Biology and conservation of the Harpy Eagle in Venezuela and Panama

MISSING IMAGE

Material Information

Title:
Biology and conservation of the Harpy Eagle in Venezuela and Panama
Physical Description:
xvii, 212 leaves : ill. ; 29 cm.
Language:
English
Creator:
Alvarez-Cordero, Eduardo
Publication Date:

Subjects

Subjects / Keywords:
Eagles -- Geographical distribution   ( lcsh )
Eagles -- Habitat   ( lcsh )
Eagles -- Venezuela   ( lcsh )
Eagles -- Panama   ( lcsh )
Wildlife Ecology and Conservation thesis, Ph. D
Dissertations, Academic -- Wildlife Ecology and Conservation -- UF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1996.
Bibliography:
Includes bibliographical references (leaves 193-211).
Additional Physical Form:
Also available online.
Statement of Responsibility:
by Eduardo Alvarez-Cordero.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 026161450
oclc - 36781867
System ID:
AA00025760:00001

Full Text










BIOLOGY AND CONSERVATION OF THE HARPY EAGLE
IN VENEZUELA AND PANAMA















BY

EDUARDO ALVAREZ-CORDERO


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1996














ACKNOWLEDGMENTS

Many private and government institutions and a great number of individuals

contributed to the realization of this study. However, before any and all, I wish to

recognize and thank Lucie E. Zaugg, friend and wife, for the commitment and loving

support and dedication invested to ensure that our family survived these long years as

we became students again. We both knew that it would be a period full of growth,

sacrifices, and trials. Lucie has done more than her share every time that I disappeared

into the jungle (once for 3 months), and has tolerated (barely) having my work office at

home. Our two sons, Jorge David (12) and Juan Michael (9), have grown with the

project and are waiting and wondering if there is a life after the Ph.D.

We came to Gainesville, Florida, thanks to the encouragement of Poly, Dr.

Eleanor Storrs, who convinced Dr. Kent Redford to give me an audience. He and Dr.

John Robinson provided me with the opportunity to become a "Gator," after a 10 year

sabbatical from being an "Aggie," and I thank them for it. At various times during the

difficult adjustment to university life, word processors, and economic challenges, their

Program for Studies in Tropical Conservation (PSTC) came to the rescue (even when

they knew that I was not from Mexico or Brazil).

Dr. Michael W. Collopy perhaps did not deserve to get stuck with me and this

project, but such is life. He approached this challenge with his always positive and

generous attitude. Here was a Latino, with ZERO experience working with raptors,

saying that he would tackle the biggest one of them all. He has been a patient mentor,

answering my every question, carefully editing my documents, and kindly respecting my

frequent ignorance about his field of expertise. He was genuinely committed to see me








through, even after realizing that funding would be very difficult to obtain. I cannot

thank him enough for trusting me to get the job done, and for the constant

encouragement during the many months it took me to shape this dissertation. The one

thing I know that I must do to show him my respect is to turn this document into peer-

reviewed publications.

Drs. Melvin E. Sunquist, Susan K. Jacobson, Marianne C. Schmink, and Richard

A. Kiltie served on my committee and were extremely supportive while I prepared this

dissertation. They each provided a different insight and careful editing to improve this

document. I appreciate their patience and interest in this project that potentially had

many "loose ends" and few guarantees of being attainable. I enjoyed the opportunity to

take classes from Dr. Schmink and Dr. Jacobson; I learned much from those

experiences and both of these super-teachers.

I recognize the support of the University of Florida, and its Department of

Wildlife Ecology and Conservation for the opportunity to train and learn on this campus,

for the equipment loans, the wonderful libraries, and flawless access to the Internet so

that I could track the Harpy Eagles from my PC.

In 1988, I visited The Peregrine Fund (Boise, Idaho) and met Dr. Bill Bumham,

President of that organization. We discussed my plans to work with the Harpy Eagle.

After I moved to Florida, he stayed in touch and often offered his support. In 1991, as

soon as I was ready to go into the field, it became evident promised funding was not

forthcoming, posing an additional challenge for this project. Through a "gentlemen"

agreement between Dr. Collopy and Dr. Burnham, I was presented with the unique

opportunity of developing the field project for the Harpy Eagle Conservation Program of

The Peregrine Fund. They employed me and supported all the needs of the project.

They have treated me as a professional and a friend, while also allowing me the

latitude of being a student. I could not have hoped for anything better, but the








Bumhams also have provided my family with the constant assurance that even if we

were not in Boise, we belonged. The wholehearted and ongoing commitment of the

"Peregrine Fund Family" to support this project is largely credited for what we have

been able to accomplish so far. I extend my gratitude to all my colleagues, coworkers,

directors, and the many members that continue to support this work.

In 1989, I began training a young Venezuelan who climbed the first active Harpy

Eagle nest that Dr. Collopy and I found during the preliminary field season. Eladio

Rafael Alvarez was sponsored in 1992 by The Peregrine Fund as a full-time Field

Assistant to this study. His dedication and contribution to all aspects of the field work

have in a great measure made this study possible. Rafael is the world expert on this

species, having seen and handled more eagles, and having found and climbed more of

their nests that anyone else; whenever I write "I", please read "Rafael and I".

I wish to make special mention of the Center for the Study for Tropical Birds,

Inc., and its Director, Jack Clinton-Eitniear, who funded my first field trip in 1989. I thank

them for believing it could be done!

In Panama we worked under a 5-year cooperation agreement established in

1991 between our sponsoring organization, The Peregrine Fund, and the Protected

Areas branch of the Natural Resources Institute (Instituto de los Recursos Naturales

Renovables-INRENARE). We made numerous friends in this country, where we always

felt at home. Indra S. Candanedo was a constant collaborator and served as the

national counterpart to our investigation. Karla Aparicio and Belkys Jimenez also gave

us support; Karda collected valuable information during her internship with the project.

Jacobo Lacs provided many important insights and logistics, and lead us to a nest in

the Atlantic Basin. We also thank our collaborators in ANCON for their cooperation. Dr.

Nathan Gale and his wife also dedicated many hours to help us and demonstrated their

commitment to help the eagles. We dedicate a special mention of gratitude to the








people of the Darien, particularly the park rangers headed by "Sr Chicho" who adopted

us and took us everywhere in a timely and safe fashion.

In Venezuela our research program was backed by CVG-EDELCA and the

Venezuelan Audubon Society (SCAV-Sociedad Conservacionista Audubon de

Venezuela), and authorized through annually renewed Research Permits (Lisencia de

Caceria Cientifica) issued by PROFAUNA, the Wildlife Programs Office of the Ministry

for the Environment (MARNR-PROFAUNA. Ministerio del Ambiente y de los Recursos

Naturales Renovables, Servicio Aut6nomo Programas de Fauna). Since 1988,

EDELCA's staff have supported this investigation of Harpy Eagles. This company has

provided a housing facility to base the project, space for a large flight-pen to hold

captive eagles, technical and logistic assistance, and even the occasional use of

helicopters.

The management of the Venezuelan government's regional development

corporation (L. Campos in litt., November 30, 1995), Corporaci6n Venezolana de

Guayana (CVG), granted access to an extensive database on regional natural

resources held by TECMIN (CVG-Tecnica Minera). We thank them for this privilege.

We acknowledge the logistic support received from CVG-Ferrominera del Orinoco,

particularly from their aviation department.

Our friends in Guri deserve a special mention, for their constant encouragement

and timely help--in particular my coworkers at Ecologia,Teo, Martin, Asterio, Leon, Ali,

who still treat me as family. The Balbas family was our hotel during the first year,

providing 5 star treatment for all the numerous guests that got involved with this project.

Luis A. Balbas as been a constant companion in this work, basically solving any

problems. I regret that he has not been able to see much of the wild eagles, but he has

done an excellent job caring for the live birds we have rescued and brought to Guri. I

also owe him for all the time he has spent administrating the project and providing








guidance to Rafael. Orlando Borrome introduced me to the use of GPS and provided

critical support to our field survey in Venezuela.

From the start of this study, cooperation with a group of Forestry Management

Units was crucial to the investigation of Harpy Eagles. Along the Imataca Range we

worked with SOMAGUA, CODEFORSA, Maderas Nuria I, INTECMACA, Aserradero

Hermanos Hemrnandez, and MADERORCA. In the Supamo River Basin our research

was supported by Aserradero Matamoros, Aserradero El Manteco, and Aserradero

Yocoima. Luis Matamoros, and his wife Verenice (my sister), were among the first to

enlist on our badwagon and have provided all sorts of support.

The Fuerzas Armadas de Cooperaci6n (FAC), the National Guard in Venezuela

deserve special mention for tracking down the hunters that shot two of the eagles that

we were monitoring, and more importantly, for retrieving NASA's satellite trasmitters so

that we could use them again. Misi6n cumplida!

Numerous private ranches allowed us to conduct research and provided logistic

support and important background information on the nests within their properties. We

are grateful for the cooperation and support received from all of these entities and their

representatives.

To compile and update the existing records for the Harpy Eagle we worked in

Mexico with Eduardo I1igo-Elias and Guido Rhar; in Costa Rica with Susan Wilson and

Belkys Jimenez; in Colombia with Cesar Marquez; and in Ecuador with Peter Jenny and

Mauricio Guerrero. We began work in Peru with Renzo Piana. Femrnando Antonini,

another student from that country, embarked under our guidance on a preliminary field

survey of the Tambopata region in 1996. In Guayana, we collaborated closely with Neil

Rettig who had already accomplished an impressive field survey and produced films on

the Harpy Eagles in that country. Sara Gilbert agreed to collect records for Suriname

during her stay in that country. Steve McGehee, a respected friend and colleague has








provided an endless stream of priceless references that only he can locate. Mike

McMillian visited the field project in Venezuela to deliver needed equipment and provide

additional telemetry training to Rafael.

We were trained to work with the eagles by Dr. David Ellis (NBS, Patuxent, MA),

one of the pioneering experts in the field of satellite telemetry applied to large avian

species. He provided valuable advise and encouragement, serving as the invisible

committee member that every grad student should have. Between 1992-96, Dr. Ellis

arranged the supply of ARGOS-TIROS satellite system transmitters to our project

through a worldwide effort funded by NASA; four additional transmitters were

purchased by The Peregrine Fund (1993-95). Dr. Paul Howie, of Microwave Telemetry,

not only provided the best satellite transmitters available, but also gave us the best

personalized attention and advise. I recognize the large contribution of NASA in

supplying 10 satellite transmitters to date, and the constant and reliable support of the

people that handle the incoming data, particularly Charlie and Jon.

We acknowledge the support of the following companies: Trimble Navigation

loaned top of the line GPS surveying equipment for the collection of data layers used in

our Geographical Information System. Ours GIS was develop using ARC VIEW versionn

2.1 by ESRI) donated to The Peregrine Fund by Environmental Systems Research

Institute (Redlands, CA). Horton Manufacturing Company, Inc., (Tallmadge, Ohio)

donated a crossbow used to get our climbing ropes over the nest trees; this avoided

the damage done to trees when climbing with spikes and allowed us to use safer rope

climbing methods. AVINET recently donated Pesola scales to the project.

Dr. Stefan Gorzula and Glenda Medina deserve credit for giving me the first

training I ever received in Field Biology. Stefan has been a valuable and supportive

colleague since I met him in the early 1980s. Glenda introduced me to the world of

birds and field techniques. Dr. Lee Harper also contributed to my early training.








Dr. Omar Linares opened the doors of his laboratory at the Universidad Simon

Bolivar to help process and identify the prey remains collected from nests. One of his

students assisted in this task.

Neil Rettig, the other world expert on the species, has demonstrated a Latino

like heart and willingness to share. He provided many important supplies and the

crossbow to shoot lines that enabled us to climb the nests. He also was kind enough to

guide me into Guyana and show me his nests. The support of his partner, Kim Hayes is

also recognized, as well as the funds provided to the project by Neil Rettig Productions.

We owe our lives to Luis Enrique Arnal, Kike, masterful climber who trained us

to reach the nests. ALWAYS on rope! SAFETY first! I remember when I was first given

enough climbing rope and no training, and told to just do it! Also, Kike has been the

project's photographer since I met him on a Cuipo in the Darien in 1993.

I must mention a group of long-time friends who adopted the cause of the

eagles and poured their hearts into making this project a reality. Peter E. Kung, for

getting me hooked on field biology >20 yr ago, teaching me to work with birds, and for

his total dedication to the project's success. He alone deserves credit for the GPS

surveys, an enormous accomplishment, and for believing that we could create a GIS on

a shoe string budget. Zan Merrill provided useful advise and unending support

throughout the study, always sharing the vision. Gustavo A. Martinez G. dedicated his

all to the needs of the project, serving as the in-country clearing house for the

Peregrine Fund during many of our trips. Gus and Elba, his wife, poured their kindness

and humor on myriads of people to make sure that they supported our goals. Gus was

instrumental in securing the GIS information and base maps for the study. He arranged

plane and helicopter rides, provided office support, and even gave our presentation at

a scientific meeting when we could not be there. Milardos de gracias!! Gus also








traveled many times to Florida, spending nights and days making my PC run flawlessly

so that I could complete this portion of the study.

My families in Venezuela and Canada were always behind me during this

endeavor. In Caracas, my parents Margot and Felipe, and my sister Belen, have been

a constant help and have run many errands to help me accomplish this work. My

mother basically donated her used car for the use of the project. Her 1962 Ford Falcon

made innumerable trips into the logging trails in the rainforest until it was totaled in an

unfortunate accident. My brother Cesar Alvarez has been extremely supportive, and

also arranged important logistic cooperation from CVG affiliated companies in the

Guayana Region. Ruth and Ferries in Alberta also supported this accomplishment in

every way possible.

Finally, I acknowledge the wonderful efforts of M. J. Schaer in formatting and

pulling together the drafts and final version of this dissertation, always with a positive,

friendly, and professional attitude.














TABLE OF CONTENTS

paqe

ACKNOW LEDGMENTS ............................. ................... ii

LIST O F TA BLES ..................... .......... ..................... xiii

LIST O F FIG URES ................................. .................. xiv

ABSTRACT ......................................... ................ xvi

C H A PT E R S .................................. ....................... 1

1 GENERAL INTRODUCTION ....................................... 1

Introd auction ............................ .......... .............. 1
B background ........................... .. ....................... 2
W hy Study the Harpy Eagle? ...................................... 3
Related Research ................................ ............... 6
Fact Versus Folkloric Notions ...................................... 8
Institutional Setting .................................. ............ 9
Organization of the Dissertation ................................... 11
Lim stations of this Study .......................... ............... 13

2 GEOGRAPHIC AND ECOLOGICAL RANGE OF THE HARPY EAGLE ..... 14

Introduction ................................................... 14
The Importance of Historical and Recent Records .................. 16
The Landscape Approach to Habitat Evaluation .................... 17
Methods ..................................................... 21
Project Areas and Site Selection ................................ 21
Sources and Search for Records ............................... 24
Field Surveys in Venezuela and Panama ......................... 27
Habitat Evaluation ........................................... 30
Explorations in Other Countries ................................. 31
Results ...................................................... 31
Historical and Recent Records in Venezuela ...................... 31
Nest Survey in Venezuela ..................................... 36
Historical Records in Panama .................................. 39
Nest Survey in Panam a ....................................... 41
Records from Other Countries .................................. 42
Discussion .................................................... 47








3 NESTING BIOLOGY OF THE HARPY EAGLE ........................ 53

Introduction ................................................... 53
M ethods ............................... ..................... 55
Definition of Term s .......................................... 55
Breeding Area and Nest Site Characteristics ...................... 57
D ispersion ................................................. 58
Monitoring Nests and the Breeding Cycle ......................... 60
Trapping, Banding and Tagging Procedures ....................... 61
Results ...................................................... 64
Nest Site Characteristics ...................................... 64
Dispersion of Breeding Pairs ................................... 64
Density Estim ation ........................................... 68
Breeding Cycle and Nesting Stages ............................. 76
Courtship and Nest Building ................................... 79
Egg Laying, Incubation and Hatching ............................ 82
Nestling Development and Branching Stage ....................... 84
First Flight and Extended Fledgling Dependency ................... 87
D discussion .................................................... 89

4 THE HARPY EAGLE AS A PREDATOR ............................. 95

Introduction ............................................... .... 95
The Evolutionary Context of Large Eagles ........................ 95
M odem Eagles ............................................. 97
Limitations in Diet Studies of the Harpy Eagle ...................... 98
M methods .................................................... 100
R results ................................... ................ 102
Early Exploration of Mammalian Fauna in the Guayana Region ....... 102
Regional Assemblage of Non-volant Terrestrial Mammals ........... 105
Prey Spectrum of the Harpy Eagle ............................. 106
Discussion ................................................... 112
Prey Specialization ......................................... 112
Rainforest Predators ........................................ 119

5 THREATS, CONSERVATION AND MANAGEMENT .................. 121

Introduction .................................................. 121
Regional Setting and Objectives ............................... 122
Threats to the Habitat ....................................... 123
Threats to the Eagles and their Nests ........................... 125
Management and Conservation ............................... 126
M ethod s .................................................... 127
Results and Discussion ........................................ 129
The Role of Roads in Threatening Eagles and Nests ............... 129
Losing the Eagles: Profiles of the Poachers ...................... 140
Development in the Venezuelan Guayana ....................... 144
H habitat Loss ............................................. 149
Approaches to Conservation .................................. 151








Local Cooperation ........................... ............... 153
Protection of Nests .......................... ............... 154
Recommended Baseline Surveys .............................. 155
Monitoring Young Eagles .................................... 156

6 SUMMARY AND GENERAL IMPLICATIONS ........................ 159

Introduction .................................................. 159
Surveys and Mapping .......................................... 159
Breeding Areas and Nest Sites ................................ 160
Regional and Local Mapping .................................. 161
Nesting Biology ...................... .......... .............. 162
Estimating Spacing and Density of Breeding Pairs ................. 162
N est S ites ................................................ 163
Breeding Cycle and Seasonality ............................... 164
Nesting Stages and Port-fledgling Dependency ................... 165
Site Fidelity, Nest Abandonment and Breeding Failure .............. 166
Predators of the Forest Canopy .................................. 166
S elective D iet ............................................. 166
O their Predators ............................................ 167
Conservation ............................................. .... 168
T hreats .................................. .. .............. 168
Surveys, Monitoring and Protection ............................. 170

A PPEN D IC ES ...................................................... 172

A LIST OF CODES FOR MUSEUMS AND COLLECTIONS CITED ......... 172

B REMOVAL OF HARPY EAGLES FROM THE
W ILD IN VENEZUELA ....................................... 173

C HARPY EAGLES CAPTURED IN VENEZUELA ...................... 176

D HARPY EAGLES OBSERVED OR MONITORED IN VENEZUELA ....... 177

E HARPY EAGLES REMOVED FROM THE WILD IN PANAMA ........... 178

F HARPY EAGLES OBSERVED IN PANAMA ......................... 180

G HARPY EAGLES REMOVED FROM THE WILD IN OTHER COUNTRIES.. 182

H HARPY EAGLES OBSERVED IN OTHER COUNTRIES ............... 186

I POTENTIAL MAMMALIAN PREY IN EASTERN BOLIVAR-AMACURO
(>500 g) .................................................. 189

J WORLDWIDE INVENTORY OF CAPTIVE HARPY EAGLES (>100) ...... 191

REFERENCES ..................................................... 193

BIOGRAPHICAL SKETCH ............................................ 212














LIST OF TABLES


Table pape

2-1. Nest Sites of Harpy Eagles in Venezuela .............................. 40

2-2. Nest Sites of Harpy Eagles in Panama ................................ 43

3-1. Characteristics of Harpy Eagle Nests in Venezuela (nests # from Table 2-
1; indicates commercially logged species; .......................... 65

3-2. Inter-nest Distances among neighboring nests in Venezuela and Panama
(nest pairs numbers from Tables 2-1 and 2-2) ........................ 67

3-3. Dispersion of Harpy Eagle Pairs in Venezuela and Panama ............... 72

3-4. Characteristics (nest and locality, cumulative number and code name,
estimated sex based on age and weigh at time of capture) and
Monitoring Status of Harpy Eagles Banded and Telemetered in
Venezuela and Panama ......................................... 78

3-5. Characteristics of Eggs and Hatchlings of the Harpy Eagle ................ 83

3-6. Causes of Breeding Failure in Venezuela (n=15) ........................ 91

4-1. Records of Mammalian Prey Species of the Harpy Eagle ................ 107

4-2. Records of Bird and Reptile Prey of the Harpy Eagle .................... 111

4-3. Comparison of Prey Records for Harpy Eagles in Venezuela, Panama, and
G uyana ..................................................... 113

4-4. Habitat of sloths (Edentata) in Venezuela (Handley 1976) ................ 116

5-1. Impacts of Human Activity on the Habitat and Breeding of Harpy Eagles in
Venezuela ................................................... 131














LIST OF FIGURES


Figure paa e

2-1. Project Areas in Venezuela and Panama .............................. 23

2-2. Main Project Sector in the Darien Region of Panama ..................... 25

2-3. Layout of Harpy Eagle Breeding Areas and Nest Sites (numbered in order
found) in the SE Guayana Region of Venezuela ...................... 37

2-4. Nest Survey in Venezuela and Panama ............................... 38

2-5. Potential habitat of the Harpy Eagle in Central and South America .......... 49

3-1. Layout of Neighboring Breeding Areas of Harpy Eagles Pairs (inside large
rectangle, including 2 overlapped alternate nests), and an isolated Nest
at Rio Grande, Imataca Range, Venezuela (circles around nests= CAR
or Circular Available Range, with .................................. 69

3-2. Layout of Harpy Eagle Nests (X, at center of Circular Available Range) in
the Darien Region of Panama (upper rectangle shows overlapping CARs
of 2 alternate nests) ............................................ 70

3-3. Layout of Harpy Eagle Nests in the NW Imataca Range, Venezuela (inter-
nest distance 7 km; upper nest active in 1995, lower nest abandoned
since 1994; large circle=potential breeding habitat of 154 km2) ........... 73

3-4. Layout of Harpy Eagle Nests in the Supamo River Basin, Venezuela (inter-
nest distance 5.6 km; nest in upper right active in 1995; large circle=
potential breeding habitat of 127 ................................... 74

3-5. Layout of Harpy Eagle Nests (rectangle on the left= pairing at Bochinche 1;
Bochinche 2, on the right) SE Imataca Range, Venezuela .............. 75

3-6. Duration of Breeding Cycle of Harpy Eagles in Venezuela ................. 80

3-7. Seasonality of Breeding (Estimated Hatch Month) of Harpy Eagles in the
Guayana Region of Venezuela .................................... 81

3-8. Growth of a Captive Bred Harpy Eagle (ZIH94) From Hatch to 131 Days of
Age (C. Sandfort pers. comm.) .................................... 85








3-9. Comparison of Growth between 2 Captive Bred (ZIH 94 and SDZ94) and 4
W ild Hatched Harpy Eaglets ..................................... 86

3-10. Status of Harpy Eagle Nestlings (n=23) Produced in Venezuela ........... 93

3-11. Status of Harpy Eagle Nestlings (n=11) Produced in Panama ............. 94

5-1. Human Settlements and Harpy Eagle Habitat in the Rio Grande Sector of
Imataca, Venezuela ........................................... 137

5-2. Overview of Logging Roads and Associated Activities
in the Habitat of Harpy Eagles in Venezuela ......................... 139

5-3. Satellite-tracking the dispersal of a Harpy Eagle fledgling (CHUTI, Nest # 9)
to a sub-adult range (C-2) in Venezuela, 1993-96 .................... 158














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

BIOLOGY AND CONSERVATION OF THE HARPY EAGLE
IN VENEZUELA AND PANAMA

By

Eduardo Alvarez-Cordero

December 1996

Chairperson: Michael W. Collopy
Major Department: Wildlife Ecology and Conservation

From 1989 to 1996, I studied the Harpy Eagle (Harpia harpyvia) in two regions at

similar latitude in Central and South America to learn more about the biology and

habitat of this species and evaluate its conservation needs. I collected approximately

248 records covering the whole geographic range of the Harpy Eagle; 103 entries were

collated for Venezuela, and 52 for Panama. Assisted by local people I located 10

nesting sites of these eagles in the Darien Region of Middle America, all in the

emergent tree species (>40 m to the first branch) called "Cuipo" (Cavanillesia

plantanifolia Bombacaceae); one additional site was found in the Atlantic Region. In the

Venezuelan Guayana I mainly worked in logging concessions, and located 29 nest

sites; the eagles built their nests 25-40 m high in emergent trees, some reaching 50-52

m. They used seven different tree species in four botanical families, mainly in the

Bombacaceae and Lecythidaceae.

In 1995, I used Global Positioning System (GPS) to survey roads and trails

(>2,500 km) over a 300 x 300 km study area in SE Venezuela, and map locations of








nests (n=29) and other features related to forest management. These data were

integrated in a Geographic Information System (GIS) with existing information for

vegetation, drainage network, and centers of human activity. Most nests sites had

some sort of tree felling activity within 1 km, and 7 nest trees have been destroyed.

While most nests in Venezuela were close to human settlements (<20 km, n=18),

measurements of nearest pair spacing (range 3-7 km) for nests in Venezuela, Panama,

and Guyana suggest that density of this raptor greatly surpasses previous estimates.

Pair dispersion ranged from 45 to 79 km2 per pair in Venezuela and was 10-63 km2 per

pair in Panama.

Sloths (Bradypus spp. and Choloepus spp.) predominated in the diet of the

eagles studied in Venezuela, as well as various primates and others arboreal mammals

(such as kinkajou, large and small porcupines, and opossums). The eagles also

captured a few species of birds and reptiles like the Green iguana and the Tegu lizard.

Harpy Eagles persist for years in selectively logged areas near landscapes

dominated by human activity. During this investigation I have equipped 16 eagles (only

2 were adults) with both satellite and radio transmitters. Usually only one egg hatched

after 56-58 days of incubation (average 56 d), nestlings were ready for flight at age

160-180 days; young eagles had an extended period of post-fledgling dependency (>2

years after fledgling, n=4) before dispersal from the nest site. Shooting of eagles

(including 2 of the young birds I telemetered), and live poaching of nestlings were major

threats to the local population of eagles that I studied.














CHAPTER 1
GENERAL INTRODUCTION

Introduction

Large birds of prey are a relatively specialized taxon, typically feeding in the

higher trophic levels. Thus, they appear to meet many biological criteria generally

ascribed to species which vanish when they become isolated on habitat islands (Harris

1984, Thiollay 1989a). It also has been claimed (Thiollay 1984:121) that these birds

were ". . among the first non-game species to disappear in the process of human

population growth and exploitation of the rainforest." Many of the larger neotropical

raptors have been widely proclaimed as sensitive indicators of disturbance, even

though supporting data are limited (Reichholf 1974; Olendorff et al. 1980; King 1981;

Burnham et al. 1988; Thiollay 1989a and b, 1991; Albuquerque 1995).

While reliable, current surveys are lacking, large-sized forest raptors are

generally believed to have small populations, a trait assumed to be typical of extinction-

prone species, and their home-range sizes are presumed to be large, a factor that also

becomes important when studying or managing areas of reduced size (Robinson and

Wilcove 1989, Thiollay 1991, Robinson 1994). In contrast, mid-sized forest raptors may

show attributes of a "syndrome of insularity" (i.e., wide distribution and habitat niche as

well as rather high densities), so that sizeable populations can survive in relatively small

patches (Thiollay and Meyburg 1988, Thiollay 1993).

A study of rain forest mammals in Australia (Laurance 1991:86) concluded that

"large size, low fecundity, high longevity were significant predictors of extinction

proneness, but only for the small subset of species that avoided modified habitat."

1






2

According to Canaday (1991:53) some Amazonian forest birds may not require

conservation measures, and "these generally are large birds that are opportunistic in

their feeding habits and uninhibited from flying, and making use of the novel

anthropogenic habitats separating fragmented tracts of forest." However, it is yet to be

determined if these characteristics also apply to Neotropical raptors (Robinson 1994,

Alvarez et al. 1996).

Background

The National Research Council (NRC 1990:35) issued an urgent mandate for

expanded research on forest landscape phenomena "for developing the theoretical

basis of landscape ecology and for strengthening the applications of its current

concepts." The priorities set for ecological research included determining the natural

scale of various processes by exploring how their operation is coupled from the local to

the landscape level, and how the spatial effects of specific disturbances are integrated

over the landscape (Brown and Roughgarden 1991).

Landscape ecology has advanced the broader perspective needed to evaluate,

manage, and preserve habitat for species with extremely large ranges, such as raptors

and large-sized mammals (Olendorff 1980, Harris 1984, Gosselink and Lee 1989, NRC

1990). But the approach requires the integration of the details of resource use within

vegetation types with evidence of species distribution both in specific districts or

locales, and over large regions (Heltne and Thorington 1976, Thiollay 1991). This

procedure has rarely been attempted with Neotropical raptors, a group of birds that until

recently has received little study (Robinson and Wilcove 1989, Vannini 1989, Eitniear

1990a, Bierregaard 1996).

Over a decade ago, Olendorff et al. (1980:37) concluded that "we have not

quantitatively illustrated through basic research the practical applications (other than

pesticide monitoring) of the concept of using birds of prey as indicators of the long-term








collective impacts of the many other major degrading environmental factors." Calling

for comprehensive, integrated strategies to raptor management, these investigators

stressed the need for multi-faceted, in-depth research to support policies that protect

these birds from extinction. Although experts in many fields are recommending an

ecosystem view for habitat protection and rehabilitation, this approach remains a

largely unmet challenge for most species of Neotropical raptors.

Why Study the Harpy Eagle?

The most powerful bird of prey in the world, the Harpy Eagle (Harpia harpvia

Linnaeus 1878) is restricted to the tropical forests of America. Found from southeastern

Mexico to northern Argentina and southern Brazil (Hilty and Brown 1986), it is generally

considered to be a rarity, seldom if ever seen by ornithologists or birding aficionados

(Robinson and Wilcove 1989). It is suspected to be thinly distributed over extensive

areas, thus it was categorized as "rare and at risk" in the Red List of Threatened

Animals compiled by the International Union for the Conservation of Nature (IUCN

1986).

Oberholser (1919:282) reviewed the historic nomenclature of the Harpy Eagle,

which was originally named Vultur harpyia Linnaeus (Syst. Nat., ed. 10, I, 1878, P. 86;

Mexico), and later changed to the generic name Thrasaetos Gray, which is "merely a

manuscript name of Gray's, originally published by Bonaparte (Thrasaetos Bonaparte,

proc. Zool. Soc. Lond., 1837 (June 14, 1883), p. 108 [ex G. R. Gray MS.], type by

monotypy, Vultur harpyia Linnaeus)." Vieillot's generic name was first spelled Harpia,

thus the scientific name of the Harpy Eagle stands as Harpia harpvia (Linnaeus).

Since the 1980s, conservationists have become increasingly concerned for the

survival of Harpia harpyia in the wild. This species heads a group of scantily studied

large-sized Neotropical raptors with macrosympatric distribution (Klein et al. 1988); this

group includes the Crested Eagle (Morphus guianensis), the Black-and-white Hawk








Eagle (Spizastur melanoleucus), the Ornate Hawk Eagle (Spizaetus ornatuss, and the

Black Hawk Eagle (S. tyrannus). Among these, the Harpy Eagle and Crested Eagle

were listed in the "Threatened Birds of the Americas" (Collar 1989).

I have compiled records of the Harpy Eagle since the 1980s, soon realizing that

these eagles often were encountered by local people living or working in or near

forests; a mounting number of poaching incidents documented in southeastern

Venezuela not only pointed to a species decline, but also suggested that these eagles

were more widespread than previously reported.

Shootings of Harpy Eagles have been widely reported: in Venezuela by Gorzula

and Medina (1986), and summarized by Alvarez and Collopy (1989); in Guyana by

Hanif (1970); in French Guiana by Thiollay (1984 and 1989); in Brazil by Pinto (1966)

and Colinvaux (1988); in Argentina by Foerster (1972); and by various authors in

Mexico and Central America (Inigo et. al. 1987, Eitniear 1987 and 1988). A well known

ecologist (Colinvaux, 1989:108) wrote that in the Amazon Basin "one person with a 16-

gauge shotgun can remove all the Harpy Eagles and less motile primates within 10 km

of a camp in a year", and in a seemingly exaggerated claim insisted that "thousands of

people have done just that." Given the rural setting of the allegedly rampant shooting

and the spotty nature of reports, it seemed that tangible evidence would continue to go

undocumented until the bird became truly rare, and conservation concerns were

heightened. It had not been determined if people regularly eat these eagles, of if there

were other incentives to shoot them.

Young of the Harpy Eagle may have one of the longest post-fledging

dependency periods among raptor species. Offspring remain close to the nest for

months after the first flight, depending on food delivered by parents to the nest (Rettig

1977). Raptor populations levels may be more sensitive to adult and subadult mortality

than they are to reproduction levels (Brown 1976, Steenhof 1987). Knowledge of the






5

biology of the Harpy Eagle, as it relates to this problem and ultimately to the survival of

this species, ranked among the principal objectives of this study from its very

beginnings.

Despite the seemingly high incidence of Harpy Eagle shootings, and an

apparent increase in reported sightings from some areas (Eason 1989, Chebez et al.

1990, Peres 1990, Emanuel 1991), knowledge of this raptor remains far from complete.

Its habitat requirements have not been studied in detail (Collar 1989). The processes

relating change in the habitat of the Harpy Eagle to past and current human activities

are still poorly understood, and conservationists lack sufficient information to define the

needs of this species.

Key questions about the biology and conservation status of this species remain

unanswered. There is a need to describe the regional patterns of distribution; the

density and conservation status of local populations over the entire geographical range;

the size and characteristics of areas used by breeding pairs; and the length of the

breeding cycle (Brown 1980, Robinson and Wilcove 1989, Newton 1979). Clearly,

further study is critical for the protection of these eagles and the management of the

complex ecosystems that they occupy.

Captive breeding efforts with Harpia harpvia had started in the United States

(Cade and Dague 1987, Burnham per. comm.) and Mexico (Eitniear 1989), and were

well underway in the Old World (Minneman et al. 1987). Stated plans for future

reintroduction of the species reinforced the urgent need for additional investigation of

the wild populations of this raptor.

Rather than embarking on a detailed observation of a limited number of nests, I

chose to take a broader ecosystem approach. I started by collecting data on distribution

both over large regions, and in specific districts or locales. These data eventually led to






6

the multi-scale integration of the details of resource use within vegetation types (Heltne

and Thorington 1976, Thiollay 1991).

Newton (1979) recommended the use of maps to examine the dispersion of

raptor nests in relation to habitat and landscape features. From the onset of this study I

proposed the utilization of digital maps, using a Geographic Information System (GIS)

as one of the best available tools to manage the information we hoped to obtain, and to

analyze and display the results. Other researchers have recently applied the GIS

approach to field ornithology (Collopy and Wood 1995; Witham and Kimball 1996).

The combination of GIS, remote sensing techniques, and field assessment

holds immense potential for the investigation of rare but widespread species that

occupy large tracts of land (NRC 1990). Because it is one of the species

"unaccountably rare within their specific widespread forest type" (Thiollay 1989a: 135),

the Harpy Eagle also is an excellent subject to explore the merits of this consolidated

approach, which remains largely untested in the Neotropics.

Related Research

The Harpy Eagle has an almost continent-wide range in South America,

including the countries of Colombia, Venezuela, the Guianas (Cayenne, Guyana and

Suriname), Ecuador, Peru, Bolivia, Brazil, Paraguay, and Argentina (Altman and Swift

1986). The primary work on the nesting ecology of the Harpy Eagle was conducted in

the Rupununi Region of Guyana during the early 1960s (Fowler and Cope 1964).

The information most frequently cited on this species was primarily derived from

a single nesting site, which was intermittently visited by investigators between 1961-92

(Fowler and Cope 1964; Rettig 1977, 1978, 1995; Izor 1985). Previously, Brock (1972)

made observations in the same region and worked with two Harpy Eagles captured

locally. Additional contributions have been made by Rettig, through his hallmark








documentary films and striking photographs (Rettig 1995, Vene 1993, Alvarez 1994)

documenting the natural history of the Harpy Eagle.

The prey species of the Harpy Eagle are mainly mammals (Fowler and Cope

1964; Rettig 1977,1978; Izor 1985), but include some birds (Rettig unpubl.

observations) and reptiles (this study). In general, these fauna have an equally broad

distribution in the Neotropics; interestingly, details on the environmental requirements

of most of these species are also largely unknown. Many of the vertebrates sharing the

same habitats as the Harpy Eagle also face similar conservation problems; all depend

on a ecosystem-level approach to management if they are to remain as viable

communities.

A preliminary status report for the Harpy Eagle already called attention to the

rapid loss of its habitat over most of its known range (Eitniear 1987). In addition, it also

appeared to be exposed to high shooting mortality. These are the same factors that

precipitated the classification of the Philippine eagle (Pithecophaga iefferyi) as an

endangered species (Fitter 1986).

The future of the Harpy Eagle is linked to the fate of lands now widely affected

by forest degradation (Bisbal 1988, Alvarez et al. 1996). The detrimental effects of

humans on its habitat stem from many of the activities associated with the dominant

economic development: these include road construction; urban growth; agricultural

expansion and forest land colonization schemes; logging; the exploration and mining of

oil and other mineral reserves; and the construction of large hydroelectric dams and

associated reservoirs.

Development programs currently underway in many Latin American countries

will continue to impinge heavily on the few regions that retain forests in relatively

undisturbed conditions. Given the advanced technology currently applied in the search

for resources, we can no longer consider any of the remaining habitat as remote,








inaccessible, or secure for the Harpy Eagle. This study documents the seriousness of

this situation for a large region in the northern Neotropics.

Fact Versus Folkloric Notions

Drawing from the original work conducted with the Harpy Eagle in Guyana

(Fowler and Cope 1964; Rettig 1977, 1978, and 1993), many authors have created

what amounts to an ornithological mythology regarding the habits and habitat of this

species. In the literature, the prevailing opinion holds that the Harpy Eagle depends

solely on large tracts of continuous forest (Robinson and Wilcove 1989; Robinson

1994; Thiollay 1989a, 1989b,1991). However, the available records revealed its

presence in a wide variety of landscape-mosaics, many of them containing sizable

proportions of both open-terrain, and closed-canopy vegetation (Alvarez et al. 1996).

The amount of land area used by a breeding pair of Harpy Eagles is unknown.

The best technique available to provide this important information is radio telemetry,

which has not been attempted with Harpy Eagles prior to this study. In French Guiana,

Tostain et. al. (1992) stated that the Harpy Eagle requires habitats of 10,000 ha, and

proposed that the density of this large eagle was diminished from exclusion by the

Guianan Crested Eagle (Morphnus quianensis). It is noteworthy that the distance

between the nests studied by Fowler and Cope (1964) in Guyana was about 10-15 km,

apparently contradicting the perspective provided by Thiollay (1989a). The general

habitat associated of nests in Guyana was described as being near the forest edge,

within 5 km of the Rupununi Savannas, and close to a fairly large river (Fowler and

Cope 1964).

The preferred habitat of the Harpy Eagle has been described as the forest

canopy, the horizontal or undulating foliage-air interface situated well above the ground

over the forest interior habitat (Stiles 1985). Again, existing records suggested that this

raptor used many if not all habitats available in the forest landscape. However, no








eagles of this species have ever been marked or followed prior to this study, and there

were no consistent data available to ascribe the species to (or exclude from) particular

habitat layers.

Reported sightings suggested that Harpy Eagles may prefer to hunt in the

natural openings of the forest (Eason 1989, Tostain et al. 1992). Johns (1986) contends

that the increased visibility of prey in human-made clearings and logged environments

facilitates prey capture by this raptor and the Guianan Crested Eagle. However, these

observations may only reflect a situation where there is greater detectability of eagles,

as also is likely to be the case with sloths (Edentata), one of the eagle's major prey.

Evidently, Harpy Eagles frequented and even nested in habitats that were

inhabited or intensively used by humans (Hanif 1972, Rettig 1978, Johns 1986, Alvarez

et al. 1996). Whether some of these situations will prove suitable to them over the long

term, or are simply relictual habitats remaining after disturbance, remains unknown and

needs investigation.

Although Harpy Eagles may fly and glide just above the forest canopy for

relatively short distances, Rettig (1977 and per. comm.) asserts that they do not soar.

The reports by Beebe (1949:181-83) of a flock of Harpy Eagles soaring high over the

Central Coastal Range of Venezuela, while migrating north and westwards across

Portachuelo Pass (Rancho Grande, Aragua state), have not been confirmed by any

additional observations. However, Albuquerque (1995) recently published a similar

report describing Harpy Eagles soaring and "still hunting" from the air in the coastal

mountains (Atlantic forests) of Brazil.

Institutional Setting

A favorable combination of biological, political, and logistic factors justified the

choice of the lowlands of the Guayana Region of Venezuela as the primary setting for








this study. Initial contacts with key institutions that supported this study were made in

1988-89, during a preliminary survey of primates in the region (Norconk et al. 1988).

In eastern Bolivar state, the 95,000 square kilometer basin of the Caroni River

contains one of the largest national parks of the Neotropics (Canaima National Park)

and a sizeable forest reserve. EDELCA (Electrificacion del Caroni, Compahia Anonima)

is the government company charged with developing hydroelectric resources and

protecting this immense basin. To fulfill its role, this company has produced a wealth of

information by supporting long-term biological, environmental, and watershed

management research, including this study.

The setting of the forestry program in the Guayana Region provided a unique

opportunity for investigations applied to the policy making process. To avoid the

negative perception of forest stewardship that has characterized forestry programs in

the slopes of the Andes, the Venezuelan government has opted for a long-term

management program to extract timber from very large Forest Management Units (FMU)

in the Guayana Region. Government lands were allocated in tracts ranging from

150,000 ha to more than 200,000 ha, and assigned to private operators through a

bidding process. After signing a contract to manage a forest concession or FMU, the

logging companies then become responsible for producing and implementing

management proposals that meet government guidelines. Resource protection, applied

research, silvicultural planning, and controlled extraction of forest products are all part of

the strategy to develop long-term management.

The wildlife management and biodiversity components in this unique forestry

program were very limited in scope, and generally their implementation has been slow

and largely neglected. However, the parties involved have expressed interest in

addressing the lack of integration of wildlife research within the forestry program. This

study of the Harpy Eagle provided an opportunity to contribute to the basic faunal








surveys mandated by the Forestry Program. Thus, it was supported by FMU managers

and government inspectors in Venezuela.

In Panama, the Protected Areas branch of the Natural Resources Institute

(Instituto de de los Recursos Naturales Renovables-INRENARE) sponsored this field

study of the Harpy Eagle in the Darien National Park. The director and staff at the park's

headquarters in El Real were cooperators in the surveys, providing key logistic support

in conducting the field work.

Organization of the Dissertation

This study was designed to address four general hypotheses. It was postulated

that:

1) The Harpy Eagle is not solely a species dependent on forest interior habitats.

Local populations of eagles are adapted to use mosaics of open terrain and closed

canopy communities.

2) Local populations of the Harpy Eagle survive near anthropic environments

(i.e., in drastically modified habitat where induced forest-edge communities

predominate).

3) The distribution and density of Harpy Eagles among different forest types is

dictated by prey availability. Anthropic actions on the local environment, such as wildlife

hunting and the modification of the vegetation, have direct and indirect effects on the

availability of prey animals for the eagles and may determine patterns of habitat

utilization and overall suitability.

4) The major threat to the survival of local populations of the Harpy Eagle in

forests frequented by humans is their direct removal by humans, and can be specifically

linked to shooting mortality.

This study examined local populations of the Harpy Eagle in regional surveys

encompassing two countries. A focus on managed forests in Venezuela provides








access to an experimental scale that is not only of utmost importance, but would

otherwise be too costly and logistically complicated to be independently funded for

research purposes. In Panama, a rural setting contrasts an environment practically

devoid of the influence of roads and the resulting fragmentation.

Chapter 2 reviews the overall geographic range of the Harpy Eagle through a far-

reaching search of historical and recent records, and extensive field surveys. The

historical breeding records are reviewed and complemented with the results of this

survey. The ecological boundaries of the species are explored, describing the major

habitat types and representative locations where the eagles occur. The description of

environmental conditions is complemented by a landscape view of the habitat and the

breeding areas encountered.

Chapter 3 deals with the breeding biology of the Harpy Eagle. A comparison of

the spacing of nesting pairs is presented for Venezuela and Panama along with a

discussion of estimates of breeding pair densities. Characteristics of the nesting sites

also are described and the chronology and duration of the nesting cycle is discussed.

The seasonality of breeding, in terms of the estimated hatch dates of the young eagles

observed during this study is examined in relation to the rainfall regime of the regions.

The role of the Harpy Eagle as a predator is discussed in Chapter 4. The

composition of prey remains collected during this study are compared with other

accounts and discussed in relation to the prey spectrum of other large predators using

the same habitat.

In Chapter 5 the conservation status of the Harpy Eagle is evaluated. I review the

historical setting of the region in South America where the main field investigation was

conducted. The recent history of human presence and resource management among

selected sites is related to effects on the suitability of the habitat for local populations of

Harpy Eagles. The main threats to the species and its habitat also are discussed,






13

relating biological data to management issues that involve local people and national and

international institutions.

Finally, the approach and techniques integrated in this study for surveying and

monitoring Harpy Eagles and their habitat are summarized and discussed in the context

of designing conservation-oriented measures. Several alternatives are proposed for the

management of forested lands as functional ecosystems, emphasizing the conservation

of the Harpy Eagle habitat and the critical features of their habitat.

Limitations of this Study

This investigation faced many challenges, particularly the lack of a single known

nesting site to begin the investigation. There were many reasons to expect that the

Harpy Eagle had a lengthy breeding cycle. As the study advanced, these assumptions

were confirmed: breeding pairs were very difficult to locate, and once found, it took a

long time to study them. Given that these two factors worked against the time frame of

this dissertation project, my strategy was to take advantage of all opportunities to collect

data on Harpy Eagles. In the amount of time allotted for the study, it would had been

impossible to follow a linear approach and attempt a more traditional, sequential study

of nesting biology. As a result of these limitations, work often began with fledged

juveniles; opportunities to document the early stages of nestling development often

came later.

Only after a reasonably large number of nests had been located, and most of

them incorporated into some sort of monitoring scheme, was it possible (given the

available time and funds) to accumulate the knowledge needed to address the

questions posed by this dissertation. It is hoped that the knowledge produced in this

study will provide the basis to take needed action so that this species and the potential

for its further investigation remain viable.














CHAPTER 2
GEOGRAPHIC AND ECOLOGICAL RANGE OF THE HARPY EAGLE


Introduction

An outstanding raptor biologist, Leslie Brown (1979:217), observed that "it is only

man who creates the need to conserve species .... and stressed that we must not

only become aware, but take responsibility for solving these problems. To be successful,

conservation biologists must learn from and deal with the people associated with the

issue in question. In addition to knowledge of the system of concern, the basic

requirements to design conservation management include detailed information of the

biology of the species involved and knowledge of their historical and extant distribution.

The Harpy Eagle is generally considered the most powerful bird of prey in the

world. Attaining adult weights close to 9 kg, females have the largest claws and thickest

legs of any living raptor (Voous 1969). This raptor is commonly believed to be a prime

indicator species of largely undisturbed, or "virgin" tropical lowland forest (Alvarez del

Toro 1964, ICBP Red Data Book 1979, Ruschi 1979, Stotz et. al. 1996). However, after

extended sampling of the hilly lowlands (below 600 m) of French Guiana, Thiollay (1989)

concluded that it was among the unaccountably rare species in the extensive rainforests

of that country. Stotz et. al. (1996:7) questioned the usefulness of the Harpy Eagle as

an ecological indicator because they "are difficult to detect in the dense forest canopy."

According to Parker et al. (1996), the Harpy Eagle is missing from all the (8)

island nations in the Caribbean/Atlantic, and is a resident breeder in 18 of the 21

countries in the mainland Neotropics (i.e., with the exception of El Salvador in Central

America, and Uruguay and Chile in South America). Prior to this study, the historical








records for Harpia harpyia had not been compiled or summarized for most countries

over its vast geographic range (Eitniear 1987). For instance, in spite of far-reaching

ornithological exploration in Venezuela since the 1930s that made its avifauna "one of

the better known in the Neotropics" (Paynter 1982), no records of nests or reports of

breeding areas of the Harpy Eagle had been compiled prior to this investigation. The

geographical range of the Harpy Eagle in this country was documented only by a few

specimens deposited in museums (Phelps and Phelps 1958, Phelps and de

Schauensee 1978) and evidence was lacking for large forested regions.

When this study was expanded to Central America in 1991, all the countries in

that region had a situation similar to Venezuela in terms of lack of information on the

Harpy Eagle. Even though the species is the national bird of Panama, neither local

biologists nor government offices had knowledge of specimens from Panama that were

deposited in museums abroad. Museum specimens and captive eagles in national and

private collections also were scattered and their information largely incomplete. No nests

had been reported anywhere in Central America prior to this study.

The available descriptions for the habitat of Harpia harpyia are in general of little

use for management considerations. Generalized terms predominate in the descriptions,

such as selvas or rainforests (Grossman and Hamlet 1964), tropical forests (Brown,

1976), heavy forest (Blake, 1977), thick forest with tall trees (Dunning, 1982), primary

forest without disturbance (Tostain et. al. 1992), and "large area of pristine forest with

abundant arboreal mammals." (Stotz et. al. 1996:7). Typically, reported habitat was in

lowlands and middle elevations up to 2,500 m (Dunning, 1982), or lowland forests below

900 m elevation (Parker et. al. 1996). Phelps and de Schauensee (1978) in Venezuela

provide somewhat more detail, indicating that this raptor uses forests and occasionally

open terrain near forest and river edges.









The Importance of Historical and Recent Records

Although some zoologists will only use records from museum specimens or from

their own direct sampling, this narrow view may limit the scope and validity of their

results. For instance, a review of the distribution of Venezuelan primates based solely

on museum records (Bodini and Fernandez 1987) was already outdated by the time it

was published, as new information was produced by independent field surveys (Kinzey

et. al. 1988). Current approaches in ornithology strongly encourage consideration of

multiple sources of information (Stotz et. al. 1996), particularly new and unpublished

data from skilled observers (Parker et. al. 1996).

The Harpy Eagle was generally considered a wide ranging but difficult to find

raptor, rarely encountered by observers. These eagles have eluded even the most

competent field observers (Stotz et. al. 1996). The records from specimens in museums

were few for most regions, and alone provided insufficient data to delineate a detailed

distribution (Parker et al. 1996). However, the difficulties involved in finding and studying

this species could be very taxing in terms of time, spatial scale, and field logistics.

Long-term efforts (1987-1996) by accomplished observers studying raptor

communities in Guatemala and Belize failed to locate Harpy Eagles at the fringe of their

northern distribution (Whitacre per. comm.). In adjacent habitat in southern Mexico, a

single Harpy Eagle was sighted in 1989, but only after 9 months of intensive sampling

(Eduardo Inigo per. comm.).

In South America, Robinson (1994) considered the Harpy Eagle very difficult to

detect. Thiollay (1984, 1989a) reported finding Harpy Eagles in 4 of 10 areas sampled

between 1981-1987 in French Guiana, and in 1 area in Surinam. He consolidated 2 of

those sites (Thiollay 1989a) to expand his raptor community study into a 10,000 ha area

with a variety of forest habitat and a high overall number of raptor species; however,

after sampling for 72 days he failed to detect any Harpy Eagles.








In Guyana, Rettig (1993) focused incentives on people who were most likely to

come in contact with these raptors, achieving excellent results. The International Union

for the Conservation of Nature (IUCN) also has developed an innovative approach to

evaluate the conservation needs for a wide array of taxa, through a workshop process

known as the Conservation Assessment and Management Plan or CAMP (Ellis and

Alvarez 1994). This technique also emphasized first hand knowledge from people at all

professional levels who would routinely be in contact with the targeted species, and

would be most familiar with their habitats.

The importance of combining historical records with current local information has

been demonstrated in the past with the California Condor (Wilburg 1978). A recent

example illustrated the usefulness of collating data from historic records, interviews, and

published accounts to evaluate breeding success of Greater (Phoenicopterus ruber) and

Lesser (Phoeniconaias minor) Flamingo populations in southern Africa (Simmons 1996).

This study of the Harpy Eagle applied an approach similar to Parker et. al. (1996) by

emphasizing extensive networking with collections, researchers and skilled observers.

However, I also relied on data from informal field interviews and information provided by

the local residents closest to the potential habitat of the Harpy Eagles.

The Landscape Approach to Habitat Evaluation

Although several of the larger tropical vertebrates (e.g., carnivores,

myrmecophages, and raptors) may serve as indicators of the health and stability of

managed systems, these animals also present challenges to investigators because they

usually are slow-breeding, hard to observe, and numerically rare (Olendorff et al. 1980,

Eisenberg 1983, Klein et al. 1988, Ellis et al. 1990, Thiollay, 1991, Albuquerque 1995).

Most past efforts to describe the habitat of large-sized Neotropical vertebrates have

failed to satisfy the information needs of conservationists, managers, and policy makers

(Strahl 1987, Parker et al. 1996). Many studies have been limited to producing habitat









classifications that only reflect the complexity and interdependency of environmental

factors, rather than providing information on how these factors may impact conservation

management for a particular organism or assemblage of species (Levins and Lewontin

1985).

Management plans for wildlands targeted for development in the Neotropics

should be based on evaluations that reflect the ecological complexity and

interrelatedness of landscapes and the biota they integrate (Stotz et. al. 1996). An

example using this approach involved an assessment of cumulative impacts in

bottomland hardwood forests of the United States that viewed raptors and large

mammals with extremely large ranges as integrators of the impacts over space; their

presence was a sign of large, isolated forest stands with healthy food webs, whereas

the absence of endangered endemic species was considered indicative of degradation

(Gosselink and Lee 1989).

Thus far, practitioners of landscape ecology have not yet taken full advantage of

the major opportunities that exist for the conservation of biodiversity in the Neotropics.

Some critics claim to have been limited by the description of landscape heterogeneity

(Westoby 1987), while others have faulted their focus on surveying for adverse

environmental impacts and the formulation of oversimplified ecological guidelines for

economic development (Eden 1990).

Franklin and Forman (1987) provide the theoretical foundation for the landscape-

level evaluation of the cumulative effects of development on native flora and fauna. This

emphasis permits the simultaneous treatment of a single species and of the whole array

of organisms that accompany it in its environment. This multiscale approach also

involves site-specific research but addresses the landscape scale where management

consideration are perhaps most important and also practically absent in Neotropical

settings (Stotz et al. 1996).








It was not until very recently that the potential benefits of landscape ecology

started to be realized in Latin America. The land system database developed by

Cochrane and coworkers (1985) provided the basis for comparisons of agricultural

potential at the landscape level. The Map of Biodiversity Reservoir Areas for the

Amazon also was a useful preliminary attempt to integrate conservation planning for the

entire basin (Backus 1990).

Investigations in the tropics have begun to identify valuable environmental

indicators of raptor well-being. The primary productivity of a habitat, in terms of the

biomass produced by the vegetation, was a key factor determining utilization dynamics

and breeding success for some tropical raptor populations in Africa (Hustler and

Howells, 1988). Attempts to describe the suspected critical habitat of organisms, in

terms useful not only to ecologists but to managers and policy makers, remains one of

the more demanding challenges in the conservation field.

In terms of activity and behavior, the habitat of raptors is separable into feeding,

breeding, dispersal, and (in some cases) a migration component; all are key elements to

the overall survival of its population. There are at least four important aspects to the

habitat and resulting distribution of any organism (Cabrera and Willink 1973): 1) spatial

or structural characteristics; 2) time considerations as to its duration in geological and

ecological time, and temporal variability such as fluctuations at different time scales

(seasonality); 3) biotic characteristics, determined by the presence (or absence) of other

organisms; and 4) disturbing agents affecting the previous features.

Vegetation is a primary element in habitat descriptions. Its study tends to flow

from small to large scale (i.e., using remote sensing techniques to cover large areas

before any field work begins). However, in Neotropical settings most basic vertebrate

work starts at a one-to-one or local scale (Cerda et al. 1979, Stotz et al. 1996).

Consequently, wildlife biologists encounter difficulties in generalizing their information to






20

broader scales. Mapping wildlife data is perhaps the ultimate challenge in making faunal

research useful; in dealing with large assemblages of vertebrates, the difficulties are

compounded (Cerda et al. 1979, Stotz et al. 1996).

In the last decade, broad scale multidisciplinary approaches have been applied

in the Orinoquian and Amazonian regions to evaluate agricultural potential (Cochrane et

al. 1985) and to inventory natural resources (Zinck 1986) and ecosystems (Gentry 1990,

Berry 1995). Other examples of the kind of tools becoming available to natural resource

managers and conservation biologists are a series of vegetation maps for Venezuela

(Huber and Alarcon 1988; Huber 1995a, 1995b) and databases of avifaunal distribution

(Stotz et al. 1996).

The landscape-level units in these studies were fashioned by integrating the

vegetation with other environmental components. The databases were planned to

provide not only technical support, but appropriate philosophical and political frames of

reference for development. Some of these long-term research programs have the

potential for constant information updating in computerized geographical information

systems (GIS); data access and evaluation was facilitated by condensing results in

cartographic form or by making available in digital media.

Application of spatial information technology is emerging as a useful approach in

the management of natural resources (Fox et al. 1996). However, its contribution to

mainstream landscape ecology and the field of wildlife management has largely gone

unnoticed. The methodologies being developed, and the results already available must

now be adapted and applied to problems of conservation biology (Stutheit 1989, Stotz et

al. 1996, Witham et al. 1996).

Today's GPS/GIS technology provides useful tools for many natural resource

applications, accomplishing in a manner of days, if not hours, complicated mapping

tasks that formerly would have taken months. Much of the emphasis placed on







21

ecosystem modeling, satellite imagery and remote sensing, has overlooked the need to

ground-truth the ensuing data bases. Often, an inordinate amount of time and money

are spent digitizing, and transforming old, and often inaccurate information.

Moreover, a surprising number of newcomers to this system of data collection

erroneously assume that a high resolution (sub-meter) GPS receiver insures perfect

information. A map may look impressive, but if it is not representative of the real world,

its value is questionable. The utility of GPS/GIS techniques lie in the rapid and accurate

collection of current field data which can be transformed into a workable map, in a much

more cost effective and amendable manner (Fox et al. 1996).

The information system developed for the Harpy Eagle project needed to be

dynamic and capable of continuous revision and refinement. Diverse user requirements,

lack of an accurate geographic data base, and an overriding compressed time frame

dictated that a GPS/GIS approach be utilized. Data management and storage in a

readily reproducible format was critical to allow complete integration of continuing

research and conservation efforts, such as producing workable forest management

alternatives in the future that could be supported by accurately geo-referenced

information.

Methods
Project Areas and Site Selection

Logistics and the availability of complementary information often impose

limitations in the selection of study areas and sampling sites for a project. I based the

selection of project areas in Venezuela and Panama on a combination of biological and

socio-political criteria. These included the presence of different types of forested

environments with varying degrees and types of human use; reasonable accessibility;

the availability of infrastructure and on-site logistics for field work; and opportunities to

contribute to management strategies. A basic selectioncriteria was the potential









presence of Harpy Eagles, suggested by reliable accounts or museum records within

the last few decades.

In Venezuela, my research concentrated on the Guayana Region, a geopolitical

entity comprising 468,000 km2, or over half of the national territory. The project was

restricted to the lowlands (from sea level to 600 m in elevation) south of the Orinoco

River in the eastern portion of the Guayana Region (Figure 2-1). The total area covered

of this region is in accordance with the range from 1 to 2 million ha recommended by

Gosselink and Lee (1989) for the cumulative impact assessment scale. I followed their

recommendation to draw boundaries along natural landscape breaks to minimize

external influence and control. Thus, I defined nested operational units from broad to

fine scales.

The project area extended east from the lower Caroni River basin along the right

bank of the Orinoco River Delta to the border with Guyana (Bolivar and Amacuro

states). The southern border was defined by the logging concessions situated between

the junction of the Paragua and Caroni rivers and the southern edge of the Imataca

Forest Reserve. The project encompassed the lands surrounding the Guri Lake in the

west, and the Imataca Forest Reserve in the east (from the edge of the Orinoco,

including the Nuria escarpment and the forests along the road Tumeremo-Bochinche to

the border with Guyana). I also worked in forestry concessions situated to the south of

the road joining El Manteco with El Callao, including portions of the Yuruari, Hacha,

Supamo, and Guariche river basins. In the center of the study area, mainly occupied by

mixed forests and savannas, field work was limited to areas relatively close to the main

roads. The research sites among these geographic sectors were defined during the

course of the investigation as nest sites were located.

In Panama I agreed with INRENARE to concentrate my research within the

buffer area of the Darien National Park (Darien Province), where park staff provided the






23


....... -- *.. ,
C.C














C14
I <-



C













|- ---- -- -
-- -------- -'' I
ji y,. ._ <



*+ S+1+'' .+ '' ;5 '+"'' ; '
* z + i '^ -T ^ *t, '* ++' *.';+.
>2 Y ? ; i o'" **- ... ^ "'-'**
^3~ ~ ^ +. "

i. r-if t + ^ "' I
S* *.o V-^'u --1*
. .. .. -
- "- --...+ F + + >i +- .+*++ j jNM ,a.l'
0 ...







I A -+++ ... + ++








necessary logistic support to conduct the surveys. The project was based at the park's

headquarters in El Real, and the sectors investigated where initially defined by the

location of stations regularly visited by rangers to patrol the park boundary and rotate

personnel.

As in Venezuela, research sites in Panama were defined by the location of nests.

In the Darien Region, I covered a comparatively smaller project area extending from Lepe

(near the Colombian border) to La Palma and the Pacific area of Punta Patiho (Figure 2-

2). In 1994,1 included a site in the Atlantic slope (Figure 2-1), in the northern edge of

Chagres National Park where other collaborators reported a nest (Baitel per. comm.).

Sources and Search for Records

In most cases, the data available for eagles in other parts of the world were

derived from multiple long-term studies. Thus, it seemed reasonable that an extensive

survey conducted over a limited period of time (< 5 yr) would likely yield only a modest

number of records for the Harpy Eagle. For the purposes of this study, I traced and

collated any reasonable evidence of contact with the species, historical or recent

published accounts, field notes from observers, and information obtained during

interviews; I also included accounts of encounters and sightings, specimens shot or

poached alive, and nesting sites discovered.

This survey borrowed from other disciplines the techniques to elicit useful

information from the people involved with the Harpy Eagle and its environment. Every

opportunity for face-to-face contact with local people during the fieldwork was used to

exchange information about experiences and encounters with eagles, creating a network

of informers.

The task of compiling records for the Harpy Eagle began in 1989, and was based

on a preliminary list of specimens in the U.S. and European museums provided by























E
(a.
0
C
.o-
0

c
0





t5
'6
C
0
'0)


C




0



U-
U)
.C'
C



I








E. Iligo (pers. comm.). The records were collated and managed in a database

containing the usual museum label data (i.e., collection number and museum code, date

of collection, locality, collector) and other useful information when available.

In July 1989, I inspected the Harpy Eagle specimens existing in national and

private museums in Venezuela. In September 1991, a similar search was conducted in

the city of Panama, capital of the Republic of Panama, to identify and visit diverse

institutions potentially holding records of Harpy Eagles. Although it was not possible to

personally inspect collections in many other countries, an effort was made to contact

them or to obtain their information through collaborators. Appendix A lists the institutions

holding specimens or records of the Harpy Eagle; it is the key to the references included

in the text and the databases of records resulting from this survey.

In the course of this study, no live Harpy Eagles were removed or collected from

the wild by project personnel. I attempted to collect information and rescue live birds

found captive (and sometimes wounded) by notifying the local authorities and the main

government office of each case detected. In a few cases it was possible to inspect the

remains of specimens shot locally or dead from unknown cause; a series of specimens

were thus salvaged by the author and collaborators (mainly EDELCA personnel in

Venezuela, and INRENARE staff in Panama) for national museums during this study.

Our probe also disclosed cases of nest trees that had been destroyed. In a few

instances a large area of forest had been cleared with heavy mechanized equipment,

and I relied on the accounts of witnesses to determine a location for the breeding area

that had been obliterated.

I also added to the database the records of the live birds handled during the field

work. These included non-releasable eagles that were found caged and had to remain

in captivity, and also those birds captured during the study to be banded (and usually

telemetered) and returned to their native habitat. We also investigated other accounts of






27

young or subadult eagles extracted from the wild, including some accounts supported by

photographs. There was even one case of an adult eagle captured (and released

unmarked) by untrained workers at a logging camp.

As the search continued, the data set grew to encompass information from a

wide variety of sources including field interviews and information supplied by informers

and collaborators, published accounts (including photos appearing in the literature),

museum records (obtained personally, through correspondence, or from digital

databases available online in the Internet), and the results of field surveys conducted

during this study. Using these sources, preliminary reports on the distribution and status

of the Harpy Eagle were assembled from the onset of the study (Alvarez and Collopy

1989, Alvarez and Espinoza 1989) and periodically updated.

Field Surveys in Venezuela and Panama

Attempting to find Harpy Eagles in a South American and a Central American

setting was no small challenge. It required the assistance of local inhabitants and forest

resource professionals to conduct surveys for nests and documented encounters with

the eagles. An earnest search for nest sites to conduct the study began in Venezuela in

March 1988, in conjunction with a field survey for primates (Norconk et al. 1989). In

1991, I resumed work in Venezuela and expanded the study to Panama.

This comprehensive search in Venezuela capitalized on my extensive knowledge

of the Guayana Region where I had been raised and conducted previous field work in

the area during more than a decade (Alvarez et al. 1986, Kinzey et al. 1988, Alvarez et

al. 1996). Widespread development in the lowlands of that region facilitated access and

provided other conditions necessary to maintain a long-term study. In Panama, where I

had no previous experience or contacts, similar search procedures were applied. Field

work in Central America began in 1992, in cooperation with staff of the Darien National






28

Park (DNP-INRENARE), with a trip to inspect a nest reported near El Real (C. Brandaris

per. comm.).

With the assistance of Rafael Alvarez, this project built an extensive network of

informants in southeastern Venezuela based mainly on field personnel of the logging

industry and local farmers. Our survey in that region took advantage of the forestry

inventories (Luy 1992) conducted in conjunction with large-scale logging of Forest

Management Units (FMU) in forest reserves and other government lands. In Panama,

the survey relied primarily on park rangers and other local residents, including colonists

and indigenous people which also were sometimes involved in small scale logging.

The Forest Management Units where most of the field work was conducted in

Venezuela were extensive areas (>150,000 ha) of government-owned lands contracted

to private enterprises to develop and implement long-term forestry plans. The extraction

of logs progressed by logging in tracts of approximately 5,000 ha designated by

evaluations of detailed pre-permitting and post-logging inventories, field inspections, and

reviews of required reporting. Field crews laid out a grid of trails and inventory plots in

blocks of forest proposed for logging, where they had to tally and mark harvestable trees

according to pre-established size and quality limits for the approved commercial species.

In preparation to cutting and extracting logs from a permitted area, a network of road

and log yards (requiring a > 1 ha clearing or "patio") had to be built. Sorting, inspecting,

and stamping of the logs for transporting to sawmills or markets was closely regulated

and had to take place primarily within the approved harvest block. Each log segment

was painted with a code identifying the FMU and a sequential number assigned to each

individual tree of a recognized species. In some cases two or more similar taxonomic

species were lumped under a single commercial tree name and species code. Each log

segment was tabulated as belonging to a particular tree harvested from a specific plot in






29

the inventory grid and had to wear these markings to pass National Guard checkpoints

along the main roads leading to markets.

In Venezuela, the study benefitted from systematic sampling of the habitat

conducted by foresters in the FMU. The forestry crews essentially had to look at, identify

and evaluate the quality of most of the larger trees in the inventoried areas. By gaining

the cooperation of these workers it was possible to learn of any large nests or eagles

that had been encountered. Their reports often provided directions to eagle breeding

sites based on the grid coding system (i.e., addressed by plot number, in a particular

strip of a specific logging compartment). The informants usually knew the type of tree

involved and provided a common-name equivalent from the approved list of species.

Sometimes they also specified the inventory number assigned to the tree where a nest

had been detected.

Our approach to finding nests differed in one main respect from the survey

conducted by Rettig (1993) during the early 1990s in southwestern Guyana. Rettig hired

local assistants and offered substantial cash rewards for each nest report that could be

confirmed. This advertisement created great expectations since the prize (about U.S. $

300) exceeded a typical yearly income in the local economy.

In our case, those involved directly in the search avoided a priori promises of

cash rewards during the surveys. A token gift, usually a pocket knife (cost ranging from

US$ 15-30) was given to the guide leading us to a new nest. The intent was to avoid

encouraging any sense of individual ownership over the sites, and to prevent jealousy

and potential competition among potential collaborators aware of the same site which

could prove detrimental to our work or to the well-being of the eagles.

Reports were screened and the most reasonable ones investigated to establish a

location and to attempt to confirm a nest used by Harpy eagles. A potential site that had

a large nest structure on an emergent tree was indicative of the species. If eagles were








not directly observed at a site, we looked for indicators, i.e., their molted feathers and

the presence of prey remains of fairly large mammals, particularly Bradypus and

Choloepus sloths (chapter 4). I continued to periodically monitor nests that appeared

abandoned to determine if eagles would use them again.

Habitat Evaluation

The sites where I studied the Harpy Eagle in southeastern Venezuela and

western Panama covered a wide geographical area and encompassed a variety of land

uses. These included undeveloped lands, national parks, forest reserves, watershed

protection sectors, indigenous people's lands, and other public and private lands under

agricultural, urban, or industrial development. I limited this preliminary evaluation to the

core area of the distributional range of Harpy Eagles in South America.

I reviewed a series of publications and related their classification units to the

records of this raptor. These sources varied considerably in geographical scope, scale,

and detail of the information. I started from the broadest, continental approaches to the

regional level. Detailed coverage and site-specific studies were limited to the project

areas in southern Venezuela.

In 1995, I conducted four trips in Venezuela using surveying-quality GPS

equipment and software to map study sites, roads, trails, and other salient features of

the project area. In collaboration with P. Kung, I resampled most of the breeding sites

known at that time, to differentially correct their locations to within a few meters. Results

of the survey were incorporated as a digital mapping layer in a Geographic Information

System (PC ARCVIEW GIS, ESRI). Due to logistic limitations it was not possible to

extend the detailed mapping survey to Panama.

I shared survey data generated by our study with TECMIN (Tecnica Minera-

Corporacion Venezolana de Guayana), and in return received GIS information

(coverages) for the vegetation, drainage, main roads, and population centers of our








project area in Venezuela (Zinc 1986, TECMIN 1991). I used the highest quality GPS

data from my surveys to control and check the quality of the layers exchanged from

other sources.

Explorations in Other Countries

I conducted this study in a collaborative fashion, pursuing opportunities for

productive collaboration with other researchers, prospective students, and organizations

interested in expanding this initiative to other countries. This was a necessary and

important component of my broad effort to compile and update the existing records for

the Harpy Eagle over its entire range.

I also cooperated with a breeding consortium for the Harpy Eagle (led by The

Peregrine Fund, Boise, ID), making a concerted effort to locate and inventory captive

eagles and obtain their records maintained in zoos and private collections.

Results

Historical and Recent Records in Venezuela

I compiled over 100 records of Harpy Eagles for Venezuela based on preliminary

work conducted in 1979-1989 (Alvarez and Collopy, 1989), museum inspections, field

interviews, published accounts, and surveys carried out between 1989-1996. These

data were sorted and tabulated into databases for the following categories: 1) Records

of Harpy Eagles removed from the wild in Venezuela, mostly shootings, but also

including museum collecting and other human related causes of mortality (Appendix B);

2) Records of Harpy Eagles taken into captivity (live poaching, Appendix C); and 3)

sightings and observations, where I also included birds tagged and released during this

study (Appendix D).

Overall, the records for Venezuela by geographic regions confirmed an

abundance of Harpy Eagles south of the Orinoco River, where I concentrated my

efforts. The collection at the Universidad de Los Andes had no specimens of the Harpy






32

Eagle (Pefaur per. comm.), and I could not locate any published evidence of the eagles

ranging into western Venezuela. However, a single report from Mondolfi (per. comm.)

was substantiated by a photograph of a fledgling shot in 1959 near Santa Barbara in

the southern portion of Lake Maracaibo (Zulia state). Six of the seven specimens that I

inspected in 1989 at other national museums had been collected in Bolivar state; the

remaining one presumably came from a zoo or a private collector in Caracas, and its

source was unknown (Appendix B). In the northern portion of the country, local

populations of the Harpy Eagle were known along the Coastal Range from sightings in

Aragua (Grajal per. comm., Pantin per. comm.) and Miranda states (Yerena per. comm.).

South of the Orinoco River, the records I compiled (Appendix A-D) covered the

entire Guayana region (Amazonas, Bolivar, and Amacuro states). In Amazonas, a state

which encompasses the largest tract of land under continuous forest cover in all of

Venezuela, Schwartz (Cerda et al, 1979) had suspected, but was unable to confirm the

presence of Harpia. However, my literature search revealed a photographic record of a

juvenile Harpy Eagle shot early in 1969 during an anthropological expedition to the

headwaters of the Orinoco River (Goetz 1969:35). It seems that the oldest existing

Venezuelan specimen of the Harpy Eagle also was collected in this region in 1928, and

deposited at the American Museum of Natural History in New York (AMNH #272336).

Correspondence with the curator led us to correct its source as the Lower Ocamo River

in Amazonas state; the tag had stated "Base Camp", which had erroneously been

assumed to be in the foothills of the Duida Mountain (A. Andors pers. comm). Although

Harpia harpvia was included in the annotated list of birds of the 1951 French-

Venezuelan expedition to the Upper Orinoco (Anduze 1960:331), that account and the

records collated for the Venezuelan Amazonas (Appendices B,C,and D) have eluded






33

reference in the foremost ornithological literature (Phelps and Phelps 1958, Phelps and

De Schauensee 1978, Cerda et al. 1979, De Schauensee and Phelps 1979).

I found evidence indicating that the distribution of the Harpy Eagle exends into

the left bank of the Orinoco. The National Museum of Sciences in Caracas listed an

eagle collected in the Upper Apure River in 1942 (MCN #33, original No. 730 by T.

Avila), but the specimen itself could not be located for examination. At the end of 1991, I

investigated a caged bird held at the port Cabruta on the Orinoco River (across from

Caicara, western Bolivar), and found a Harpy Eagle in adult plumage that had reportedly

been captured in 1987 near the mouth of the Apure River. A photograph proved that this

bird was captured as a white-plumaged fledgling; in 1994, this eagle was transferred to

a private zoo in San Femrnando de Apure (Apure state).

Across the Orinoco River from Apure, I found records from the granite mountains

of Los Pijiguaos, near the mouth of the Suapure River (Bolivar state). In the late 1980s,

the Venezuelan government commissioned environmental impact studies for the post-

development phase of their bauxite mine. Field teams from two different museums

conducted this work, and each group shot an adult Harpy Eagle in 1987 (EBRG #

10134) and 1988 (MCNLS # 10021). My investigation of these incidents, including an

inspection of the area in 1990, led me to believe that these specimens were members of

a breeding pair.

At least four Harpy Eagles have been shot in the Caura River Basin (central

Bolivar). In the past 10 years, nearly every biologist that has worked in the Caura (L.

Balbas, per. com., 1995) or gone into the Nichare River (e. g., Strahl, Castellanos, Silva,

Goldstain, Taphorn, and Desenne in 1988-89; T. Milleron in 1996) had the opportunity to

observe a Harpy Eagle at least once (Appendices B and C).

My first direct contact with the species dates to 1980, when the National Guard

confiscated a young eagle on the main road south of Ciudad Piar and brought it to me in








Guri (Bolivar). Apparently this bird had been held at La Paragaua (just south of the

junction of the Caroni and Paragua Rivers), no one could tell exactly where this bird had

been captured. After about a year of care, I sent this bird to the Loeflin Zoo in Puerto

Ordaz, where it died 2 years later.

In 1982, a hunter told me that he had shot a large eagle on a large ceiba

(silkcotton) tree (Ceiba pentandra) close to the left bank of the Guri lake (Bolivar state).

Field investigation of his report led me to discover the first nest of the Harpy Eagle ever

observed and photographed by biologists in Venezuela (Alvarez and Gorzula,

unpublished observation). A few months later, I saw my first Harpy Eagle in the wild: an

adult female observed perching overnight at a ceiba tree, <5 km away from the inactive

nest mentioned above.

The director of the animal rescue operation during the initial stage of the Guri

dam shot a Harpy Eagle in 1968 (EBRG # 6853). In 1986, a resident of Guri shot a

young fledgling just east of the Guri Dam powerhouse and I salvaged its skin for a

museum (MCNG # 1560). This definitely confirmed the nesting of Harpy Eagles along

the Lower Caroni (Appendix B). Unfortunately, as the Guri dam was again expanded in

1983-1986, much of this habitat flooded; only the elevated portions of the landscape

(>270 m in elevation) persisted as man-made islands in the Lower Caroni River (Alvarez

et al. 1986). S. Gorzula (per. comm.) provided 2 additional records from the vicinity of

the Guri Lake. In 1975, he received a fatally wounded Harpy Eagle shot while it perched

on a mango tree (Mangifera spp.) behind a ranch house at the outskirts of the town of El

Manteco. He also provided photographs of a second bird, shot and wounded in 1977

about 20 km south of El Manteco in the Yuruari River Basin. He treated this bird for a

few days and released it at the site where it had been collected by a local farmer; the

eagle was reportedly seen in the area up to two weeks after its release.








In the Supamo River Basin, southeast of El Manteco, Harpy Eagles were

repeatedly sighted and photographed by loggers (L. Matamoros per. comm.) in the past

15 years; in that period at least five of these birds also were shot and killed, and a

tractor driver reportedly destroyed a large nest accidentally while clearing a logging trail

near the river (no exact location available). While conducting road mapping in 1995, I

found the carcass of an adult male Harpy Eagle almost completely buried in the mud;

according to the loggers I interviewed at a nearby camp, this bird was apparently shot by

one of the many gold prospectors that frequented the area.

East of the Caroni River Basin, I compiled numerous records along the Imataca

Range. A specimen (EBRG # 7252) was collected in 1969 east of El Palmar, and

Gochfeld and Kleibaun (1978) recorded a vocalizing pair. There were many additional

sightings in the forests near Rio Grande and Curaima (Farrera, Grajal and Strahl per.

comm.). On the main road heading south towards the Brazilian border, M. R. Cuesta

(per. comm.) saw an eagle near Betania in 1985.

I obtained new records extending the distribution of the Harpy Eagle into the

lowlands of Amacuro State (Appendices B and C). Lubin (per. comm.) provided

photographs of a nestling captured in 1978 in the lower fringes of the Orinoco River

Delta; this evidence suggests that the species nests near mangrove estuaries and

coastal forests. In 1983, in the same general area visited by Lubin, a very large adult

eagle was shot by a local Warao man in a mangrove forest at the edge of one of the

river channels; the bird's skin was salvaged by S. Fayola for the reference collection I

maintained previously at Guri for EDELCA. In 1995, I deposited this specimen at the

Phelps Ornithological Collection in Caracas (CWHP # 77434), documenting the

easternmost location and the lowest elevation registered for the Harpy Eagle in

Venezuela (Appendix B).








Nest Survey in Venezuela

There are three historical records indicating that the Harpy Eagle formerly bred in

forests across Venezuela. Among the earliest breeding records is the 1959 photograph

of a dead fledgling shot near Santa Barbara del Zulia supplied by E. Mondolfi (per.

comm.). I also identified a Harpy Eagle fledgling (shot dead) in a photograph (Goetz,

1969:24) taken in the mid 1960s during an anthropological expedition to the Upper

Orinoco River (Amazonas). In 1978, a colleague (Lubin pers. comm.) photographed a

downy nestling captured by the Warao people in the Orinoco River Delta near Guayo

(Appendix D).

A site discovered by loggers in Bolivar state in 1989 provided the first opportunity

to study a nesting pair of Harpy Eagles. I learned of this site by interviewing a 50+ yr old

farmer, selected at random while I was driving on a rural road east of El Palmar. This

man had seen Harpy Eagles (local Spanish name: "gavilucho") at his farm in Curaima;

when I asked about nests, he sent me to a logging camp across the Rio Grande bridge

(> 40 km drive,) where his nephew had recently seen a pair of eagles nesting.

From 1991-1996, my search for nests of Harpy Eagles expanded to cover much

of the lowlands in the eastern portion of the Venezuelan Guayana. The Imataca Range

became the core of our study, with almost half of the 29 breeding areas located in

Venezuela. Figure 2-3 shows the geographic location of these sites, numbered in the

order that I found them. Finding these many nests was a cumulative process that took

many years (Figure 2-4). It is stressed here that all the nests sites located in the forestry

areas were first found by locals inside intact forest, prior to or while the habitat was

being logged for the first time. For the most part I learned of these breeding areas later,

after the habitat had already been logged. It often took months (over a year in one case)

to locate a willing guide that could take us to a reported nest site. Nests also were

reported in areas slated for logging, and in some cases months or even years passed

before approval was granted to proceed with the tree harvest.










.. 1: %200 km

S; .. ... .
\V. Ir
. .J ......... .:. I ^ ^ -Y >
"N^ *>> ^t\ Cri ro
T7 \,"' A
t'. '..' 0 |.

..,.,- N ... <"
.A k "A. E. C ^ ~ 1L



(nn






M CU
r-4-
CD
i.,, ^ ) ^ nf i"-
... ' .' ^1, --.? ; -* I '" -: ....' .,, "* ---\ *







. ... . . ..



^... 4 f.[\ ^S- ^ -*^ Itt^'- :' y X -'" S>
"v ~ ~ ~ ~ ~ ~ ~ ~ .,.. ,** ***~ '\'''i^i ' .*' ,.. *.....-,7
"( ".... ...... . .... ... -






INI
M ^ / :.... i,,, :- r^ :"' "" J / I





iSfe OfL\i
, ^ ;"^ ..i.. ,/ ^" ,--. .... ,." '< ,,' ; >' \' ",

.. .... ...- _
"% :* \ ; -i ^ : \ .. '*.< .,. .;**' i- **** / **



C .
.-114! V
] ; ... ...i .
,. ,i -. .::. .i3:


























LU T, (
o F3






J N
-i

i-











Z>
C4 U
00,
0
4m CU


z .2m

CL 0

0


zz








By compiling local accounts and observations by other visitors, which rarely

were published (Homrnbuckle 1993, Wilcove 1995, Manson 1996), this study attempted

to piece together the history of each nest from the time it was first seen by locals

(Appendices B-D, Table 2-1). These records supported the assumption that I was

dealing with a natural distribution of breeding pairs, largely undisturbed until recent

years when the habitat has become extensively affected by human activity that also led

to finding the sites.

This survey of the Venezuelan Guayana confirmed locations for 29 nesting sites

of the Harpy Eagle (Table 2-1), and implemented for the first time long-term monitoring

of nests of this elusive species. Among them, I documented at least five cases in which

the nest trees have been destroyed by logging or flooded.

The Imataca Range in the Venezuelan Guayana, as with the Kunuku Mountains

in Guyana (Rettig 1995) appear to be a stronghold where Harpy Eagles have nested

regularly for decades and perhaps centuries (Table 2-1). I found a large "cluster" of

nests (n= 6) east of El Palmar in the Rio Grande area. At least 16 other breeding sites

were inside or adjacent to the Imataca Forest Reserve, including a site reported on top

of the Nuria escarpment and 4 additional nests to the southeast near the border with

Guyana (Bochinche area). Two other sectors were >100 km to the west of Imataca; I

confirmed 3 nests in the basin of the Guri Lake, and 3 nests in the Supamo River basin.

I also found a nest in the narrow and drier forest galleries along the Guarichapo River,

inside a linear habitat surrounded by a matrix of open and scrubby Curatella tree

savannas.

Historical Records in Panama

The project area in Venezuela is larger than the entire country of Panama.

However, I did not have the resources to conduct a country-wide survey in Central

America. Even so, I compiled over 50 records and accounts of Harpy Eagles for

Panama. These data were sorted and tabulated into the following categories:





U U U U U U U U U U U U 9 U P 9-


FIN
C 4
10"o

*z
%OU-0

W0
U~- <>


0
CjO 0o


S<
.g0
.rA


C.)
0 0

(U
GIn

|ji|


15
c 0
C: c


C(




c %_
0 U)



0n

'o>a


o)) o)
(oN CF) 0) 0 -40) co co
I CD Iml I ) 0) 0)SVa) 1 1)10 01 0)0)
c 4) >~ c 0) C D 0) ac 0)0 4) y)4 1 (y
0 v(I' 0~ 'D ~'o ~ > CC ZZ



0- T-04)
'a ; > 0c !
4 OCOIO 0fl=jflU= (a(7U~ .




c (7 " S n 0 0 > Win I0 CD (0 > (0i0l
o N) '0) 4) 000) oi0 c
g0) Ni? ? m m VS to Lo ?n ? n i n oS ? ? ^ ? !S
'- -- IC) r- OD 0) ) 00







L..L
(Uh





E E
0u_ r




ui I
a. -6: E C(0U00 a

cWEEE E W E OW
0 6U(U' = I



Co UJI E<<(1(Oir 0< .C0 )W v W G 1
'@ -a M we O~C
E k'I CO( CU L z



L) E
0. QQ(U:ol<03O 0- z 0' *S 'W^- *- (


Mu .E E 0 W( '0K wC<(a)







Oo
ch D u) ao 0 0) 0 U



< xi w M0 0 o 0U
<<0 E




Slg n (U(U
.< 00 W "o _


o_ o C 2
L) o ll U_ QL ( U_ mLL 0
w w w > 0,0 2g
LL2 0 w 2 fO m m U
(Uwm DOW EU C WV -
Zr- (U 0 m o







c 0.00 0 0 0 2! cr
(IU0 ca < 'd E < <
> > c C



0 E


- 0
'a
(U it)
'60)
0) 1 r
_j Lm
di t
z1C
cn.a








Appendix E contains records of eagles removed from the wild (mostly killed by

shooting, but also including museum collecting and birds in captivity); and Appendix F

documents sightings and observations (including eagles tagged and released in this

study).

An eagle skin, collected on or near the area of Panama over 100 years ago, is

among the oldest museum specimens known of the Harpy Eagle (BMNH

#1887.5.1.622). Sturgis (1928:137) mentioned a sighting west of the Panama Canal

entrance in the Atlantic, where an eagle was "seen over Monkey-cap palm grove near

Fort Sherman, in May, 1924." The Harpy Eagle is mainly known from the Darien

Province and the western region of La Amistad National Park (encompassing Bocas del

Toro and Chiriqui Provinces), where eagles have been reported over the last 10 years

(Ridgely and Gwynne 1993, Gale per. comm.).

The number of specimens from Bocas del Toro province in western Panama

suggests that local populations of Harpy eagles should still exist in that region

(Appendix E). It appears that after the 1970s, collecting by naturalists was stopped in

Panama; however, local removal of eagles by poachers has continued (Appendices E

and F).

Nest Survey in Panama

In 1992, guided by staff of the Darien National Park (DNP-INRENARE) at El

Real, I visited a nest reported at the park boundary. The nest was located in the

mountains of Cerro Pirre and I found it deteriorated and apparently abandoned. My

efforts during subsequent field visits were mostly devoted to checking reported sites,

and to begin studying any young eagles discovered at nests. In June 1992, the rangers

found a second nest close to El Real, in the lower portion of the Uruseca River. On

their second visit to this site they found the fresh carcass of a Harpy Eagle below the






42

nest, which had apparently been shot. Interestingly, the next day the creek flowing next

to the nest tree flooded the area as a result of heavy rains; had the rangers come a day

later, any evidence of the dead bird would have been washed away.

In January and February 1993, I worked for three weeks in the upper reaches of

the Uruseca River, camping with a filming crew and a group of scientists near the site

were I had seen the Harpy Eagle the previous year. Two expert climbers had been

assigned to built a tree platform and a bridge to film the forest canopy, and my

suggestion to work on the tree were I had seen the eagle perch was accepted. A few

days after work began atop the tree, a Harpy Eagle flew close to the platform carrying

prey (observed by L. Arnal, P. Ruiz and 0. Linares). The following afternoon I searched

the valley below and located the suspected nest within an hour. The new nest was

occupied by a large nestling that appeared ready to fledge. Based on GPS estimation,

this breeding site was <2 km from the old nest up the hill (PND boundary), which I had

rechecked that same week and found still inactive. This close proximity suggested that

this site was an alternate nest.

Over the next two years, the survey in Panama progressed as the park rangers

spread word of our project and investigated reports of nests. By 1996, I had reliable

reports on nine nests for the Darien Province and had began work with an active site

reported in the Atlantic slope north of Panama City (Table 2-2, Figure 2-2). I also was

trying to confirm a report from the San Bias region.

Records from Other Countries

In addition to the detailed studies conducted in Venezuela and Panama, I

compiled numerous records for other countries within the range of the Harpy Eagle.

This information was separated into two categories: Appendix G includes records of

Harpy Eagles removed from the wild population (shootings, museum collection and










W4 a) K 0

00 0 0. V
z


o G) 'o - -
U) ce)'o 0 a
0 0)
W C 0| .CI 0 0) C
D) 01 (1) -- jY) C1)
w co C Zo 0z *c _
o aI )ec Za
w 0 zc Vr) 0)
w- w > < U)






w I. I
M :) mY~ < C, <
m "D <- m U) _

l= IE () i m (|
W D; o | Q 0.2
^ d ^ ic (D cc|5_ K

CU a)C
1 z 1<1 I W Ij <^
Q1 (D '3U



mJ- W4 <
U MM CO ( o'r- M i 0



N 0C o 0) 0 ) 00
W. T-T- 3 T c n- TG T '0 co T- Tl





cN 0 0 V)
z c T c a
ll > c1 4) E II
m~ c







R aV 0 V E f

SC 5f a a) o E c
o Ld w U) WC -j 0







m a) a) 0) 0g 0 a) a 0) 0)
a) 0) 0) a) -o 0( 0) 0







/r *C*CC *C: C*C*C C *C _-
o or r ( CD CO .o Ci r

_ a a

[^'tI '- fM M'-tn t ^ co o
z _____
- 0. .








captive eagles); Appendix H documents sightings and observations. Again, a list of

institutions holding specimens is presented in Appendix A.

In 1993, in collaboration with Eduardo I1igo-Elias and Guido Rahr (unpublished

report), I compiled and revised the records of Harpy Eagles for Mexico. Records existed

for only 10 museum specimens, the oldest skin dating back to 1868 (USNM #4330 at

the Smithsonian, Washington DC). Accounts of sightings were even fewer, mostly from

the Montes Azules Biosphere Reserve in the state of Chiapas (1Iigo et al. 1987, I1igo

per. comm.). According to information from the files of the Instituto de Historia Natural

de Chiapas (B. Alvarez per. comm.), a Harpy Eagle was captured alive at Selva

Lacandona in 1988 and donated to the ZOOMAT (Zool6gico Miguel Alvarez del Toro in

Tuxtla GuLtierrez, Chiapas).

In neighboring Belize (formerly British Honduras), Russell (1964:50) mentioned

a sighting by Lancaster of a Harpy Eagle catching a snake "in tall rainforest at Gallon

Jug on March 11, 1958." In Honduras, Vince Murphy photographed a Harpy Eagle in

1992 in the Rio Patuca (near the Rio Platano Biosphere Reserve), and David Anderson

(unpublished) recorded the species inside the reserve during raptor surveys conducted

January-June 1996.

Few records of Harpy Eagles were available for Guatemala. I visited the Museo

de Historia Natural in October 1992, and inspected two mounted specimens inside a

diorama; unfortunately there was no information about these birds. Jorge Ibarra,

Director of the museum, told me of an eagle shot in 1952 at a ranch in the valley of Rio

Ischkan, between Cerro Cuache and Cerro de San Jorge.

In Costa Rica, a country widely held as an example of progressive conservation

in Latin America, the general feeling is that the Harpy Eagle is doomed if not already

extinct. According to Julio Sanchez, the four specimens at the Museo Nacional in San

Jose apparently were birds which died in captivity and there was no information about

their place of origin (B. Jimenez per. comm.). During an interview conducted by B.






45

Jimenez (per. comm.) in 1995, a man by the name of Morales showed her a photo and

gave her feathers of a Harpy Eagle that he had shot in Karate over 20 years ago.

Slud (1964) reported a formerly broad distribution of the Harpy Eagle in Costa

Rica, with records for the central highlands. Stiles (1985) reported a local population of

eagles in the Osa Peninsula, where there have been additional recent sightings.

According to Stiles and Skutch (1989) the species was previously widely distributed

among humid forests in the lowlands and foothills below 2000 m elevation, in the

Caribbean region and the southern Pacific slope. These authors reported that until

1989 perhaps a few individuals survived in Peninsula Osa, the Talamanca Range, and

in pockets of forest north of the Llanura de San Carlos near the border with Nicaragua.

In 1994, I interviewed Dan Janzen at the Santa Rosa National Park (close to the

Nicaraguan border). In his opinion, perhaps less than 20 Harpy Eagles remained in

Costa Rica and there no longer existed a viable population. According to G. Rahr (per.

comm.), some Costa Rican biologists believe that the last Harpy Eagle in the Osa

peninsula was shot in 1985. However, additional reports were documented during this

study in collaboration with B. Jimenez (per. comm.). C. Foerster (per. comm.) saw a

Harpy Eagle at Cerro Chirmpo in 1990, and knew of other sightings near Corcovado

National Park 1991-92. An eagle was observed in February 1993 at Puesto de

Vigilancia Pittier, a park post in La Amistad (H. Chaves pers. comm.); another one was

reported in 1994 at Rio Lapas, 42 kilometers south of Puerto Jimenez in Corcovado

National Park. Paulino Valverde, a park ranger at Corcovado, reported an eagle near

Sirena in 1995, and also knew of a nest near Rio Oro (Jimenez pers. comm.). These

recent sightings suggests that local populations still exist in at least two parks in Costa

Rica.

Hilty and Brown (1986) report the Harpy Eagle in Colombia as known west of

the Andes from northern Choco (Rio Salaqui) and Baudo Mountains; from Tolima in the

middle Magdalena Valley; and as widespread east of the Andes from northwestern






46

Meta and Vaupes Departments and southwards. During a field visit in 1994, our group

sighted an eagle in the Los Katios National Park, suggesting a continuous distribution

of eagles to the nests sites I located near Lepe in the Darien (in Panama, < 50 km

across the border). In 1996, a rehabilitation center in Bogota (C. Marquez per. comm.)

received a young female fledgling and later an adult male eagle captured in the upper

Sinu River Basin (Department of Cordoba).

In Guyana, Beebe and coworkers (1917) include the Harpy Eagle in their list of

birds for the Tropical Research Station in Bartica, near the junction of the Cuyuni and

Esequibo rivers. Neil Rettig received >12 reports and confirmed 9 nests in 2 areas of

western Guyana (Rettig 1993). In 1992, Rettig invited me on a brief inspection of the

sites where he had filmed and studied Harpy Eagles in the Kanuku Mountains. Using a

portable GPS receiver, we obtained the first map locations for some of the nests near

Napi. There are a number of records for Suriname (Appendix G), and the distribution

for French Guyana is summarized in Tostain et al. (1992).

There are museum specimens of Harpy Eagles from Ecuador dating back to

1880 and the 1920s (Appendix G), and recent records of captive birds from the Pacific

slope (Esmeraldas) and the Amazonian lowlands (P. Jenny per. comm.). In 1991, M.

Guerrero (per. comm.) conducted a preliminary survey in the Yasuni National Park

(Napo Province), locating 2 nests of Harpy Eagles and compiling records of shootings

and live poaching of eagles. E. Asanza (per. comm.) also provided sighting records for

the Laguna Grande of the Cuyabeno reserve.

A striking aspect of the records from Peru is the large number of Hrapy Eagles

collected or salvaged for museums in the U.S. in 1963-1964 in the Rio Curanja

(Loreto). In 1996, a survey conducted by A. Fernandini (per. comm.) in the area of

Tambopata (Madre de Dios) confirmed a similar abundance of nesting sites to that

encountered in Venezuela, Panama and Guyana.






47

I found few records for Bolivia (Appendices G and H). The records for Argentina

are summarized in Chebez (1992) and Chebez et al. (1990). Although Paraguay is

mentioned in the geographic range of the Harpy Eagle( Blake 1977), no recent records

are available from that country.

The record summarized for Brazil (Appendices G and H) should be considered

only as a small sample of what would result from a more complete search. In addition to

>40 live birds in zoos for which I am attempting to establish the source localities

(Appendix J), the information on numerous museum specimens was not readily

available at the time this document was prepared (Albuquerque per. comm.).

Discussion

Our inventories confirmed that Harpy Eagles occur in some of the world's

wettest forests (area of Parque Los Katios, Colombia-Panama border), which receive

up to 10,000 mm of rain annually (Cabrera and Willink 1973). Yet they also are found in

areas of relatively low annual rainfall (>1,000 mm/year), i.e., dry semi-deciduous gallery

forests mixed with savannas in the lowlands of the Venezuelan Guayana (TECMIN

1991).

My records show that Harpy Eagles are generalized in the central lowlands of

Tropical South America throughout the forests of the vast Amazonian Province, which

are included in the biogeographic Amazonian Dominion (Cabrera and Willink 1973).

The habitat of the species also extends into the Pacific Province (which includes much

of their Central American range), and the Venezuelan, and Atlantic Biogeographic

Provinces. The core area of habitat for this raptor in northern South America extends

east from the border between Panama and Colombia, across Venezuela, Guyana,

Suriname, and French Guyana; and south along both sides of the Andes in Ecuador,

Peru, and Bolivia; a large portion of Brazil on both sides of the Amazon river completes

this huge territory, encompassing a broad gradient in forest types, elevation, and

climate.








Parker et al. (1996) placed the Harpy Eagle in the following seven

zoogeographic regions, from north to south: Gulf-Caribbean Slope (Central America),

Choco Lowlands, Northern South America, Northern Amazonia, Southern Amazonia,

Central South America, and Atlantic Forest. The records and observations compiled

during this study provide a basis for expanding the range of this species into additional

regions in the Neotropical Birds Databases, i.e., the Chiriqui-Darien Highlands

(Tacarcuna subregion in Cerro Tacarcuna and Cerro Pirre, Panama), the Northern

Andes (Venezuelan Coastal Mountains), and the Tepuis (Gran Sabana, Venezuela).

The broad geographical (and ecological) range of the Harpy Eagle presents a

challenge to the investigators wishing to compare habitats or attempting to define the

habitat requirements of this raptor (Figure 2-5). Parker et al.(1996) assembled a

database of ecological and distributional information for Neotropical birds, which

considered 15 types of forest habitat (including second-growth forests). They assigned

the Harpy Eagle to the Tropical Lowland Evergreen Forest type, broadly characterized

as <900 m in elevation and >2,000 mm rainfall per year (Stotz et al. 1996). Although

these authors recognized 4 additional kinds of forests associated with lowland forests

(i. e., the flooded forests and river-edge forests, the palm forests, and the rarer white-

sand forests), these forest types were assigned to a particular bird species only if it had

a proven predilection for the specific habitat (Parker et al. 1996). The records for

Venezuela (i.e., Caura River and Orinoco Delta), and other places (i.e., Peru), indicate

that Harpy Eagles routinely utilize flooded and river-edge forests; they also provide

evidence of a much broader range in habitats over their geographic distribution than

reflected in Parker et al. (1996).

A multi-volume database prepared for agricultural planning in the Central

Lowlands of Tropical South America (Cochrane et al. 1985) provides a complementary

source of detailed information on the potential habitat of the Harpy Eagle. This

reference describes a vast macro-region (819 million ha) of diverse landscapes and





49










0
E
4 4-


0
(A5
c



Iuu










Cu
i w- .
r' '+ .. .. ....O

' 'J -o11

-- 3 C





/-.. "+4...0
* ~~ .,. A t
_._^^ ^ Cu" Sl ?"' A3

I. CSr;-1 ^".j^J^ g E^ S '"








4. *. I






50

habitats, summarizing and mapping its climate, vegetation, and soils characteristics; it

also includes detailed database printouts for 855 land-systems, landscapes, and their

land-facet profiles.

Because the established climatic classifications had limited usefulness in

describing vegetational patterns, Cochrane et al. (1985) developed a novel system

based on estimated dependable precipitation and potential evapotranspiration;

assuming no soil-moisture stress, the wet-season potential evapotranspiration (WSPE)

approximated the total annual energy available for plant growth. This new approach to

climatic classification zoned five broad climatic subregions which closely described the

boundaries of recognizable vegetation. It accounted for the Semi-deciduous forests

(<900 mm WSPE, 103 million ha), Semi-evergreen seasonal forests (1,061-1,300 mm

WSPE, 308 million ha), and Tropical rain forests (>1,300 mm WSPE, 217 million ha);

and separated those forests types from the two main types of savanna (Cochrane et al.

1985). These extensive forests comprise a total of 628 million ha which can be

considered as the core of the potential habitat for the Harpy Eagle in South America.

Information on the use of distinct habitats by Harpy Eagles is still preliminary.

However, a landscape level transect in the eastern portion of the Venezuelan Guayana

illustrates the complexity and variation of the habitat where this raptor is known to

occur. Such a profile spans the known altitudinal gradient of the species and its broad

utilization of landscapes, from the Highlands of the Gran Sabana were the eagle

perhaps surpasses 1,000 m elevation, to the mangrove forest in the Lower Orinoco

Delta at sea level. The salient characteristics of the vegetation along this transect were

first described in the broad scale units mapped by Huber and Alarcon (1988), and

discussed in detail by Huber (1986, 1995a, 1995b). According to Huber (1995b), the

largest continues tracts of upland and highland landscapes (500-3,000 m elevation) of

the Venezuelan Guayana are in southcentral Bolivar State and include the montane,

submontane, and upland plains forests of the Gran Sabana and adjacent tepuis. To the






51

north, these forests grade into the submontane and basimontane forests (500-650 m)

of the Sierra de Lema and the Serrania de Imataca; the forests of the Cuyuni-Caroni

lowlands (< 500 m); and into the inland and coastal forests (<150 m elevation) of the

Orinoco Delta and Amacuro plains (Huber 1995b).

It is remarkable that the Harpy Eagle ranges into the unique Guayana Province,

described by Cabrera and Willink (1973) as part of the Guayano Dominion. At the

Phelps Collection in Caracas I inspected the only known museum specimen from this

region, the skin of an adult Harpy Eagle salvaged in 1942 near the settlement of the

Kavanayen Mission (southern Bolivar). In 1995,1 also visited the general location where

this eagle was shot in the Upper Caroni basin and confirmed that the Harpy Eagle is

generally known by the local Pemon people. This submontane band of habitat is

dominated by a complex altitudinal gradient of high to very high forests with emergent

trees up to 45-50 m tall (Huber 1986). Above this elevation is the Pantepuy region of

southern Venezuela (1,000-1,800 m), ranking among the oldest formations on earth

and famous for its unique high-montane ecosystems of very low productivity and highly

endemic biota (Huber 1995a). Many common groups of lowland terrestrial and arboreal

mammals, including sloths and monkeys, may be entirely missing from the flat tops of

these high mountains known as tepuis (Gorzula 1987); and the avifauna, with a range

from 52-82 species, is relatively poor (Haffer 1990).

I compiled evidence (i.e., specimen shot in Aripichi, a museum specimen from

Kavanayen, a live eagle confiscated near Ciudad Piar in 1994) that suggests a

continuous distribution of the Harpy Eagle from the lower slopes of the tepuis and

grading into the lowlands to the north (nests and records from the Imataca Range and

Cuyuni River, Supamo and lower Caroni-Paragua basins, Paragua Forest Reserve),

were the tree cover may be much denser (Huber 1995b). My survey also located nests

in the more open and drier forests mingled with savannas of the Lower Caroni River,








where >48 types of Tropical Dry forest communities were described for the area of

affected by Guri Lake (Alvarez et al. 1986, Pernia 1985).

Although most of the nests of Harpy Eagles that I located in Venezuela were

situated along the hilly continuous forests of the Imataca Range, I also found a nest

between Guri and Imataca in a linear forest habitat, formed by narrow tree galleries

along the Guarichapo River and surrounded by a matrix of open and scrubby Curatella

tree savannas (nest # 18). The records I collated also confirmed that the species

extends into the Orinoco River Delta, where the eagles nest in a swamp forest with

many palms intermixed with mangrove forest (Rhyzophora spp.) and brackish

estuaries, and subjected to strong tides.

I found remarkable similarity in the boundaries of descriptive units drawn for the

Venezuelan Guayana. This uniform perception of the underlying landscape patterns by

different authors holds for a wide variety of subjects, including vegetation (Ewel et al.

1968, Veillon 1977, Huber and Alarcon 1988), land systems (Cochrane et al. 1985),

and diversity of vertebrate fauna (Roze 1966, Cerda et al. 1979, Eisenberg and

Redford 1979, Rudran and Eisenberg 1982, Bodini and Perez 1983 and 1987).

Moreover, contemporary socio-political and management divisions in our study area

also follow similar regional features (Eden 1990). Thus, they can be seen as defining

broad units with land-forms and biota representative of the regional landscapes (Huber

and Alarcon 1988).

The records I compiled demonstrate that the Harpy Eagle is generally

distributed in Central and South America, among the tropical forests of the

biogeographic regions described for the Neotropics by Cabrera and Willink (1973). The

overall distribution of this raptor indicates that it is adapted to a wider array of forest

environments than previously described (i.e., Storz et al. 1996).















CHAPTER 3
NESTING BIOLOGY OF THE HARPY EAGLE


Introduction

The nesting biology of many species of birds of prey inhabiting tropical forests

remains largely unknown (Newton 1979, Brown 1976, Bierregaard 1995). We are told

that raptors, especially secretive species, may be more easily studied at their nests

(Brown 1980; Newton 1979). Yet the nests and details of the reproductive biology of

some locally conspicuous raptors, such as the forest species of Caracara (Daptrius

spp.), remain practically unknown (Bierregaard 1995). Some birds of prey may be

mistakenly considered rare because their limited exposure to open spaces and

secretive behavior make them very difficult to detect. Pairs of some forest falcons

species at Tikal National Park (Guatemala) were found to nest only 300 m apart (R.

Thorstrom per. comm.), other studies also have recently reported high densities of

other little known forest falcons (Robinson 1994, Thiollay 1993, Thiollay and Meyburg

1988).

Setting the question of rarity aside, most biologists agree that it is very difficult

to observe Harpy Eagles, let alone find their nests. Chance encounters are rare and

usually report single birds (Pinto 1964, Goetz 1969, Stanley 1972, Rettig 1977,

MacCreagh 1985, Inigo et. al. 1987, Eason 1989, Peres 1990, Emanuel 1989 and

1991, Sherman 1991, Tostain et. al. 1992, Albuquerque 1995, Alvarez et. al. 1996),

with few exceptions (Gochfeld and Kleibaun 1978). Except for the detailed

observations by Rettig (1977, 1978), little is known about the nesting biology of Harpy

Eagles in the wild.






54

Among the earliest observers of breeding Harpy Eagles, Bond (1928) collected

two eggs from a nest in Brazil. However, the bulk of the published information on this

species comes from intermittent work begun in the Kanuku Mountains of Guyana in the

early 1960s and continued in the 1990s. Fowler and Cope (1964) were the first to study

nesting Harpy Eagles, mainly at a tree nests close to Moco-Moco Creek near Lethem

(Rupununi Region). Planning to film the Harpy Eagle, Rettig and his coworkers (1977)

conducted their first search for nests in Suriname in 1973. During March-April of that

year, they sighted three adult eagles along the Langa Solea stretch of the Coppename

River. In 1974, still lacking a film location, they settled for the known site in Guyana

recommended by Jim Cope (Rettig 1977, 1978). Rettig (1995) would return to film at

the same site in the early 1990s.

Rettig's film footage and photographs of Harpy Eagle, and the publications

derived from his detailed observations of their nesting habits (Izor 1985, Rettig 1977,

1978, 1995; Vene 1993) summarize the knowledge available on the species. In

addition to the field work in Guyana, there were notes of early success with captive

breeding (Hanif 1970, Minneman et. al. 1987); however, with the notable exception of a

nest survey in Argentina (Chebez 1992, Chebez et. al. 1990), original observations of

the breeding cycle of these eagles in the wild were few until this study.

Newton (1979) reviewed the arguments and presented evidence supporting the

view of density dependent (rather than randomly fluctuating) population regulation of

raptors. "In continuously suitable habitat, the nests of pairs of a species are often

separated from one another by roughly equal distances" (Newton 1979:60), and this

spacing is largely the result of territorial behavior (defense) or other limiting factors that

cause the birds to partition the habitat orderly and not at random.








Spacing in raptors reflects how food resources and density are adjusted

(Newton 1979), and the distance between adjacent pairs breeding simultaneously could

be indicative of the "available territory size" or range in a uniform habitat (Garget 1990).

Raptor studies are usually conducted in areas where distribution is already

known in some detail, and the target species commonly found. However, if the spatial

pattern of distribution of a species (i.e., its nests or home ranges) is not homogenous

across the landscape, then the estimates of density and habitat associations should not

be extrapolated over areas larger than the conventional study site (Smallwood 1995).

This limitation appears to become critical when a study area is small in size and

preselected to include the known (or suspected) areas of aggregation of the target

species or its nests (Smallwood 1995).

Ours was the first formal survey for a large Neotropical forest eagle which had

proven very elusive. Consequently, even the basic information required to plan this

evaluation was lacking at the onset of this study (i.e., knowing locations and spacing of

breeding pairs and their nests). Thus, I considered this study largely free from the

potential biases mentioned above. In 1991, I selected Panama for the comparative

project to the Venezuelan study; however, I lacked any prior familiarity with that

country. In 1992, after visiting some of the sites Rettig studied, I declined an offer to

move my study to Guyana due to logistic difficulties. This area of the Kunuku

Mountains was said to harbor one of the greatest known concentrations of Harpy

Eagles (Rettig 1995:42). I also felt that field work on the Harpy Eagle should continue

elsewhere to find out if that particular situation was the norm for the species.

Methods
Definition of Terms

In general I based terminology on the Raptor Management Techniques Manual

(Steenhof 1987) and classic works by Brown (1976) and Newton (1979). Because most






56

raptor investigations have been conducted in temperate areas, the resulting terms and

approaches established by researchers were based on species with well-defined

annual breeding cycles, and did not fit our particular conditions and needs.

I defined "breeding area" as the locality where a pair of Harpy Eagles historically

used one or more nesting sites to build one or more nests (Brown 1976, Steenhof

1987). This term is equivalent to the "nesting station" used by Newton (1979).

Generally, the nest situation or "nest site" is a smaller area that contains the nest

(Newton 1979). Brown (1976:137) applied a broader meaning to "nest site", as when he

stated that "most eagles occupying the same nest site over many years have several

nests in the same small area, group of trees, or even the same tree." Under more

restricted usage, nest site (sensu Cumutt and Robertson 1994:19) becomes limited to a

unique nest structure. In our case, because the trees used by the Harpy Eagle for

eyries have quite extensive canopies (Rettig 1977), I set the nest site area to a 100 m

radius from the main trunk of the nest tree. Modem satellite surveying techniques make

it possible to "fix" a nest location within a range of 100 m.

In this study, a "nest" is the structure of sticks built on a tree by the eagles and

used to "attempt breeding" in a particular nesting site. I followed Newton's (1979)

definition of "home range" as the overall area used during a defined time span. For a

breeding pair it may includes a nesting territory (if an area is found to be defended from

other Harpy Eagles), and all hunting and roosting areas.

Steenhof (1987: 158) recommended that a pair of raptors should be considered

breeding only if eggs were laid, but this could be broadened to also include ". . the

presence of a nest that has been recently built, repaired, or decorated" as evidence

that the site is occupied (or "active") and the pair "breeding." Because Harpy Eagles

were expected to take longer than one season (>1 year) to complete a breeding cycle

(Rettig 1977; Newton 1979), I considered a nest site abandoned only if there were no






57

sightings or other evidence of usage by the eagles (adults or dependent fledglings) for

>18 months.

Although the Harpy Eagle was suspected to have an extended period of post-

fledgling dependency (Rettig 1977 and 1978), the length of its breeding cycle was

unknown prior to this study. However, limited information from two known-age Harpy

Eagles (in captivity at the World Center for Birds of Prey, Boise, ID) indicated that

females of this species could lay their first egg at 4.5 yr of age. Crowned Eagles

(Stephanoaetus coronatus) in Kenya had a fledgling period of 105-125 days and

continued to care for their young for nine to eleven months after they were capable of

flight (Brown 1976). Therefore, I estimated the age of independence at 27 months

(approximately 810 days), or 50% of the 4.5 yr to reach potential breeding age. A

nesting attempt was not considered successful unless it produced a nestling that

survived to independence, estimated to occur after 2 yr of age.

Breeding Area and Nest Site Characteristics

In this study, location was considered one of the most important attributes of a

breeding site. I used a portable receiver (a 5-channel Magellan) for the earth-orbiting

satellites of the Global Positioning System (GPS). As a rule I obtained averaged

locations from 10 fixes (without differential correction), but sometimes this was not

possible. By physically visiting every site possible, even if the nest tree had been

destroyed or the area cleared, it was possible to obtain a precise geo-referenced

location (waypoint), often under heavy canopy.

Using guides to the local flora, species lists with common names, and other

references (Van Roosmalen 1985), I identified the species of trees supporting nests.

For each nest tree I measured the circumference at breast height (CBH) and later

calculated the diameter at breast height (DBH). However, with species that had wide








buttresses a measuring tape was used to visually project the DBH excluding lateral

support structures.

In most cases, the height from the forest floor to the nest was measured directly

by a climber. Crown height (above the nest) was obtained as an ocular estimate by an

observer, either from the ground by extrapolating the distance measured to the

branches supporting the nest, or by a climber working in the canopy. Total height of the

nest tree was calculated by adding the previous two measurements.

The extent of the nest tree canopy was estimated for some trees from an

average of 4-6 radial measurements taken on the ground, from the tree trunk to a point

below the edge of the upper tree canopy, in at least three of the main compass

directions (N, E, S, and W azimuths). The observer also estimated the predominant

height of the surrounding forest and visually appraised the presence and maximum

height of nearby emergent trees.

Dispersion

Dispersion (sensu Newton 1979) or the distribution of birds over a landscape,

deals with the spacing of their nests, their social organization, and their behavior

towards each other. To quantify the home range or total space used by a bird (including

nesting and hunting areas), an individual must be followed and its movements tracked

for an adequate period of time to pinpoint the extent of its movements. Spacing

between nests may serve as indicators of the use of space, at least for the breeding

portion of the range (Newton 1979).

I surveyed widely separated geographic areas and differing forest types to

establish the spacing between breeding pairs of Harpy Eagles monitored in Venezuela

and Panama. I used a GIS to map the locations and sample the distribution of nests

located in Venezuela (Table 2-1) and Panama (Table 2-2), measuring the spacing






59

(linear distances) between neighboring pairs. I also obtained one set of measurements

(using a GPS) between adjacent active nests in Guyana, but these were not mapped.

To estimate the spacing I only considered adjacent pairs <15 km apart; I

excluded any nesting areas where the exact location of a nest had not be determined.

Using the digital map at each of the selected sites, I measured inter-nest distances

using the measurement tool of the GIS. I measured the distances between all possible

nest pairs in a sector, and determined which pairings were nearest neighbors. The

resulting distances, measured in meters, were rounded to the nearest tenth of a km

and tabulated to calculate the average inter-nest distance between neighboring pairs

and the average nearest neighbor distance.

Home ranges and breeding ranges are expected to have somewhat irregular

shapes in some species of raptors (Garget 1990; Grim and Kallemeyn 1995). However,

based on Allen's work with Black Eagles (Aquila verreauxii) (discussed in Gargett

1990), I visualized the breeding areas of Harpy Eagles by defining a minimum "circular

available range" (CAR) centered on each of the nest sites. I set the CAR diameter to

approximate the distance between active pairs in a particular habitat.

To estimate dispersion, I used the GIS to fit simulated habitat plots around the

known groups or pairs of nests in a sector. I fitted the smallest circular plot that would

contain all the previously generated breeding ranges (CARs) surrounding each of the

adjacent nest sites <15 km apart. The area of the circular sampling plot was used to

obtain an estimate of the area of habitat allocated to each breeding pair of eagles.

Because I chose to represent simulated ranges for each pair using a circle

(CAR), habitat plot shapes would contain small "empty areas" of space between

adjacent (touching) CARs. However, in most cases there also were larger empty

sectors, suggesting that additional (not yet located) breeding pairs might be present, or

at least that breeding habitat was potentially available in the sample plot.








I have not surveyed most of these sectors, beyond confirming the nests

reported by locals, to find all the nests that may exist. Thus, I projected additional CARs

to saturate the simulated habitat plots (circular frames) encompassing known nest

groupings. This potential maximum habitat occupancy (simulating a habitat packed with

breeding areas) was used to generate an additional set of values, contrasted with the

area allocated to the "observed" pairs.

Monitoring Nests and the Breeding Cycle

I documented the development of young in Harpy Eagle nests discovered in

Venezuela and Panama and described stages in the breeding cycle following

nomenclature from Brown (1980) and Newton (1979). When eggs were found, extreme

care was taken to avoid undue disturbance or shock to potential embryos (every nest

where eggs were manipulated hatched at least one eaglet). Every egg was marked to

be returned to the same position in the nest, and carefully handled to obtain its mass

and measurements. Wild nestlings were captured by hand in the nest and measured

and weighed periodically approximately once a month until they fledged. I also

measured and weighed any wild fledglings trapped to be marked and released. I

supplemented the observations and measurements with photographs, used later for

comparison between birds.

To describe the nesting phenology of Harpy Eagles, I pooled the records of

breeding activity documented for the project area in Venezuela. I included all the

documented nesting attempts, even if they failed to produce eggs or hatch a nestling;

however, three reports of young birds (discussed elsewhere) were excluded because

they lacked sufficient documentation and the breeding area likely fell outside the

boundaries of our regional study. I developed timelines for these breeding events,

estimating the time of hatch for each young bird based on their observed characteristic,

and the duration of the complete nesting cycles monitored during the study.








For all birds produced (even those that did not survive or had been removed

from the wild), I estimated time of hatch to the nearest year and month. The most

precise estimates of hatch date (day or week) occurred when eggs were found, and a

short time later a small nestling was found on the nest. Older birds were aged by

comparing their characteristics (i.e., body measurements, weight, plumage, and

behavior) with those of nestlings and fledglings of known age that were monitored

during this study; additional descriptions were available for wild birds filmed (Rettig per.

comm.), or described and illustrated elsewhere (Fowler and Cope 1964; Rettig 1977,

1978) and captive-produced eagles (C. Sandfort per. comm.). Body mass and

measurements were routinely obtained from wild-caught eagles found in captivity, and

from specimens that were captured, marked and released during the study.

Trapping, Banding and Tagging Procedures

Because little was known about the biology of the Harpy Eagle, I initiated long-

term monitoring of individual birds. My aim was to monitor the dispersal of fledged

eagles and to describe the movements of adult eagles caring for young. Starting in

Venezuela in 1992, I combined visual observations at the breeding sites with long-term

remote tracking of birds by means of satellite telemetry. In 1993, I expanded the

telemetry study to Panama.

Initially I only used ARGOS satellite transmitters, but later complemented these

with conventional (VHF) radio transmitters, as suggested by other investigations (Rose

et. al. 1993). Basic descriptions of the ARGOS system can be found in a number of

publications (Ellis et. al. 1990, Weimerskirch et. al. 1993, and Fuller et. al. 1995).

I conducted preliminary tests of backpack harness configuration by

instrumenting caged Harpy Eagles. I worked with two adult eagles at the World Center

for Birds of Prey (Boise, ID) and with a juvenile eagle at a flight cage in Guri

(Venezuela). Using a single or continuous loop configuration to attach the transmitters






62

presented a serious risk of entanglement, thus I settled on a permanent harness design

with four strands of strong, flexible tubular Teflon ribbon sewn and tied at the breast of

the bird (see Ellis et. al. 1991, Buehler et. al. 1995).

The harness configuration described as "a breast loop and a body loop (behind

the wings, around the abdomen)" was custom fitted to each bird (Fuller et. al. 1995). I

assembled a dual transmitter backpack which had the radio transmitter (VHF signal)

fastened with epoxy-glue to the side of the 95 g PTT box (platform transmitter terminal).

The VHF antenna pointed forwards, and I fed it at a top corner of the PTT through the

inside of the tubular harness strand so that it passed over the bird's shoulder.

Satellite transmitters were programmed to attempt to regularly sample the

locations of tagged birds. All the tags used on young birds (transmitters with harness

and other hardware) weighed <150 g and were programmed to last 5 years. They

sampled the bird's locations during a 24-hour "transmitting window" on roughly a

monthly basis. Three satellite transmitters were retrieved and used successively on

different birds, after changing the batteries or making any necessary repairs (i.e., in one

instance replacing the antenna that was chopped off by a hunter).

In an attempt to collect more intensive data on adult eagles, in 1996 I used

smaller and short-lived satellite tags (<30 g "Nano PTT" by Microwave Telemetry, Inc.)

with expected 3-4 months battery life (depending on programmed sampling rate).

These were programmed to transmit data for a continuous 3-4 day period every week.

In 1992,1 field tested our original design for a manually-triggered trap using a

single-noose of utility cord to safely and efficiently capture fledged and adult eagles.

This independent design was similar to the power snare developed at about the same

time in Scotland by McGrady and Grant (1996). Our set was built with a base or

platform that could be placed on any perching surface, and I did not attempt to

incorporate any electronics or powering device to achieve tension (Jackman et. al.

1994).






63

Initially I developed a procedure to deploy the trap from the ground, using ropes

to raise and operate the set on a lateral branch near the nest or on a perch preferred

by the targeted bird (usually > 35 m high); sometimes the trap was placed and adjusted

by a climber. The trap was baited with a small piece of meat, held to the center of the

snare with metal wires to prevent a flyby pickup by the eagles.

Most times domestic chickens were purchased locally as bait, but on some

occasions I baited the trap with portions of game mammals donated by local hunters; in

one case a small (<20 cm) dead lizard (Kentropics spp.) was used. This trap was

operated manually. When birds were snared they attempted to fly and usually ended

up hanging by one or both legs to one side of the branch. The trap and bird were then

quickly lowered to the ground by the trapper, from the side of the branch were the bird

had fallen.

To avoid the more difficult and time consuming use of the snare-trapping,

nestlings were tagged prior to their first flight by climbing the nest tree and trapping

them by hand or using a noose on a short pole.

All the birds handled were marked with an aluminum leg band fastened with a

riveter. A scheme combining different band widths, leg placement, band colors, and

locations was developed to provide unique identification of individual birds and facilitate

their individual recognition in the field. The banded birds were weighed with a Pesola

spring scale (for their size range), and a series of body measurements were obtained

(wing cord, extended wings, tarsi diameter, tail length, girth, and total length). In most

cases a bird was processed, tagged, banded, and released within an hour of the time

of capture. In two cases, adult birds were held captive overnight, and then processed

and released.








Results

Nest Site Characteristics

In general, Harpy Eagles living in the lowlands (<500 m elevation) prefer large

emergent trees in the rainforest canopy in which to build their nests (Rettig 1977, 1995;

Table 3-1). These nests are among the largest built by any avian species (usually >1 m

across the largest axis). As a rule the structure is built in the crotch between the main

trunk of the tree and 1-2 horizontally-oriented main branches.

Sometimes the nests were wedged between two vertical branches, where the

canopy of the tree branched out. Even more rarely (2 cases), I have seen nests built

atop secondary branches away from the main trunk. In 1992, I observed one such site

with Rettig in Guyana; I found another nest placed distally (>10 m from the main trunk)

in the Atlantic forests of Panama (reported by A. Baitel per. comm.).

One apparent disadvantage to distally placed nests lies in the weight that such

bulky structures can attain under the frequent rains of a wet tropical climate. In

December of 1994, when I first visited the nest in the Atlantic area of Panama, I found it

collapsed to the ground; its remains, accumulated from many years of usage,

resembled an enormous compost pile of organic matter. Presumably it became

waterlogged and extremely heavy during a rainstorm, which combined with gusty winds

may have caused the branch (>1 m circumference) to break off at the point of

attachment to the tree. A fledgling present at this site escaped this incident unharmed,

and continued to perch on the storm-damaged tree for many more months. It remains

to be seen if the eagles will rebuild their nest on the same tree.

Dispersion of Breeding Pairs

I have rarely seen Harpy Eagles away from nests, and never in pairs. I have not

been able to document any behavior between adjacent pairs; tracking of adult eagles is

still preliminary, and I do not yet know if they forage in exclusive territories. At this stage










4- 0
-- C
mCO

- 4- '00i-
Lo '-

CO~
0 C >%





|E,






.--"
SN 4)



E 0 -C



















-rl-
o (0 ,
0



" ).- '
C'J



0) E
1- E~9












CU)

4- 4)C(
Ua)
E M T


U) C
C 0
'-C




4)-o .. -.
> C U* )
(U^' 4)
M


E U) c Z
4)4)(UC
N




(U<1 t
(u 4- E 4



0 c (U

5*54)


U) CD 0
2-1





CX U)

0 (UO4)
-) -
Z a- a) C

.~E
'402







w ) 0 4
4C CU)





M 0) -U
0 C m
.~a .-


*fIIIIIIII*--- - --I- -
oo o- r- I'
E 5 J 0 (o-U 0) S 8^

I___-___-_____ ------------


E 00 C>
CN C4 Co I,- -) 6) Cf n


ao C> I- C I

cc $ E CO uloc ) l ,= a |




-IP p r i ^
C0




CO C { O M ) vD C o m) 0






a). I- wwEEE ^ ^i f
im - ca (n
CN V C4 CI CN C

U)VI
V 0
wx lE -aU) IT ) Co uT
0D (D. CV) X
x >1 -' (N)





ni > M 41 C o c,4 oD o C14 a >o o
>< -* 0 I -: f-C


a) X) /^ U L






^ io iS EEE =
E X0 0 0U (U (U CC




Cx (U < '-' < U<
5 o uj. o a Ju .u a. uJ >J 0. (U0 < i]

<0< LU LU I-'


~~~~ Ln <





wz (D W W W LU mL LmmmU LUmoLjUi i

C C 0 Z M -<0 0 0 o o 0 0 m C C
oS~co S'0 ^ o 0




x L L L 0 0 0 0 0 C LU LU H LU 0 0 0
=aa) : EE










C 0 ( r v D r0 0 0
0m Cmm (D 0 0 .. C C > 1 M M m r ) - 0 .
0a- W = t l J= C m m C m M 01 w M :M
E0 U U0o m 0
0 0(fl U a)0 0 ~ wU m.0 0 (D 4) a(( (U M M U >14) 1 0
I a.UU_000 L M 1U
0-< M :010~ NO 0.U 0 0 0 0 00 NO < 0

Uj i LUL U UL
ol 00 0.U( U(Ui-.E
< 00000000000<- OOO< C <(U (U (Uw
0 ~ ~ ~ L Uj < < < (U (U (U (U (U (U (U (U 0 w
LU z U z z- < (U () (U (U (U (U (U (U (U (U (U <

E EE1'- <- <- -
In~C U)(o) (N >- >-N)ID








in the ongoing investigation, I have only acquired scant preliminary data on the

movements of one pair of adult Harpy Eagles tagged in February and March of 1996 in

Venezuela. Since a nesting territory could not be distinguished from other areas used

by the members of a pair, I relied on nest spacing to estimate "available territory size"

and dispersion of breeding pairs (Gargett 1990).

The minimum distance measured between nest sites used by breeding pairs of

Harpy Eagles (straight-line distance between known closest neighbors) ranged from <3

km for nest sites in the Darien Province of Panama (n=5) to just under 6 km in the

eastern Guayana Region of Venezuela (Table 3-2). Comparison of GPS locations for

two adjacent nests in the Kunuku Mountains of Guyana, showed that nearest neighbor

breeding pairs were separated there by about 3 km. Furthermore, recent preliminary

observations of five nests located in the Peruvian Amazon also were estimated to fall

within the 6 km range (F. Antonini, unpublished report, Tambopata, Madre de Dios

Province).

As there appears to be no shortage of potential trees for nesting (especially in

the Darien Region), the observed spacing of active breeding pairs of Harpy Eagles in

Venezuela (Figures 2-3 and 3-1) and Panama (Figure 3-2) likely reflects the carrying

capacity of these environments (Newton 1979). Mean inter-nest distance between sites

in Venezuela was approximately 9 km if I considered all possible pairings (n=9) of

nests <15 km apart, and 6.3 km if calculated only for those sites <8 km apart (n=5)

considered to be nearest neighbors (Table 3-2). In Panama, the known nests in Darien

(n=5) were closer (Figure 3-2), averaging 5.4 km for all possible pairings, with the

closest pairs of neighbors as close as 2.3 km. The nearest neighbor inter-nest distance

averaged 3.8 km (Table 3-2).








Table 3-2. Inter-nest Distances among neighboring nests in
(nest pairs numbers from Tables 2-1 and 2-2)


Venezuela and Panama


COUNTRY AND SECTOR Nest Pairs Inter-nest Nearest Alternate
(<15 km) distance neighbors nests
____________________________(<8 km) (<2 km)
VENEZUELA (14 nests, Table 2-1) Nest #s km km km
Imataca Center, Rio Grande 8-10 5.850 5.850______
Imataca Center, Rio Grande 4-8 6.650 6.650______
Imataca Center, Rio Grande 20-4 6.380 6.380______
Imataca Center, Rio Grande 20-8 8.430 __________
Imataca Center, Rio Grande 20-5 13.430I
Imataca West, San Ramon 14-26 7.000 7.000______
Imataca SE, Bochinche 9-13 14.400_______
Imataca SE, Bochinche 28-29 12.400_______
Supamo, Rio Hacha 21-24 5.700 5.700______
Imataca Center, Rio Grande 20-7 ___________ 1.20
AVERAGE DISTANCE _____ 8.916 6.316_____
NUMBER OF MEASUREMENTS______ n=9 n=5 n=1

PANAMA (6 nests, Table 2-2) Nest #s km km km
Pirre, Darien- Uruseca Lower 3-2 2.300 2.300
Pirre, Darien-Qda. Honda 3-7 4.540 4.540
Pirre, Darien-Qda. Cochino 3-6 10.600
Pirre, Uruseca Lower-Qda. Honda 2-7 5.200______
Lepe, Qda. Cochino-Tigre 6-4 4.710 4.710______
Pirre, Darien- PND boundary 3-1 ____1.70
AVERAGE DISTANCE 15.470 3.850
NUMBER OF MEASUREMENTS______ n=5 n=3 n=1








Density Estimation

Spacing of established pairs of raptors in linear habitats is often reported as the

distance between nests; breeding density is expressed as the number of pairs per unit

area or the area allocated per pair (Newton, 1979). To help visualize the spacing of

breeding areas I assigned simulated "circular available territories" or CARs (Gargett

1990) around the known nest sites mapped in the GIS. For nests in Venezuela, I set

this circular breeding range to 6 km diameter or a 3 km radius, approximating the

midpoint distances between nearest neighboring nests (Table 3-2, Figure 3-1).

The resulting size for a CAR in Venezuelan sites was much larger than in

Panama. For example, the CAR for Venezuela was sufficiently large to contain all the

nest sites (i.e., four nests in three breeding areas, one with an alternate nest site) of the

only known group of neighboring pairs in Panama (Figure 3-2). Consequently for the

Central American study area, simulated CAR areas were set to 2 km in diameter

around the mapped nests.

Estimates of breeding density of pairs of Harpy Eagles, in the two countries

where I studied this raptor, were based on the observed spacing between nests. I

calculated separate estimates for the two project areas, because the limited data

collected for Panama suggest a denser packing of breeding pairs in that habitat as

compared with results from the broader region covered in Venezuela.

Survey results in both countries suggested a regular spacing of nests in the

continuous habitats where I found the eagles (Figures 2-3 and 3-2). I fitted a circle to

encompass the largest known groupings of neighboring nests; arriving at a 20 km

diameter plot to represent the area for density calculations in Venezuela (Figure 3-2).














0) 4
4 -)
.v ..... '.. \ .)
*y ..* "- ,

as~ 0).-4
\ -_ E c

....i ........ .. "T '* '! "
\C M C r

/ _1
U) (-0




\ / "6^"**
.' w, :' "o c <
ECD


,. ,, /.\ ^"

. ... < *"*. \ u ;i S % 5
:. L. 4

/~ mi
m


r- 0
. . . ("D 4 )
4) 0 4

Q0) CL)
w4) 0



6 4
^ ~~ ~ ~ ~ ) \ -A*E5i\*- ?
& ... o., / C...- ....... 4) 0 4)
E 4) x 2-
E ,.m..
to r VU
El )U)0

\~~~~r El c-> o o"


\,,' ;" /.* x \ \ aw
V 0 U)
0~~~~~. 0. u\ 2 .\ M^
4 a)

_Cl
CL a)
o co)





C
4)
(U


Q





0

<











.-0
. c

( U0






C
-(U
5U Z












T "'
.CL



-0
~~0
0 0)




z

Sm.




t5
CL.
-0


o E
% m


0) C




.Q



C6 0.
cE


T> 0
; C
*I (0
.0Q

c10w
SIg
I 0)






71

For the largest observed grouping of nests in Venezuela (Rio Grande, Imataca),

I estimated 1 pair per 79 km2. Using this approach, there remained sufficient potential

area to support at least three additional breeding pairs (CARs) inside the sample plot.

Thus, I suggest that the area of breeding habitat allocated to a breeding pair would

range between 1 pair per 45 to 79 km2 (Figure 3-1 and Table 3-3).

I located four additional sets of paired nests spaced <15 km apart in Venezuela.

Only two of these clusters of nests, in widely separated areas (> 140 km), were

considered nearest neighboring pairs (i. e., nests <8 km apart, similar in spacing to

those in the main study area above). The resulting estimates of dispersion using the

same sampling procedure ranged from 1 pair per 51 to 1 pair per 77 km2 (Table 3-3)

according to the size of the circular plot fitted to contain the paired breeding areas

(Figures 3-3 and 3-4).

The other two sectors in Venezuela with paired nests were in the same type of

habitat, and in close proximity to each other (only about 20 km apart). However, their

inter-nest distances were more extreme (>12 km, Figure 3-5) suggesting that these

were not nearest neighbors (i.e., there should be a closer breeding pair to either one of

these nests). Estimates of breeding densities ranged from 157 km2 per eagle pair if

only the known pairs were assumed to be present, and 1 pair per 63 km2 if I allocated

the empty habitat to additional potential pairs. Although the area allocated per pair for

Venezuelan nesting sites exceeded 150 km2 in one of the habitat studied, a more

realistic range based only on the observed and potentially present nearest neighbor

pairings would be 1 pair per 45-79 km2 (Table 3-3).



















= -^


o d) ? :-N to I I 6 6 A 6
< i-


0 1
LO 0O T| 114 (0 10 D r*^







I .-
*j o


0 C
a I- CU)4
4) 4 CN (N N CO




E o










W- u.
z _




ro ) E o'TT LOo o N
.d -


0i- -.









S^n c (oocn^ or1







(D E Z -i v <0
ci o E 5 (dC u iN' ^co ^
0




[0 m 0 u


..j -" LC ',,

W 1
-C
0 D (n
C <


o m -D 6 F- < O(
C LLJ E6 rg Cg a u c -c -c
0 W E E m 0 0 -j > -aw m cD w C 0 10 0 0 l






73








.... ...... .... _.
"..... ............... .


> )
L)
co~
"- .. 4 0
S0)
0)


zcO
c0)
>0)
... . ... ..




0) ~


C6I

wr o
a "CL\
CO' '


I ,4 U) 0)


/ /. \ -


/ :/\) .c: .
-" *! Q. 0 .i' / u





\ CU -
X1' / / -jE0)
^ / \ 0)
's. / '**:0 a)-
l N ^ -" l \I1 I





















...........


4-'

c a




ro .
C





C4)
(D Q
0



w0)-


0 0)
CL
o ID
IX.0





U,3n
00

4-
z(D




)0).0


. Q.
(0:3
I C
<- 4)
0

0 .-



m m
-js-
.ci a)

IT TW
A1.8 C-4
LL*0E
















C '4-




C
O'2
o 0


. $D
mG-
V- M
i2
0
T-'.



Co

c oJ
ki00


o.GO
- C



CU
0U c

Ic o
OCT
'a




) 0
c 0)


o c
. ,





cCU>
0






4- C
?Z o





M_00
-o C)



0CU
m0 M
0^
Cc

0 M
5, E


*I *uj-
Iw .
C0m
-. W
.U)
ir 6 c






76

In Panama, I had only one area with what I considered nearest-neighbor pairs;

they were in two discrete groupings separated by approximately 10 km (Figure 3-2).

Inter-nest distances measured among nearest-neighbor pairs in Panama were below

the overall minimum distance observed in Venezuelan sites (5.7 km). The closest pair

in Panama was only 2.3 km apart, approaching the boundary I had set for alternate

nests (Table 3-2).

As the closest pairing of nests was only 2.3 km, CARs were much smaller in

Panama. Thus, a 20 km in diameter circular plot was more than ample to encompass

all 6 nests known between El Real and Lepe in Panama and resulted in a spacing of 1

pair per 63 km2. However, if I assumed maximum packing of the potential habitat with

the closest possible nearest-neighbor pairs, the allocated area dropped to

approximately 1 breeding pair per 21 km2.

Comparing study areas in Panama and Venezuela, in each country I found an

analogous clustering of breeding sites, with one of the three breeding areas having an

alternate nest (Figures 3-1 and 3-2); however, in the Panamanian habitat, about 2 pairs

were found in the area (CAR) allocated to a single pair in Venezuela. Given the spacing

among the analogous 4-nest clusters in the two study areas, the average inter-nest

distance for the group in Panama (3.4 km) approached 60% of the average inter-nest

spacing in the Venezuelan group (6.3 km). The overall observed and potential breeding

densities considered realistic for Harpy Eagle pairs in Panama was 1 pair per 10-20

km2, compared to 1 pair per 45-79 km2 for Venezuela (Table 3-3).

Breeding Cycle and Nesting Stages

I adapted the terms used by other authors for sequential series of breeding

phases. Brown (1980), in describing the breeding behavior of eagles in Africa, included

a courtship and display stage, nest building and nesting, egg laying and incubation, a






77

fledgling period, and a post-fledgling period. Similar details were derived from Newton

(1979), in a sequence starting with the occupation of territories and pair formation,

followed by a pre-laying period, laying, incubation, nestling, and a post-fledging

dependency period, followed by the attainment of independence.

I documented the nesting cycle of the Harpy Eagle in Venezuela and Panama,

recording evidence of >27 nesting attempts among the 40 nest sites located in the

course of this study; I found neighboring nests with young of widely differing ages (i.e.,

downy nestlings and fledged birds). These records included some failed attempts, and

also 14 young birds banded and monitored during the study (Table 3-4). It would be

difficult to determine if the reproductive cycle of Harpy Eagles has a seasonal character

based on the few published accounts of nesting available for the species (Bond 1927;

Ruschi 1979; Chebez et al. 1990; Fowler and Cope 1964; Rettig 1978, 1995); however,

these records indicate that Harpy Eagle pairs can be found raising young in just about

any month of the year, and this is likely due to the extended dependency of fledglings.

In Venezuela, I recorded 20 eggs hatched at 17 of the known sites between

1989-1996; except for the most recent record of 2 chicks hatched at a nest, the rest

were single birds (n=18). For Central America I documented 6 single young hatched at

6 of the 11 sites reported for Panama 1992-96; one additional nestling reportedly

produced in 1995 has not yet been confirmed (Table 2-2). These figures are only

indicative of potential production because some of the nests have only been monitored

once, or for a short period during the course of the study. Among areas sampled only

once, the exact nest site location could not be determined for some (i.e., location of a

single nestling was the record), others had the nest site destroyed, and in a few cases

regular monitoring was not possible for logistic reasons.









'5-
o0

to
E



4-
0)

*0
m
(0
0)
(U
o


mOAE
M M
c



.E M
0
(D .
(U(
n cU


Q-
()C


-4- (U)


a)



E'
*E (D

cU








"m
(D g





E|
C .
a )




> (D
Vo a?





m c
0"'


0)O



>;
CE



,.. "0





.0I
U) r








- 0
IO
3 "
CUCj


00

0 Ox

C-
CU

C -'
c CU

0 (0
CU5




C
cv,0
i L
COm0


cl- "E- e f -
we gl ^ l '' 06- "|
co -.2 )


Lj O cm 0 f 0 U
EL co) E! VS .2cE ~


VVVE- CfU-f o ) ^
Co
W







o o >>I Z, R Z>,Z Z > o . Z ,
"'lo o c'n o I Eo Z
0













111 "o > ? "o '6 o o* o >o o> o b "o '6 "o
v R Nr V) 04 'a L A a a a o a
0 M 0 0
F- Z" Co CO Z__ Z( Z zz z0
0~< 0 CE 0 0 E o 00 09E E 0





























o o o .o.o oo
C cro















o 0) I
W CO C
<0 0 o o 0) 00 0
OQ 0) o ) E 0) E > E' E > E ) E ))0 0 0 0 4) O) 0)

.ST _n 0 a) .E c.o Lo ia
$l a > C -4 ;wA co co .=a
T. V

U-. D )( 0 4)










ax a a a a a U


=^i LL 2L 0 0 8 ^ 2 "22 g
-000 00 oo 0) 0^00 00 0
O ~ O) )0 0)0Z


















W t ooo. o- o'bo o o
(/3)













0) c o co a CM CM CD O N a


0 0__ _ _ _ 4) (DC' ) t4)
aw ~ a a)G a) =0a 7

m~ ~ ~ 0 v ~mm0 0~ N ~ M

<~~~0 OO909)a0 (9) 09 0




w~J 0 o L 0




z <-e z ei I L






79

I used a graphical approach to facilitate comparisons between nesting attempts

documented over the entire study period in Venezuela. By developing a timeline for

each of the breeding events documented in Venezuela (Figure 3-6), I obtaining an

approximation of the chronology of the "typical" breeding cycle for this species and its

seasonal variation. In pairs that nested at the same site, after producing a fledgling that

survived >2yr (n=4), nesting was spaced >2.5 yr from their previous successful attempt

(i.e., renesting was delayed until the fledgling left the nesting area to attain

independence). These cumulative records indicate that the timing of egg hatch in

eastern Venezuela is seasonal, showing an apparent relationship with the prevalent

rainfall regime in the eastern Guayana Region (Figure 3-7).

In my limited sample, no birds hatched in the driest (and hottest) portion of the

year, between February and April. The data show two peaks in hatching during the

locally variable transition periods between the wet and dry seasons (known as the

entrance and departure of the rains). This trend suggested that the wettest portion of

the rainy season also may be unfavorable for the most crucial portion of the

reproductive cycle.

Courtship and Nest Building

The early phase of the nesting cycle of Harpy Eagles remains practically

unobserved. Detailed behavioral observations, such as dawn to dark watches (Brown

1980), were not a priority of this study when an insufficient number of nests were

known. Initially, most of the active nesting sites that I located contained fledged birds.

Few opportunities were available to observe the earlier phases of the breeding cycle

until the latter part of the study.

In 1993, a pair of eagles were discovered as they rebuilt a nest, but soon they

disappeared from the site without laying; the site (nest #12) remained inactive when

last checked in August of 1996. Presumably this failure was caused by disturbance
























#17 (2 chicks) nest# EAGLE NAMEj y LE
94 9_5 96 yearnestling-fledgling stages
#10 __ SARA ___ PAUL^
92 93 94 95 96
#8 amo" TAYA"
92 93 94 95 96
#7 PLANO NEIL #20
91 92 93 94 95 alternate nest
#4 ALBERT __ Inactive
89 90 91 92 93 94
J AM AS dJ AM AS dJ AM AS d.TAMAS dJ AMAS dJ AMAS




Note: Nest # from Table 2-1, Eagle names from Table 3-4, >2.5 yr cycle
in nests #4, 7,8,10, and 17)


Figure 3-6. Duration of Breeding Cycle of Harpy Eagles in Venezuela





























Hatching events n=


4 3 4
(.) 1 2 2 NO 2
m m -


month Feb. Mar. April May June July Aug. Sep. Oct. Nov. Dec. Jan. Feb.


Rainfall s z .:..
dry


transition wet


transition dry


Figure 3-7. Seasonality of Breeding (Estimated Hatch Month) of Harpy Eagles
in the Guayana Region of Venezuela.


I = a -








from heavy mechanized equipment clearing a log yard under the nest tree. This

observation suggests that the initial period of the nesting cycle appears to be most

critical as far as risk of abandonment in response to disturbance. It was not until 1995

that I was able to observe nests with eggs and young chicks in Venezuela.

Existing nests deteriorated soon after the nestling fledged; some nests

eventually disintegrated and their materials fell completely to the ground (Table 3-1). I

have not witnessed the nest rebuilding process, but periodic observations indicate that

the nest structure can be completed or refurbished over a period of a few weeks. It is

only during this early stage that nests have greenery (i.e., fresh vegetation) and a fairly

distinctive cup. Although the adult eagles continue to add branches to the structure

throughout the nesting cycle, it soon becomes a fairly flat platform, usually hosting

termites and ant nests underneath, which rapidly consume the remains of prey

(including bones) interred under the added nest material.

Egg Laying. Incubation and Hatching

Harpy Eagles usually lay 1-2 eggs; Bond (1928) collected a 2-egg clutch in

Brazil over a 12 day period. Eggs range in weight from 125-150 g (Table 3-5).

Published accounts (Rettig 1978, Ruschi 1979), captive breeding records (C. Sanfort

pers. comm.), and our own observations indicate that incubation ranges from 53-58

days. In 2-egg clutches, incubation apparently stops when the first egg hatches; the

remaining egg decomposes and presumably breaks or becomes buried under the new

nest materials added by the nesting pair.

Ruschi (1979:42) found a nest with two nestling in 1944 in Brazil; I also received

some unconfirmed reports of nests with two grown nestlings during this study.

However, I only confirmed a single case where a second egg also hatched (Rafael

Alvarez per. comm.). The first egg to hatch out of this 2-egg clutch was slightly bigger


























01
0 0 0 0 to 0 ) 0
T ( D r N 0) 5 )' C0
SC0 004








mo Lo Q

< -s
w(O M ~ U' O c-(
0ii





0) r-0 O 0) U') 0) )


.H r'- 0 -zz O( N00CN N


0)
0)0) 0

o >0 0rc 0 0 0
m2 (A CN CN)Q OD)c0)



0~~~~~~~ N .- -- -- *
V) N -N .- N co (N N N -- -
6- 1 V i CD IC1 -








U) m 0 ))
ZZ 0 00 0cl (V
0>- -0 C0 N0 CM 0N








C tm CMN ( N -M -0 i n




0)0
w0
c 04 0000000
) 0) 0)) )0 )a 0 )
> I









0 0 0 00000
E z' z 0 ( 0 0-* CT0 C M m' r M- mq 4) w^* m^*
m s n o ) m- C of) ofi m <^ f LL mD z-< D O
Cf)-s CN CN 0)1-0 o 0












*N NNNNN













1 .2 0) 0 L M
E-. < < < < < < C/) 53: 1- 1 '- C -jC IC -j CM -j^ i j -
0) i M