1. INTRODUCTION

Cockpit Country of west-central Jamaica is recognized nationally and globally for its unique biological diversity and for its ecological importance as being one of the largest contiguous tracts of wet limestone forest remaining in Jamaica and in the West Indies. With over 100 bird species recorded for the area, including 27 of Jamaica’s 28 endemics and 36 Neotropical migrant species (the latter of which includes the globally-threatened Swainson’s Warbler (Limnothlypis swainsonii) and Bicknell’s Thrush (Catharus bicknelli)), Cockpit Country is recognized by BirdLife Jamaica (BLJ) and BirdLife International (BLI) as a major “Important Bird Area” (IBA) for the insular Caribbean. The diverse avian species which occur in the core 25,000 hectare forest block and peripheral habitats are not only maintained as large, viable populations, but their numbers, in turn, ensure that bird-dependent plants continue to thrive with the presence of their pollinators and seed dispersers. Maintaining (or, in certain instances, restoring) these functional relationships, with all of the unique components -- species, communities, and ecosystems -- is critical for ensuring the long-term persistence of a healthy Cockpit Country ecosystem.

Recognizing the importance of Cockpit Country and the need for comprehensive management, partnerships between Government and local and international NGOs have been developing over the past several years. One program, The Nature Conservancy's (TNC) USAID-funded "Parks-in-Peril" (PiP) program, assessed the current status of biodiversity and the human activities that threaten the Cockpit Country ecosystem using the adaptive management tool of Conservation Area Planning (CAP). This assessment identified important conservation targets, including the Jamaican Blackbird (Nesopsar nigerrimus) and its forest habitat, and is developing strategies to abate threats and improve the health of conservation targets. By maintaining extensive, closed-canopy forest, it is hoped that parasitism of Jamaican Blackbird nests by the Shiny Cowbird (Molothrus bonariensis) will be minimized. All birds nesting in Cockpit Country will benefit from this conservation umbrella species and the protection of its habitat.

Effective management of Cockpit Country requires the ability to monitor and evaluate conservation activities for their outcomes and to adapt management as necessary. This ability requires not only reliable and accurate information on focal conservation targets (or indicators of these targets) but also incorporating this information into management plans to achieve maximum benefit. To meet these needs, Windsor Research Centre, through the support of the National Fish and Wildlife Foundation (NFWF), the Environmental Foundation of Jamaica (EFJ), and TNC established an avian monitoring project, which:

• is collecting systematic data on resident and migratory bird distributions and their seasonal habitat usage across the gradient of land use types present in Cockpit Country;
• has identified areas that support Jamaican Blackbirds so that additional directed research may be undertaken for this PiP conservation target;
• is operating a permanent bird banding station for long-term monitoring of avian demography and physical condition of birds during critical stages of their life cycles (e.g., breeding vs. non-breeding seasons, arrival and departure of migrants);
• is training a core group of research assistants and professional natural resource managers to implement their own avian monitoring projects and to have skills to assist other researchers
• is integrating the results of this avian monitoring project with Forestry Department's national biodiversity database to ensure all data are accessible to resource managers and policy makers

The bird banding occurs in an area where WRC will be encouraging scientific research on forest restoration. Changes in avian health parameters and demography will serve as one type of indicator variable to assess long-term efforts of rehabilitating degraded wildlife habitat in Cockpit Country.

This report summarizes the results of this project during the NFWF funding period from May 1, 2002 through April 30, 2004.

2. CONSERVATION OBJECTIVES

The over-all goals of this project were to develop local capacity to conduct reliable research, to establish baseline data on habitat use and the effects of habitat quality on resident and migratory bird populations in Cockpit Country and to develop the capacity to use this information for effective adaptive management of this important ecosystem. To achieve these goals, the project implemented activities for long-term monitoring of Cockpit Country avifauna by:

• · Establishing permanent geo-referenced bird monitoring points, with standard, replicable methodology to ensure valid future comparisons
• · Documenting bird use in different habitats (including human-modified) and integrating this with other activities that assess the Cockpit Country forest ecosystem
• · Establishing a permanent bird banding station
• Training local Jamaicans in avian monitoring techniques
• Making available results of research activities to decision-makers and deriving further research/monitoring activities from dialogue with resource managers.

3. PROJECT FUNDING

WRC received a NFWF Challenge Grant of US$45,000 for an original project period of May 1, 2002 - April 30, 2003. WRC was granted an extension through October 31, 2003 after Challenge Funds were provided by TNC for the period of October 31, 2002 through October 31, 2003. On September 17, 2003, WRC signed a contract with the Environmental Foundation of Jamaica (EFJ) for a grant of US$100,00 to expand the bird banding component of the project. NFWF approved a project extension through April 30, 2004 to enable WRC to access the grant balance with these Challenge Funds and to complete activities of the NFWF proposal. Funding also was provided by Forestry Department's Trees-for-Tomorrow Project, which is funded by the Canadian International Development Agency (CIDA).

4. PROJECT ACTIVITIES & THEIR EVALUATION

As delineated in the proposal approved by NFWF (Table 1), WRC's Cockpit Country Bird Monitoring Project encompassed the following activities for training professional resource managers, monitoring the composition and health of bird populations, and incorporating these research findings into management plans:

Table 1. Project Objectives & Evaluation

Objectives	Indicators	Verifying Media
1. Establish permanent bird banding station	1a. Acquire equipment 1b. Initiate banding program	1a. Banding records submitted to BLJ and NEPA
2. Trained field research assistants	2a. Identify bird species 2b. Handle/band bird 2c. Examination on research ethics	2a. > 90% accuracy 2b. < 1% mortality 2c. > 90% test score 2d. WRC/BLJ certificate
3. Establish permanent bird monitoring points	3a. Acquire topographic maps 3b. Liaise with FD-TFT 3c. Replicate points identified	3a. WRC GIS established 3b. Georeferenced map
4. Generate baseline map on bird-habitat associations	4a. Data analysis and computer-generated mapping	4a. Report submitted and data available through national databases
5. Effective dissemination of results to resource managers and decision-makers	5a. Participation in Adaptive Management Seminar 5b. Research activities affirmed or modified	5a. Attendance record of seminars 5b. Report submitted to CC-PiP and IBA programmes

4.1. WRC Bird Banding Demonstration Laboratory

4.1.1. Infrastructure and Protocols

A permanent bird banding demonstration laboratory was established on the five-acre organic farm of Windsor Great House (T1). The farm supports a mix of coffee, citrus and fruit trees adjacent to sugar cane and closed-canopy, regenerating wet limestone forest. Two additional satellite stations were established, also in Windsor. One station (T2) is in a five-acre area of regenerating pasture adjacent to wet limestone forest; this area is available for research on forest regeneration processes and is owned by Michael Schwartz, owner of Windsor Great House and a director of WRC. The second station (T3) is also in an area of regenerating pasture surrounded by wet limestone forest; this area is leased by a local resident and we pay a small monthly rent (US$25 for two days per month), which encourages the owner not to return the site to cattle pasture.

We are currently operating 12 nets (2.5 x 9 m; 32 mm; 110-d; 5 shelves) at each of the three stations, for two days per month per station. Twelve nets are operated safely by three assistants, who have been trained in extracting birds safely from nets, banding, and conducting physical assessments, and one data transcriber, who is not required to have skills in handling birds. At a minimum, the two resident directors of WRC, S. Koenig and M. Schwartz, would be able to operate six nets to maintain a monthly constant effort. Because of this permanent station and the establishment of the baseline data sets, WRC is now attracting other researchers interested in avian studies, including one doctoral student from Berkeley (CA) who will be studying hummingbirds in July 2004, one student from the University of Maryland, who will be studying the effects of deforestation and patch size on bird communities in 2005-2006 and one faculty member from Humboldt State University (CA) who will be following-up on his doctoral research in Jamaica on the effects of shade and chemical coffee on bird communities and expanding his research to new topics of avian conservation and habitat protection in Jamaica. Unfortunately, our colleagues in the Life Sciences Department of the University of the West Indies have yet to identify students with a committed interest in avian research associated with the types of data generated from banding / mark-recapture studies.

Under this project, we developed data sheets to collect demographic information on all resident and migrant birds captured (Appendix 1). The objective was to create a standardized form to record information that would lead to accurate conclusions about age, sex, reproductive status and physical condition (e.g, fat deposition) of birds. The data form ensures that all birds are examined systematically and, for banders who may not have experience with certain species, such as researchers coming from overseas, the data form was designed to guide them through an assessment to reach correct conclusions about the age and sex of an individual bird (or to allow someone with experience to reach the same conclusion). Additionally, all banding apprentices are trained in understanding the scientifically rigorous principle that a blank box on the form is quite different from a null condition of "no" or "0" i.e., they confirmed the absence of a certain condition versus not knowing the information because they failed to look. We developed the data sheet from a template provided by Bird Studies Canada (BSC) and made revisions through literature review (e.g., Ralph et al. 1993) and in collaboration with Certified Banding Instructors from Klamath Bird Observatory (KBO). We believe our data forms could serve as a template for standardized data collection throughout Jamaica.

4.1.2. Banding Results

Since implementing the bird banding in May 2002, we have netted / handled 3024 birds: 1946 were newly banded (1519 resident birds, 427 migrants), 819 were recaptured, and 259 were released without bands (mostly hummingbirds) (Table 2). Our species list includes 40 resident species (including two introduced resident species) and 22 migratory species. By the end of 2004, we will have sufficient data to conduct preliminary analyses on survivorship (as a function of age class and sex and among habitat types (stations) for the most common resident species: Jamaican Tody (Todus todus), Banaquit (Coereba flaveola), Black-faced Grassquit (Tiaris bicolor), Greater Antillean Bullfinch (Loxigilla violacea), and Orangequit (Euneornis campestris).




Table 2. Birds banded from May 2002 - 2004. See Appendix 2 for scientific names.

Resident Banded		Migrant Banded		Resident Unbanded
Zenaida Dove Common Ground Dove Caribbean Dove Ruddy Quail Dove Crested Quail Dove Yellow-billed Parrot Mangrove Cuckoo Jamaican Lizard Cuckoo Chestnut-bellied Cuckoo Smooth-billed Ani Jamaican Mango Red-billed Streamertail Jamaican Tody Jamaican Woodpecker Jamaican Elaenia Jamaican Pewee Sad Flycatcher Rufous-tailed Flycatcher Stolid Flycatcher Loggerhead Kingbird Jamaican Becard Rufous-throated Solitaire White-chinned Thrush White-eyed Thrush Jamaican Vireo Blue Mountain Vireo Yellow Warbler Bananaquit Jamaican Euphonia Yellow-faced Grassquit Black-faced Grassquit Yellow-shouldered Grassquit Greater Antillean Bullfinch Orangequit J'can Stripe-headed Tanager Greater Antillean Grackle Jamaican Oriole Chestnut Manakin TOTAL	1 14 1 19 1 1 1 4 2 18 3 10 106 2 27 1 14 29 1 13 1 12 25 15 24 1 8 543 29 46 176 56 137 154 1 8 14 1 1519	Belted Kingfisher Grey Kingbird Black-whiskered Vireo Gray-cheeked Thrush American Redstart Black-and-White Warbler Blackpoll Warbler Black-throated Blue Warbler Common Yellowthroat Hooded Warbler Magnolia Warbler Northern Parula Prairie Warbler Swainson's Warbler Western Palm Warbler Worm-eating Warbler Northern Waterthrush Louisiana Waterthrush Ovenbird Rose-breasted Grosbeak Indigo Bunting	1 1 214 3 21 4 2 50 24 3 1 7 6 1 4 9 28 2 43 1 2 427	Green Heron^1 Common Ground Dove^1 Caribbean Dove^1 Ruddy Quail Dove^1 Olive-throated Parakeet^1 Green-rumped Parrotlet^1 J'can Lizard Cuckoo^1 Chestnut-bellied Cuckoo Smooth-billed Ani^1 Jamaican Mango^1 Red-billed Streamertail^1 Jamaican Tody Sad Flycatcher^2 Rufous-throated Solitaire White-chinned Thrush^3 Bananaquit^2 Yellow-faced Grassquit^2 Black-faced Grassquit^2 Orangequit Greater Antillean Grackle^1	2 3 1 3 2 2 1 1 2 48 134 4 3 1 7 25 2 9 3 3 (256)
				Migrant Unbanded
				American Redstart^2 Black-throated Blue Warbler^2 Chestnut-sided Warbler^2	1 1 1 (3) 259

¹Correct band size not in-stock at WRC.

²Released / escaped from net or station before band applied.

³Legs damaged by scaly mite (Knemidocoptes spp.) infestation, unable to band.




With the establishment of stations in three different habitat types in January 2004, we are beginning to compare species composition, health, and demography in a subset of landuse types. All stations support similar avian communities, as determined both by the species netted and by species we detected during operation of nets but which we did not capture (Appendix 2). In the next three months, as the project continues through the support of EFJ, we will be establishing permanent vegetation monitoring points within and adjacent to the banding sites to quantify habitat characteristics, including structure and fruiting and flowering phenology (timing of food availability).

The numbers of individuals of migratory species, particularly of Black-throated Blue Warblers (Dendroica caerulescens) and Common Yellowthroats (Geothlypis trichas) enabled WRC to participate in the Institute for Bird Populations' Monitoreo de Sobrevivencia Invernal (MoSI) programme, which is establishing a network of North American, Mexican, Central American, and Caribbean banding station to monitor populations of migrant birds on the summer breeding and winter grounds <http://www.birdpop.org/MoSI/MoSI.htm>. We began operating as a MoSI station in October 2003 and will continue to participate in this programme.

Data on migratory birds were given to Catherine Levy, who holds the US Geological Survey (USGS) Master Station permit for BirdLife Jamaica (permit number: 22743) and who is responsible for reporting to USGS Bird Banding Laboratory. Dr. Koenig is a subpermittee of the BLJ permit (USGS no. 22743A). A list of local birds that were handled and banded is provided annually to the National Environment and Planning Agency (NEPA), which issues local research permits. WRC currently is developing an on-line database of monthly banding results, complete with search capabilities for birds banded in Windsor but recaptured by others elsewhere in Jamaica: <http://wrc.cockpitcountry.com/bbdatahome.php>. We currently are resolving technical issues related to the search capabilities.

Problems encountered: None

4.2. Bird Banding Skills Development

With the support of this project, WRC hosted two training workshops to introduce local Jamaicans to the techniques and ethics of bird banding. The first workshop was held April 30 - May 17, 2002 in collaboration with BLJ and BSC. Six persons participated in this workshop: one employee from the Institute of Jamaica's Natural History Division, four affiliated with Community-based Organisations (CBO), and one US Peace Corps volunteer (Table 3). From this group, one trainee from the Dolphin Head National Trust (DHNT) CBO, along with WRC's Dr. Koenig, went to BSC's Long Point Bird Observatory, Ontario, Canada, in the fall of 2002 for advanced training. Based on the continued local interest in skills development, WRC hosted a second 3-week banding workshop in January 2004 in collaboration with Klamath Bird Observatory (KBO), Oregon. For this second workshop, we focused training on government natural resource managers from Jamaica's Forestry Department (FD; 4) and NEPA (1) and several community residents (3, including the apprentice from DHNT). The focus on training government employees arose from a request by FD and NEPA for professional skills training to improve their capabilities in wildlife research and monitoring. Additionally, the instructors of the second workshop from KBO were both USGS Certified Banding Instructors and will be evaluating Dr. Koenig in the next year for certification as an instructor.

Of the 13 participants in the training workshops, five continue with their training apprenticeships under Dr. Koenig. Another three Jamaicans, who had banding experience prior to the WRC workshops, also participated in WRC banding on an irregular basis. A fourth trainee from the Jamaica Conservation and Development Trust, (JCDT) who had pre-workshop experience in both banding and bird surveying techniques, emigrated to the United States in January 2004.

Eleven of the 13 participants completed the full training workshops. One trainee decided not to return after the first week, citing a lack of interest and conflict with other responsibilities. The second trainee lacked the manual dexterity to safely handle birds and poor literacy skills prevented his participation in data recording.

Table 3. Summary of participants in Windsor Research Centre’s Bander Apprentice Programme. LM=Long Mountain Workshop 28 Jan-14 Feb 2002 (BSC-sponsored workshop held before period of NFWF funding); LPBO=Long Point Bird Observatory (BSC, Ontario, Canada) Latin American Training Programme; WRC=Windsor Research Centre workshop 30 April - 17 May, 2003; KBO=WRC workshop sponsored in collaboration with Klamath Bird Observatory 5-23 Jan 2004. With the exception of the two banders who currently hold USGS and NEPA permits, all apprentices theoretically need training in report preparation; however, most do not have the literary skills to achieve this level and only those with potential are identified for further training needs

Organizational Affiliation	Workshop Training	Able to Band Unsupervised	Further Training Needs
Windsor Research Centre	LM LPBO WRC KBO	Yes, currently holds USGS subpermit and NEPA permit	a. Certification as instructor
BirdLife Jamaica	LM WRC	Yes, currently holds USGS Master Station Permit and NEPA permit	a. Hummingbird banding b. Analyses of mark-recapture data
BirdLife Jamaica	LPBO	No, needs more experience in demonstrating ability to resolve problems and arriving at e to banding on-time	a. Hummingbird banding b. Using Pyle Guide quickly and correctly c. Ageing species not commonly encountered d. Analyses of mark-recapture data e. Report preparation for USGS and NEPA
Dolphin Head National Trust	WRC LPBO KBO	No, needs more experience in demonstrating ability to resolve problems and maintaining a banding station	a. Hummingbird banding b. Using Pyle and other reference guides quickly and correctly c. Accurate data recording d. Correct and complete maintenance of daily log of activities and data summaries e. Data entry
Dolphin Head National Trust	WRC	No, voluntarily left training workshop due to poor handling skills
Forestry Department	LM KBO	No, requires assistance with difficult extractions	a. Hummingbird banding b. Physical assessment of birds, ageing all species c. How to use Pyle Guide d. Data management
Forestry Department	KBO	No, needs more experience in demonstrating ability to resolve problems	a. Hummingbird banding b. Using Pyle Guide quickly and correctly c. Ageing species not commonly encountered d. Data management & analyses of mark-recapture data e. Report preparation for USGS and NEPA
Forestry Department	KBO	No, has not banded since the training workshop	a. Needs supervision with difficult extractions b. Hummingbird banding c. Using Pyle Guide quickly and correctly d. Ageing species not commonly encountered
Forestry Department	KBO	No, very nervous in handling birds	a. Data management & analyses of mark-recapture data b. Report preparation for USGS and NEPA
Institute of Jamaica	WRC	No, has not banded since the training workshop	a. Will require complete review before resuming banding
Jamaica Conservation & Development Trust	LM LPBO	No, has not participated in banding for one year and will require review of skills	a. Hummingbird banding b. Using Pyle and other reference guides quickly and correctly c. Data recording d. Data management e. Report preparation for USGS and NEPA
Jamaica Conservation & Development Trust	LM	No, disproportionate number of “wing strains” associated with extracting birds from mist nets; Has emigrated to US and no longer participates in training	a. Review of extraction techniques b. Physical assessment of birds for aging c. How to use Pyle and other reference guides d. Accurate data recording e. Data management
National Environment and Planning Agency	KBO	No, needs more experience in demonstrating ability to resolve problems and arriving to banding on-time	a. Hummingbird banding b. Using Pyle Guide quickly and correctly c. Ageing species not commonly encountered d. Data management & analyses of mark-recapture data e. Report preparation for USGS and NEPA
Southern Trelawny Environmental Agency	WRC	No, has not banded since the training workshop	a. Will require complete review before resuming banding
Southern Trelawny Environmental Agency	WRC	No, has not banded since the training workshop	a. Will require complete review before resuming banding
Southern Trelawny Environmental Agency	WRC	No, voluntarily left training workshop
US Peace Corps	WRC	No, but resumed banding in June 2004 after not participating since May 2002 workshop; is expected to be qualified for unsupervised banding in 3 months	a. Hummingbird banding b. Using Pyle Guide quickly c. Ageing species not commonly encountered d. Analyses of mark-recapture data e. Report preparation for USGS and NEPA
Member of local community	KBO	No, needs more experience in demonstrating ability to resolve problems and maintaining a banding station; training needs are no longer relevant as trainee "disappeared" in the US en-route to summer apprenticeship with KBO	a. Hummingbird banding b. Using Pyle and other reference guides quickly and correctly c. Accurate data recording d. Correct and complete maintenance of daily log of activities and data summaries e. Data entry f. Report preparation

In the absence of banding guidelines for Jamaica, WRC used trainee evaluation scorecards (Appendix 3) developed by BSC and KBO, which adhere to North American Banding Council (NABC) guidelines and adopted the descriptions and skills required for the grades of permits issued by the British Trust for Ornithology (Ringer's Manual 2001, pp. 63-68; Appendix 4). At the end of the 3-week workshops, 10 of the 11 participants achieved minimum skill levels comparable to BTO's "T" (Trainee) permit, which includes basic handling and elementary skills of:

1. recognition of common species in the field and in hand

2. an appreciation of the effect of handling time on birds

3. holding a bird in the bander's grip and demonstrating dexterity in handling in order to examine a bird

4. fitting and butting a conventional split ring on a bird correctly

5. reading numbers from USGS-BBL and local bands correctly

6. carrying bird bags correctly

7. writing clearly and legibly (But see Problems encountered below)

8. demonstrating an ability to learn ageing and sexing techniques

These 10 participants also were competent is safely extracting from mist nets birds that were not badly entangled and recognizing when they needed to request assistance from an instructor. The 11^th workshop participant had a slight shake to his hands which limited his abilities to safely handle birds, but his data recording skills and theoretical understanding of aging birds make him a valuable scribe. It should be emphasized that a 3-week workshop provides insufficient time to train any inexperienced bander to operate at a level unsupervised.

Of the five workshop trainees who continue with their skills development (four Jamaican, one US Peace Corps volunteer) during monthly banding activities, four are reliable in their assessments of physical condition, taking basic biometrics and, for the commonly-encountered species, ageing and sexing. While all are reliable in describing and recording data for routine characteristics of birds (i.e., completing the standardized columns on the data forms), only the Peace Corps volunteer is reliable in noticing or recording unusual details as supplemental comments. It should be noted that the Peace Corps volunteer holds a bachelor's degree in biology. Primary difficulties for the Jamaican trainees are with spelling, working knowledge of avian topography / anatomy, and a failure to remember that field identification guides with diagrams of standard terminologies are always available and to be used for reference during banding. At present, Dr. Koenig is the only bander in the WRC programme who is able to using proficiently the book Identification Guide to North American Birds, Part I (Pyle et al. 1987), a reference for identifying, ageing, and sexing species that migrate to Jamaica from North America and which is required by USGS-BBL for persons holding Master Bander permits. Four of the 5 apprentices understand how to use the guide, but need more practice, particularly with species encountered infrequently or those which can be difficult to identify and age.

Problems encountered: Nearly all of our Jamaican trainees and apprentices mastered the technical skill of extracting birds from mist nests and collecting standard morphometric data (e.g. wing length) Most were competent in aging and sexing birds which they had seen frequently (i.e., received descriptive instruction), but had difficulties transferring this knowledge to evaluate a new species. With the exception of the BLJ, NEPA and FD trainees, our biggest challenge related to weak literacy skills -- a result of the Jamaican primary and secondary education system. Jamaican culture tends towards aural learning; reading and spelling skills are frequently deficient. Because of this, our apprentices have not progressed as quickly towards establishing their own stations. For example, all apprentices still have difficulties remembering to record "unusual" data (i.e., describing observations that are not collected and coded in a standardized format on the data sheet) or noting physical abnormalities, managing data, and preparing reports. These are minimum requirements for any apprentice hoping to operate a station with minimal supervision by an instructor (i.e., to hold a USGS Subpermit) or to operate independently (e.g to hold a USGS Master Bander permit).

Secondly, there are no field guides of resident Jamaican birds to assist banders in accurately ageing birds as a function of plumage characteristics (i.e., we lack the type of reference material organized by Pyle et al. in the Identification Guide to North American Birds). This made for many uncertainties by the instructor of the first two workshops. However, with our continuing banding, WRC is resolving this problem and is working towards creating a guide for all local and visiting banders / researchers.

A final major problem for this project, and indeed, any NGO's professional training development programme, is the limited employment opportunities and subsequent emigration of apprentices to the United States. In one instance, an apprentice and park ranger for JCDT in the Blue and John Crow Mountain National Park chose to emigrate to the US to earn higher wages as a welder. The second "emigration" is of cause for greater concern for WRC's development of international collaborations and efforts to find short-term overseas field projects for which we can recommend our apprentices to improve their skills as field biologists and research assistants. In this instance, a trainee who demonstrated keen interest and skills during the January 2004 workshop was invited to work with KBO in Oregon for the summer field season, 2004. KBO, in collaboration with Southern Oregon University, sponsored his visa for their international apprentice programme. The trainee, however, used this opportunity to "go underground for a better life" in New York rather than continuing on his flight to San Francisco. Despite verbal promises via telephone with KBO from New York that he would arrive for his apprenticeship and verbal warnings from KBO that the visa would be terminated if he failed to arrive at KBO, our Jamaican chose to remain in New York. KBO had no choice but to notify US Homeland Security and we await further details. This situation has serious consequences for the other apprentices we had hoped to send to KBO for field experience, with the expectation that the US embassy in Jamaica will be reluctant to issue visas.

4.3. Habitat Use by Birds - Permanent Point Counts

4.3.1. Survey Point Descriptions and Methods

Permanent bird monitoring points were established in seven localities across Cockpit Country (Figure 1) and surveyed quarterly by two knowledgeable observers and, in 2004, by two accompanying trainees. The localities encompassed major land use / cover type classifications of: (1) sugar cane; (2) bamboo (Bambusa vulgaris); (3) non-native fern (Nephrolepis spp.); (4) coffee / mixed fruit trees; (5) Yam cultivation; (6) forestry plantation of Caribbean pine (Pinus caribaea) and Blue Mahoe (Hibuscus elatus); (7) herbaceous cockpit bottom; (8) disturbed broadleaf - edge forest; (9) closed broadleaf interior forest (Table 4). It should be noted, however, that no survey point was located further than 300 m from regenerating edge forest . These survey points were established to enable long-term monitoring of seasonal and annual patterns of bird species composition, abundance, and population trends within each habitat type and for comparison of avifauna associations between habitat types; they were not established to provide exhaustive surveys of avifauna throughout Cockpit Country. Additionally, these baseline data points were established in areas where forest restoration research would be appropriate (i.e., relatively accessible).

Figure 1. Locations of permanent bird survey points in Cockpit Country. All points are geo-referenced using WGS84 reference datum.






Table 4. Number of fixed-radius point counts completed in the major land use classifications of Cockpit Country and peripheral environs, October 2002 - April 2004. Point count surveys were conducted quarterly (January, April, July, October) to coincide with the arrival and departure of Neotropical migrant species and to the beginning and end of resident bird breeding events in order to assess presence / absence and potential detection biases throughout the year.

Habitat Type	No. of Points Surveyed
1. Sugar cane	8
2. Non-native bamboo monoculture	6
3. Non-native fern monoculture	5
4. Yam cultivation	14
5. Coffee / mixed fruit tree	8
6. Forestry plantation	25
7. Herbaceous cockpit bottom (regenerating abandoned agriculture)	16
8. Disturbed broadleaf - edge	40
9. Closed broadleaf - interior	19
Total	141

Pilot surveys conducted in 2002 led to standardized survey sampling protocols as follows:

(a) surveys were conducted quarterly, in January, April, July, October, to monitor for patterns of breeding vs. non-breeding season habitat use by resident birds and seasonal use by winter and summer migratory species;

(b) points along transects were spaced at least 100 m apart to minimize duplicate counting of individual birds;

(c) all birds seen or heard within a radius of 25 m of an established point were recorded during a 10-minute period (see Figure 2); birds detected beyond the 25-m radius were recorded for an overall species list but their presence and abundance data were not incorporated into the habitat association analyses;

(d) morning point counts were initiated at sunrise and conducted for no longer than three hours;

(e) evening point counts were initiated three hours before sunset and conducted for two hours.




Figure 2. Species discovery curve. The graph shows the cumulative total of the number of species identified during a 10-minute sampling period, with data pooled for all seven habitat types.



4.3.2. Bird - Habitat Associations

Eighty three bird species were recorded during the quarterly point counts from October 2002 through April 2004: 62 resident species (including 27 of Jamaica's 28 extant endemic species) and 21 migratory species (Table 5; Appendix 2). Data from 59 of these 82 species were analyzed for a comparison of avifaunal communities and abundances among habitat types. The excluded 23 species either occurred > 25-m from the observer during the fixed-radius point counts (n = 13), occurred in fewer than three points (n = 9), or, as was the case for Cave Swallows (Pterochelidon fulva), their soaring and foraging activities occurred over multiple habitat types (n = 1). Turkey Vultures (Cathartes aura) were not excluded if they used the habitat for roosting or for sunning their wings.

Table 5. Summary total of bird species detected in Cockpit Country environs, October 2002 through April 2004. Abundance data were included for analyses when the species was detected in a minimum of three fixed-radius point counts.

Species status	No. observed in the area	No. included in analyses
Endemic	27	26
Resident	31	21
Introduced resident	2	2
Natural range expansion	2	1
Summer migrant	3	2
Winter migrant	18	7
Total	83	59

The nine habitat types varied both in species richness (i.e., number of species recorded as "present" in at least one point count) and in the abundance of individuals (i.e. percentage of detected species which exceeded three point count occurrences and, consequently, were included in abundance estimates to discriminate habitat associations from transient movement through an area) (Table 6). Both the native vegetation habitats of disturbed broadleaf forest of edge habitat and closed broadleaf forest in the interior and the agriculture habitat of coffee and mixed fruit trees supported significantly greater numbers of species and individuals (which correlates highly to the number of species) compared to all other habitats (Table 7). These habitats are noted for their multi-species plant communities and structural complexity. In contrast, the three habitats of mono- or near-monoculture vegetation -- sugar cane, fern, and bamboo --, with their corresponding uniform physical structure, supported avian communities of least diversity and of lowest abundance of individuals.

Table 6. Numbers of species detected during fixed-radius point counts in each habitat. For those species which occurred in at least three point counts, abundances were analyzed to compare species communities among habitats.

Habitat	Presence Detected	Abundance Analyzed	% of Detected Analyzed
1. Sugar cane	23	8	35
2. Non-native bamboo monoculture	9	6	67
3. Non-native fern monoculture	17	7	41
4. Yam cultivation	39	26	67
5. Coffee / mixed fruit tree	50	38	76
6. Forestry plantation	56	43	77
7. Herbaceous cockpit bottom	52	38	73
8. Disturbed broadleaf - edge	63	52	83
9. Closed broadleaf - interior	46	39	85

The two native vegetation habitats of edge and interior forest also were the only two habitats in which the numbers of resident species varied significantly between winter / non-breeding season surveys and summer / breeding season surveys (Table 7). A seasonal comparison of migrant species, obviously, was not valid. Greater numbers of resident species and individuals were detected during the summer. Quite likely this corresponded not only to an increase in population sizes because of the presence of young birds but also because of greater detection capabilities due to the more conspicuous behaviors associated with breeding (e.g., territorial males in song; fledglings giving begging calls). The lack of difference in species numbers in the other habitats between seasons raises a call for further research to explore whether these habitats are used for nesting, whether nesting is successful (i.e., source-sink dynamics) or whether they serve predominately as foraging habitats. These questions are important for forest managers, particularly for evaluating the quality of agro-forestry plantations as suitable bird habitat. A further management implication of these observed seasonal differences relates to the timing of birds surveys which are included for environmental impact assessments. Surveys conducted outside of breeding seasons may fail to detect cryptic, secretive species and underestimate the importance of an area.




Table 7. Mean (+ SE) number of individuals and total number of species in fixed-radius point counts in the six dominant landuse types of Cockpit Country, October 2002 - April 2004. Comparisons of seasonal changes in species number and abundance, by habitat type, were made for resident species only.

		Winter/Non-breeding season				Summer/Breeding season
	No. of species		Indiv. per point count		No. of species		Indiv. per point count
Habitat	Resident	Migrant	Resident	Migrant	Resident	Migrant	Resident	Migrant
1. Sugar cane	1.7 + 1.1	0.1 + 0.3	3.4 + 2.8	0.1 + 0.3	1.9 + 1.0	0.0 + 0.0	3.2 + 2.0	0.0 + 0.0
2. Bamboo	1.2 + 0.9	0.0 + 0.0	1.5 + 1.3	0.0 + 0.0	1.6 + 1.1	0.0 + 0.0	1.8 + 1.3	0.0 + 0.0
3. Fern	1.2 + 1.1	0.2 + 0.5	1.8 + 1.7	0.2 + 0.5	2.1 + 0.9	0.0 + 0.0	2.5 + 1.1	0.0 + 0.0
4. Coffee / mixed fruit trees	5.0 + 2.0^a	0.8 + 0.7	7.5 + 5.5^a	0.8 + 0.8	5.8 + 2.4^a	1.3 + 1.0	8.7 + 5.0^a	1.7 + 1.4
5. Yam cultivation	3.1 + 1.1	0.6 + 0.7	4.5 + 2.3	0.6 + 0.7	3.0 + 1.6	0.6 + 0.7	4.2 + 2.9	0.8 + 1.1
6. Forestry plantation	4.0 + 1.9	0.8 + 0.9	5.4 + 3.0	0.9 + 1.2	4.9 + 2.1	0.8 + 0.6	6.9 + 3.6	1.0 + 0.9
7. Herbaceous cockpit bottom	3.6 + 1.9	0.7 + 0.7	4.9 + 3.2	0.8 + 0.7	4.2 + 1.9	0.7 + 0.7	5.4 + 2.9	0.9 + 0.9
8. Disturbed broadleaf - forest edge	5.6 + 2.4^a, b	0.7 + 0.8	7.4 + 3.9^a, b	0.7 + 0.9	6.9 + 2.7^a, b	0.9 + 0.7	9.2 + 3.9^a, b	1.0 + 0.9
9. Closed broadleaf - interior forest	5.3 + 1.8^a, b	1.2 + 1.2	6.4 + 2.6^a, b	1.1 + 1.0	8.0 + 2.2^a, b	1.0 + 0.6	10.8 + 3.7^a, b	1.2 + 0.7
^aP < 0.02 in a comparison of habitats 4, 8 and 9 to all other habitats.
^bP < 0. 003 in a seasonal comparison within habitat type.




Not only did the nine habitats vary in the average number of species detected in each point count but they also varied in species composition and the density at which individuals occurred, as assessed by the percentage of survey points in which individuals were detected (Figures 3 and 4). Again, the habitats of regenerating edge and interior forest were notable in that 16 and 18 of the species detected, respectively, occurred in greater than 20% of the points surveyed while only two species occurred "commonly" in bamboo and three in fern monocultures (Table 8).

Table 8. Number of species detected as "common" in points counts. (n) = number of species detected within point count surveys for the habitat

Habitat	Occurred in > 20% of point counts	Occurred in > 25% of point counts
1. Sugar cane (n=23)	6	3
2. Non-native bamboo monoculture (n=9)	2	2
3. Non-native fern monoculture (n=17)	3	3
4. Yam cultivation (n=50)	9	7
5. Coffee / mixed fruit tree (n=39)	9	7
6. Forestry plantation (n=56)	8	7
7. Herbaceous cockpit bottom (n=52)	10	7
8. Disturbed broadleaf - edge (n=63)	16	13
9. Closed broadleaf - interior (n=46)	18	13

Of importance for forest resource managers, 6 endemic species and two of the island's resident large columbids, the Plain Pigeon (Patagioenas inornata) and the Ruddy Quail Dove (Geotrygon Montana), displayed strong affinities for closed-canopy forested habitat (Figure 5). Two non-native species, the introduced Chestnut Manakin (Lonchura malaccaI) and the Cattle Egret (Bubulcus ibis), a species whose range has been expanding naturally through the conversion of forested habits to pastoral agriculture, were restricted entirely to agriculture landuse types; their presence indicates extreme conversion of habitat. Also of particular concern and relevant for forest managers was a sighting of one Shiny Cowbird (Molothrus bonariensis) (either juvenile male or juvenile/adult female) along the Barbecue Bottom Road, believed to have originated from the Campbell's District on the north eastern edge of Cockpit Country. While Jamaican Blackbirds (Nesopsar nigerrimus), a species believed by ornithologists to be the most endangered of all birds on Jamaica, were not recorded along the Barbecue Bottom / Burnt Hill Road, this sighting serves to highlight concern about the maintenance of gaps through the closed canopy forest that may serve as corridors for cowbirds. Shiny Cowbirds have been confirmed brood parasites on Jamaican Orioles (Icterus leucopteryx), which are in the same family (Icteridae) as the Jamaican Blackbird.

Problems encountered: None




Insert tons of figures from Excel

Resident hab 1&2




Res Hab 3 & 4




Res hab 5&6




Res hab 7 & 8




Res hab 9




Mig hab 1 2 3




mig hab 4 5 6




Mig hab 7 8 9




Figure 5. Birds of Cockpit Country demonstrating the strongest affinities for open, agricultural areas and closed-canopy forest-dependence. The presence of the former is an indication of extreme habitat disturbance. Species are recognized as endemic (E), resident (R), a natural range expansion (facilitated by human alteration of the landscape) (N), or introduced by humans (I). Habitat types: 1=sugar cane; 2=bamboo; 3=fern; 4=mixed coffee/agriculture; 5= yam cultivation; 6=forestry plantation; 7=herbaceous cockpit bottom; 8=regenerating edge; 9=interior forest.

			Habitat Type
Family	Common Name	Status	1	2	3	4	5	6	7	8	9
ARDEIDAE	Cattle Egret	N
CUCULIDAE	Smooth-billed Ani	R
MIMIDAE	Northern Mockingbird	R
EMBERIZIDAE	Chestnut Manakin	I
COLUMBIDAE	Plain Pigeon	R
	Ring-tailed Pigeon	E
	Ruddy Quail Dove	R
	Crested Quail Dove	E
COTINGIDAE	Jamaican Becard	E
MUSCICAPIDAE	Rufous-throated Solitaire	R
	White-eyed Thrush	E
VIREONIDAE	Blue Mountain Vireo	E
EMBERIZIDAE	Jamaican Blackbird	E

4.4. Map Reading and Orienteering

Critical to any type of biological or physical survey and long-term monitoring are the needs to (1) establish survey points in areas of interest and (2) return to the same points at determined sampling intervals. With the advent of satellite-guided geo-referencing of one's position anywhere on the Earth, the determination of a position with respect to the equator and the North and South Poles (i.e., latitude and longitude) is straightforward. However, we are finding that, with the improved technology, basic skills in interpreting topographic maps with respect to one's current position and being able to navigate using a compass to pre-determined points in a logical way (e.g., following contour lines, avoiding cliffs) are declining. Jamaicans, in general, do not know how to interpret maps of any type. Few natural resource professionals and students in the natural sciences have experience working with topographic maps and being able to communicate accurately where they have worked, save a "regurgitation" of numbers obtained from their hand-held Global Positioning System (GPS) device. This "regurgitation" also will lack precision if users fail to note their reference datum: Jamaican topographic maps and the national Geographic Information System (GIS) are aligned to either JAD69 or JAD2001 datums, in contrast to the more commonly used World Geodetic System of 1984 (WGS84). The resulting risk is that follow-up surveys cannot be repeated for exact localities or, worse, field researchers will get lost.

To address local short-comings and before we established permanent points for bird surveys, WRC, in collaboration with Forestry Department's CIDA-funded Trees-for-Tomorrow Project conducted a week-long workshop in map reading and orienteering in November 2002. Supplemental training in wilderness first aid was provided by the Red Cross of Jamaica. Theoretical (lecture) and practical (field) training modules included:

• accurate recording of field data;
• use of an orienteering compass;
• identifying features of topographic maps;
• developing an orienteering route;
• using a map to orient in the field.

The Windsor environment presented a diverse landscape, from well-defined, easy to locate road junctions, small bridges, and isolated hills in an alluvium plain through a complex cockpit karst topography.

Ten participants, representing four institutions on Jamaica (Forestry Department, Institute of Jamaica, JCDT, Dolphin Head National Trust), participated in the workshop. Nine of the ten successfully completed all components of the course. These nine demonstrated consistently high levels of interest and enthusiasm in both the lecture and practical field exercises. All participants easily mastered the use of an orienteering compass. Eight of the ten quickly mastered recognizing topographic features on the map (e.g., hilltops vs. depressions; saddle-corridors). In teams of two, all participants successfully completed a pre-determined orienteering road that took them along a road, trail, and then traversing off-trail around a cockpit hill. The one individual who failed to pass the course did develop an understanding of basic principals but requires additional supervision in the classroom to review map features. However, he, along with the other nine participants, was able to develop a suitable orienteering route on a map and provide written instructors to navigate this route. The final field examination, ground-truthing a route created by one of the workshop participants, proved very useful in highlighting that maps often will not align perfectly in the “real world. The exercise also reinforced the need to be constantly aware of one’s surroundings and accurately record activities when traversing a topographically complex terrain. Participants evaluated the workshop as being extremely useful for their professional activities.

Based on the success of the first workshop and from expressed interest by institutions for training of other staff, we conducted repeat of the workshop in March 2004. Participants were from Forestry Department (4), National Environment and Planning Agency (NEPA) (1), and one local resident. All participants in this workshop had participated in the January 2004 bird banding workshop. All 6 successfully completed the course and should competently be able to locate themselves anywhere on Jamaica, including locating all points of our bird surveys and banding stations.

Problems encountered: None

4.5. Information Sharing

4.5.1. Adaptive Management Seminar

As originally planned, the results of this bird monitoring project were to be presented to the management committee of the USAID-funded Cockpit Country Parks-in-Peril (CC-PiP) programme. Results were to be evaluated within the framework of Site Conservation Planning and management strategies and monitoring activities revised, as necessary. The management committee was comprised of the initiating four principals of the CC-PiP programme: TNC-Jamaica, Forestry Department, WRC, and Southern Trelawny Environment Agency (STEA) In 2003, TNC-Jamaica unilaterally disbanded the management committee and assumed full responsibility of project management. TNC-Jamaica planned a follow-up management meeting with WRC and STEA in December 10^th, 2003, which was postponed at the last minute to December 12^th, then to the "New Year." A nebulous period of "July 2004" was proposed by TNC-Jamaica, but at the time of this report, no confirmation has been received by WRC. WRC submitted a final report to TNC-Jamaica in November 2003 to complete contractual obligations.

4.5.2. Community Outreach

With the solid foundation of information acquired in the past two years from bird banding and surveying, WRC is developing educational materials and undertaking community outreach. Through the Jamaica Environment Trust's Schools Environment Programme (JET - SEP), we introduced our project to three schools: two schools are located on the periphery of Cockpit Country, one is 15 miles away, on the coast and is located at the mouth of the Martha Brae River, whose headwaters develop in Windsor, Cockpit Country. Thus far, we have introduced 15 students (ages 10-14 yrs) and 4 teachers to the bird banding, giving students a chance to see their Jamaican birds in-hand, to learn about their habitat needs, and to learn about the conservation concerns for the island. As an aid to learning the many new concepts, we are modifying educational materials developed by Colorado Bird Observatory for resident species (Bonfield and York 1996). With these pilot schools, we also are developing a questionnaire to assess students' knowledge and attitudes towards wildlife. Thus far, the questionnaire has been administered after students have observed the banding and interacted with the birds, but in future the finalized form will also be administered before presentations to see can assess efficacy of our outreach.

WRC also was invited by Forestry Department to participate in their Annual Exhibitions in 2003 and again in 2004 (Appendix 5). These exhibitions are held in community centers in the capitals of parishes across Jamaica and provide Forestry Department and their partners in the NGO and private sector communities an opportunity to inform the public about activities and products associated with Jamaica's forests.

Problems encountered: The change in management structure of the CC-PiP programme resulted in delays in developing management strategies and implementing conservation activities. Because of these programmatic changes, the management meetings at which we would planned to present results and use as training opportunities for our apprentices to gain experience in giving presentations were not convened. We will present at future management meetings as appropriate.

5. CONTINUATION OF ACTIVITIES

As noted under section 3. PROJECT FUNDING, the 2-year grant recently approved by EFJ will enable WRC to continue to develop its bird monitoring, skills training, and community outreach activities. Our planned activities include:

• Two-day-per-month banding and training at WRC expanded to six days at three local sites; emphasis will continue on obtaining high quality data and providing training opportunities for apprentices;

• Continue collaborations with Klamath Bird Observatory and Bird Studies Canada to offer banding training workshops for Jamaicans and other Caribbean nationals and to promote exchange apprenticeships for carefully selected participants;

• Establish a fourth banding station near the community of Duanvale in an area known as Linton Park, northern Cockpit Country, to be operated quarterly in collaboration with Forestry Department; this site will be developed for WRC's environmental community outreach, which will target a minimum of five Trelawny Parish schools in 2004 and will be incorporated into Jamaica Environmental Trust's Schools Environment Programme (JET-SEP);

• Develop forest restoration research in the aforementioned Linton Park -- an area of approximately 50 hectares which Forestry Department recently learned is part of the Cockpit Country Forest Reserve and requires active management for restoration from pasture; we have identified Linton Park as a Conservation Demonstration Area and, with financial support from BirdLife International and UNEP-GEF, we will work with colleagues from Forestry Department and the University of the West Indies to evaluate multiple strategies to promote natural forest regeneration, with evaluation of the roles of birds as pollinators and seed dispersers

• With our foundation of banding and survey data, strengthen institutional collaborations to facilitate new research on resident and migratory landbirds, including gaining a better understanding of the effects of fragmentation and forest patch size on bird communities (University of Maryland / Smithsonian Institution 2005 - 2006), evaluating forest restoration dynamics on floral and faunal communities (Humboldt State University, California 2005 - ), and assessing the effects of shade vs. sun coffee on bird communities (Humboldt State University, 2005 - ). These collaborations also provide employment opportunities for our apprentices.

6. ACKNOWLEDGEMENTS

Windsor Research Centre would like to thank the directors and staff of the National Fish and Wildlife Foundation for the grant which served as the cornerstone of this project. We also thank The Nature Conservancy-Jamaica for administering funds from the Orvis Foundation, which served as our preliminary Challenge Funds. Recent funding from the Environmental Foundation of Jamaica provided additional support of Challenge Funds and will enable us to continue with our bird conservation efforts for at least two more years. A recent grant from UNEP-GEF, administered by BirdLife International's Important Bird Areas Programme, will enable us to develop pilot demonstration plots for restoring bird habitat in the globally-recognized IBA of Cockpit Country.

Many individuals contributed to the success of this project: Catherine Levy worked tirelessly with colleagues from Bird Studies Canada to set up the programme of bird banding training workshops in Jamaica and gave her full devotion to David Wege of BirdLife International and the IBA programme. Master Banders Paul N. Prior (Bird Studies Canada), Keith Larson (Klamath Bird Observatory), and Robert Frey (Klamath Bird Observatory) inspired all with their dedication, knowledge, and respect for bird banding, not to mention their extraordinary patience with trainees at the nets. We also thank Steve Wilcox, John McCracken, and Michael Bradstreet of Bird Studies Canada for their support of local bird banding and for their supporting apprenticeships of Susan Koenig, Christopher Samuels and Ryan Love at Long Point Bird Observatory. Thanks also to John Alexander of Klamath Bird Observatory for his support of Keith and Bob and attempting to provide a summer apprenticeship to a Jamaican -- sorry he was so ungrateful and disappeared in New York.

A special thanks to Herlitz Davis for braving the mosquitoes and rain during the bird surveys and to Chris Samuels for his continued dedication to the birds of Jamaica. To Sugarbelly, for keeping the field crew well-fed, to Danny Brown for not converting his regenerating pasture back to tick-infested cow pasture and to Michael Schwartz, who deals with all of our technical glitches, from manufacturing the best mist net poles imaginable to keeping the solar electricity burning for the magnifying lamp, Susan Koenig gives a special thank you.

7. LITERATURE CITED

Bonfield, S. and S. York. 1996. Birds in Hand and Field: Activities for Exploring Birds. Colorado Bird Observatory

North American Banding Council. 2001. The North American Banders' Study Guide. Point Reyes Station, California. http://nabanding.net/nabanding

Raffaele, H., J. Wiley, O. Garrido, A. Keith, and J. Raffaele. 1998. A Guide to the Birds of the West Indies. Princeton University Press, Princeton, New Jersey.

Ralph, C.J., G.R. Geupel, P. Pyle, T.E. Martin, and D.F. DeSante. 1993. Handbook of Field Methods for Monitoring Landbirds. Gen. Tech. Rep. PSW-GTR-144, Albany, California: Pacific Southwest Research Station, Forest Service, US Department of Agriculture.

Redfern, C.P.F. and J.A. Clark. 2001. Ringers' Manual. BTO Thetford, UK.