and entrap the insects which could then be identified. Since mosquitoes have a preference for the type of host they feed on, there must be specific factors that elicit a host location behavior. Researchers are now looking at how effective artificial attractants are when put in competition with an animal in a natural state. In 2003 Campbell stated that a large variance was found between mosquito species and numbers aspirated directly from a horse compared to various mosquito traps. There are significantly more mosquitoes attracted to the horse than all of the mosquito traps that

PAGE 38

25 were tested. Campbells (2003) research has lead to more investigations looking at animal-baited traps and traditional baited mosquito traps. Summary The utilization of host odors as attractants for trapping mosquitoes has received increased interest in the past few years for population management and surveillance (Kline, 1994). Historically mosquito control has been performed with chemical insecticides and topical repellents, however, these methods will be greatly restricted in the future because of potential health and environmental hazards. It has been stated by Kline (1994) that there is a need for new, safe and effective ways to kill pest/vector species of mosquitoes, and to deter blood seeking mosquitoes from biting humans and other hosts. During a research investigation conducted in 2002, an interesting skew was observed in the mosquito species collected directly from a horse compared to those being captured by various mosquito traps. The primary species found on the horse was Mansonia titillans, which comprised 40% of the total catch; while Culex nigripalpus comprised 85-91% trapped in the Mosquito Magnet Pro (Campbell 2003). In Florida there are two common species of Mansonia, M. dyari and M. titillans. Until recently Mansonia spp. has been thought of as only a nuisance mosquito with a minor impact on livestock production or humans. In the United States it has been found that Mansonia spp. primarily prefer to feed on mammals (Cupp and Stokes, 1973), and mosquito infestations can cause a significant weight gain reduction in cattle (Steelman, 1979). In 1999, the Center for Disease Control reported that Mansonia spp. has been found a positive carrier of West Nile Virus in Ethiopia (Hubalek and Halouzka, 1999). It has also been reported that Mansonia spp. can be positive carriers of Venezuelan Equine Encephalitis Virus (Mendez et al., 2001) and Japanese Encephalitis Virus (Arunachalam

PAGE 39

26 et al., 2002). There could be a disastrous effect on the United States equine industry if Mansonia spp. are able to carry and transmit these encephalitis viruses because efforts to control them through trapping have been largely unsuccessful (Campbell 2003). The objectives of this study is to determine if traditional mosquito traps using artificial attractants such as carbon dioxide, heat and moisture are effective when put into competition with a horse. Is the horse eliciting a response from the mosquitoes that is much stronger than what the trap is doing?

PAGE 40

27 Figure 1-1. CDC 1012 (John W. Hock Company, Gainesville, FL) mosquito trap.

PAGE 41

28 Figure 1-2. Mosquito Magnet X (American Biophysics Corporation, East Greenwich, RI) mosquito trap

PAGE 42

29 Figure 1-3. Mosquito Magnet Pro (American Biophysics Corporation, East Greenwich, RI) mosquito trap.

PAGE 43

30 Table 1-1. Classification of family Culicidae Subfamily Tribe Genera Anophelinae Anopheles, Bironella, Chagasia Culicinae Aedeomyiini Aedeomyia Aedini Aedes, Ochlerotatus, Verrallina, Ayurakitia, Armigeres, Eretmapodites, Haemagogus, Heizmannia, Opifex, Psorophora, Udaya, Zeugnomyia Culicini Culex, Deinocerites, Galindomyia Culisetini Culiseta Ficalbiini Ficalbia, Mimomyia Hodgesiini Hodgesia Mansoniini Coquillettidia, Mansonia Orthopodomyiini Orthopodomyia Sabethini Sabethes, Wyeomyia, Phoniomyia, Limatus, Trichoprosopon, Shannoniana, Runchomyia, Johnbelkinia, Isostomyia, Tripteroides, Malaya, Topomyia, Maorigoeldia Uranotaeniini Uranotaenia Toxorhynchitinae Toxorhynchites The classification of all mosquitoes into 3 subfamilies, 10 tribes of Culicinae, and 38 genera is based on Knight and Stone (1977).

PAGE 44

31 Table 1-2. Systematic list of mosquitoes found in Florida Genus Species Anopheles atropos, barberi, bradleyi, crucians, diluvialis, georgianus, grabhamii, inundatus, maverlius, perplexens, punctipennis, quadrimaculatus, smaragdinus, walkeri, nyssorhynchus, albimanus Aedes cinereus, vexans, aegypti, albopictus Ochlerotatus atlanticus, Canadensis, mathesoni, dupreei, fulvus pallens, infirmatus, mitchellae, scapularis, sollicitans, sticticus, taeniorhynchus, thelcter, thibaulti, tormentor, tortilis, hendersoni, triseriatus, bahamensis Psorophora columbiae, discolor, pygmaea, cyanescens, ferox, horrida, johnstonii, mathesoni, ciliata, howardii Culex bahamensis, nigripalpus, quinquefasciatus, restuans, salinarius, tarsalis, atratus, cedecei, erraticus, iolambdis, mulrennani, peccator, pilosus, biscaynensis, territans Deinocerites cancer Culiseta melanura, inornata Coquillettidia perturbans Mansonia dyari, titillans Orthopodomyia alba, signifera Wyeomyia mitchellii, smithii, vanduzeei Uranotaenia lowii, sapphirina Toxorhynchites rutilus rutilus, rutilus septentrionalis The classification of Florida mosquitoes by Genus species based on Public Health Entomology Research and Education Center 2002

PAGE 45

CHAPTER 2 SEASONALITY OF MOSQUITOES AT AN EQUINE FACILITY IN NORTH CENTRAL FLORIDA Introduction Mosquitoes cause two significant issues in Floridas equine industry including a nuisance factor and potential for disease transmission. The economic loss of biting mosquitoes on horses has not been studied, but cattle have been found to have significant weight gain reduction, milk production reduction and a compromised immune system due to stress (Steelman, 1979 and Byford et al., 1992). In 1999, West Nile Virus started spreading throughout the United States and has caused the state of Floridas equine industry millions of dollars every year in disease prevention, health care and overall morbidity of infected horses (Porter et al., 2003). Extensive research has been conducted to develop and refine trapping systems that increase mosquito collection and enable efficient surveillance of local mosquito populations. The ability of monitoring seasonality population trends enables efficient mosquito control and knowledge of when to take protective measures for horses against mosquitoes. For example, Fort Dodge Animal Health suggests, in Florida, vaccinating horses against West Nile Virus in March and August. In Gainesville Florida, mosquitoes are generally present twelve months a year, but the abundance and species composition of mosquitoes varies depending on the season. Mosquito population variations are closely linked to rainfall and temperature. The highest numbers of mosquitoes are generally 32

PAGE 46

33 trapped in August, and the lowest are generally in January. Population peaks usually occur 2-3 weeks following a heavy rainfall. The following experiment is an evaluation study of the seasonal population trends of mosquitoes at an equine facility in North Central Florida, using the Mosquito Magnet Pro (American Biophysics Corporation, North Kingston, RI). Meteorological conditions, such as maximum and minimum temperature and rainfall, were collected for comparison of mosquito population trends. Materials and Methods Adult mosquito seasonality studies were conducted in Gainesville, Florida at the University of Florida Horse Teaching Unit, a 60-acre equine facility, housing approximately 40 head of quarter horses varying in sex and age. The study began September 26, 2003 and concluded September 2, 2004. The trapping devices used were four Mosquito Magnet Pro traps, which were placed in predetermined locations at the equine facility. The Mosquito Magnet Pro (MM-Pro) (Figure 1-3) uses patented technology that catalytically converts propane into carbon dioxide, heat and moisture, which attracts mosquitoes to the counterflow of a suction fan and released attractants. The mosquitoes are vacuumed into a collection net where they dehydrate and die. The trap is self-powering when provided with propane through a thermoelectric module, which is supplied by a standard 20-pound commercial propane tank. MM-Pro stands 2 meters high and is built of stainless steel with a PVC shell. A preexisting weather station was utilized to measure maximum and minimum temperature (C) and rainfall (cm) during the seasonality study (Figure 2-1). The station

PAGE 47

34 is positioned on a wooden post measuring 2.4 meters tall, located in a predetermined position. A rain gauge is mounted at the top of the station and a waterproof thermometer is located below the rain gauge. Experimental Design The mosquito seasonality study was conducted from September 26, 2003 until September 2, 2004. The four MM-Pros were placed in the same location for the duration of the study (Figure 2-2). The MM-Pro is powered by propane and the 9-kilogram tank was replaced approximately every three weeks. The nylon collection nets were collected twice a week on Tuesday and Friday mornings around 10am. Mosquitoes were stored at 0C until identification and counting was conducted. When the nylon mosquito collection nets were retrieved on Tuesdays and Fridays, maximum and minimum temperature and total rainfall were recorded from the weather station. Data Analysis Count data from the studies described above were analyzed with the General Linear Model (GLM) and Proc Means Programs; Tukey's Studentized Range Test was used for separation of means (SAS Institute, Version 8.2, Copyright 1999-2001). The significance level was set at P 0.05. Results Total mosquitoes captured using the Mosquito Magnet Pro plotted against months and as related to rain measured in centimeters can be seen in Figures 2-3 and 2-4 respectively. Maximum and Minimum temperatures were recorded at the University of Florida Horse Teaching Unit and are plotted against time during the seasonality study (Figure 2-5). Temperature ranges were similar to previous years; there were no notable

PAGE 48

35 differences in maximum or minimum temperatures. Rain measured at the UF Horse Teaching Unit is shown for years 2001-2004 in Figure 2-6. A total of 42,077 mosquitoes were trapped during this study (Table 2-1). The highest populations of mosquitoes were trapped during the months of August, September, and October (Table 2-2). The lowest populations of mosquitoes were trapped during the months of December, January, February, and March (Table 2-2). An increase in mosquito populations was seen approximately 2-3 weeks following a high amount of rainfall. A significant difference (P 0.05) was found between MMPro traps 2 and 3 when compared to traps 1 and 4. Traps 2 and 3 were placed in similar environments during this study, as were traps 1 and 4. Seasonality trends of specific mosquito species were analyzed and plotted in Figures 2-7 through 2-13. Only species with trapped numbers higher than 1000 over the course of the year were plotted individually for seasonality trends. Discussion Mosquito populations fluctuated throughout the study period, but populations were at their highest during the warm, wet summer months of July through September. Lower counts were seen during the cool, dry months of January through April. With the first major rainfall of the year in late March, a peak of mosquitoes resulted approximately 2-3 weeks later during early April. This population quickly decreased because temperatures were still fairly cool. Data suggest that temperature and rainfall appear to be major factors affecting mosquito seasonality trends. Gentrys (2002) research supports this observation that was conducted at the UF Horse Teaching Unit. Temperature in Gainesville, Florida fluctuates greatly during the cool seasons and become quite regular during the summer months. Even though nighttime temperatures may increase enough

PAGE 49

36 during cool season months for mosquitoes to develop, it quickly drops and disables these mosquitoes from greatly increasing population numbers. Mosquitoes are found in north central Florida 12 months a year, but they are present in much more significant numbers during the warm, wet seasons of summer and fall (Campbell, 2003). Data from this seasonality study suggest that Anopheles spp., Culex erraticus and Culex salinarius mosquitoes were found all 12 months of the year, with high peaks in populations during cool, dry months. The majority of mosquitoes trapped significantly decreased in November and did not increase until April. There was one species, Coquillettidia perturbans, which peaked suddenly in April and never again achieved such high numbers throughout the rest of the year. Other species such as Mansonia titillans, Psorophora spp. and Culex nigripalpus population peaks did not occur until late summer. Mosquito surveillance is a requirement to achieve an effective and environmentally friendly mosquito control program. Surveillance is most effective when combined with monitoring meteorological and environmental factors that may influence mosquito population change: For example, rainfall, temperature, relative humidity, wind direction, and extreme weather such as hurricanes. This year, 2004, Florida had a record number of hurricanes make landfall, and caused severe flooding. The standing water created mass numbers of mosquito populations. Conclusions Identifying local mosquito seasonality trends is a valuable tool for mosquito control agencies and the general public. During this study in Gainesville, Florida a total of 42,077 mosquitoes were trapped encompassing 17 different species. The most prominent genus trapped was Culex including Cx. nigripalpus, Cx. salinarius, and Cx. erraticus. Culex is one of the main vectors of West Nile Virus in Florida, and was trapped all 12

PAGE 50

37 months of the year at the UF Horse Teaching Unit; therefore a threat of this virus is high. Mosquito populations peaked during the warm, wet months of July through September. Mosquitoes were present during the cold months of winter, but in very low numbers. Data suggest that low nighttime temperatures greatly affect mosquito populations, by inhibiting larval development. Generally 2-3 weeks following a heavy rainfall, mosquito populations increased. Therefore, temperature and rainfall are major factors that affect mosquito populations. Surveillance of mosquitoes emerging from the floodwater could aid in predicting dangerous disease outbreaks. Ideally, through surveillance officials potentially could detect a period of risk in advance of the appearance of a disease. This measure could provide control agencies the opportunity to spray for population control. The medical community made aware of potential human cases, and the public warned so that protective measures can be made. Knowing the times of the year that specific disease vector mosquitoes are prevalent, can aid in developing a suitable pest management program for a facility. In 1999 West Nile Virus greatly affected the horse industry in the United States. By 2002 a vaccine was approved for horses to protect against this disease. The recommendations made by veterinarians for time of year when this vaccination should be given correspond with population peaks of mosquito vectors capable of transmitting WNV. Demographics where WNV vectors are only prevalent for short periods of time only need one vaccine and one booster shot per year. Places like Florida where WNV vectors are present 12 months a year require 3 total shots. This example demonstrates how important monitoring mosquito populations are and how this information can greatly benefit the general public.

PAGE 51

38 The general public can greatly influence the population peaks and prevalence of vectors for a number of diseases such as WNV, EEE, SLE, LAC and Yellow Fever. Personal protection and education are very important aspects of mosquito control. It may not be mosquito control in a population sense, but it is personal mosquito control and helps protect from nuisance bites and potential disease infection. The most popular insect repellent is called DEET (N, N-diethyl-metatoluamide). DEET is one of the few repellents that can be applied to human skin or clothing. DEET does not kill insects, but repels them from treated areas. Scientists are not completely sure how it repels insects, yet it is believed that DEET affects the insects ability to locate animals to feed on by disturbing the function of receptors in the mosquito antennae that sense host location chemicals (Rose, 2001). Education is key to protection against mosquitoes, avoid outside activities during peak mosquito times, if outside activities are necessary, and wear proper clothing and insect repellent. Reducing areas of mosquito breeding, such as tires and water filled containers will decrease mosquito populations. Proper education of horse-owners will help protect horses from the nuisance biting and potential disease infection. Keep horses inside under fans during mosquito peak times, vaccinate horses against mosquito borne diseases prevalent in the area, and use mosquito spray regularly.

PAGE 52

39 Figure 2-1. Weather station at south side of UF Horse Teaching Unit, used to collect rainfall, minimum and maximum temperature readings for the mosquito seasonality study conducted September 26, 2003 through September 2, 2004.

PAGE 53

40 Figure 2-2. Aerial photograph of the UF Horse Teaching Unit show the four MMPro trap sites for the mosquito seasonality study conducted September 26, 2003 through September 2, 2004.

PAGE 54

05001000150020002500300035004000910101011111212121112223334455666777888MonthTotal # Mosquitoes Trapped 41 Figure 2-3. Total mosquito counts as related to months during the seasonality study conducted at the UF Horse Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e.1=January).

PAGE 55

05001000150020002500300035004000910101011111212121112223334455666777888MonthMosquito Counts0246810121416Rain (cm) Rain (inches) Mosquito Count 42 Figure 2-4. Total mosquito counts from the MMPro, related to rainfall during the seasonality study conducted at the UF Horse Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e. 1=January).

PAGE 56

-20.0-10.00.010.020.030.040.050.0910101011111212121112223334455666777888MonthsTemperature (C) Max Temperatures (C) Min Temperatures (C) 43 Figure 2-5. Maximum and Minimum temperatures recorded at the meteorological station during the seasonality study conducted at the UF Horse Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e. 1=January).

PAGE 57

0.005.0010.0015.0020.0025.0030.0035.0040.00ONDJFMAMJJAMonthRain (centimeters) 2001-2002 2002-2003 2003-2004 44 Figure 2-6. Total rainfall (centimeters) at the UF Horse Teaching Unit from October 2001 through August 2004.

PAGE 58

0500100015002000250030003500910101011111212121112223334455666777888MonthCulex nigripalpus Trapped 45 Figure 2-7. Total numbers of Culex nigripalpus trapped by MMPro during the seasonality study conducted at the UF Horse Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e. 1=January).

PAGE 59

020406080100120910101011111212121112223334455666777888MonthCulex erraticus Trapped 46 Figure 2-8. Total numbers of Culex erraticus trapped by MMPro during the seasonality study conducted at the UF Horse Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e. 1=January).

PAGE 60

050100150200250300350400450910101011111212121112223334455666777888MonthCulex salinarius Trapped 47 Figure 2-9. Total numbers of Culex salinarius trapped by MMPro during the seasonality study conducted at the UF Horse Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e. 1=January).

PAGE 61

020040060080010001200910101011111212121112223334455666777888MonthMansonia spp. Trapped 48 Figure 2-10. Total numbers of Mansonia spp. trapped by MMPro during the seasonality study conducted at the UF Horse Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e. 1=January).

PAGE 62

020406080100120140160910101011111212121112223334455666777888MonthAnopheles spp. Trapped An. crucians An. quadrimaculatus 49 Figure 2-11. Total numbers of Anopheles spp. trapped by MMPro during the seasonality study conducted at the UF Horse Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e. 1=January).

PAGE 63

0100200300400500600910101011111212121112223334455666777888MonthCoquillettidia perturbans Trapped 50 Figure 2-12. Total numbers of Coquillettidia perturbans. trapped by MMPro during the seasonality study conducted at the UF Horse Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e. 1=January).

PAGE 64

050100150200250300350910101011111212121112223334455666777888MonthPsorophora spp. Trapped Ps. columbiae Ps. ferox Ps. cyanescens 51 Figure 2-13. Total numbers of Psorophora spp. trapped by MMPro during the seasonality study conducted at the UF Horse Teaching Unit from September 26, 2003 through September 2, 2004. Note: Month number equals corresponding calendar date (i.e. 1=January).

PAGE 65

52 Table 2-1. Total count, and percent of total count of mosquito species trapped by MMPro traps during seasonality study at the UF Horse Teaching Unit conducted September 26, 2003 until September 2, 2004. Mosquito Species Total Count Percent of Total Culex nigripalpus 18213 43.3 Mansonia spp. 7398 17.6 Culex salininarius 5374 12.7 Coquillettidia perturbans 3854 9.2 Psorophora columbiae 2216 5.3 Culex erraticus 1566 3.7 Anopheles crucians 1218 2.9 Anopheles quadrimaculatus 1110 2.6 Aedes vexans 372 0.88 Psorophora ferox 365 0.87 Uranataenia sapphirina 144 0.27 Ochlerotatus infirmatus 103 0.24 Psorophora cyanesens 84 0.20 Uranataenia lowii 44 0.11 Ochlerotatus trisariatus 6 0.01 Culiseta melanura 5 0.01 Ochlerotatus mitchellae 3 0.01 42077

PAGE 66

53 Table 2-2. Mean standard deviation of mosquitoes trapped each month during seasonality study at the UF Horse Teaching Unit conducted September 26, 2003 until September 2, 2004. Month Mean Catch ( Std) n Total September 2003 666 ( 221) 4 3566 October 2003 357.9 ( 354.1) 32 10552 November 2003 139.1 ( 173.3) 28 3895 December 2003 19.59 ( 15.90) 29 568 January 2004 13.03 ( 11.46) 35 456 February 2004 11.97 ( 11.06) 31 371 March 2004 29.31 ( 20.03) 26 762 April 2004 114.0 ( 78.9) 23 2623 May 2004 37.53( 23.03) 32 1201 June 2004 96.9 ( 166.3) 30 2908 July 2004 192.2 ( 267.2) 32 6150 August 2004 178.8 ( 154.0) 31 5542 September 2004 1742 ( 1538.0) 2 3483 42077 Note: n=number of observations

PAGE 67

CHAPTER 3 MOSQUITO TRAPPING STUDIES AT AN EQUINE FACILITY IN NORTH CENTRAL FLORIDA Introduction Mosquito traps are very effective as a surveillance tool to monitor seasonal prevalence and the species composition in a specific area. Traps are also effective in reducing numbers of mosquitoes in the location of the trap. Over the years there have been lots of different kinds of mosquito traps developed, varying greatly in effectiveness and usefulness (Kline, 1999). Many commercial traps have been advertised to control mosquitoes up to 1 acre and are commonly found around livestock facilities. Most of these traps use stimuli that mimic behavior such as heat, carbon dioxide, kairomones, and moisture (Kline and Mann, 1998). But how effective are mosquito traps when they are competing with a natural host? Campbell (2003) conducted a series of experiments evaluating mosquito traps and a horse, and found a significant difference between species and total quantity of mosquitoes collected from mosquito traps compared to mosquitoes aspirated directly off a horse. If adult mosquito trapping is the main technique for surveillance, is this the true representation of the actual mosquito species population? Trap catches could potentially be improved by adding odors directly from the horse to the specified trap. Little research has been conducted which directly compares the efficiency of mosquito traps with a natural host in the immediate proximity. Therefore, the main objectives of the following studies are: (1) to conduct trapping competitive studies with 2 traps and a horse; (2) to evaluate mosquito location profiles 54

PAGE 68

55 around the University of Florida Horse Teaching Unit, (3) an evaluation was conducted looking mosquito species feeding on a horse at a given time by the technique of aspiration; (4) to determine the distance required between 2 horses to achieve separate entities. Materials and Methods Six studies were conducted in Gainesville, Florida at the University of Florida Horse Teaching Unit; a 60-acre equine facility housing approximately 40 head of quarter horses varying in sex and age. Three CDC 1012 traps, a portable vacuum aspirator, a Mosquito Magnet Pro, a gelding paint Quarter horse, and an Appaloosa mare were placed throughout the property. Traps Centers for Disease Control (CDC) trap. The CDC trap model 1012 (John W. Hock Company, Gainesville, FL) (Figure 1-1), was used during this experiments with slight modifications. Generally the trap uses a 6.3-volt light as an attractant, but this was eliminated to decrease attractive variables of the trap. The CDC 1012 uses a 6-volt battery to provide energy to a fan, which creates a downward airflow to capture mosquitoes attracted to the top of the trap. Carbon dioxide was supplied from a compressed gas cylinder at a rate of 500 ml/min. The carbon dioxide line was attached with rubber bands to the side of the trap, with the gas being released directly above the fan. A polypropylene container with screen on the bottom was used as the collection device and connected to the bottom of the trap. Mosquito Magnet Pro. The Mosquito Magnet Pro (MM-Pro) (American Biophysics Corporation, North Kingston, RI) (Figure 1-3) uses patented technology that catalytically converts propane into carbon dioxide, heat and moisture, which attracts

PAGE 69

56 mosquitoes to the counterflow of a suction fan and released attractants. The mosquitoes are vacuumed into a collection net where they dehydrate and die. The trap is powered through propane by catalytically converting the gas into carbon dioxide, heat, and moisture. The MM-Pro stands 2 meters high and is built of stainless steel with a PVC shell. Portable Vacuum Aspirator. Mosquitoes that landed on the horse were collected using a portable vacuum mosquito aspirator (DC Insect Vac. BioQuip, Rancho Dominguez, CA) (Figure 1-1). A 12-volt AC supplied by a modified plug, which fits into a car cigarette lighter socket, powered the vacuum. A plastic collection cup fits into the vacuum and contains the mosquitoes while the vacuum in turned on; when the vacuum was turned off a cap was placed on the container. Horses. The horses used during this experiment were a gelding Paint (Figure 3-2) and an Appaloosa mare (Figure 3-3), both approximately 7 years of age. The horses were housed at the University of Florida Horse Teaching Unit and remained there after the experiment concluded. Experimental Design CDC 1012 Trapping Study The CDC trapping efficiency study was conducted from July 19, 2004 until August 21, 2004. Starting 30 minutes after sunset, a CDC 1012 mosquito trap was placed directly next to the gate on a stall of a horse barn (Figure 3-4). The position of the barn and the stall where the study was conducted in shown in Figure 3-5. The trap was baited with only CO2 and the net was changed every 20 minutes to include five 20-minute trapping periods. There were two treatment groups: no horse present and horse present, each chosen at random. During nights when the horse is present, a Paint gelding was

PAGE 70

57 placed in the stall during the second 20-minute trapping period, then removed. Mosquitoes were stored at 0C until identification and counting was conducted. Each treatment, horse and no horse, was repeated six times (see Table 3-1 for schedule). MMPro Trapping Study The MMPro trapping efficiency study was conducted from October 1, 2004 until October 21, 2004. Starting 30 minutes after sunset a Mosquito Magnet Pro trap was placed directly next to side of a horse feeding slip stall (Figure 3-6). The position of the barn and the stall where the study was conducted in shown in (Figure 3-7). The trap was baited with CO2, moisture and heat; the net was changed every 20 minutes to include 5 20-minute trapping periods. There are two treatment groups: no horse present and horse present. The treatment horse present was conducted coordinating with when the facilities were available. During nights when the horse is present, a Paint gelding was placed in the stall during the second 20-minute trapping period, then removed. Mosquitoes were stored at 0C until identification and counting was conducted. Each treatment, horse and no horse, was repeated six times (see Table 3-2 for schedule). Horse Odor Study Field data for this study were collected beginning August 27, 2004 and concluded September 24, 2004, examining mosquitoes captured by a modified CDC 1012 trap with the addition of equine odors to the attractive air stream. A modified hand held vacuum was constructed for this portion of the experiment. A 2-inch shop vacuum hose was connected to a 2-inch pvc pipe, with the open end placed near the CDC 1012 trap airstream. There was an electric powered in-line fan in the pvc pipe which created a suction and air through the hose and out the open end. Starting 30 minutes after sunset, a

PAGE 71

58 CDC 1012 mosquito trap was placed directly next to the gate on a stall of a horse barn. The net was changed every 20-minutes to include 4 20-minute trapping periods. The gelding Paint horse was placed in a stall of a horse barn (Figure 3-5) for 1 hour during intervals 1, 2 and 3. The mosquito trap was only baited with CO2 during interval 1, and then again during interval 3. During interval 2, the modified hand held vacuum was attached to the CDC 1012 trap, so the trap was baited with horse odor and carbon dioxide. The horse was vacuumed all over the body (Figure 3-8) and the odors traveled through the PVC pipe and were released directly above the fan close to the carbon dioxide source. The horse was removed from the stall during interval 4 and collection continued until for 20 more minutes. This protocol was conducted six times. All collection intervals were 20-minute intervals. See Tables 3-3 and 3-4 showing time schedule and treatment date schedule respectively. Mosquitoes were stored at 0C until identification and counting was conducted. A similar protocol was conducted to monitor mosquito activity from August 27, 2004 through September 24, when no horse was present. During intervals 1-4, a CDC 1012 trap was set up in the same position as mentioned above, baited only with carbon dioxide. All collection intervals were 20-minute intervals. Mosquitoes were stored at 0C until identification and counting was conducted. Location Profile Study The mosquito location profile through trapping study using CDC 1012 traps was conducted from July 13, 2004 until August 26, 2004. Three CDC 1012 mosquito traps were placed in three positions throughout the University of Florida Horse Teaching Unit (Figure 3-9). The first position was close to a swamp area and the other two positions

PAGE 72

59 were on each side of a horse barn. All three CDC 1012 traps were baited only with carbon dioxide and allowed to run from 8pm until 10:30pm. The traps were set up 30 minutes before sunset and collected after mosquito peak activity had concluded. Mosquitoes were stored at 0C until identification and counting was conducted. Horse Vacuuming Study The quantity of mosquitoes feeding on a horse at a given time period of time was examined from June 15, 2004 until July 12, 2004 and then repeated October 4, 2004 and concluded October 22, 2004. During the first trial, mosquitoes were aspirated from the skin of a gelding Paint horse (Figure 3-10) located in a stall of a barn using 2 portable vacuum mosquito aspirators (Figure 1-1) for a 1 hour time period starting 30-minutes after sunset. During the second portion of this experiment, a mare Appaloosa and a gelding Paint horse were vacuumed simultaneously starting at for 1 hour and 40 minutes starting 30-minutes after sunset. Mosquitoes were aspirated from all over the body of the horse. Mosquitoes were stored at 0C until identification and counting was conducted. Separate Entity Study The distance required for the mosquito to distinguish two horses as separate entities was examined from October 4, 2004 until October 22, 2004. Two horses, a gelding Paint and an Appaloosa mare, were placed in adjacent feeding slip stalls starting 30-minutes after sunset (Figure 3-11). The horses were vacuumed simultaneously using 2 portable vacuum mosquito aspirators for 20 minutes. The collection cups on the vacuum aspirators were emptied and new collection cups were placed in the aspirators. The Appaloosa mare remained in the original stall, and the Paint gelding was moved to

PAGE 73

60 another stall depending on the randomly chosen assignments for the night. The horses were then vacuumed again for 20 minutes. This procedure was repeated until the horses were vacuumed in all five stalls. Five distance intervals were set up, D1=3.05 meters, D2=6.09 meters, D3=9.14 meters, D4=12.19 meters, D5=20.42 meters. See the schedule of stall assignments for each date of the trial in Table 3-5. The mosquitoes were stored at 0C until identification and counting was conducted Data Analysis Count data from the studies described above were analyzed with the General Linear Model (GLM) and Proc Means Programs; Tukey's Studentized Range Test was used for separation of means (SAS Institute, Version 8.2, Copyright 1999-2001). The significance level was set at P 0.05. Results CDC 1012 Trapping Study The total number of mosquitoes collected by CDC 1012 mosquito trap during this trial was 2,249 with the most prominent species being Mansonia titillans comprising 76% of the total catch, followed in descending order by Coquillettidia perturbans (11%) and Culex salinarius (7%). The distribution and means of mosquitoes trapped during the two treatment groups are shown in Figure 3-12. Total values of mosquitoes that were trapped during each interval by treatment are shown in Figure 3-13. The mean number of mosquitoes captured during each interval by treatment group is shown in Figure 3-14. There was a significant difference (P 0.05) in the mean numbers of mosquitoes between the treatment groups horse and no horse (Table 3-6).

PAGE 74

61 During the treatment horse present, there was a significant difference (P 0.05) in the mean numbers of mosquitoes caught between intervals (Table 3-6). There was a significant reduction (P 0.05) in the mean mosquito catch in interval 2 when compared to intervals 1 and 3. There was no significant difference between intervals 1 and 3. The greatest mean number of mosquitoes was captured during interval 1. The fewest mean of mosquitoes was captured during interval 2, which coincided with when the horse was located in the stall next to the CDC 1012 mosquito trap. During the treatment no horse present, there was a significant difference (P 0.05) in the mean numbers of mosquitoes caught between intervals (Table 3-6). Interval 2 was significantly greater (P 0.05) when compared to all of the other intervals. There was no significant difference between intervals 1 and 3 or between intervals 4 and 5. The greatest mean number of mosquitoes was captured during interval 2. The fewest mean of mosquitoes was captured during interval 5. MMPro Trapping Study The total number of mosquitoes collected by MM-Pro mosquito trap during this trial was 5,533 with the most prominent species being Culex nigripalpus comprising 36% of the total catch, followed in descending order by Mansonia titillans (31%) and Psorophora columbiae (23%). The mean number of mosquitoes captured during each interval by treatment group is shown in Table 3-7. The distribution and mean numbers of mosquitoes trapped during the two treatment groups are shown in Figure 3-15. Total values of mosquitoes that were trapped during each interval by treatment are shown in Figure 3-16. The mean number of mosquitoes captured during each interval by treatment group is shown in Figure 3-17. There was a significant difference (P 0.05) in the mean

PAGE 75

62 numbers of mosquitoes captured between the treatment groups horse and no horse (Table 3-7). During the treatment horse present, there was a significant difference (P 0.05) in the mean numbers of mosquitoes caught between intervals (Table 3-7). There was a significant difference (P 0.05) in the mean mosquito catch in interval 2 when compared to 1 and 3. There was no significant difference between interval 1 and 3. The greatest mean of mosquitoes was captured during interval 1. The fewest mean of mosquitoes was captured during interval 2, which coincided with when the horse was located in the stall next to the MMPro mosquito trap. During the treatment no horse present, there was a significant difference (P 0.05) in the mean numbers of mosquitoes caught between intervals (Table 3-7). Intervals 1 and 2 were significant greater than intervals 4 and 5. There was no significant difference between intervals 1, 2 and 3, or between 3, 4, and 5. The greatest mean number of mosquitoes was captured during interval 2. The fewest mean number of mosquitoes was captured during interval 5. Horse Odor Study The total number of mosquitoes collected by the CDC 1012 mosquito trap during this trial was 1,109 with the most prominent species being Mansonia titillans comprising 76% of the total catch, following by Culex nigripalpus (28%). The distribution and mean numbers of mosquitoes trapped during the two treatment groups are shown in Figure 3-18. Total values of mosquitoes that were trapped during each interval by treatment are shown in Figure 3-19. The mean number of mosquitoes captured during each interval by treatment group is shown in Figure 3-20. There was no significant

PAGE 76

63 difference between the mean numbers of mosquitoes trapped during treatments Horse Present and Horse Odor Vacuumed (Table 3-8). There was a significant difference (P 0.05) between the mean numbers of mosquitoes trapped of No Horse Present when compared to all other treatments. Location Profile Study The total number of mosquitoes collected by the CDC 1012 mosquito traps during this trial was 7,184 with the most prominent species being Mansonia titillans comprising 66% of the total catch, followed by Culex nigripalpus (16%)(Figure 3-21). The distribution of mean numbers of mosquitoes captured from each trap position is shown in Figure 3-22. There was a significant difference (P 0.05) between the mean numbers of mosquitoes trapped in position 2 when compared to positions 1 and 3 (Table 3-9). There was no significant difference between the mean numbers of mosquitoes trapped in positions 1 and 3 (Table 3-9). There was no significant difference between the mean numbers of Cx. nigripalpus and Cx. erraticus between trap positions (Table 3-10). There was a significant difference (P 0.05) in mean numbers of Cx. salinarius trapped in trap position 2 when compared to positions 1 and 3 (Table 3-10). There was a significant difference (P 0.05) in mean numbers of Ma. titillans and Cq. perturbans trapped in trap position 1 when compared to positions 2 and 3 (Table 3-10). Horse Vacuuming Study The total numbers of mosquitoes aspirated from the Paint horse from June 15, 2004 until July 12, 2004 was 1,946. The total number of mosquitoes aspirated off of the Paint and Appaloosa horses from October 4, 2004 until October 22, 2004 was 7,197. During the first trial, the most prominent species trapped was Cq. perturbans comprising

PAGE 77

64 40% of the total catch, followed in descending order by Cx. salinarius (37%) and Ma. titillans (12%) (Figure 3-24). During the second trial, the most prominent species trapped was Cx. nigripalpus comprising 41% of the total catch, followed in descending order by Ma. titillans (27%) and Ps. columbiae (17%) (Figure 3-25). There was a significant difference (P 0.05) between the mean numbers of mosquitoes and the species composition aspirated from the Paint and Appaloosa horses. Almost 50% of the total mosquitoes vacuumed from the Appaloosa mare were Cx. nigripalpus, followed by Ma. titillans (21%) and Ps. columbiae (15%) (Figure 3-26). Approximately 40% of the mosquitoes vacuumed from the Paint were Ma. titillans, followed by Cx. nigripalpus (25%) and Ps. columbiae (21%) (Figure 3-27). Separate Entity Study The total numbers of mosquitoes aspirated from the Appaloosa and Paint horses from October 4, 2004 until October 22, 2004 was 4,810 and 2,387 respectively. The most abundant species was Culex nigripalpus comprising 41% of the total catch, followed in descending order by Mansonia titillans (27%) and Psorophora columbiae (17%). There was a significant difference (P 0.05) between the mean numbers of mosquitoes and species composition aspirated from the Paint and Appaloosa horses. Almost 50% of the total mosquitoes vacuumed from the Appaloosa mare were Cx. nigripalpus, followed by Ma. titillans (21%) and Ps. columbiae (15%) (Figure 3-26). Approximately 40% of the mosquitoes vacuumed from the Paint were Ma. titillans, followed by Cx. nigripalpus (25%) and Ps. columbiae (21%) (Figure 3-27). There was no significant difference between the difference of mean numbers of mosquitoes vacuumed off horse 1 and horse 2 per distance interval (Table 3-10). There was no significant

PAGE 78

65 difference between the mean of total mosquitoes vacuumed from horse 1 and horse 2 per distance (Table 3-11), the means and standard deviations (shown in Figure 3-28). Discussion CDC 1012 and MMPro Trapping Studies Over the years there have been many different kinds of mosquito traps developed, varying greatly in effectiveness and usefulness. The most effective traps use a combination of factors that attract mosquitoes such as light, heat, moisture, carbon dioxide and synthetic chemicals for host attraction, such as octenol. Trap efficiency and species composition have been researched thoroughly, yet little research has been conducted testing trap efficiency when placed in a competitive situation with an actual host. Before this study was conducted, it was suggested that a host located near a mosquito trap would cause an increase in the catch of the trap. However, the data in studies 1 and 2 suggest that the traps (CDC 1012 and MM-Pro) were both significantly out competed when a natural host was in the vicinity. During the 100-minute trapping sessions, when there was no horse present the mosquitoes trapped followed a natural increase the hour after sunset and steadily decreased for the remaining trapping period. This mosquito activity pattern has been well documented, and it has been stated that most mosquito species possess a bimodal flight activity pattern during the night, with the larger peak occurring soon after sunset and the smaller peak just prior to dawn (Schmidt, 2003). When a horse was added to the system for a 20-minute interval out of the total 100-minute session, the natural increase did not occur and virtually no mosquitoes were trapped, but only when the horse was present. As soon as the horse was removed, the mosquitoes captured in the traps increased significantly.

PAGE 79

66 This study was conducted twice with two different trap types during different months of the year. The CDC 1012 trap was tested along with the MM-Pro during the trapping studies. Previous data suggest that the MM-Pro is a superior mosquito trap when compared to the CDC 1012 (Kline, 1999; Burkett et al., 2001), yet Campbell (2003) suggested the CDC 1012 trap performed as well or better than the MM-Pro. Since it has been suggested that both traps perform similarly, their mosquito yield capabilities were not studied during this experiment. Only their performance when placed in a competitive situation with a host was examined. The CDC 1012 trapping study was conducted during the months of July and August, and the MMPro trapping study was conducted in the month of October. The mosquito species composition changed between study 1 and study 2 most likely because of change in season. During CDC 1012 trapping study the most prominent mosquito species trapped were Mansonia titillans, Coquillettidia perturbans and Culex salinarius, respectively. During the MMPro trapping study the most prominent mosquito species trapped were, Culex nigripalpus, Mansonia titillans, and Psorophora columbiae, respectively. The change in seasons between the studies provided a wide array of mosquito species present, which enabled the examination of preferences of host or trap. The traps were placed approximately 1 meter from the horse, but the horse could move around in the 3 x 3 meter stall. So at times the trap may have been as far away as 3 meters from the horse. This distance between horse and trap has been shown to reduce the catch of the trap by almost 95%. This is a substantial reduction of mosquitoes, and could suggest that the traps are inefficient when a host is within 3 meters of the trap.

PAGE 80

67 Further studies should be conducted examining the exact distance between horse and trap when the mosquito catch begins to decrease. Horse Odor Study The next study took the information learned from the previous studies and elaborated by adding equine odors to a CDC 1012 trap. It was believed that by adding an odor attractant to the carbon dioxide airstream of the trap, the mosquito catch would increase. During this study when the horse was being vacuumed to pull off skin odors, the mosquito catch did not significantly increase. This data may suggest that host location of mosquitoes may not be from equine skin odors, or that the added attractant to the trap was still not sufficient enough to out compete the natural host in the vicinity. The widely used mosquito attractant octenol was isolated from compounds in oxen breath, so further studies could be conducted examining equine breath for attractiveness. Only one horse (Paint) was used during this study, and even though it was found that the Paint was an attractive host, another horse could have potentially caused an increase in mosquitoes trapped with odors. Another experiment that could be conducted would remove the visibility of the horse from the mosquitoes, and therefore only have odors as the attractant. Location Profile Study Another study was conducted examining trap location differences at the University of Florida horse teaching unit. This facility is 60-acres and CDC 1012 mosquito traps were distributed around the facility in various types of environments. One trap (position 2) was located near a swamp, and the other two traps were positioned on each side of a 10-stall barn. Previous data collected suggested a difference in mosquito quantities found on the west side (position 1) of the barn when compared to the east side (position 3). The

PAGE 81

68 most prominent mosquito species trapped during this study was Mansonia titillans and Culex nigripalpus. Position 2 trapped the highest number of mosquitoes, which is feasible since this trap is located close to a standing body of water. Position 3 trapped a higher number of mosquitoes than position 1, but this difference was not found to be significant. Both position 1 and 3 were located the same distance from the nearest body of water, but their surroundings varied slightly. There was a stall enclosed by white material adjacent to the CDC 1012 trap at position 1, earlier it was believed that this enclosure was deterring mosquitoes from the area. Position 3 was adjacent to an open stall and had no visible structures surrounding it that suggested being any kind of deterrence. Even though the difference in mean mosquitoes was not found to be significantly different during this study, the total numbers were different. Position 1 trapped a total of 1134 mosquitoes while position 3 trapped 1801 mosquitoes. These positions are only 30 meters apart, so what caused this slight difference in total mosquitoes? It has been documented that mosquitoes are attracted or repelled by different colors and light intensities (Muir et al., 1992), the white material covered stall adjacent to the trap at position 1 could have affected the trap totals during this study. Horse Vacuuming Study High numbers of mosquitoes feeding on horses in Florida can cause significant health issues ranging from a direct impact like West Nile Virus to an indirect nuisance effect causing reduced feed conversion efficiency, weight gain reductions, and decreased milk yield (Steelman, 1979 and Byford et al., 1992). During this trial high quantities of mosquitoes were aspirated off of the skin of a Paint gelding and an Appaloosa mare. Trapping occurred over 11 nights and a total of 9,143 mosquitoes were vacuumed from the horses. These numbers should cause concern about the sheer abundance of

PAGE 82

69 mosquitoes routinely feeding on horses in Gainesville, Florida. Horses are too busy fighting off biting mosquitoes that they may neglect grazing or drinking adequate water. There were common localities where the mosquitoes were most likely found during this vacuuming study including the neck, legs, heels, barrel and stomach. An interesting note was made that the Appaloosa mare always had mosquitoes congregating around her cornet band; this was not seen with the Paint horse. Both horses were very adequate at shaking off the mosquitoes by twitching their skin or knocking them off with their tail. Some times the horses response to the biting mosquito was faster than I was at vacuuming the mosquito. The species composition of mosquitoes vacuumed during this study varied by season and by horse. During the first trial conducted in June and July 2004, the most abundant species collected was Coquillettidia perturbans (39.5%) and Culex salinarius (36.9%). Cq. perturbans are primarily mammal feeders, but are opportunistic feeders and will feed on a bird if available (Sardelis et al., 2001). Cx. salinarius are primarily avian feeders, but if the population of birds is insufficient or unavailable, they will contentedly feed on mammals (Braverman et al., 1991). During the second trial the three highest abundant mosquito species were Culex nigripalpus, Mansonia titillans and Psorophora columbiae. There was a difference in the percentage of these species found on each horse. The Appaloosa mare had almost 50% of the total mosquitoes vacuumed being Cx. nigripalpus, followed in descending order by Ma. titillans (21%) and Ps. columbiae (15%) The Paint gelding had 40% of the mosquitoes vacuumed being Ma. titillans, followed in descending order by Cx. nigripalpus (25%) and Ps. columbiae (21%). The difference in species composition between horses is interesting. As stated earlier, a

PAGE 83

70 congregation of mosquitoes around the cornet band was found on the Appaloosa mare that was not present on the Paint gelding, data suggests that these mosquitoes were Cx. nigripalpus. Why was Cx. nigripalpus more attracted to the Appaloosa than the Paint is unknown, it could be the difference in sex or color of hair. Ma. titillans and Ps. columbiae are primarily mammal feeders, but are opportunistic feeders and will feed on a bird if available (Edman, 1971). Cx. nigripalpus is an opportunistic feeder and shifts host selection based on the season, feeding on avian hosts in the winter and spring and on mammalian hosts in the summer and fall (Sardelis et al., 2001). It has been stated that male hosts are more attractive to mosquitoes than female hosts (Khan et al., 1965), but the opposite was seen during this study. More mosquitoes were vacuumed from the mare Appaloosa than the gelding Paint horse. Since the male horse was a gelding (castrated), the lack of testosterone could have affected the normal behavior of mosquito preference. During trial 2 the species composition was much more diverse when compared to trial 1. There were 11 different mosquito species collected during the MMPro trapping study and 6 species during the CDC 1012 trapping study. The difference in species composition is probably due to seasonal differences. When the CDC 1012 trapping study was conducted there was little rainfall and the mosquito populations were still low, the MMPro trapping study was conducted during optimal meteorological conditions, warm and wet. Separate Entity Study The distance needed between two hosts for a mosquito to view them as separate entities is valuable information. This information could be utilized by mosquito trap manufactures when determining the range of effectiveness with natural hosts, such as livestock, situated around the trap. The closest distance examined during this study was 3.05 meters and the farthest was 20.42 meters between horses. There was no significant

PAGE 84

71 difference found between mosquitoes vacuumed at any of the distance intervals, yet there was a significant difference between total mosquitoes vacuumed off of each individual horse. This data could suggest a couple of conclusions, either the mosquitoes recognized the two horses as separate entities immediately at a distance of 3.05 meters apart, or the goal of separate entities was not accomplished at a distance of 20.42 meters apart. Since there was a significant difference in mosquitoes vacuumed off of each individual horse, I believe that the mosquitoes immediately recognized the two horses as separate entities at 3.05 meters apart. Another study could be conducted by placing two horses side by side and measuring mosquitoes landing through vacuuming, which would clarify at what distance separate entity status is achieved. Conclusions Mosquito traps are thought to be effective tools for surveillance and reducing numbers in a specific area. Data from this study suggests that the CDC 1012 and Mosquito Magnet Pro are effective at trapping mosquitoes as long as they are not placed in a competitive situation with a natural host. If adult mosquito trapping is the main technique for surveillance, this may not be a true representation of the actual mosquito species population. There seems to be a different composition of mosquito species being caught in mechanical traps when compared to mosquitoes vacuumed directly off of a horse. Therefore, it is suggested that these mechanical traps are only representing a fraction of the total mosquito population in an area. Mechanical traps could potentially be improved by adding odors directly from the host to the trap. Even though this technique was not successful during this trial, more field research should be conducted on the subject. Hall et al. accomplished isolating a host odor (octenol) that attracted Tsetse flies in 1984. Octenol was isolated from oxen breath and is currently used as a mosquito

PAGE 85

72 attractant (Kline, 1994). Since the mosquitoes are attracted to horse instead of the traps tested, this suggests that there is some kind of attractant overriding the standardized attractants used in traps. This stimulant has not been identified, it may be a combination of visual and chemical odor attractants. Further studies should evaluate these stimulants to alleviate such high numbers of biting mosquitoes on horses. If the traps are being out competed by a natural host, then trap placement is important to effectively reduce adult mosquito populations around horses. The distance between a natural host and trap where the competitive state decreases needs to be investigated. This information could provide mosquito trap manufactures beneficial data to make suggestions about trap placement.

PAGE 86

73 Figure 3-1. Portable vacuum aspirator (DC Insect Vac. BioQuip, Rancho Dominguez, CA) used for collection of mosquitoes during vacuuming studies.

PAGE 87

74 Figure 3-2. Paint gelding that was sampled for mosquitoes during trapping studies.

PAGE 88

75 Figure 3-3. Appaloosa mare that was sampled for mosquitoes during trapping studies.

PAGE 89

76 Figure 3-4. CDC 1012 trap placement during CDC 1012 trapping study at the UF Horse Teaching Unit from July 19, 2004 through August 21, 2004

PAGE 90

PASTURE PASTURE 77 CDC 1012 trap Horse PASTURE PASTURE PASTURE Figure 3-5. CDC 1012 trapping study placement used for mosquito trapping.

PAGE 91

78 Figure 3-6. MMPro trap placement during MMPro trapping study at the UF Horse Teaching Unit from October 1, 2004 through October 21, 2004

PAGE 92

79 5.7 acres 5.0 acres MMPro trap Covered Arena and Stalls Figure 3-7. Mosquito Magnet Pro trap placement used during mosquito trapping study.

PAGE 93

80 Figure 3-8. Modified CDC 1012 mosquito trap used during trapping horse odor trapping study at the UF Horse Teaching Unit from August 27, 2004 through September 24, 2004.

PAGE 94

81 PASTURE PASTURE PASTURE Figure 3-9. CDC 1012 trap layout used for mosquito location profile study. 64.92 meters 72.51 meters 161.54 meters 32.31 meters PASTURE PASTURE 81

PAGE 95

82 Figure 3-10. Technique of aspirating mosquitoes off of horse during horse vacuuming study and separate entity study.

PAGE 96

83 Figure 3-11. Feeding slips at UF Horse Teaching Unit where the 2nd trial of horse vacuuming study and separate entity study was conducted.

PAGE 97

84 Total # Mosquitoes Interval54321No HorseHorseNo HorseHorseNo HorseHorseNo HorseHorseNo HorseHorse 120100806040200 11.521.233.86.35130.532.752.381.354.2 Figure 3-12. Distribution and mean numbers of mosquitoes trapped with CDC 1012 for treatments, horse and no horse present, and intervals of testing from July 19, 2004 through August 21, 2004, at an equine facility in Gainesville, Florida.

PAGE 98

85 3063820312769325488314196183010020030040050060012345Time intervalsTotal # mosquitoes Horse No Horse Figure 3-13. Total numbers of mosquitoes captured with CDC 1012 trap when a horse is present during interval 2 and when no horse is present in interval 2; testing from July 19, 2004 through August 21, 2004, at an equine facility in Gainesville, Florida.

PAGE 99

86 51.06.333.821.211.554.281.252.332.730.50.010.020.030.040.050.060.070.080.090.012345Time intervalsAvg # of mosquitoes Horse No Horse Figure 3-14. Mean numbers of mosquitoes captured with CDC 1012 trap when a horse is present during interval 2 and when no horse is present in interval 2; testing from July 19, 2004 through August 21, 2004, at an equine facility in Gainesville, Florida.

PAGE 100

87 Total # Mosquitoes Interval54321NoHorseHorseNoHorseHorseNoHorseHorseNoHorseHorseNoHorseHorse 4003002001000 5779.3333157.66756.8333163.537.666750.586.1667129.833103.667 Figure 3-15. Distribution and mean numbers of mosquitoes trapped with MMPro for treatments, horse and no horse present, and intervals of testing from October 1, 2004 through October 21, 2004, at UF Horse Teaching Unit in Gainesville, Florida.

PAGE 101

88 98234194647634162277951730322602004006008001000120012345Time intervalsTotal # mosquitoes Horse No Horse Figure 3-16. Total numbers of mosquitoes captured with MMPro trap when a horse is present during interval 2 and when no horse is present in interval 2; testing from October 1, 2004 until October 21, 2004, at an equine facility in Gainesville, Florida.

PAGE 102

89 163.756.8157.779.356.8103.7129.886.250.537.702040608010012014016018012345Time IntervalsMean # of mosquitoes Horse No Horse Figure 3-17. Mean numbers of mosquitoes captured with MMPro trap when a horse is present during interval 2 and when no horse is present in interval 2; testing from October1, 2004 until October 21, 2004, at an equine facility in Gainesville, Florida.

PAGE 103

90 TreatmentTotalMosquitoes NoHorseHorseVacuumOdorHorse 1101009080706050403020 79.828.841.534.6 Figure 3-18. Distribution and mean numbers of mosquitoes trapped with CDC 1012 for treatments, horse, vacuumed odor, and no horse present; testing from August 27, 2004 through September 24, 2004, at an equine facility in Gainesville, Florida.

PAGE 104

91 34.6741.528.8379.830102030405060708090HorseVacuum OdorsHorseNo HorseTreatmentMean # of Mosquitoes Figure 3-19. The mean numbers of mosquitoes trapped with CDC 1012 for treatment groups horse, vacuumed odors, and no horse present; testing from August 27, 2004 through September 24, 2004, at an equine facility in Gainesville, Florida.

PAGE 105

92 34.6741.528.8379.83111.5153.087.366.00204060801001201401601801234Time IntervalMean # of Mosquitoes Horse Treatment No Horse Treatment Figure 3-20. Mean numbers of mosquitoes captured with CDC 1012 trap when a horse is present during interval 2 and when no horse is present in interval 2; testing from August 27, 2004 through September 24, 2004, at an equine facility in Gainesville, Florida.

PAGE 106

93 Cx. nigripalpus16.37%Cx. erraticus2.35%Cx. salinarius6.60%Ma. titillans65.85%Ma. dyari0.35%Cq. perturbans8.09%Ae. vexans0.04%Ps. ferox0.33%Ps. cyanescens0.01% Figure 3-21. Mosquito species (represented as a percent of total mosquitoes collected) by CDC 1012 traps during position study conducted from July 13, 2004 through August 26, 2004, at an equine facility in Gainesville, Florida.

PAGE 107

94 Trap PositionTotal Mosquitoes 431 1400120010008006004002000 200.1531.1103.1 Figure 3-22. Distribution and mean numbers of mosquitoes captured per trapping position with CDC 1012 traps from July 13, 2004 through August 26, 2004, at an equine facility in Gainesville, Florida.

PAGE 108

95 103.1531.1200.10100200300400500600123Trap PositionsMean Values of Mosquitoes Figure 3-23. Mean numbers of mosquitoes captured per trapping position with CDC 1012 traps from July 13, 2004 through August 26, 2004, at an equine facility in Gainesville, Florida.

PAGE 109

96 Cx. nigripalpus9.4%Cx. erraticus2.2%Cx. salinarius36.9%Ma. titillans11.9%An. crucians0.3%Cq. perturbans39.4% Figure 3-24. Mosquito species (represented as a percent of total mosquitoes collected) collected by a vacuum aspirator during study conducted from June 15, 2004 through July 12, 2004, at an equine facility in Gainesville, Florida.

PAGE 110

97 Cx. nigripalpus41.7%Cx. erraticus5.3%Cx. salinarius2.4%Ma. titillans27.6%An. crucians0.8%An. quadrimaculatus2.3%Cq. perturbans1.8%Ps. columbiae17.4%Oc. infirmatus0.1%Ps. ferox0.3%Ps. ciliata0.3% Figure 3-25. Mosquito species (represented as a percent of total mosquitoes collected) collected by a vacuum aspirator during study conducted from October 4, 2004 through October 22, 2004, at an equine facility in Gainesville, Florida.

PAGE 111

98 Cx.nigripalpus49.96%Cx.erraticus6.40%Cx.salinarius2.37%Ma.titillans21.10%An.crucians0.69%An.quadrimaculatus1.87%Cq.perturbans1.50%Ps.columbiae15.51%Oc.infirmatus0.06%Ps.cilitia0.25%Ps.ferox0.29% Figure 3-26. Mosquito species (represented as a percent of total mosquitoes collected) collected from Appaloosa mare by a vacuum aspirator during study conducted from October 4, 2004 through October 22, 2004, at an equine facility in Gainesville, Florida.

PAGE 112

99 Cx.nigripalpus25.01%Cx.erraticus3.10%Cx.salinarius2.43%Ma.titillans40.64%An.crucians1.09%An.quadrimaculatus3.31%Cq.perturbans2.30%Ps.columbiae21.11%Ps.cilitia0.34%Ps.ferox0.42%Oc.infirmatus0.25% Figure 3-27. Mosquito species (represented as a percent of total mosquitoes collected) collected from Paint gelding by a vacuum aspirator during study conducted from October 4, 2004 through October 22, 2004, at an equine facility in Gainesville, Florida.

PAGE 113

100 Total # Mosquitoes Distance54321Horse 2Horse 1Horse 2Horse 1Horse 2Horse 1Horse 2Horse 1Horse 2Horse 1 4003002001000 184.67159.83175.17202172.3368.6773.3393.3392.83104 Figure 3-28. Distribution and mean numbers of total mosquitoes captured from horse 1 and horse 2 per trapping distance with vacuum aspirator from October 4, 2004 through October 21, 2004, at the UF Horse Teaching Unit in Gainesville, Florida.

PAGE 114

101 Table 3-1. Schedule of CDC 1012 trapping study showing treatment groups from July 19, 2004 through August 21, 2004 at an equine facility in Gainesville, Florida. Date Treatment 07/19/2004 Horse Present 07/21/2004 Horse Present 07/29/2004 Horse Present 08/02/2004 No Horse Present 08/09/2004 No Horse Present 08/11/2004 No Horse Present 08/14/2004 Horse Present 08/15/2004 No Horse Present 08/16/2004 Horse Present 08/19/2004 No Horse Present 08/20/2004 Horse Present 08/21/2004 No Horse Present

PAGE 115

102 Table 3-2. Schedule of MMPro trapping study showing treatment groups from October 3, 2004 through October 21, 2004 at an equine facility in Gainesville, Florida. Date Treatment 10/03/2004 Horse Present 10/04/2004 No Horse Present 10/05/2004 No Horse Present 10/06/2004 Horse Present 10/07/2004 No Horse Present 10/10/2004 No Horse Present 10/11/2004 Horse Present 10/12/2004 No Horse Present 10/13/2004 Horse Present 10/18/2004 Horse Present 10/19/2004 Horse Present 10/21/2004 No Horse

PAGE 116

103 Table 3-3. Interval schedule of horse odor study showing treatment groups from August 27, 2004 through September 24, 2004 at an equine facility in Gainesville, Florida. Interval Treatment 1 Horse Present 2 Horse Odor Vacuumed 3 Horse Present 4 Horse Not Present

PAGE 117

104 Table 3-4. Schedule of horse odor study showing dates of treatment groups from August 27, 2004 through September 24, 2004 at an equine facility in Gainesville, Florida. Date Treatment 08/27/2004 No Horse Present 08/29/2004 No Horse Present 08/30/2004 Horse Present 08/31/2004 No Horse Present 09/01/2004 Horse Present 09/02/2004 Horse Present 09/03/2004 Horse Present 09/20/2004 Horse Present 09/21/2004 No Horse Present 09/24/2004 Horse Present

PAGE 118

105 Table 3-5. Schedule of separate entity study showing stall assignments from October 4, 2004 through October 22, 2004 at an equine facility in Gainesville, Florida. Date Horse 1 Horse 2 7:45 8:05 8:25 8:45 9:05 10/04/2004 1 2 3 4 5 10/05/2004 1 2 3 4 5 10/07/2004 1 2 4 5 3 10/12/2004 1 2 5 3 4 10/14/2004 1 5 4 3 2 7 10/21/2004 1 4 2 5 3 7 10/22/2004 1 7

PAGE 119

106 Table 3-6. Mean numbers ( standard deviation) of mosquitoes captured per trapping interval and total numbers of mosquitoes captured per trapping interval with a CDC 1012 trap from July 19, 2004 through August 21, 2004, at an equine facility in Gainesville, Florida. Treatment Interval Mean Catch ( Std) n Total Catch Horse Present 1 51.0 ( 32.2)a 6 306 2 (Horse in Stall) 6.33 ( 5.82)c 6 38 3 33.83 ( 13.56)a,b 6 203 4 21.17 ( 5.23)b,c 6 127 5 11.50 ( 4.18)b,c 6 69 No Horse Present 1 54.17 ( 16.70)e 6 325 2 81.33 ( 22.21)d 6 488 3 52.3 ( 25.5)e 6 314 4 32.67 ( 14.39)f 6 196 5 30.50 (15.08)f 6 183 Note: Means followed by the same letter are not significantly different (P 0.05), and n= numbers of observations.

PAGE 120

107 Table 3-7. Mean numbers ( standard deviation) of mosquitoes captured per trapping interval and total numbers of mosquitoes captured per trapping interval with a MMPro trap from October 1, 2004 through October 21, 2004, at an equine facility in Gainesville, Florida. Treatment Interval Mean Catch ( Std) n Total Catch Horse Present 1 163.5 ( 98.7)a 6 982 2 (Horse in Stall) 56.8 ( 51.4)b 6 341 3 157.7 ( 131.6)a 6 946 4 79.3 ( 38.2)b 6 476 5 57.0 ( 35.5)b 6 341 No Horse Present 1 103.7 ( 51.6)c 6 622 2 129.8 ( 80.5)c 6 779 3 86.2 ( 46.9)c,d 6 517 4 50.5 ( 19.6)d 6 303 5 37.67 ( 18.94)d 6 226 Note: Means followed by the same letter are not significantly different (P 0.05), and n= numbers of observations.

PAGE 121

108 Table 3-8. Mean numbers ( standard deviation) of mosquitoes captured per trapping treatment and total numbers of mosquitoes captured per trapping treatment with a CDC 1012 trap from August 27, 2004 through September, 2004, at an equine facility in Gainesville, Florida. Treatment Mean Catch ( Std) n Total Catch Horse Present 34.67 4.68b 6 208 Horse Odor Vacuumed 41.50 5.17b 6 249 Horse Present 28.83 6.31b 6 173 No Horse Present 79.83 18.66a 6 479 Note: Means followed by the same letter are not significantly different (P 0.05), and n= numbers of observations.

PAGE 122

109 Table 3-9. Mean numbers ( standard deviation) of mosquitoes captured per trapping position and total numbers of mosquitoes captured per trapping position with CDC 1012 traps from July 13, 2004 through August 26, 2004, at an equine facility in Gainesville, Florida. Trap Position Mean Catch ( Std) n Total Catch 1 103.1 ( 114.4)b 10 1134 2 531.0 ( 538)a 8 4249 3 200.1 ( 167.8)b 9 1801 Note: Means followed by the same letter are not significantly different (P 0.05), and n= numbers of observations.

PAGE 123

110 Table 3-10. Mean numbers of mosquito species trapped per position during location profile study from July 13, 2004 through August 26, 2004, at the UF Horse Teaching Unit, Gainesville, Florida. Species Trap Position Mean Culex nigripalpus 1 46.90a 2 60.88a 3 24.44a Culex erraticus 1 3.40b 2 9.12b 3 6.67b Culex salinarius 1 11.80d 2 33.00c 3 9.89d Mansonia titillans 1 30.90f 2 378.0e 3 143.0e,f Coquillettidia perturbans 1 7.80h 2 45.75g 3 14.78g,h Note: Means followed by the same letter are not significantly different (P 0.05), and n= numbers of observations.

PAGE 124

111 Table 3-11. Mean numbers ( standard deviation) of difference of mosquitoes vacuumed from horse 1 and horse 2 per distance and total difference of mosquitoes vacuumed from horse 1 and horse 2 per distance with a vacuum aspirator from October 4, 2004 through October 22, 2004, at the UF Horse Teaching Unit, Gainesville, Florida. Distance Mean Difference Vacuumed ( Std) n Difference Total Catch 1 68.3 ( 84.5)a 6 410 2 109.2 ( 103.6)a 6 655 3 81.8 ( 78.2)a 6 491 4 86.5 ( 74.8)a 6 519 5 116.0 ( 22.5)a 3 348 Note: Means followed by the same letter are not significantly different (P 0.05), and n= numbers of observations.

PAGE 125

112 Table 3-12. Mean numbers ( standard deviation) of mosquitoes vacuumed from horse 1 and horse 2 per distance and total mosquitoes vacuumed from horse 1 and horse 2 per distance with a vacuum aspirator from October 4, 2004 through October 22, 2004, at the UF Horse Teaching Unit, Gainesville, Florida. Horse Distance Mean Vacuumed ( Std) n Total Catch 1 1 172.3 ( 87.8)a 6 1034 2 202.0 ( 124.9)a 6 1212 3 175.2 ( 85.1)a 6 1051 4 159.8 ( 86.4)a 6 959 5 184.7 ( 38.6)a 3 554 2 1 104.0 ( 24.8)b 6 624 2 92.8 ( 45.5)b 6 557 3 93.3 ( 26.9)b 6 560 4 73.3 ( 25.0)b 6 440 5 68.8 ( 23.9)b 3 206 Note: Means followed by the same letter are not significantly different (P 0.05), and n= numbers of observations.

PAGE 126

CHAPTER 4 SUMMARY Mosquito traps are a very effective surveillance tool to monitor seasonal prevalence and the species complex in a specific area. Traps are also effective in reducing numbers of mosquitoes in the location of the trap. The ability of monitoring seasonality population trends were evaluated which enables efficient mosquito control and knowledge of when to take protective measures for horses against mosquitoes. Therefore, control programs should encompass a range of methods to successfully combat diverse mosquito species. The seasonality study conducted from September 26, 2003 through September 2, 2004 at provided important information on nuisance mosquito population trends. The mosquito populations peak during warm, wet seasons and therefore protective measures should be taken by horse owners should protect their animals the most during these seasons. The trapping studies conducted during this revealed important information pertaining to the effectiveness of mosquito traps when placed in a competitive situation with a natural host, such as a horse. During this study, the traps were out competed when a horse was present, even when horse odors were fed into the trap airstream. If traps are the main tool of adult mosquito surveillance, then detecting the quantities of a given mosquito species feeding on horses could be skewed. The mosquito species trapped, regardless of whether a horse was present or not, did not significantly change. So using current traps to monitor the presence of species may be adequate. When the horses were vacuumed to directly collect mosquitoes feeding, an alarming number of mosquitoes were present. Some of the species captured were known vectors of arboviruses present in 113

PAGE 127

114 Gainesville, Florida. With such high biting rates, the potential of disease transmission is high. By improving trap efficiency when horses are present, this potential of disease transmission could be reduced. Knowing at what distance the traps become ineffective when a natural host is present is valuable information that should be examined. The data obtained during the separate entity study revealed that when two horses are as close as 3 meters and as far as 21 meters there is no difference in mosquito numbers between the horses. This indicates that either mosquitoes can determine the presence of two horses even when they are close together, or they require more than 21 meters to achieve this. If a similar study is conducted using a trap and a horse, the information obtained could aid trap manufactures with trap placement recommendations.

PAGE 128

LIST OF REFERENCES Allan, S.A., Day, J., and Edman, J.D. (1987). Visual ecology of biting flies. Ann. Rev. Entomol. 32, 297-316. Appleton, C.C., and Sharp, B.L. (1985). A preliminary study on the emergence of Mansonia uniformis (Diptera: Culicidae) from swamps at Richards Bay, Natal, South Africa. J. Ent. Soc. Sth. Afr. 48(1), 179-184. Arunachalam, N., Samuel, P.P., Hiriyan, J., Thenmoxhi, V. (2002). Vertical transmission of Japanese Encephalitis virus in Mansonia species, in an epidemic-prone area of southern India. Annuals of Tropical Medicine & Parasitology 96(4), 419-420. Belkin, J. N. (1954). Simple larval and adult mosquito indexes for routine mosquito control operations. Mosquito News 14, 127-131. Bhuyan, M., and Das, S.C. (1985). Field trials of colour affinity of host seeking Mansonia mosquitoes. Indian J. Med. Res. 82, 139-140. Bidlingmayer, W.L. (1974). The influence of environmental factors and physiological stage on flight patterns of mosquitoes taken in the vehicle aspirator and truck, suction, bait and New Jersey light trap. J. Med. Entomol. 11, 119-146. Bidlingmayer, W.L. (1985). The measurement of adult mosquito population changes-some considerations. J. Am. Mosq. Control Assoc. 1(3), 328-348. Bradley, G.H., and McNeel, T.E. (1935). Mosquito collections in Florida with the New Jersey light trap. J. Econ. Entomol. 28, 780-786. Braverman, Y., Kitron, U., and Killick-Kendrick, R. (1991). Attractiveness of vertebrate hosts to Culex pipiens (Diptera: Culicidae) and other mosquitoes in Israel. J. Med. Entomol. 28(1), 133-138. Brown, A.W. (1951). Factors which attract Aedes mosquitos to humans. Proceedings 10th Internat'l Congress of Entomol. 3, 757-763. Burkett, D., Lee, W., Lee, K., Kim, H., Lee, H., Lee, J., Shin, E., Wirtz, R., Cho, H., Claborn, D., Coleman, R., and Kline, T. (2001). Light, carbon dioxide, and octenol-baited mosquito trap and host-seeking activity evaluations for mosquitoes in a malarious area of the Republic of Korea. J. Am. Mosq. Control Assoc. 17, 196-205. 115

PAGE 129

116 Byford, R.L., Craig, M.E., and Crosby, B.L. (1992). A review of ectoparasites and their effect on cattle production. J. Anim. Sci. 70, 597-602. Campbell, C.B. (2003). Evaluation of five mosquito traps and a horse for West Nile vectors on a north Florida equine facility. Thesis, University of Florida, Gainesville, Florida. Centers for Disease Control and Prevention. (2001). Division of vector-borne infectious diseases: arboviral encephalitides. Retrieved August 20, 2004 from http://www.cdc.gov/ncidod/dvbid/arbor/arbdet.htm Centers for Disease Control and Prevention. (2003A). Division of vector-borne infectious diseases: Dengue Fever. Retrieved June 18, 2004 from http://www.cdc.gov/ncidod/dvbid/dengue/index.htm Centers for Disease Control and Prevention. (2003B). National center for infectious diseases. Travelers health: Lymphatic filariasis. Retrieved June 18, 2004 from http://www.cdc.gov/travel/diseases/filariasis.htm Centers for Disease Control and Prevention. (2004). Malaria SurveillanceUnited States, 2002. MMWR Morb. Mortal. Wkly. Rep. 53, 412-413. Crowley, J.C. (2003). Determining seasonality of nuisance flies and evaluating stable fly pests on horses at an equine facility in North Central Florida. Thesis, University of Florida, Gainesville, Florida. Cupp, E.W. and Stokes, G.M. (1973). Identification of bloodmeals from mosquitoes collected in light traps and dog-baited traps. Mosquito News 33, 39-41. Day, J.F., and Curtis, G.A. (1989). Influence of rainfall on Culex nigripalpus (Diptera: Culicidae) blood-feeding behavior in Indian River County, Florida. Ann. Entomol. Soc. Am. 82(1), 32-37. Defoliart, G.R., Grimstad, P.R., Watts, D.M. (1987). Advances in mosquito-borne arbovirus-vector research. Annu. Rev. Entomol. 32, 479-505. Duryea, R., Donnelly, J., Guthrie, D., O'Malley, C., Romanowski, M., Schmidt, R. (1996). Gambusia effectiveness in New Jersey mosquito control. Proceedings of the Eighty-Third Annual Meeting of the New Jersey Mosquito Control Association, Inc. Lindenworld, New Jersey, 95-102. Dwinell, S.E., Baker, R.H., Barnard, D.R., Barnett, B., Beidler, E.J., Conklin, E.J., Dominy, R., Harden, P.G., McCullagh, L.N., Milio, J., Mulrennan, J.A., Robinson, J.W., Smith, J.P., Wiersma, S., Wells, R.A. (1998). Florida Coordinating Council on Mosquito Control. In: Florida Mosquito Control: The State of the Mission as Defined by Mosquito Controllers, Regulators, and Environmental Managers. University of Florida, Vero Beach.

PAGE 130

117 Edman, J.D. (1971). Host-feeding patterns of Florida mosquitoes: Aedes, Anopheles, Coquillettidia, Mansonia and Psorophora. J. Med. Entomol. 8(6), 687-695. Fletcher, M.G., Turner, E.C., Hansen, J.W. (1988). Horse-baited insect trap and mobile insect sorting table used in a disease vector transmission identification study. J. Am. Mosq. Control Assoc. 4(4), 431-435. Floore, T. (2003). Mosquito information. In: The American Mosquito Control Association. Retrieved June 2, 2004 from http://www.mosquito.org/info.php Gentry, A.B. (2002) Evaluation of protection systems and determination of seasonality for mosquito and biting flies at the University of Florida Horse Teaching Unit. Thesis, University of Florida, Gainesville, Florida. Gibson, G., and Torr, S.J. (1999). Visual and olfactory responses of haematophagous Diptera to host stimuli. Medical and Veterinary Entomology 13, 2-13. Gillies, M.T. (1980). The role of carbon dioxide in host finding by mosquitoes (Diptera; Culicidae): a review. Bull. Ent. Res. 70, 525-532. Gillies, M.T., and Wilkes, T.J. (1972). The range of attraction of animal baits and carbon dioxide for mosquitoes. Studies in a freshwater area of West Africa. Bull. Ent. Res. 61, 389-404. Grimstad, P.R., and DeFoliart, G.R. (1975). Mosquito nectar feeding in Wisconsin in relation to twilight and microclimate. J. Med. Entomol. 11, 691-698. Haddow, A.J. (1942). The mosquito fauna and climate of native huts at Kisumu, Kenya. Bull. Ent. Res. 33, 91-142. Hall, D.R., Beevor, P.S., Cork, A., Nesbit, B.F. and Vale, G.A. (1984). 1-Octen-3-ol: a potent olfactory stimulant and attractant for tsetse isolated from cattle odours. Insect Sci. Applic. 5, 335-339. Harwood, R.F. and M.T. James. (1979). Entomology in Human and Animal Health. MacMillan Publishing Co., New York. Howlett, F.M. (1910). The influence of temperature upon the biting of mosquitoes. Parasitology 3, 479-484. Hubalek, Z., and Halouzka, J. (1999). West Nile fever-a reemerging mosquito-borne viral disease in Europe. Emerging Infectious Diseases 5(5), 1-25. Jones, R.H., Hayes, R.O., Potter, H.W., Francy, D.B. (1977). A survey of biting flies attacking equines in three states of the southwestern United States. J. Med. Entomol. 14, 441-447.

PAGE 131

118 Khan, A.A., Maibach, H.I., Strauss, W.G., Fenley, W.R. (1965). Screening humans for degrees of attractiveness to mosquitoes. J. Econ. Entomol. 58, 694-697. Kline, D.L. (1994). Olfactory attractants for mosquito surveillance and octenol: 1-octen-3-ol. J. Am. Mosq. Control Assoc. 10, 280-287. Kline, D.L. (1999). Comparison of two American Biophysics mosquito traps: the professional and a new counterflow geometry trap. J. Am. Mosq. Control Assoc. 15(3), 276-282. Kline, D.L. (2002). Evaluation of various models of propane-powered mosquito traps. J. Vec. Ecol. 27(1), 1-7. Kline, D.L., and Mann, M.O. (1998). Evaluation of butanone, carbon dioxide, and 1-octen-3-ol as attractants for mosquitoes associated with north central Florida bay and cypress swamps. J. Am. Mosq. Control Assoc. 14, 289-297. Kline, D.L., Wood, J.R., and Morris, C.D. (1990). Evaluation of 1-octen-3-ol as an attractant for Coquillettidia perturbans, Mansonia spp. and Culex spp. associated with phosphate mining operation. J. Am. Mosq. Control Assoc. 6, 605-611. Knight, K.L., and Stone, A. (1977). A Catalog of the Mosquitoes of the World (Diptera: Culicidae). Entomological Society of America, College Park. Kumm, H.W. and Zuniga, H. (1944). Seasonal variation in the numbers of Anopheles albimanus and A. pseudopunctipennis caught in stable traps in Central America. Amer. J. Hyg. 39, 8-15. Lanciotti, R.S. (1999). Origin of the West Nile virus responsible for an outbreak of encephalitis in the northeastern United States. Science 286, 2333-2339. Magoon, E. (1935). A portable stable trap for capturing mosquitoes. Bull. Entomol. Res. 26, 363-369. Malkinson, M., Banet, C., Weisman, Y., Pokamunski, S., King, R., Drouet, M.T., and Deubel, V. (2002). Introduction of West Nile virus in the Middle East by migrating white storks. Emerging Infectious Diseases 8, 392-397. Mboera, L.E.G. and Takken, W. (1997). Carbon dioxide chemotropism in mosquitoes (Diptera: Culicidae) and its potential in vector surveillance and management programmes. Rev. Med. and Vet. Ent. 85, 355-368. McNelly, J.R. (1989). The CDC trap as a special monitoring tool. Proceedings of the Seventy-Sixth Annual Meeting of the New Jersey Mosquito Control Association, Inc. Lindenworld, New Jersey, 26-33.

PAGE 132

119 Mendez, W., Liria, J., Navarro, J.C., Garcia, C.Z., Freier, J.E., Salas, R., Weaver, S.C., and Barrera, R. (2001). Spatial dispersion of adult mosquitoes (Diptera: Culicidae) in a sylvatic focus on Venezuelan Equine Encephalitis virus. J. Med. Entomol. 38(6), 813-821. Mitchell, C.J., Darsie, R.F., Monath, T.P., Sabattini, M.S., and Daffner, J. (1985). The use of an animal-baited net trap for collecting mosquitoes during Western Equine Encephalitis investigations in Argentina. J. Am. Mosq. Control Assoc. 1(1), 43-47. Muir, L.E., Thorne, M.J., Kay, B.H. (1992). Aedes aegypti (Diptera: Culicidae) vision: spectral sensitivity and other perceptual parameters of the female eye. J. Med. Entomol. 29(2), 278-281. Nasci, R.S., Savage, H.M, White, D.J., Miller, J.R., Cropp, B.C., Godsey, M.S., Kerst, A.J., Bennett, P., Gottfried, K., and Lanciotti, R.S. (2001). West nile virus in overwintering Culex mosquitoes, New York City, 2000. Emerg. Infect. Dis. 7(4), 742-745. Nayar, J.K. (1968). Effects of larval and pupal environmental factors on biological status of adults at emergence in Aedes taeniorhynchus. Bull. Entomol. Res. 58(4), 811-827. Ogg, C.L. (2002). Mosquito biology and mosquito-borne diseases. In: Public Health Pest Control Pesticide Safety Education Program, 1-12. Retrieved June 28, 2004 from http://pested.unl.edu/catmans/public/chapter2.pdf Porter, M.B., Long, M.T., Getman, L.M., Giguere, S., MacKay, R.J., Lester, G.D., Alleman, A.R., Wamsley, H.L., Franklin, R.P., Jacks, S., Buergelt, C.D., Detrisac, C.J. (2003). West Nile virus encephalomyelitis in horses: 46 cases (2001). J. Am. Vet. Med. Assoc. 222(9), 1241-1247. Provost, M.W. (1952). The dispersal of Aedes taeniorhynchus. Preliminary studies. Mosq. News 12, 174-190. Public Health Entomology Research and Education Center (PHEREC). (2002). Retrieved July 1, 2004 from http://www.pherec.org/training/mosquitolistflorida.html Rose, R.I. (2001). Pesticides and public health: integrated methods of mosquito management. Emerg. Infect. Dis. 7(1), 17-24. Rowley, W.A., and Graham, C.L. (1968). The effect of temperature and relative humidity on the flight performance of female Aedes aegypti. J. Insect Physiol. 14, 1251-1257. Sardelis, M.R., Turell, M.J., Dohm, D.J., OGuinn, M.L. (2001). Vector competence of selected north American Culex and Coquillettidia mosquitoes for West Nile virus. Emerg. Infect. Dis. 7(6), 1-5.

PAGE 133

120 Schmidt, R.F. (2003). Relationship of landing count observations to the time of sunset. Wing Beats-of the Florida Mosquito Control Association 14(3), 12-19. Slaff, M., and Haefner, J.D. (1985). Seasonal and spatial distribution of Mansonia dyari, Mansonia titillans, and Coquillettidia perturbans (Diptera: Culicidae) in the central Florida, USA, phosphate region. J. Med. Entomol. 22(6), 624-629. Spielman, A., and M.D'Antonio. (2001). Mosquito: A natural history of our most persistent and deadly foe. Hyperion Press, New York. 255-256. Stamm, D.D., Chamberlain, R.W., Sudia, W.D. (1962). Arbovirus studies in south Alabama, 1957-1958. Am. J. Hyg. 76:61-81. Stark, L.M., and D. Kazanis. (2003). Arbovirus surveillance: annual summary report 2003, pp. 1-25. Florida Department of Health, Bureau of Laboratories, Tampa, FL. Retrieved August 28, 2004 from www.doh.state.fl.us/Environment/hsee/arbo/pdf_files/2003_arboannual.pdf Steelman, C.D. (1979). Economic thresholds in mosquito IPM programs. Mosquito News 39(4), 730-742. Takken, W. and Kline, D.L. (1989). Odors as mosquito attractants. J. Am. Mosq. Control Assoc. 5, 311-316. Taylor, L.R. (1963). Analysis of the effect of temperature on insects in flight. J. Anim. Ecol. 32, 99-117. TenBroeck, C., Hurst, E.W., and Traub, E. (1935). Epidemiology of equine encephalomyelitis in the eastern United States. J. Exp. Med. 62, 677. Waters, B. T. and M. Slaff. (1987). A small habitat, larval mosquito sampler. J. Am. Mosq. Cont. Assoc. 3, 514. Wood, P.W., Wright, R.H. (1968). Some responses of flying Aedes aegypti to visual stimuli. Can. Entomol. 100, 504-13. Woodbridge, A.F., and Walker, E.D. (2002). Mosquitoes (Culicidae). In: Medicial and Veterinary Entomology. Elsevier Science, San Diego, 203-257. Zyzak, M., Loyless, T., Cope, S., Wooster, M., and Day, J. (2002). Seasonal abundance of Culex nigripalpus Theobald and Culex salinarius Coquillett in north Florida, USA. J. Vec. Ecol. 21(1), 155-162.

PAGE 134

BIOGRAPHICAL SKETCH Sarah Courtney Dilling was born in Glen Cove, New York, in 1979 to Robert and Susan Walsh. She moved to Pine Island, Florida, in 1984 where she remained until 1996. Sarah attended and graduated from Venice High School in Venice, Florida, in 1997. She obtained a Bachelor of Science degree in environmental science and policy from the University of South Florida in May 2001. Sarah took a year off between undergraduate and graduate school to work in an environmental lab as a chemist in Tampa, Florida. In November 2002, Sarah was accepted into a graduate program in equine health management under Dr. Saundra TenBroeck. During her graduate program she served as a teaching assistant for equine reproduction management, techniques in equine science, and sales preparation for thoroughbred yearlings under the supervision of Mr. Joel McQuagge. She graduated with a Master of Science degree in animal science from the University of Florida in December 2004. Sarah will pursue a career in the biological science field. 121