UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Population dynamics of the Ruddy-capped nightingale thrush (Catharus frantzii) in the central highlands… Rangel-Salazar, José Luis 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2006-200728.pdf [ 5.53MB ]
Metadata
JSON: 831-1.0075067.json
JSON-LD: 831-1.0075067-ld.json
RDF/XML (Pretty): 831-1.0075067-rdf.xml
RDF/JSON: 831-1.0075067-rdf.json
Turtle: 831-1.0075067-turtle.txt
N-Triples: 831-1.0075067-rdf-ntriples.txt
Original Record: 831-1.0075067-source.json
Full Text
831-1.0075067-fulltext.txt
Citation
831-1.0075067.ris

Full Text

POPULATION DYNAMICS OF THE RUDDY-CAPPED NIGHTINGALE THRUSH (CATHARUS FRANTZll) IN THE C E N T R A L HIGHLANDS OF CHIAPAS, MEXICO by Jose Luis Rangel Salazar B.Sc, Universidad Nacional Autonoma de Mexico, 1990 M. Sc., Universidad Nacional de Costa Rica, 1995 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Forestry) THE UNIVERSITY OF BRITISH COLUMBIA August 2006 ©Jose Luis Rangel-Salazar Z0O& ABSTRACT I examined local population dynamics and regional distribution of the threatened Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Central Highlands of Chiapas, Mexico. Pair density, nesting success and daily nest survival were higher in the moist forest habitat with more understory vegetation than in the dry-disturbed forest habitat. Nest-site selection and daily nest survival were associated with concealment attributes at the nest-site and patch levels. Parental visitation rates to nests did not differ across habitats, but were lower for successful than failed nests. Female nest attentiveness was higher in the moist forest and for successful nests than in the dry-disturbed forest and failed nests. Despite a higher carrying capacity for breeding pairs and captured individuals in the moist forest than the dry-disturbed forest, I found no variation in productivity (female offspring/breeding female/year) or adult survival (cp = 0.79) across habitats. Both survival ( r = 0.93) and productivity (r- -0.33) correlated significantly with population growth (A). Overall, both moist and dry forest types supported stable populations of C. frantzii (A = 1.09), and thus I found no evidence for habitat-specific demography. In occupied forest habitats in the region around the Huitepec Reserve, C. frantzii showed a high probability of detection with repeated sampling. Abundance of singing males tended to be lower in younger and disturbed forest than in old-growth broadleaf forest. Habitat attributes such as elevation, slope, tree height, and shrub and herb densities varied across sites, but showed no consistent relationship to abundance. Although moist forests with more understory vegetation may favour better breeding success, adjacent dry-disturbed forests also supported stable, albeit lower, densities of C. frantzii in the Huitepec reserve. Both habitat types also supported non-territorial individuals that may be important contributors to stable populations. Thus, old growth and second-growth broadleaf forests with low to moderate vegetation harvesting can provide suitable habitats for C. frantzii in montane forests of the Central Highlands of Chiapas, Mexico. TABLE OF CONTENTS ABSTRACT n TABLE OF CONTENTS in LIST OF TABLES v LIST OF FIGURES vn ACKNOWLEDGEMENTS ix DEDICATION xi CHAPTER 1: 1 GENERAL INTRODUCTION 1 THE CONSERVATION OF AVIAN DIVERSITY FROM NEOTROPICAL MONTANE FORESTS 1 VITAL RATES AND LIFE-HISTORY TRAITS OF TROPICAL MONTANE FOREST BIRDS 3 POPULATION DYNAMICS AND HABITAT QUALITY 5 THREATS TO TROPICAL MONTANE FOREST BIRDS 6 THE STUDY SPECIES - CATHARUS FRANTZII (RUDDY-CAPPED NIGHTINGALE THRUSH)..... 7 THESIS OBJECTIVES 7 STUDY AREA 8 OVERVIEW OF THE THESIS 9 REFERENCES 13 CHAPTER 2: 19 INFLUENCES OF HABITAT VARIATION, NEST-SITE SELECTION AND PARENTAL BEHAVIOUR ON THE BREEDING SUCCESS OF CATHARUS FRANTZII (RUDDY-CAPPED NIGHTINGALE THRUSH) IN CHIAPAS, MEXICO 19 INTRODUCTION 19 SPECIES AND STUDY AREA 21 METHODS 23 RESULTS 28 DISCUSSION 31 CONCLUSION 36 REFERENCES 47 CHAPTER 3: 53 POPULATION DYNAMICS OF THE MEXICAN ENDANGERED CATHARUS FRANTZII (RUDDY-CAPPED NIGHTINGALE THRUSH): HABITAT-SPECIFIC DENSITY, PRODUCTIVITY AND SURVIVAL 53 INTRODUCTION 53 METHODS 55 RESULTS 60 DISCUSSION .. . .62 CONCLUSION 65 . REFERENCES 72 CHAPTER 4: . . .78 OCCUPANCY AND DETECTABILITY RATES OF CATHARUS FRANTZII (RUDDY-CAPPED NIGHTINGALE THRUSH) IN THE CENTRAL HIGHLANDS OF CHIAPAS, MEXICO 78 INTRODUCTION 78 METHODS 80 RESULTS : 83 DISCUSSION 85 CONCLUSION 88 REFERENCES 94 iii CHAPTER 5: 98 CONCLUSIONS 98 POPULATION ECOLOGY OF A TROPICAL MONTANE FOREST BIRD 98 CONSERVATION ECOLOGY OF CATHARUS FRANTZII IN THE MONTANE FORESTS OF CHIAPAS 101 REFERENCES .- 104 iv LIST OF TABLES Table 2 .1 . Frequencies of main plant species (6 of 18) used as nest-substrata by Catharus frantzii (Ruddy-capped Nightingale Thrush) in moist and dry-disturbed forest habitat types at the Cerro Huitepec Biological Reserve, Central Highlands of Chiapas, from 2000 to 2003. Frequencies included all nests found (N=193), and percent of successful and depredated nests was restricted to fate-known and fate-inferred nests 37 Table 2.2. Habitat attributes at nest-site and nest-patch levels of Catharus frantzii (Ruddy-capped Nightingale Thrush) nest and random sites found in moist and dry-disturbed forest types at the Huitepec Biological Reserve, Central Highlands of Chiapas. Selected variables were used for the final discriminant analysis (DA) model for nest-site selection in the overall study site (a), moist forest (b), dry-disturbed forest (c), and for the Mayfield logistic regression model (d) 38 Table 2.3. Parameter estimates for selected habitat attributes that reflect the maximum effect on daily nest survival rates from a Mayfield logistic regression model for Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Cerro Huitepec Biological Reserve, Central Highlands of Chiapas, during 2000-2003. Estimates are from a single global model that included fate-known nests (n = 78). Positive estimates indicate higher nest survival 40 Table 3.1. Model selection results from program MARK for adult apparent survival ($) and encounter rates (p) for Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas, between 1995 and 2003. Only the five best models and the fully parameterized model are presented (f = year, h = habitat, s = sex, AlCc = Aikaike's Information Criterion, Deviance = model fit). Sample size = 196 individuals 66 Table 3.2. Estimates of apparent adult apparent survival (<(>) and encounter rates (p) for Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas, from 1995 to 2003. Sample sizes are shown in parentheses 67 Table 3.3. Demographic parameter estimates over a four-year period (2000-2003) for Catharus frantzii (Ruddy-capped Nightingale Thrush) in two forest-habitat types (moist forest and dry-disturbed forest) at the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas. Seasonal fecundity represents the number of female offspring produced per breeding female within a breeding season. Seasonal productivity (3) represents the number of female offspring produced per successful female by the end of the breeding season. The intrinsic population growth rate (X) £ 1 may indicate a source population, while < 1 may indicate a sink population 68 Table 4 .1 . Habitat attributes at 5 sites for surveys of Catharus frantzii (Ruddy-capped Nightingale Thrush) in 2002 and 2003 (n = 11 transects, 55 sampling points) in the Central Highlands of Chiapas. Habitat attributes were measured in 2003. Sites are: Huitepec (Cerro Huitepec Biological Reserve), CVSJCH (Camino Viejo a San Juan Chamula), Arcotete (Sendero Arcotete), Florecilla (Camino a La Florecilla), Callejon (Sendero Callejon). Dominant Forest Type: old-growth broadleaf (OGB), second-growth broadleaf (SGB), and coniferous - second-growth broadleaf (C-SGB). Disturbed Ratio: proportion of sampling points with disturbance by anthropogenic activities over the points in the site (n = 11) 89 v Table 4.2. Multiple regression analysis examining the effect of habitat attributes on relative abundance of singing Catharus frantzii (Ruddy-capped Nightingale Thrush) at the site and regional levels in the Central Highlands of Chiapas, in 2002-2003. Columns give site and variable attributes, estimated coefficients, standard error (SE), ANOVA-F values, degrees of freedom (df) and probability (P). Generalized Linear Model (GLM-ANOVA) estimates are shown beside each site and for the overall region. Bolded values indicate significant values at a = 0.05 90 vi LIST OF FIGURES Figure 1.1. The Cerro Huitepec Biological Reserve (CHBR) in the Central Highlands of Chiapas, southern Mexico. Chiapas state is shown in the upper left corner. Map modified from Ramirez-Marcial e ta l . (1998) 11 Figure 1.2. Moist and dry-disturbed forests and open habitats in the Cerro Huitepec Biological Reserve, Central Highlands of Chiapas. Map obtained from Ramirez-Marcial et al. (1998). Habitats: GRS-grassland, SHR-brush, SGF-second growth (disturbed) forest, DOF-dry-oak (disturbed) forest, OF-oak-wet forest, and ECF-evergreen cloud forest. Note that no riparian forest habitat is shown on the map. Contour lines are in meters 12 Figure 2.1. (a) Number of territories, (b) pair density (pair per ha), and (c) nest density (nests per ha) from 2000 to 2003 of Catharus frantzii (Ruddy-capped Nightingale Thrush) in moist forest (77 ha, white bars) and dry-disturbed forest (70 ha, black bars) habitat types at the Huitepec Biological Reserve, Central Highlands of Chiapas 41 Figure 2.2. Estimated onset of incuation from fate-known nests in (a) moist forest (n = 42 nests) and (b) dry-disturbed forest (n = 36 nests) of Catharus frantzii (Ruddy-capped Nightingale Thrush) at the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas, from 2000 to 2003. The vertical lines within the quantile boxes represent the sample median and boxes encompass the 25 t h to the 75 t h percentile..... 42 Figure 2.3. Fate of 176 nests of Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Cerro Huitepec Biological Reserve, Central Highlands of Chiapas, from 2000 to 2003. Fates were unknown for 12 nests in the moist forest and 5 in the dry-disturbed forest habitat types 43 Figure 2.4. Biplot of canonical discriminant functions for the habitat attributes at nest-sites and random sites of Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Central Highlands of Chiapas, from 2000 to 2003. Selected variables for both models are listed in Table 2.2. (a) Moist forest (Wilk's X, u = 0.72, F 5 2 0 1 = 15.5, P < 0.001), and (b) Dry-disturbed forest (Wilk's X, u = 0.68, F 3 , 1 4 0 = 21.8, P < 0.001). Internal circles correspond to 95% confidence limits for the group mean, and the external circles contain 50% of the normal contours. Biplot arrows from the grand mean (centroid) show those significant attributes selected by forward-stepwise procedure with P < 0.1 44 Figure 2.5. Daily nest survival rates (± Johnson's SE) for 78 fate-known nests from moist forest (white bars) and dry-disturbed forest (black bars) habitat types for Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Central Highlands of Chiapas, from 2000 to 2003 45 Figure 2.6. Parental activity (a, mean number of trips to the nest per hour ± 1SE) and parental nest attentiveness (b, ratio time per hour on-bouts ± 1SE) during incubating and brooding stages of successful (Moist forest - successful) and failed (Moist forest -failed) nests in the moist forest, and successful (Dry-disturbed forests - successful) and failed (Dry-disturbed forest - failed) nests in the dry-disturbed forest in the Central Highlands of Chiapas, for Catharus frantzii (Ruddy-capped Nightingale Thrush). Nest observations: 18 nests for moist forest and 15 for dry-disturbed forest habitat types 46 vii Figure 3.1. Total number of individuals (N= 221) of Catharus frantzii (Ruddy-capped Nightingale Thrush) banded in the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas each year from 1995 to 2003, by (a) sex, (b) age, and (c) habitat 69 Figure 3.2. Estimates for Catharus frantzii (Ruddy-capped Nightingale Thrush) of (a) adult apparent survival (§) and (b) encounter rates (p) of birds initially banded as adults (males = 93, females = 53, and unknown = 75) at the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas, from 1995 to 2003. Point estimates were derived from the fully time-dependent Cormack-Jolly-Seber (CJS) Model (•kiabitaftiPhabitaft) 70 Figure 3.3. Finite rate of population growth (X) vs. (a) pair density, (b) seasonal productivity (female offspring per pair per season), and (c) apparent survival of Catharus frantzii (Ruddy-capped Nightingale Thrush) at the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas, from 2000 to 2003. Finite rate of population growth (X) is plotted as a residual where the effect of one of the three independent variables in the overall model was removed through regression. Pair density: r = -0.05, F 1 t 4 = 0.79, P = 0.42; season productivity: r= -0.33, 4 = 7.91, P- 0.05; apparent survival: r = 0.93, F, 4 = 310.7, P < 0.001; Model P? = 0.98, F 3 4 = 103.6, P < 0.001 71 Figure 4.1. Box-plots of average relative abundance of singing male Catharus frantzii (Ruddy-capped Nightingale Thrush) across five mixed forest sites during 2002 and 2003 in the Central Highlands of Chiapas. The horizontal line within each box represents the sample median and boxes encompass the 25 t h to the 75 t h percentile. Bars represent the 10 t h and 90 t h percentiles, and the grand horizontal line represents the average abundance of singing males per sampling point = 1.43 ± 0.79. Dominant Forest Type: old-growth broadleaf (OGB), second-growth broadleaf (SGB), and coniferous -second-growth broadleaf (C-SGB) 92 Figure 4.2. Estimates of the total number of survey points (TSP) required for achieving a specific level of precision (p*) for estimating occupancy (IN) and detection probability (p) parameters for Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Central Highlands of Chiapas. Initial parameter values were estimated in MARK. Estimates were based on a standard design of occupancy (MacKenzie and Royle 2005) 93 viii ACKNOWLEDGEMENTS Accomplishment a PhD is only possible through the interaction of experiences and confluence of ideas of many people. First, I would like to thank my supervisor Kathy Martin, from whom I learned critical process in science, education and communication. I also would like to thank my committee advisors Peter Marshall, Bob Elner and Jamie Smith ( t ) , for their confidence and constant support. I thank also Raleigh Robertson, Judy Myers and Bruce Larson for their comments to the final draft of the thesis. I also learned from discussions with faculty members of Forestry and Zoology, and members of the Canadian Wildlife Service, Simon Fraser University and "El Colegio de la Frontera Sur". Interesting discussions happened within the Martin lab group: Nancy, Brett, Lisa, Scott W., Katie, Peter, Scott H., Tanya, Lesley, Kristina, Andrea, Mark, Brad, Susan, Hilary, Kelly, Susan P., Janet, Stephany, Devon, Heather, and Xiang. In the field I had the enthusiasm and dedication of P. Enriquez, T. Will, F. Bolom, E. Pineda, L. Rubio, E. Santiz, M. Hiron, C. Chavez-Zichinelli, J. Fernandez, K. Elliot, D. Bradley, A Licona, G. Chavez, J. and R. Santiz, and N. Rangel in data collection. I also thank M. Martinez-Ico for identifying plant species at Huitepec and elsewhere at "Los Altos". I would like to thank also P. Enrfquez, T. Will, B. Sandercock, N. Mahony, S. Wilson, B. Hassani, K. Hazier and L. Cayuela for their advice in ornithology, population ecology, statistical modelling and landscape ecology, and L. Rubio, C. Martinez and M. Mossop for help with data management and editing. I would like to thank also P. Liedo, M. Rojas, D. Ramos, G. Islebe, A. Moron, E. Naranjo, G. Escalona, and C. Lorenzo from ECOSUR for their kindly support along my studies at UBC. The "Consejo Nacional de Ciencia y Tecnologia" Fellowship program (Grant No. 84108), The "El Colegio de la Frontera Sur" and The University of British Columbia have supported me during my program. The Canadian Wildlife Service-Latin American Program and The Centre for Applied Conservation Research provided funding and equipment for my fieldwork. Pronatura issued a permit for undertaking most of my research at the Cerro Huitepec Biological Reserve, and the communities of San Juan Chamula, La Florecilla and El Callejon also provided permits. ix Finally, I will never forget Silvia, Victor, Gigi, Badre, Nazip, Liu, Juanita, Dafne, Katia, Juarez, Antonio, Adriana, Fernando, Hilda, Francisco, Christelle, Tanya, Eduardo, Isidro and Sandra for making Vancouver a closer place from home. x "The precious in our life are our grandchildren and the far they go" Mixtec anonymous / dedicate my thesis to my partner and colleague Paula, for her love and conscious making possible our stable exhilaration, and to our daughters Natalia Isabel and Paula Rebeca whose love, happiness and patience made this journey enjoyable. I always will owe a debt of gratitude to my mother Rebeca and my grandmother Isabel, whom always made their children's education their priority at any situation and condition. My brother Victor Manuel who went ahead on the trail I followed, and Eduardo and Dolores, whom planted in myself the seed of respect and curiosity for Nature and Humanity. My parents in law, Genaro and Bertha, always have been providing me constant support and encourage for continuing and reaching the goal. To my family's children Fernanda, Genaro, Alejandro, Laura, Celeste and Maria Isabel, and all the children whom will became interested in conservation of nature, with respect and honest curiosity. To Jamie, an outstanding example. xi CHAPTER 1: GENERAL INTRODUCTION Understanding the effects of habitat loss and fragmentation on demographic processes is critical for conserving bird populations. A third of the global avian biodiversity (36% of the species) occurs in the Neotropics (Stotz et al. 1996), and montane forest habitats are key elements because they maintain most of this biodiversity. However, habitat loss and fragmentation have altered the distribution and abundance of avian species from tropical montane forest ecosystems (Watson 2003, Martinez-Morales 2005). Human-induced disturbance in forest ecosystems likely affects demographic rates of birds by reducing habitat quality at different spatial scales (Hanski and Gaggiotti 2004). Thus, the persistence of populations may relate to its response to external threats (Githiru and Lens 2006) and to habitat-based variation in population vital rates (Murphy 2001, Chase et al. 2005). However, information is lacking on population size, survival and productivity for most bird species restricted to tropical montane forests. Catharus frantzii (Ruddy-capped Nightingale Thrush), is a species restricted to montane forests of Mesoamerica (Clement et al. 2000). The aim of my study was to examine how population dynamics of C. frantzii at the local level and abundance across suitable habitats varies with vegetation disturbance in the Central Highlands of Chiapas, Mexico. THE CONSERVATION OF AVIAN DIVERSITY FROM NEOTROPICAL MONTANE FORESTS Abiotic and biotic factors characterizing montane forest ecosystems in the Neotropics promote a rich but threatened avian diversity. First, the accumulation of new species throughout adjacent habitats (i.e., Fj-beta- species richness) is extremely high and contrasts with the relatively low degree of species overlap among locations (i.e., 5-gamma- species richness; Watson and Peterson 1999, Weir 2006). Second, there are more plant-bird species interactions, resulting in potentially greater speciation and endemism compared to lowland forests (Rejinfo et al. 1997). Many understory plant species depend on birds for pollination and seed dispersal (Hamilton et al. 1995). For instance, the montane forests of Mesoamerica (i.e., tropical Mexico and Central 1 America) contain more endemic bird species than the lowland forests (Peterson et al. 1998, Watson and Peterson 1999). Furthermore, forest loss across the Neotropics is greater in areas containing a high ratio of endemism and a number of habitat-restricted species than outside those areas (Manne and Pimm 2001). Finally, the presence of wind-driven cloud moisture is a unique hydrologic process among terrestrial ecosystems (Still et al. 1999). Increased moisture levels in tropical systems are associated with high species richness (Newton 2003). The primary causes of global species declines are loss and fragmentation of habitats. The montane forests in Mesoamerica have been subjected to these processes for more than ten centuries, and currently are considered among the world's most threatened ecosystems (Myers et al. 2000, Kappelle and Brown 2001). In the Central Highlands of Chiapas, forest cover has been reduced gradually to the extent that original, old growth forest tracts occur on relatively remote mountains. Even in these inaccessible sites, local people selectively harvest the forest in their daily search for firewood, timber and food (Gonzalez-Espinosa et al. 1995, Ramfrez-Marcial et al. 2001). Here, cloud and humid pine-oak forests have been reduced to less than 25 percent of their original area, and plant species richness has declined from 50 to 75 percent (Gonzalez-Espinosa et al. 1995, Quintana-Ascencio et al. 2004). Selective logging has changed tree species composition by increasing Pinus spp. throughout the region. In these modified stands, most understory trees from old-growth broadleaf forests are absent (Camacho-Cruz et al. 2000, Ochoa-Gaona et al. 2004). With the simplification of forest composition and structure, formerly shaded forest interiors become drier, brighter, and warmer, with resultant micro-environmental changes. The effects of habitat disturbance on wildlife species in the Central Highlands of Chiapas are poorly known. Current land-use patterns in the region may affect the availability of resources used by birds from old-growth forests, particularly for those species depending on understory plants (Gonzalez-Espinosa et al. 1995). Reduction in understory plant cover may result in decreased food resources, nest substrates and concealment from predators (Martin 1996, Martin et al. 2000). These negative effects of reduced understory foliage may persist for decades. Conservation guidelines for birds cannot be effective unless we identify the attributes determining 2 habitat suitability, which in turn influence life-history traits and population vital rates. This information is much needed for the conservation of tropical montane forest birds. VITAL RATES AND LIFE-HISTORY TRAITS OF TROPICAL MONTANE FOREST BIRDS Life-history theory predicts that breeding birds at low latitud and high elevations are characterised by lower reproductive rates, compensated by higher longevity, compared with breeding birds from highe latitud and lower elevations (Badyaev and Ghalambor 2001, Martin 2001). Thus, for tropical montane forest birds we might expect a low number of offspring and high adult survival. Conversely, high productivity (number of female offspring per breeding adult female per year) has been hypothesized for bird populations from isolated ecosystems, through rapid re-nesting attempts as a mechanism for increasing reproductive success (Sieving and Karr 1997, Grzybowski and Pease 2005). Hence, increasing reproductive success through re-nesting after a failed nest attempt may be a crucial life-history trait characterizing birds from tropical montane forests. Population fluctuations are caused by changes in vital rates (i.e. inputs by birth and immigrants, and outputs by death and emigrants; Hastings 1997, Sibly et al. 2003). Most research on the breeding performance in birds relies on estimates of nesting success (i.e., the probability a nest survives the nesting period and produces at least one fledgling), as a surrogate of the breeding season productivity (Farnsworth et al. 2000). Productivity, the number of female offspring surviving per reproductive female, varies in space and time (Sherry and Holmes 1995, Holmes et al. 1996, Martin 1996). However, estimation of productivity is often obscured by the difficulties of following individual female productivity for the breeding season (Nagy and Holmes 2004, Grzybowski and Pease 2005), or by offspring dispersal (Newton 1998, Sibbly et al. 2003, Kawecki 2004). Nesting success of tropical forest birds varies from 9 percent in lowland forests to more than 60 percent in montane cloud forests (Willis 1974, Skutch 1985, Morton and Stutchbury 2000, Robinson et al. 2000). This inter-specific variation may not support the hypothesis that low nesting success selects for small clutch sizes in tropical species (Martin 1996, Martin et al. 1997). 3 The nesting success of forest birds may be further reduced by habitat disturbance, because disturbance may reduce food and nest sites, and increases the risk of predation (Burke and Nol 2000, Zanette and Jenkins 2000, Easton and Martin 2002). Food is considered to be an important limiting factor for bird populations during both breeding and non-breeding periods (Martin 1987, Newton 2004). Nest type, location, and concealment may influence nest predation rates (Gotmark et al. 1995, Martin 1995). Nest predation may strongly interact with food limitation to influence bird populations (Badyaev and Ghalambor2001, Martin et al. 2006). Skutch (1949) suggested that parents might alter nest visitation to reduce predation. Nest predation may be reduced experimentally when food is added and the number of neighbours reduced (Rodenhouse et al. 2003, Sillett et al. 2004). Hence, food, nest predation and nest-site attributes may interact to determine avian productivity (Martin 2004). Data on these interactions do not exist for tropical montane forest birds. Annual survival is also a key trait for the evaluation of population persistence. However, few analyses of survival rates of bird species from the Neotropics have been conducted to evaluate latitudinal variation in survival (Ricklefs 1997). Estimates of survivorship for passerine forest birds from Panama and the eastern United States showed little difference in the median local survival (Karr e ta l . 1990). The local survival of the Green-rumped Parrotlet (Fropus passerinus) in Venezuela was comparable to local survival of small cavity nesters in temperate forests (Sandercock et al. 2000). However, in comparative studies of robins and thrushes, estimated annual survival rates increased from <|> - 0.56 in temperate populations to <> = 0.68 in the subtropics to <> = 0.85 in montane cloud forest populations (Johnston et al. 1997, Ricklefs 1997). Thus, high annual survival rates for bird populations in tropical montane forests may represent a high survival life-history strategy (Saether and Bakke 2000). Populations of some species are particularly sensitive to demographic stochasticity (i.e., variation in population growth rate arising from individual productivity and survival) and/or environmental stochasticity. These populations are more likely to be locally extirpated (Simberloff 1994, Ovaskainen and Hanski 2003). Below a certain population size, stochastic factors can be important to persistence (Holsinger 2000). The relative contributions of these factors to variation in 4 population growth can only be indirectly measured by understanding the environmental effects on the finite rate of population growth (X; i.e., birth, death, immigration and emigration; Sibly et al. 2003, Anders and Marshall 2005). Generally population persistence may be more related to its responses to habitat-specific vital rates, environmental stochasticity, and external threats such as habitat loss and degradation, than the population size (Stacey and Taper 1992, Murphy 2001). POPULATION DYNAMICS AND HABITAT QUALITY Population dynamics can be evaluated by measuring the per capita productivity and survival of individuals occupying a particular habitat (Martin 1998, Martin and Joron 2003). Assessments of habitat quality, often defined by the density of individuals, have been described for many species, but models predicting abundance of target species for other areas often perform poorly (Hanski and Ovaskainen 2003). Environmental factors that may determine habitat quality (e.g., microclimate, abundance of conspecifics, food cover and predators) interact with morphological, behavioural and life-history traits to influence vital rates of populations (Badyaev and Ghalambor 2001, Martin 2001, Carrete et al. 2005, Ferretti et al. 2005). Source-sink models have been developed to evaluate differences in population's vital rates from habitats differing in quality (Dias 1996, Anders and Marshall 2005). The source-sink metapopulation model (Murphy 2001, Kawecki 2004, Jonzen et al. 2005) applies to heterogeneous environments where the finite rate of population growth of targeted species is stable or increasing (X > 1) in source habitats, but decreasing (X < 1) in sink habitats (Wiens and Rotemberry 1981, Pulliam 1988). Populations from sink habitats are unable to persist in the absence of immigration (Pulliam 1996). Determining that a population exist in a source habitat may be difficult because these habitats may be less common compared to sink habitats (Jonzen et al. 2005), and dispersal of individuals into source populations might produce density-dependent reductions in vital rates that characterize "relative-s/'n/c" populations (Watkinson and Sutherland 1995, Kawecki 2004). Although Source-sink population structure results from differences in habitat quality and dispersing individuals are assumed generally to move from source to sink, reverse dispersal may be possible. Temporal variation in habitat quality may invert the 5 source-sink dynamics, and reverse dispersal may result in lower quality habitats becoming temporally sources, whereas higher quality habitats may act at the same time as sinks (Kawecki 2004). Competition has been suggested as an ecological mechanism that generates source-sink dynamics (Opdam and Wiens 2002). If social status interferes with habitat choice by either ideal-despotic (Fretwell and Lucas 1970) or pre-emptive habitat selection (Pulliam and Danielson 1991, Rosenzwig 1991, Greene and Stamps 2001), sink habitats are occupied because individuals in source habitats control access to resources and thus force less dominant individuals into sink habitats. Critical predictions of the source-sink model are that individual fitness varies among habitats differing in quality, and those individuals from source habitats will perform better than individuals from sink habitats (Jonzen et al. 2005). Habitat-specific vital rates and dispersal may strongly influence persistence of populations with source-sink structure and dynamics (With and King 2001, Kawecki 2004). Hence, data on habitat-specific vital rates may provide a clear index of habitat quality, and may allow identification of good and poor habitats. THREATS TO TROPICAL MONTANE FOREST BIRDS Loss and fragmentation of forest habitats, a primary cause of bird species declines in the tropics, can lead to several consequences, such as increased edge effects, decreased connectivity and reduced habitat heterogeneity (Opdam and Wiens 2002). Modifying habitats can alter the distribution and abundance of plant and animal species, including predators and invasive species. Changes in environmental factors resulting from habitat alteration can decrease reproductive success and increase mortality rates in birds (Newton 1998), leading to population declines, and extinction (Sieving and Karr 1997, Saether and Engen 2003). Habitat suitability and nest predation are important factors that can drive fluctuations in bird population growth rates. Predation accounts for the largest share of nest failure resulting in low and variable reproductive success in most temperate (Nagy and Holmes 2004, Chase et al. 2005) and tropical forest birds in North America (Arango-Velez and Kattan 1997, Morton and Stutchbury 2000, Robinson et al. 2000). Distinct types or different abundances of predators may account for variable 6 predation rates among different habitat types (Remes 2005, Mahon and Martin 2006, Schmidt et al. 2006). It is important to identify conditions under which predation is of critical concern for tropical montane forest birds. Avian populations of habitat-restricted species may be particularly susceptible to changes in habitat suitability. THE STUDY SPECIES - CATHARUS FRANTZII (RUDDY-CAPPED NIGHTINGALE THRUSH) Catharus frantzii (Ruddy-Capped Nightingale Thrush or "Zorzalito de Montana") belongs to the family Emberizidae and Sub-family Turninae (American Ornithologists' Union 1998). C. frantzii is a sedentary, discontinuously distributed, and secretive inhabitant of the understory and dense undergrowth of tropical montane forests from Western Mexico to Panama (Howell and Webb 1995, Clement et al. 2000). The species has been listed as endangered on the Mexican Red List due to the loss and conversion of tropical montane forests over its geographic range in Mexico. However, there has not been an evaluation of the population status and there is no understanding how habitat variation affects population dynamics (Rangel-Salazar et al. 2006). Variation in abundance of C. frantzii over its range has been reported. Density estimation by point counts of C. frantzii in cloud forests of Central Mexico was 1.3 individuals per square km, with the species reported only in cloud forest remnants over 50 square km (Martinez-Morales 2001). In the cloud forests of Western Costa Rica, C. frantzii is reported to be fairly common (Young and MacDonald 2000). Abundance of C. frantzii may be correlated with habitat remaining and degree of coexistence with other Nightingale Catharus species (Raitt and Hardy 1970, Watson 2003, Tejeda-Cruz and Sutherland 2004). THESIS OBJECTIVES In this study, I examined the demographic rates of Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Central Highlands of Chiapas, Southern Mexico, a threatened listed species restricted to montane forests. 7 My objectives were: 1) To examine the reproductive ecology, including nest-site selection, nesting success, nest survival and parental care, and correlate habitat attributes at three spatial levels: nest, nest-site and nest-patch; 2) To estimate productivity and survival, and correlate those to the finite rate of population growth rate and population stability; 3) To determine whether the Cerro Huitepec Biological Reserve supports a persistent populations of C. frantzii; 4) To evaluate habitat attributes associated with the distribution and abundance by estimating occupancy and detection probabilities at the Huitepec Biological Reserve and several nearby sites; I conducted this study from 2000 to 2003 in the Central Highlands of Chiapas, Southern Mexico (Figure 1). Most of my research was conducted in the Cerro Huitepec Ecological Reserve ("Estacion Biologica Cerro Huitepec", hereafter Huitepec Reserve) and in four additional sites around the City of San Cristobal de las Casas, Chiapas. The area is located in the humid pine-oak forests of Mesoamerica, which are dominated by Pinus, Quercus, Liquidambar, Alnus and other tree genera of Nearctic affinities (Gonzalez-Espinosa et al. 2004). The vegetation in the area is typically associated with high rainfall (>1000 mm annually) and cool-temperate climates (10-17 °C). I selected the Huitepec reserve because it supports a population of C. frantzii in habitats differing in vegetation composition and structure. Variation in vegetation allowed me to explore the habitat-specific demography of C. frantzii at the local and regional levels. STUDY AREA The Huitepec reserve is a semi-isolated forest in the Central Highlands of Chiapas, 4.5 km west of San Cristobal de las Casas, Chiapas (Figure 1.1). It is located on the east, north and west sides of the Huitepec Volcano (16°44'38"N; 92°40'15"W). This reserve was designated as private land in 1987 (owned and managed by PRONATURA, an environmental non-governmental organization in Mexico). The reserve is 137 ha in size and maintains a variety of natural and disturbed habitats ranging in elevation from 2230 m to 2710 m. I included also 10 ha of disturbed 8 dry-oak forest adjacent to the east side of the reserve, and, thus, increased the study area to 147 ha. The mean annual temperature is 14.5°C and the mean annual precipitation is 1300 mm. The Huitepec reserve contains 315 vascular plant species in six plant communities: grassland, shrub-land, second growth forest, dry oak forest, moist oak forest and evergreen cloud forest (Ramfrez-Marcial et al. 1998). The reserve includes 32% of the regional plant species richness above 2000 m, but many plant species occur in low densities and many of them may have suffered local extirpation. Firewood gathering occurs frequently in the reserve, particularly in the second growth and dry oak forests. This practice generates openings in both the canopy and the understory. The land surrounding the reserve is used for housing, agriculture and cultivation of exotic pine stands; thus, unsuitable secondary habitats around the reserve have increased. A detailed description of the vegetation across forest habitat types within the reserve is presented in Chapter 2. OVERVIEW OF THE THESIS In this Chapter, I introduce patterns of avian diversity of tropical montane forests and the potential effects of habitat loss and disturbance on avian populations. I then present the theoretical context relating habitat-specific demography and the source-sink population model. Also, I describe the biology of Catharus frantzii (Ruddy-capped Nightingale Thrush) and discuss my selection of this species in a semi-isolated reserve in the Central Highlands of Chiapas as a good model for understanding the variation in factors driving population stability. In Chapter 2,1 examine the influence of habitat, nest-site selection and parental behaviour in nesting success and daily nest survival of C. frantzii in both moist and dry-disturbed forests. I present data on breeding pair-density, nesting fate and daily nest survival, and identify habitat attributes at three spatial scales (i.e., nest, nest-site and nest-patch). I also present data on parental visitation rates and female nest attentiveness during the incubating and brooding periods. In Chapter 3, I explore variation in density, productivity and survival between habitat types by combining data from the reproductive success presented in Chapter 2, with survival estimates to evaluate how these parameters influence the finite rate of population growth (A) in C. frantzii. I evaluated whether survival varied across sex, age and habitat type. I also explore if variation in 9 population parameters is correlated with habitat-specific population growth, and whether populations of C. frantzii a\ the Huitepec reserve were stable. In Chapter 4,1 measure variation in abundance and habitat attributes at Huitepec and at several nearby sites where C. frantzii were present. I model occupancy [UJ] and detection probability (p), and correlated these parameters with site, year, and number of visits. Then, I assessed whether abundance varied with habitat attributes and forest types (from old-growth humid forest to coniferous-broadleaf disturbed forest). In Chapter 5,1 summarize my conclusions and discuss implications for the population stability and persistence of C. frantzii in the Central Highlands of Chiapas. I also comment on the strengths and weaknesses of my research findings and suggest future research for the management and conservation of bird populations, as well as make specific recommendations for the status of C. frantzii in Mexico. I present Chapters 2, 3 and 4 as stand-alone papers. 10 9 4 ° 0 0 ' 9 1 ° 2 0 ' 9 2 ° 4 5 ' 9 2 ° 4 0 ' 9 2 ° 3 5 ' Figure 1.1. The Cerro Huitepec Biological Reserve (CHBR) in the Central Highlands of Chiapas, southern Mexico. Chiapas state is shown in the upper left corner. Map modified from Ramirez-Marcial et al. (1998). 11 F i g u r e 1.2. Moist and dry-disturbed forests and open habitats in the Cerro Huitepec Biological Reserve, Central Highlands of Chiapas. Map obtained from Ramirez-Marcial et al. (1998). Habitats: GRS-grassland, SHR-brush, SGF-second growth (disturbed) forest, DOF-dry-oak (disturbed) forest, OF-oak-wet forest, and ECF-evergreen cloud forest. Note that no riparian forest habitat is shown on the map. Contour lines represent altitude in meters. 12 R E F E R E N C E S American Ornithologists' Union. 1998. Check-list of North American Birds, 7th. Edition. American Ornithologists' Union, Washington, DC. Anders, A. D., and M. R. Marshall. 2005. Increasing the accuracy of productivity and survival estimates in assessing landbird population status. Conservation Biology 19:66-74. Arango-Velez, N., and G. H. Kattan. 1997. Effects of forest fragmentation on experimental nest predation in Andean cloud forest. Biological Conservation 81:137-143. Badyaev, A. V., and C. K. Ghalambor. 2001. Evolution of life histories along elevational gradients: trade-off between parental care and fecundity. Ecology 82:2948-2960. Burke, D. M., and E. Nol. 2000. Landscape and fragment size effects on reproductive success of forest-breeding birds in Ontario. Ecological Applications 10:1749-1761. Camacho-Cruz, A., M. Gonzalez-Espinosa, J. H. Wolf, and B. D. Jong. 2000. Germination and survival of tree species in disturbed forests of the Highlands of Chiapas, Mexico. Canadian Journal of Botany 78:1309-1318. Carrete, M., J. A. Sanchez-Zapata, J. F. Calvo, and R. Lande. 2005. Demography and habitat availability in territorial occupancy of two competing species. Oikos 108:125-136. Chase, M. K., N. Nur, and G. R. Geupel. 2005. Effects of weather and population density on reproductive success and population dynamics in a Song Sparrow (Melospiza melodia) population: a long-term study. Auk 122:571-592. Clement, P., R. Hathway, C. Byers, and J. Wilczur. 2000. Thrushes. Christopher Helm Ltd., London, UK. Dias, P. C. 1996. Sources and sinks in population biology. Trends in Ecology and Evolution 8:326-330. Easton, W. E., and K. Martin. 2002. Effects of thinning and herbicide treatment on nest-site selection by songbirds in young managed forests. Auk 119:685-694. Farnsworth, G. L., K. C. Weeks, and T. R. Simons. 2000. Validating the assumptions of the Mayfield method. Journal of Field Ornithology 71:658-664. Ferretti, V., P. E. Llambias, and T. E. Martin. 2005. Life-history variation of a neotropical thrush challenges food limitation theory. Proceedings of the Royal Society. London B 272:769-773. Fretwell, S. D., and H. L. Lucas. 1970. On the territorial behaviour and other factors influencing habitat distribution in birds. I. Theoretical development. Acta Biotheoretica 19:16-36. Githiru, M., and L. Lens. 2006. Demography of an Afrotropical passerine in a highly fragmented landscape. Animal Conservation 9:21-27. Gonzalez-Espinosa, M., S. Ochoa-Gaona, N. Ramfrez-Marcial, and P. Quintana-Ascencio. 1995. Current land use trends and conservation of old growth forest habitats in the highlands of 13 Chiapas, Mexico. Pages 190-197 in M. H. Willson and S. A. Sader, editors. Conservation of Neotropical Migratory Birds in Mexico. Maine Agricultural and Forest Experimentation Station Misc., Maine. Gonzalez-Espinosa, M., J. M. Rey-Benayas, N. Ramirez-Marcial, M. A. Huston, and D. Golicher. 2004. Tree diversity in the northern Neotropics: regional patterns in highly diverse Chiapas. Ecography 27:741 -756. Greene, C. M., and J. S. Stamps. 2001. Habitat selection in low population densities. Ecology 82:2091-2100. Grzybowski, J. A., and C. M. Pease. 2005. Renesting determines seasonal fecundity in songbirds: what do we know? what should we assume? Auk 122:280-291. Gotmark, F., D. Blomqvist, O. Johansson, and J. Bergkvist. 1995. Nest site selection: a trade-off between concealment and view of the surroundings? Journal of Avian Biology 26:305-312. Hamilton, L. S., J. O. Juvik, and F. N. Scatene, editors. 1995. Tropical Montane Cloud Forest. Spring-Verlag, New York, NY. Hanski, I., and O. E. Gaggiotti. 2004. Ecology, Genetics, and Evolution of Metapopulations. Elsevier Academic Press, Burlington, MA. Hanski, I., and O. Ovaskainen. 2003. Metapopulation theory for fragmented landscapes. Theoretical Population Biology 64:119-127. Hastings, A. 1997. Population Biology: Concepts and Models. Springer, New York. Holmes, R. T., P. P. Marra, and T. W. Sherry. 1996. Habitat-specific demography of breeding Black-Throated Blue Warblers (Dendroica caerulescens): implications for population dynamics. Journal of Animal Ecology 65:183-195. Holsinger, K. E. 2000. Demography and extinction in small poulations. Pages 55-74 in A. G. Young and G. M. Clarke, editors. Genetics, Demography and Viability of Fragmented Populations. Cambridge University Press, Cambridge, UK. Howell, S., and S. Webb. 1995. A Guide to the Birds of Mexico and Northern Central America. Oxford University Press, Oxford, UK. Johnston, J. P., W. J. Peach, R. D. Gregory, and S. A. White. 1997. Survival rates of tropical and temperate passerines: a trinidadian perspective. The American Naturalist 150:771-789. Jonzen, N., J. R. Rhodes, and H. P. Possingham. 2005. Trend detection in source-sink systems: when should sink habitats be monitored? Ecological Applications 15:326-334. Kappelle, M., and A. D. Brown, editors. 2001. Bosques Nublados del Neotropico. INBIO, San Jose, Costa Rica. Karr, J. R., J. D. Nichols, M. Katheleen, Klimkiewicz, and J. D. Brawn. 1990. Survival rates of birds of tropical and temperate forests: will the dogma survive? The American Naturalist 136:277-291. 14 Kawecki, T. J. 2004. Ecological and evolutionary consequences of source-sink population dynamics. Pages 387-411 in I. Hanski and O. E. Gaggiotti, editors. Ecology, Genetics, and Evolution of Metapopulations. Elsevier Academic Press, Burlington, MA. Mahon, L. C , and K. Martin. 2006. Nest survival of chickadees in managed forests: habitat, predator, and year effect. Journal of Wildlife Management In press. Manne, L. L , and S. L. Pimm. 2001. Beyond eight forms of rarity: which species are threatened and which will be the next? Animal Conservation 4:221-229. Martin, J . -L, and M. Joron. 2003. Nest predation in forest birds: influence of predator type and predator's habitat quality. Oikos 102:641 -653. Martin, K. 2001. Wildlife in alpine and sub-alpine habitats. Pages 285-310 in D. H. Johnson and T. A. O'Neil, editors. Wildlife-habitat relationships in Oregon and Washington. Oregon State University Press, Corvallis, OR. Martin, K., P. B. Stacey, and C. E. Braunn. 1997. Demographic rescue and maintenance of population stability in grouse - beyond metapopulations. Wildlife Biology 3:4:295-296. Martin, T. E. 1987. Food limitation in birds: a life-history perspective. Ann. Rev. Ecol. Syst. 18:453-487. Martin, T. E. 1995. Avian life-history evolution in relation to nest sites, nest predation and food. Ecological Monographs 65:101-127. Martin, T. E. 1996. Life-history evolution in tropical and south temperate birds: what do we really know? Journal of Avian Biology 27:263-272. Martin, T. E. 1998. Are microhabitat preferences of coexisting species under selection and adaptive? Ecology 79:656-670. Martin, T. E. 2004. Avian life-history evolution has an eminent past: does it have a bright future? Auk 121:289-301. Martin, T. E., R. D. Bassar, S. K. Bassar, J. J . Fontaine, and P. Lloyd. 2006. Life-history and ecological correlates of geographic variation in egg and clutch mass among passerine species. Evolution 60:390-398. Martin, T. E., J. Scott, and C. Menge. 2000. Nest predation increases with parental activity: separating nest site and parental activity effects. Proceedings of the Royal Society. London B 267:2287-2293. Martinez-Morales, M. A. 2001. Forest fragmentation effects on bird communities of montane cloud forest in eastern Mexico. University of Cambridge, Cambridge, UK. Martinez-Morales, M. A. 2005. Landscape patterns influencing bird assemblage in fragmented Neotropical cloud forests. Biological Conservation 121:117-126. Morton, E. S., and B. J. Stutchbury. 2000. Demography and reproductive success in the Dusky Antbird, a sedentary tropical passerine. Journal of Field Ornithology 71:493-500. 15 Murphy, M. T. 2001. Habitat-specific demography of a long-distance, neotropical migrant bird, the Eastern Kingbird. Ecology 82:1304-1318. Myers, N., R. Mittermeier, C. Mittermeier, G. da Fonseca, and J. Kent. 2000. Biodiversity hot spots for conservation priorities. Nature 403:853-858. Nagy, L. R., and R. T. Holmes. 2004. Factors influencing fecundity in migratory songbirds: is nest predation the most important? Journal of Avian Biology 35:487-491. Newton, 1.1998. Population Limitation in Birds. Academic Press, London, UK. Newton, I. 2003. The Speciation and Biogeography of Birds. Academic Press, London, UK. Newton, I. 2004. Population limitation in migrants. Ibis 146:197-226. Ochoa-Gaona, S., M. Gonzalez-Espinosa, A. A. Meave, and V. Sorani-DalBon. 2004. Effect of forest fragmentation on the woody flora of the highlands of Chiapas, Mexico. Biodiversity and Conservation 13:867-884. Opdam, P., and J. A. Wiens. 2002. Fragmentation, habitat loss and landscape management. Pages 202-223 in K. Norris and D. J. Pain, editors. Conserving Bird Biodiversity: General Principles and their Application. Cambridge University Press, Cambridge, UK. Ovaskainen, O., and I. Hanski. 2003. How much does and individual habitat fragment contribute to metapopulation dynamics and persistence? Theoretical Population Biology 64:481-495. Peterson, A. T., G. Escalona-Segura, and J. A. Griffith. 1998. Distribution and conservation of birds of Northern Central America. Wilson Bulletin 110:534-543. Pulliam, H. R. 1988. Sources, sinks, and population regulation. The American Naturalist 132:652-661. Pulliam, H. R. 1996. Sources and sinks: empirical evidence and population consequences. Pages 45-69 in O. E. Rhodes, R. K. Chesser, and M. H. Smith, editors. Population Dynamics in Ecological Space and Time. The University of Chicago Press, Chicago, IL. Pulliam, H. R., and D. J. Danielson. 1991. Sources, sinks, and habitat selection: a landscape perspective on population dynamics. The American Naturalist 137 (Supplement):S50-S66. Quintana-Ascencio, P. F., N. Ramfrez-Marcial, M. Gonzalez-Espinosa, and M. Martinez-lco. 2004. Sapling survival and growth of coniferous and broad-leaved trees in successional highland habitats in Mexico. Applied Vegetation Science 7:81-88. Raitt, R. J., and J. W. Hardy. 1970. Relationships between two partially sympatric species of thrushes (Catharus) in Mexico. Auk 87:20-57. Ramirez-Marcial, N., M. Gonzalez-Espinosa, and G. Williams-Lineara. 2001. Anthropogenic disturbance and tree diversity in Montane Rain Forest in Chiapas, Mexico. Forest Ecology and Management 154:311-326. Ramfrez-Marcial, N., S. Ochoa-Gaona, M. Gonzalez-Espinosa, and P. Quintana-Ascencio. 1998. Analisis florfstico y sucesional en la Estacion Biologica Cerro Huitepec, Chiapas, Mexico. Acta Botanica Mexicana 44:59-85. 16 Rangel-Salazar, J. L , E. Santana-Castellon, J. Carrillo, and P. L. Enrfquez-Rocha. 2006. Zorzal de Frantzius (Catharus frantzii). in P. Escalante-Pliego, D. Ayala, and V. Nequiz, editors. Libro Rojo de las Aves de Mexico, Mexico D.F. In Press. Rejinfo, L. M., G. P. Sevat, J. M. Goerk, B. A. Loiselle, and J. G. Blake. 1997. Patterns of species composition and endemism in the northen Neotropics: a case for conservation of montane avifaunas. Orntihological Monographs 48:577-594. Remes, V. 2005. Birds and rodents destroy different nests: a study of Blackcap Sylvia atricapilla using the removal of nest concealment. Ibis 147:213-216. Ricklefs, R. E. 1997. Comparative demography of the new world population of thrushes (Turdus spp.). Ecological Monographs 67:23-43. Robinson, W. D., T. R. Robinson, S. K. Robinson, and J. D. Brawn. 2000. Nesting success of understory forest birds in central Panama. Journal of Avian Biology 31:151-164. Rodenhouse, N. L , T. S. Sillett, P. J. Doran, and R. T. Holmes. 2003. Multiple density-dependence mechanisms regulate a migratory bird population during the breeding season. Proceedings of the Royal Society. London B 270:2105-2110. Rosenzwig, M. L. 1991. Habitat selection and population interactions: the search for mechanism. The American Naturalist 137:S5-S28. Sandercock, B. K., S. R. Beissinger, S. H. Stoleson, R. R. Melland, and C. R. Hughes. 2000. Survival rates of a neotropical parrot: implications for latitudinal comparisons of avian demography. Ecology 81:1351-1370. Schmidt, K. A., R. S. Ostfeld, and K. N. Smyth. 2006. Spatial heterogeneity in predator activity, nest survivorship, and nest-site selection in two forest thrushes. Oecologia 148:22-29. Sherry, T. W., and R. T. Holmes. 1995. Summer versus winter limitation of populations: what are the issues and what the evidence? Pages 85-120 in T. E. Martin and D. M. Finch, editors. Ecology and Management of Neotropical Migratory Birds. Oxford University Press, New York, NY. Sibly, R. M., J. Hone, and T. H. Clutton-Brock. 2003. Introduction to wildlife population growth rates. Pages 1-10 in R. M. Sibly, J. Hone, and T. H. Clutton-Brock, editors. Wildlife Population Growth Rates. The Royal Society, Cambridge, UK. Sibly, R. M., J. Hone, and T. H. Clutton-Brock, editors. 2003. Wildlife Population Growth Rates. Cambridge University Press, Cambridge, UK. Sieving, K. E., and J. R. Karr. 1997. Evian extinction and persistence mechanisms in lowland Panama. Pages 156-170 in R. O. Bierregaard and W. F. Laurance, editors. Tropical Forest Remnants: Their Ecology and Management. Chicago Press, Chicago, IL. Sillett, T. S., N. L. Rodenhouse, and R. T. Holmes. 2004. Experimentally reducing neighbour density affects reproduction and behavior of a migratory songbird. Ecology 85:2467-2477. Simberloff, D. 1994. Habitat fragmentation and population extinction of birds. Ibis 137:S105-S111. 17 Skutch, A. F. 1949. Do tropical birds rear as many young as they can nourish? Ibis 91:430-455. Skutch, A. F. 1985. Clutch size, nesting success, and predation on nests of Neotropical birds, reviewed. Ornithological Monographs 36:575-594. Stacey, P. B., and M. Taper. 1992. Environmental variation and the persistence of small populations. Ecological Applications 2:18-29. Still, C. J., P. N. Foster, and S. H. Schneider. 1999. Simulating the effects of climate change on tropical montane cloud forests. Nature 398:608-610. Stotz, D. F., T. A. Parker III, and J. W. Fitzpatrick. 1996. Neotropical Birds: Ecology and Conservation. University of Chicago Press, Chicago. Saether, B.-E., and 0 . Bakke. 2000. Avian life history variation and contribution of demographic traits to the population growth rate. Ecology 81:642-653. Tejeda-Cruz, C , and W. J. Sutherland. 2004. Bird responses to shade coffee production. Animal Conservation 7:169-179. Watkinson, A. R., and W. J. Sutherland. 1995. Sources, sinks and pseudo-sinks. Journal of Animal Ecology 64:126-130. Watson, D. M. 2003. Long-term consequences of habitat fragmentation-highland birds in Oaxaca, Mexico. Biological Conservation 111:283-303. Watson, D. M., and A. T. Peterson. 1999. Determinants of diversity in a naturally fragmented landscape: humid montane forest avifaunas of Mesoamerica. Ecography 22:582-589. Weir, J. T. 2006. Divergent timing and patterns of species accumulation in lowland and highland neotropical birds. Evolution 60:842-855. Wiens, J. A., and J. T. Rotemberry. 1981. Habitat associations and community structure of birds in shrub-steppe environments. Ecological Monographs 51:21-41. Willis, E. O. 1974. Populations and local extinction of birds on Barro Colorado Island, Panama. Ecological Monographs 44:153-169. With, K. A., and A. W. King. 2001. Analysis of landscape sources and sinks: the effect of spatial pattern on avian demography. Biological Conservation 100:75-88. Young, B. E., and D. B. MacDonald. 2000. Birds. Pages 179-222 in N. M. Nadkarni and N. T. Wheelwright, editors. Monteverde: Ecology and Conservation of a Tropical Cloud Forest. Oxford University Press, New York, NY. Zanette, L., and B. Jenkins. 2000. Nesting success and nest predators in forest fragments: a study using real and artificial nests. Auk 117:445-454. 18 CHAPTER 2: INFLUENCES OF HABITAT VARIATION, NEST-SITE SELECTION AND PARENTAL BEHAVIOUR ON THE BREEDING SUCCESS OF CATHARUS FRANTZII (RUDDY-CAPPED NIGHTINGALE THRUSH) IN CHIAPAS, MEXICO INTRODUCTION Breeding success is an important component for understanding fluctuations in avian populations. The influence of factors on the breeding performance of birds has gained increasing attention because of the effects of habitat loss (Laurance and Bierregaard 1997, Opdam and Wiens 2002) and global climate change (Still et al. 1999, Martin 2001). Currently, the loss of tropical forests is considered the most important force of modern extinctions (Simberloff 1994, Harris and Pimm 2004). At the same time, the proximate causes driving population fluctuations and life history characteristics of tropical forest birds are poorly understood. Accordingly, identifying the effects of factors influencing the breeding performance of tropical forest birds represents a main concern in avian ecology and conservation. Habitats vary in quality across heterogeneous landscapes. Although, birds may breed in these habitats, they may not be reproducing at sustainable levels in all of them (Robinson et al. 1995, Burke and Nol 2000, Donovan and Lamberson 2001). Consequently, given the influence of habitat quality on the breeding performance of birds, individuals benefit by recognizing and settling accordingly in those habitats that represent high breeding performance. The ideal-despotic model (Fretwell and Lucas 1970, Wiens 1976), predicts that best quality individuals (older or dominant) select the best territories and sites for nesting, and through territorial behaviour, relegate lower quality individuals (younger or subordinate) to less suitable territories and nesting sites. Due to habitat heterogeneity, the progressive occupation of low quality territories as breeding density increases may cause a decline of the mean individual fecundity (i.e., density-dependent fecundity). This process has been shown to regulate populations of territorial species (Sinclair 1989, Arcese et al. 1992, Newton 2004). However, this mechanism of population regulation in avian population is still in debate (Rodenhouse et al. 2003). The combination of habitat heterogeneity, despotic 19 settlement, and density-dependence fecundity has been called site-dependent population regulation (Rodenhouse et al. 1997). Density-dependence and site-dependent population regulation predict that the progressive occupation of low quality territories as breeding density increases causes a decline in the mean per capita fecundity of a population (Murphy 2001, Sergio and Newton 2003, Chase et al. 2005). In habitats with high risk of nest predation, parental behaviour can facilitate understanding of the process of nest-site selection (Badyaev and Ghalambor 2001, Martin 2004, Weidinger 2004). The nest predation and parental visitation hypothesis (Skutch 1949) predicts that nesting success will decrease with increasing parental visitation because it increases the likelihood of detection by predators. Nest predation has been identified as the most important proximate cause of nest failure in forest birds (Martin et al. 2000, Martin 2004, Weidinger 2004). Nonetheless, birds may adopt anti-predation strategies to reduce the likelihood of a nest being found by predators, such as placing nests in sites not accessible to predators, constructing cryptic nests in thick vegetation, and maintaining cryptic behaviour and coloration (Gotmark et al. 1995, Rodriguez et al. 2006, Schmidt et al. 2006). One of the most important determinants of nesting success is concealment correlated to specific vegetative characteristics, such as plant density (Howlett and Stutchbury 1996, Boulton et al. 2003). Dense forest vegetation may confer protection by creating visual barriers, increasing the number of available nesting sites, and hindering movement of avian and mammalian predators (Martin 1995). Hence, an individual's decision regarding nest site selection can be critical for breeding performance, and have consequences for population fitness. The breeding ecology of most Neotropical forest birds is poorly known. Specifically, information on components of reproductive success, such as nesting success (i.e., the probability a nest survives the nesting period to produce at least one fledgling) and daily nest survival (i.e., the probability a nest survives a day during the nesting period), have shown a broad spatial and temporal variation within and across forest bird species (Morton and Stutchbury 2000, Robinson et al. 2000b, Willson 2004). For instance, nesting success ranged from 8 percent in lowland tropical forests to more than 60 percent in montane cloud forests (Skutch 1985, Robinson et al. 2000a). Such variation may result from a combination of proximate factors favouring a diverse suite of life 20 history and behavioural traits (Martin 1996, Ricklefs 2000, Badyaev and Ghalambor 2001, Saether and Engen 2002). Northern Neotropical montane forests contain high levels of avian diversity and endemism (Watson and Peterson 1999), as well as several threatened bird species (Long 1995, Peterson et al. 1998). In addition to the overall loss and fragmentation of montane forests in the central highlands of Chiapas (Ochoa-Gaona et al. 2004), forest harvesting for firewood and construction may have altered avian breeding habitats. Continually removing understory vegetation reduces potential nest sites and may lower reproductive success of forest songbirds by decreasing cover and concealment for both offspring and parents (Martin et al. 2000, Ghalambor and Martin 2001), as well as alter food supply (Sherry and Holmes 1995, Chase et al. 2005). In this chapter, I examine the effects of habitat variation, nest-site selection and parental behaviour on the breeding performance of C. frantzii in a montane cloud forest reserve of Chiapas, Mexico. Particularly, I estimate breeding pair and nest densities of C. frantzii; examine nest-site choice by comparing habitat attributes between nest-sites and random sites; estimate daily nest survival; and evaluate the importance of parental behaviour on nest fate by examining parental visitation rates and female nest attentiveness. S P E C I E S A N D S T U D Y A R E A The Ruddy-Capped Nightingale Thrush (Catharus frantzii) is a species listed as threatened in Mexico (Diario Oficial de la Federacion 2002). Little is known about the breeding ecology and the factors that influence population fluctuations of the C. frantzii in Mexico. C. frantzii \s a forest dwelling, cup nesting bird that breeds in the interior of montane cloud forests during the rainy season (March to August) in the northern Neotropics (Clement et al. 2000). The female usually constructs mossy cup-nests in bushes or small understory trees (Raitt and Hardy 1970). Two light-blue eggs with brown spots are laid in a single clutch per breeding season (Skutch 1960). Incubation is solely by the female and lasts 1 5 - 1 6 days, while the brooding period lasts 1 5 - 1 7 days (Clement et al. 2000). 21 This study was conducted at the Cerro Huitepec Biological Reserve (16°44'38"N; 92°40'15"W), a semi-isolated, private forest island in the Central Highlands of Chiapas, Mexico from April to August of 2000-2003. t h e reserve is 137 ha and contains a variety of natural and disturbed forest habitats ranging in elevation from 2230 m to 2710 m. The mean annual precipitation and temperature are 1300 mm and 14.5°C, respectively. The reserve contains a total of 83 vascular plant families, including 186 genera and 315 species in six plant communities: grassland, shrub-land, second growth forest, dry oak forest, oak humid forest and evergreen montane cloud forest (Ramirez-Marcial et al. 1998). The reserve includes 32% of the regional plant species richness above 2000 m, but many plant species occur in low densities or have locally been extirpated (Gonzalez-Espinosa et al. 1995). Selective forest harvesting for construction and firewood gathering was frequent before the reserve was established in 1987. These processes were more evident in the drier, flat parts of the reserve where understory vegetation had been partially removed. The land surrounding the reserve is used for agriculture, housing and cultivation of exotic pine stands. Open and secondary habitats around the reserve have increased in the previous decade. Habitat types at Huitepec have been classified into five forest types differing in plant species composition, structure, topography and aspect (Ramirez-Marcial et al. 1998). 1) Montane cloud forest (~ 35 ha) is found from 2400 to 2700 m altitude in the western and northwestern parts of the reserve. The canopy is 30-35 m high, and dominated by Quercus laurina and Q. crassifolia. Other tree species at a lower height (20-25 m) are Clethra macrophylla, Cleyera theaeoides, Persea americana and Styrax argenteus. The understory contains Drimys spp., Miconia spp., Microtropis spp., Oreopanax spp., Ostrya spp., Prunus spp., Rapanea spp., Rhamnus spp., Saurauia spp. and Ternstroemia spp. 2) Wet-oak forest (-32 ha) is located in the eastern and southeastern part of the Reserve, from 2450 to 2620 m altitude. Quercus and Arbutus xalapensis and Alnus acuminata (30-35 m) dominate the canopy. A sub canopy stratum includes Garrya laurifolia, Oreopanax spp., Prunus serotina, Rapanea juergensenii and Viburnum juncundum. The understory includes Cestrum, Fuchsia, Gaultheria, Litsea, Senecio and Xylosma, and Adiantum, Asplenium and Polypodium. 3) Riparian forest (~ 10 ha) is distributed along two small streams on 22 the east side of the reserve. This habitat is present from 2100 to 2300 m altitude. The canopy varies and there is limited vegetation information for this habitat. However, many plant species from adjacent moist and dry forests are present in this habitat. 4) Dry-disturbed oak forest (~ 40 ha) is located in the northeast part of the reserve. For this study I also included 10 ha of this forest type adjacent to the reserve. This habitat was located between 2200 and 2350 m altitude and the canopy is dominated by oak trees (25-30 high) such as Quercus crassifolia, Q. rugosa, Q. laurina, O. candicans, Q. skutchii, Q. acutifolia, and Q. crispipilis. Understory plants are Oreopanax, Rapanea, Styrax, Viburnum, Adiantum, Bidens and Chimaphila. 5) Second-growth (disturbed) forest (~ 30 ha) is located in the lower, north-northeast part of the reserve and contains 119 plant species. This forest was harvested for construction material and firewood before Huitepec became a reserve. The canopy is discontinuous and mostly dominated by Quercus trees with heights of 8-10 m. The understory contains Eupatorium, Rubus, Viburnum and seedlings of Alnus, Arbutus, Buddleia, Crataegus, Garrya, Prunus, Oreopanax, Rhamnus and Viburnum. Based on the variation in moisture, plant species composition and structure, and management history among forest types, I combined montane cloud, wet oak and riparian forests as moist forest (~ 77 ha) and dry-oak and second growth (disturbed) as dry-disturbed forest (~ 70 ha) habitat types. METHODS During the early breeding season (April-May), I used spot-mapping to identify territories and estimate pair densities. Visual observations, songs and calls were recorded and individual locations were used to identify territories (Ralph et al. 1993). Simultaneous calls of different individuals were considered to distinguish territories (Brawn and Robinson 1996). Mapping territories involved =13 days per habitat/season during the early breeding seasons of 2000 to 2003. I assumed that C. frantzii: 1) established territories during the breeding season, 2) a pair occupied each territory, and 3) maps of territories accurately reflected the spatial distribution of pairs. By using the singing male method I assumed no effect by young males that may sing sporadically and 23 influence areas used by breeding males in different habitat conditions (Mazerolle and Hobson 2004). Nest searching was conducted during the breeding season, with a consistent yearly sampling effort of =127 person hours per habitat-type. Nests were located by following territorial individuals that showed breeding behaviour and by searching within territories for potential nest-sites (Robinson et al. 2000b). Once a nest was located, it was marked with flags placed approximately 10 m from the nest. Nest-sites were GPS-located on a map. I monitored each active nest every 3-4 days by observing nests from a distance of >10 m until either the young fledged or the nest failed. During appropriate conditions and times (e.g. absence of rain and potential predators), the content of the nest was checked with a mirror. A nest was considered to have survived the interval between visits if at least one egg or nestling was present or if a least one young had fledged before the final visit. Nest visits were more often within 2-3 days of predicted fledging to increase certainty in assigning fate. Nests that fledged at least one young were considered successful. Observation of fledglings or parents feeding fledglings near the nest was used as evidence for nesting success. A nest was considered a failure if all eggs or nestlings disappeared before any fledgling could have left the nest. To explore if parental behaviour influenced nest predation risk, I recorded the parental visitation rates (i.e., the number of times the female left the nest) and calculated female nest attentiveness (i.e. percent of time sitting on the nest or on-bouts per hour) during the incubation and brooding stages from 33 nests (18 nests in the moist forest and 15 nests in the dry-disturbed forests). To record parental behaviour during nesting, observers with binoculars (10x50) or a telescope (15x20x60) sat at a distance of 10-15 m from nests. Observers selected sites that allowed visibility of the nests, but were also well camouflaged by vegetation. Observations were made during the incubation stage after confirmed clutch completion and during the brooding stage when chicks were at least 5 days old. Each nest was monitored two hours a day between 07:00 and 12:00 for four days at both the incubation and brooding stages. Habitat attributes.- After a nest failed or the young fledged, I collected data on habitat attributes that might influence nest-site selection, nesting success and daily nest survival. I 24 measured habitat attributes at three spatial levels: nest, nest-site and nest-patch. For each nest, I identified the plant species used as breeding substrata, and measured nest height, plant height and basal diameter. At the nest-site level, I measured the distance to the nearest tree > 5 cm diameter at breast height (DBH), and vertical distance in cm to the nearest plant structure covering the nest. Each nest was observed at 1 and 6 m from the four cardinal directions (N, S, E, and W), and the mean percent of horizontal visibility was estimated with a densitometer (Gbtmark et al. 1995). The mean canopy cover at 1 and 6 m distance were also estimated with a densitometer. Additional measurements at the nest-site included shrub density, tree density > 5 cm DBH, and slope and aspect in degrees. Vegetation structure was quantified at the nest-patch level by using a 12.5 m-radius circular plot centred on the nest (0.049 ha). Within the plot, I considered trees with a DBH > 5 cm and measured DBH, tree height, and tree density. Four rectangular subplots of 2 m width were established in each cardinal direction for measuring the density of shrubs, seedlings, herbaceous plants, ferns, Adiantum, epiphytes on trees and ground, mushroom breeding bodies, Licopodium, snags, woody debris, canopy cover, ground cover, and air temperature and humidity. A random point was selected by choosing one of eight compass directions, and a distance from the nest between five to 30 meters in intervals of five meters. A plant and a specific point in the plant were arbitrarily selected as a potential site for nesting substrata. For random sites, I measured the variables recorded for the nest-sites, but plant species. Statistical analysis. - Territory and nest density between habitat types and years were compared using general linear models one-way analysis of variance (GLM-ANOVA; Gotelli and Ellison 2004). To examine nest-site selection and identify habitat attributes that account for selected sites, I pooled fate-known (monitored nests, n = 78) and fate-inferred (i.e., non-monitored but able to infer fate, n = 98; see Easton and Martin 2002) nests. I tested the proportion of successful and depredated nests across plant species used as substrata for nest placement with a Wilcoxon's rank-sum tests (Gotelli and Ellison 2004). I compared the breeding performance as the proportion of successful nests over the total number of recorded nests per habitat type (N= 193). 25 For univariate comparisons of all habitat variables, I performed Mests or Wilcoxon's rank-sum tests to identify those variables differing between nest-sites and random sites. Then, a correlation matrix from Principal Component Analysis (PCA) was used to estimate the degree of association and interdependence between pairs of variables. Highly correlated variables (r > 0.7, P < 0.05) were evaluated and the selected variable included in subsequent analyses had the higher univariate value and lower P value. After standardizing selected variables, I used discriminant function analysis (DA) to test the null hypotheses that C. frantzii did not discriminate nest-sites from random sites. Predictive models for the dependent, binary response variables, defined as nest-site and random site, were developed to identify those habitat attributes associated with nest-site selection. Candidate independent variables were selected for inclusion in the final model using a forward-stepwise selection procedure with a P < 0.1 to enter in the model. Mahalonobis distances (minimum D 2) between group centroids were used for maximizing separation of nest-sites from random sites (McGarigal et al. 2000). Structured coefficients were obtained to determine correlative variables with discriminant functions (McGarigal et al. 2000, Gotelli and Ellison 2004). Positive parameter coefficients in the discriminant function indicated that an increase in the value of a variable increases the probability of a nest-site being selected. Conversely, a negative coefficient indicated that as the variable value increases, the probability of a nest-site being selected decreases (Neter et al. 1996). The parametric assumptions on 4 of 10 variables for DA were relaxed because my objectives were to identify a linear composite of covariates, detect significant differences between group centroids and determine individual contribution of attributes to the overall discrimination function (McGarigal et al. 2000). Daily nest survival rates over the nesting cycle (Mayfield 1975, Stanley 2000) were estimated with calculated standard errors (Johnson 1979, Hensler and Nichols 1981). Daily nest failure rate (r) for fate-known nests (n = 78) were estimated as the total number of failed nests divided by the total number of observer days pooled across all nests per habitat type. Daily survival rates (S) were then calculated as 1 - r (Hazier 2004). Daily nest survival probabilities among forest habitats were compared using CONTRAST (Hines and Sauer 1989). Because of potential seasonal variation in daily nest survival, I estimated the probabilities of a nest surviving through the 26 incubating, brooding, and nesting periods in days (a) per habitat type with (Hazier 2004, Rotella et al. 2004). I did not attempt to correct for the time prior to the nest being first observed (Grant et al. 2005). Because of a limited data set for fate-known nests which were found in different stages through the nesting cycle, I used the events/trials syntax (i.e., where events were either success, 0, or failure, 1, over the observation interval, and trials were the number of exposure days), and I used the Mayfield logistic regression to determine the combination of habitat attributes that best explained daily nest survival rates (Hazier 2004). For a successful nest, I assumed that the last active date was the last exposure day, while for a failed nest I used the midpoint between the last check and the last active day (rounded up to the nearest day). Previous to applying the Mayfield logistic regression, I tested each habitat variable in turn, using univariate logistic regression. All variables that were significantly different at P < 0.1 were then tested for independence before being entered into the final model. Also, when Pearson correlations between any two variables were significant at P < 0.05,1 selected the one with higher univariate value and lower probability to be included in the Mayfield logistic regression model. To perform the Mayfield logistic regression, I used PROC LOGISTIC in SAS (SAS 2004), and the events/trials syntax. Thus, the total sample size was N = 1180 (moist forest = 705, and dry-disturbed forest = 475). Because the Mayfield logistic regression estimates nest mortality (i.e., failure), and my interest was to evaluate effects on daily nest survival rates, signs of all coefficients and the intercept were reversed (Hazier 2004). The Mayfield logistic regression has five assumptions: (1) the number of exposure days is known, (2) nests with unknown fates are censored observations, (3) there was a constant survival rate over time, (4) there was a common daily survival pattern for all nests, and (5) each nest represents an independent sample unit. Assumptions (1) and (2), were either estimated or satisfied. However, nest survival may have varied between the incubation and the brooding stages, and linear trends in survival across and within stages may not be consistent (Grant et al. 2005). All measured variables were considered to constrain the common daily nest survival. Over dispersion of data was diagnosed with a goodness-of-fit test (Farnsworth et al. 2000, Hazier 2004). 27 Parental visitation rates and female nest attentiveness were analysed with GLM two- and three-way ANOVA (Zar 1999). The original models containing all two- and three-way interactions among habitat*fate*stage were reduced by sequential elimination of non-significant interaction terms (P> 0.05). When a significant interaction of categorical predictors with parental visitation rates and female nest attentiveness occurred, the analysis was partitioned for the sign interaction. If parental visitation rates and female nest attentiveness differed among nesting stages (Grant et al 2005), I separated them into incubation and brooding. Prior to statistical analysis, variables were checked for normality with a Shapiro-Wilks W test and equal variance with Bartllet's test (Gotelli and Ellison 2004). When required, data were square root-transformed (distances) or arcsine-transformed (percentages) as appropriate to meet parametric assumptions. When the data differed from parametric assumptions, non-parametric Wilcoxon/Kruskal-Wallis rank tests were used. When factor interactions were significant, independent univariate analyses were performed to evaluate the role of each factor independently. Statistical analyses were performed with SAS-JMP 5.1 (Sail et al. 2005) and SAS 9.1 (SAS 2004). All means are presented ±1 SE, and considered significant at a = 0.05. R E S U L T S A total of 270 territories of C. frantzii were recorded over four years (2000-2003; Fig. 2.1a). The mean pair density was higher in the moist forest than in the dry-disturbed forest (0.57 pairs ha 1 ±0.07 SE vs. 0.34 + 0.08; GLM-ANOVA, hf= 0.9, F 1 | 6 = 11.9 P < 0.05), and this relationship was consistent among years (Fig. 2.1.b). Conversely, the mean nest density (nests/ha) did not differ between the moist forest and the dry-disturbed forests (0.34 ± 0.09 vs. 0.23 ± 0.09; F 1 6 = 0.3, P = 0.79), though with higher densities in 2001 in both habitats (Fig. 2.1 .c). Pair density was positively correlated to nest density across habitat types (r= 0.77, F 1 i 6 = 20.5, P < 0.01). A total of 193 nests of C. frantzii were found. Seventy-eight nests were fate-known, 98 fate inferred and 17 unknown. The estimated date of starting incubation for the nest found active extended from mid April to early August, and did not vary between moist and dry-disturbed forest when year was fixed (x2- 2.68, df - 77, P- 0.1; Fig. 2.2). Combining fate-known and fate-inferred 28 nests (n = 176), 50 fledged at least one young (nesting success = 28.4%; Fig. 2.3). Predation was the major cause of nest failure (94%); and only seven nests (4%) were found abandoned, and other 2% were unknown. Nests were found in 18 plant species, and two unidentified plant substrata, with six species being the most frequently used (Table 2.1). The ratio of successful nests across plant species did not differed between moist and dry-disturbed forest habitats (Likelihood ratio test x2 - 1.9, df- 191, P = 0.16). The same pattern of nesting success was found when considering only the six most used plant species as nest substrata between moist and dry-disturbed forest habitats (Likelihood ratio test x*= 5.1, df= 143, P = 0.4). Nest-sites selected by C. frantzii significantly differed from random sites in 10 of the 31 habitat attributes measured at the nest, site and patch levels (Table 2.2). Discriminant analysis revealed clear differences among nest-sites from random sites (Wilk's X, u = 0.65, F10,34o = 18.42, P < 0.001), and correctly classified 82.4% of the selected sites and 76.7% of the random sites. The canonical discriminant function with an eigenvalue of 0.54 included canopy cover at 6 m, shrub density, horizontal visibility at 6 m, canopy cover at 1 m, and horizontal visibility at 1 m (0.69 > r> 0.33) as habitat attributes strongly correlated with the selection of nest-sites by C. frantzii. In the moist forest, canopy cover at 6 m, horizontal visibility at 1 m and shrub density (0.89 > r > 0.46) were the habitat attributes selected by the discriminant analysisi (Wilk's X, u = 0.66, F 8 1 9 8 = 12.7, P < 0.001). The canonical discriminant function, with an eigenvalue of 0.51, implied a concealment gradient influencing selected sites for nesting by C. frantzii in this habitat type (Fig. 2.4.a). In the dry-disturbed forest, shrub density, canopy cover at 6 m and mean canopy cover at patch level (1.07 > r> 0.61) were the habitat attributes selected by the DA (Wilk's X, u - 0.52, F6,137 = 20.9, P < 0.001). The first canonical discriminant function, with an eigenvalue of 0.92, also implied a concealment gradient and reflected a difference of concealment at site level, and canopy cover at the patch level between nest-sites and random sites (Fig. 2.4.b). More habitat attributes explained separation between nest-sites from random sites in the moist forest than in the dry-disturbed forest (Fig. 4a,b). The canonical distances separating point relationships from the grand centroid was shorter for the moist forest than the dry-disturbed forest for nest sites (F 1 t 3 5o = 4.19, P 29 < 0.05) and random sites (F 1 i 3 5 o= 6.97, P < 0.01), implying stronger selection for habitat attributes in the later habitat. Mayfield estimates of daily nest survival rates (S) for fate-known nests were higher in moist than dry-disturbed forests when year was fixed (0.96 ± 0.007 vs. 0.93 ± 0.013; x 2 = 4.08, df = 76, P < 0.05; Fig. 2.5). When habitat was fixed, pair density positively correlated with nesting success (r = 0.84, F 1 6 = 9.32, P < 0.05), while nest density did not correlated with nesting success (r= 0.52, Fi , 6 = 3.42, P = 0.12). Neither pair density nor nest density was associated with daily nest survival ( r< 0.09, F 1 i 6 < 2.63, P > 0.16). Daily nest survival rates were positively correlated to effects of the distance to the structure covering the nest, horizontal visibility at 6 m and the density of epiphytes on the ground (Table 2.3). Estimated nest survival probabilities for the incubation and brooding stages, and the entire breeding cycle were higher in the moist forest than in the dry-disturbed forest (SincUbation- 0.55 ± 0.1 vs. 0.36 ± 0.09, x2 = 8.0, df= 76, P < 0.05; SBr0odmg- 0.53± 0.1 vs. 0.38 ± 0.09; x2 = 8.4, df= 76, P < 0.05; SBreeding. 0.29 ± 0.1 vs. 0.13 ± 0.06; X 2 = 18.04, df= 76, P < 0.01). Parental behaviour.- Parental visitation rates across habitats and nest fate were significantly lower during the incubation stage than during the brooding stage (2.9 trips h' 1 ± 0.7 vs. 4.7 ±0.5; Welch paired f-test = 12.2, d f= 31 , P < 0.001). The GLM-ANOVA model for parent visitation rates revealed a significant habitat x nest fate interaction (F,,i28 = 19.4, P < 0.001), indicating that parent visitation rates of C. frantzii differed according to the habitat, but these differences were dependent on the nest fate. Hence, separate analyses for habitat and nest fate were performed. Parental visitation rates (number of trips per h) during incubation and brooding stages did not vary between the moist forest and the dry-disturbed forest (3.97 trips h' 1 ± 0.18 SE vs. 3.24 ±0.19; paired f-test = 1.3, df - 31 , P = 0.19). As well, parental visitation rates during incubation and brooding did not vary between successful and failed nests (3.56 trips h"1 ± 0.19 vs. 4.05 ± 0.18; paired f-test = 1.9, df= 31 , P = 0.07). However, controlling.for stage and habitat, parental visitation rates during incubation were significantly lower for successful than failed nests (3.15 trips h"1 ±0.29 vs. 4.04 ± 0.28; paired f-test = 2.18, oY=31, P < 0.05; Fig. 2.6a). 30 Female nest attentiveness (proportion of time on-bouts per h) was significantly higher during the incubation than the brooding stage when habitat and nest fate were fixed (0.78 ±0.05 vs. 0.64 ± 0.04; paired f-test = 12.06, df, = 1, P < 0.001). The variation in female nest attentiveness was independently explained by habitat and nest fate (GLM, P?=0.M, F 3 6 2 = 4.21, P<0 .01 ) . Female nest attentiveness was higher in the moist forest than in the dry-disturbed forest (0.73 ratio time ±0.012 vs. 0.69 ±0.013; F , 6 2 = 5.9, P < 0. 05), and higher for successful than failed nests (0.73 ±0.013 vs. 0.69 ±0.013; F 1 > 6 2 = 6 . 1 2 , P < 0 . 0 5 ; Fig. 2.6b). D I S C U S S I O N The breeding performance of C. frantzii in the central highlands of Chiapas varied across contiguous, but distinct habitat types, and it was correlated with habitat attributes and parental behaviour. Important differences across habitats were that breeding pair density and nest density were higher in the moist forest, suggesting that moist habitats are more likely to be preferred for reproduction. Both pair and nest densities were markedly higher in the most humid sites in the reserve, such as small creeks. These sites contained dense vegetation that may have been more suitable for breeding. Other authors have found that moist sites, particularly streams and creeks, are preferred by nightingale thrushes for reproduction, with higher breeding performance indicators, than more dry, open habitats (Beltran and Kattan 2003, Londoho 2005). Breeding performance was consistently better in the moist forest with more understory vegetation than in the dry-disturbed forest, where understory vegetation was partially removed. Removal of vegetation may reduce quantity and quality of both breeding sites and patches for C. frantzii. These findings are consistent with the implication that safe-nesting sites may be more limited in disturbed habitats, particularly those habitats modified by human activities. The reduction of safe nesting-sites may result in hyper-dispersion of breeding sites with a combination of decreasing food availability and increasing risks of nest predation (Martin 1995, Sherry and Holmes 1995). Sites with less understory vegetation may make nests more conspicuous to predators (Howlett and Stutchbury 1996, Weidinger 2002), and thus less suitable sites for C. frantzii. Less understory vegetation could also increase detectability and predation risk for adult birds, reduce 31 thermal cover and limit observational perches for vigilance (Gotmark et al. 1995). Habitat alteration also may result in differences in the predator community, food abundance, and quality of individuals in each habitat, contributing to the different observed patterns. Over one quarter (28%) of C. frantzii breeding pairs at Huitepec fledged at least one young. This level of nesting success was consistent with data reported for passerine forest birds in tropical forests of Costa Rica, Ecuador and Panama (Ricklefs and Bloom 1977, Sieving and Karr 1997, Robinson et al. 2000b). Among thrush species in temperate forests, approximately 27% of Wood Thrush (Hylochycla mustelina) nests were successful in southern Ohio (Artman and Downhower 2003). Nest predation was the main cause of nest failure at Huitepec and was lower in the moist forest (67%) than in the dry-disturbed forest (78%). The level recorded in moist forest was similar to nest predation rates reported for Wood Thrush in Ohio (66%; (Artman and Downhower 2003). However, nest predation rates for C. franttzii were high relative to those reported for other North American passerines. For instance, nest predation rates were 50% for Hooded Warblers {Wilsonia citrina) from north-eastern North America (Howlett and Stutchbury 1996) and for shrub-nesting birds (Martin 1995, Badyaev and Ghalambor 2001). Thus, nest predation was the main factor determining breeding performance of C. frantzii in the study site. C. frantzii used a variety of plant species as substratum to place their nests. Although, six of 18 plant species were more frequently used as nest-substrata for breeding, nest fate was not correlated with any particular plant species. C. frantzii used a variety of plant species in the moist forest (e.g., Cornus excelsa, Styrax argentus, and Querqus rugosa). While in the dry-disturbed forest, Q. rugosa and Oreopanax xalapensis were more used as substrata for nesting. Differences in the use of plant species may result from vegetation removal. Vegetation removal impacts the density and the structure of vegetation (Easton and Martin 2002). Periodically removing vegetation and decreasing the density and structure of vegetation reduces potential substrata for nesting, as well as modifies concealment for the nest (Martin et al. 2000, Ghalambor and Martin 2001), and reduces food availability (Gonzalez-Espinosa et al. 1995). My data suggest that nest concealment and cover were the most important determinants of nest placement by C. frantzii at Huitepec Reserve, and that daily nest survival was positively correlated to concealment. 32 Sites selected by C. frantzii \ox building nests differed from random sites principally in the structure at the nest-site and patch levels. The discriminant model showed that selected sites for nesting had lower horizontal visibility at 6 m, higher canopy cover at 1 and 6 m, and higher shrub density compared to random sites in both the moist and the dry-disturbed forests. Thus, habitat structure attributes relating to concealment have an important influence in nest-site selection. Additional covariates such as lower horizontal visibility at 1 m and tree density at the nest-site level were also important nest-site features in the moist forest. Fern density and canopy cover at the patch level influenced nest-site selection in the dry-disturbed forest. Characteristics at the patch-level rather than at nest-site may determine the nest-site choice in dry-disturbed forest. The rugged topography of the study site promoted variation in selective harvesting intensity in the dry-disturbed forest because selective harvesting was less severe along moist drainages and cooler slopes than on flatter and north-facing slopes. In the dry-disturbed forest, small patches of habitat remained relatively undisturbed. Such variation in selective harvesting intensity is important for maintaining availability of suitable sites and conditions for breeding, at least on a limited basis, within disturbed areas. Additionally, increasing canopy cover at the patch level may increase shade and reduce visibility of eggs, chicks or parents during incubating or brooding (Gbtmark et al. 1995). One of my objectives was to identify which habitat attributes influenced nesting success and also when understory removal alters these attributes. Because nest predation is a dominant factor affecting nesting success, there should be selective pressure for birds to choose less vulnerable nest sites (Martin and Li 1992). Placement of nests within a suite of features such as high vegetation density helps to reduce accessibility and search efficiency of predators (Martin and Joron 2003). The daily nest survival estimates were consistently high in the moist forest through the years compared to the dry-disturbed forest. Mean daily nest survival estimates varied from 0.96 to 0.94 in moist forest and dry-disturbed forest, respectively. These estimates were slightly higher or comparable to the values reported for a variety of passerine open-cup nesting birds such as the Western Slaty Antshrikes [Thamnophilus atrinucha; 0.91; Roper 1996), the enclosed nester Song . Wrens (Cyphorhinus phaeocephalus; 0.97; Robinson et al. 2000a), and for 10 open-cup nesting species (0.95; Robinson et al. 2000b) in the lowland forests of Panama. Also, these values were 33 similar to those reported for passerines in North America: 0.94 reported for the Wood Thrush in Ohio (Artman and Downhower 2003) and 0.96 reported for the Acadian Flycatcher (Empidonax virescens; Hazier et al. 2006). Notwithstanding slight differences in daily survival rates between the previous studies and my own, the longer nesting period of C. frantzii may increase the overall exposure time of the nest, resulting in lower nest survival. For the Acadian Flycatcher, the probability of a nest surviving the nesting period was 0.54 with its breeding duration of 29 days (Hazier 2004). For C. frantzii this probability may reach 0.39 in the moist forest and 0.19 in the dry-disturbed forest as it has a nesting duration of 33 days (incubation: 16, and brooding: 17 days). The effects of pair density and nest density on the breeding performance of C. frantzii were unclear. Only pair density positively and significantly determined nesting success when habitat was not considered. Coefficients of variation (CV) for the two components of breeding performance measured indicated lower levels of variation in the moist forest than in the dry-disturbed forest (nesting success: 16.8 vs. 43.9; daily nest survival: 1.3 vs. 1.9, respectively). Thus, as both breeding pair density decreased and variation in the components of breeding performance increased, there was a decline in the breeding performance of C. frantzii. This was the case in the dry-disturbed forest. However, different population regulation mechanisms, such as density-dependence and site-dependence, may act simultaneously at different scales, influencing the breeding performance of C. frantzii in the reserve. The simultaneous effect of diverse regulatory mechanisms on avian breeding performance has been demonstrated for several species (Rodenhouse et al. 2003, Chase et al. 2005, Carrete et al. 2006). The combined effects of canopy cover, nest concealment and the density of epiphytes on the ground significantly influenced nest fate. Canopy cover and nest concealment may offer overhead protection to reduce the likelihood of visual detection from avian predators because predatory birds, such as jays, rely primarily on visual cues for detection of nests (Martin 1995, Weidinger 2002). Vegetation that provides high concealment may also reduce the likelihood of nest detection by mammalian predators by affording visual and olfactory protection and hindering search efficiency by impeding predator movements (Martin and Li 1992, Martin 1995). A rich 34 community of mammal predators resides in the Huitepec Reserve, including 17 species of rodents (Naranjo and Espinoza-Medinilla 2001). Parental behaviour varied across breeding stages and habitats. The parental visitation rates increased significantly from the incubation stage to the brooding stage, while female nest attentiveness declined from the incubation stage to the brooding stage. Skutch (1949) hypothesized that increased parental activity increased nest predation. However, Martin et al. (2000) found that nest predation was generally higher during incubation than during brooding. When controlling for nest-site variation, an increase in parental visitation rates increased nest predation risk (Martin 2004, Muchai and du Plessis 2005, Remes 2005). Parental visitation rates for C. frantzii may influence nest fate during the incubation stage. Thus, increasing parental visitation rates during this stage could result in an increase in the risk of clutch predation. Female nest attentiveness was higher in the moist forest where most successful nests were found than the dry-disturbed forest. Female condition may influence both nest attentiveness and extend the incubation period. For instance, nesting success of females in poor body condition may be compromised because decreased nest attentiveness increases the length of nest exposure, resulting in slower embryo development and higher predation rates (Weidinger 2002). Female survival could decline during the breeding season because they are exposed to predation pressure for longer periods. Moreover, later hatching could reduce fledgling survival. Fledglings in these conditions may confront unpredictable, adverse events coupled with intense predation, a seasonal decreased of food resources and decreased parental care (Brown and Roth 2002, Styrsky et al. 2005). Female nest attentiveness probably represents an important strategy of passive nest defence for C. frantzii, which combines a vigilant behaviour with cryptic coloration. Thus, both nest concealment by vegetation and cryptic plumage in C. frantzii are probably important to the breeding performance (Gotmark et al. 1995). Variation in the behaviour of C. frantzii, such as parental provisioning and nest attentiveness, in concert with other proximate factors such as habitat attributes, food availability and female body condition, may play an important role in the interaction between ecological pressures and breeding performance (Badyaev and Ghalambor 2001, Ghalambor and Martin 2001). 35 C O N C L U S I O N The breeding performance of C. frantzii varied across habitats differing in forest type and density of understory vegetation. Removing understory vegetation in the Central Highlands of Chiapas appeared to reduce breeding pair density, nesting success and daily nest survival. Increasing pair density appeared to affect nesting success, particularly in the dry-disturbed forest where density-dependence was apparent. Habitat attributes associated with overall concealment at the patch level correlated with nest-site selection and nest survival. Parental behaviour in concert with these attributes influenced the risk of nest predation. Because long-term habitat changes may alter the breeding habitat and lead to low breeding performance in birds, I suggest that long term studies should continue to be used to assess the consequences of vegetation removal and other forest management practices in the region on population demography for C. frantzii, and other forest nesting passerines. 36 Table 2.1. Frequencies of main plant species (6 of 18) used as nest-substrata by Catharus frantzii (Ruddy-capped Nightingale Thrush) in moist and dry-disturbed forest habitat types at the Huitepec Biological Reserve, Central Highlands of Chiapas, from 2000 to 2003. Frequencies included all nests found (N=193), and percent of successful and depredated nests was restricted to fate-known and fate-inferred nests. Habitat type Plant species Moist forest Dry-disturbed forest Overall study site Frequency Successful Predated Frequency Successful Predated Frequency Successful Predated (%) (%) (%) (%) (%) (%) Cornus excelsa 25 4.15 7.25 3 0.00 0.00 28 4.15 7.25 Crataegus pubescens 13 2.07 3.63 8 1.04 3.11 21 3.11 6.74 Oreopanax xalapensis 10 1.04 3.63 14 0.52 . 6.22 24 1.55 9.84 Quercus rugosa 16 2.59 5.70 22 1.55 7.77 38 4.15 13.47 Rapanea juergensenii 7 1.55 2.07 3 0.52 1.04 10 2.07 3.11 Styrax argentus 18 3.11 5.18 5 1.04 1.55 23 4.15 6.74 Total 116 17.62 35.23 77 8.29 26.42 193 25.91 61.66 Table 2.2. Habitat attributes at nest-site and nest-patch levels of Catharus frantzii (Ruddy-capped Nightingale Thrush) nest and random sites found in moist and dry-disturbed forest types at the Huitepec Biological Reserve, Central Highlands of Chiapas. Selected variables were used for the final discriminant analysis (DA) model for nest-site selection in the overall study site (a), moist forest (b), dry-disturbed forest (c), and for the Mayfield logistic regression model (d). Habitat type Moist forest Dry-disturbed forest Nest Random Nest Random Variable (units) Average SE Average SE Average SE Average SE Selected Nest Nest height (NEHE, m) 1.97 0.93 2.01 1.21 1.99 1.17 1.84 1.15 a, b, d Plant height (m) 5.46 3.49 5.64 4.05 4.46 3.47 4.99 4.57 Basal diameter (cm) 12.72 13.96 12.78 15.32 15.00 14.27 15.53 15.79 Nest-site (6 m radius) Distance nearest Tree (m) 0.58 0.84 0.60 0.94 0.94 1.43 0.84 1.11 Distance to structure covering the nest (cm) 0.41 0.33 0.42 0.41 0.30 0.19 0.28 0.19 d Canopy cover at 6 m (CC6M, %) 79.83 19.83 65.31 20.53 78.92 10.82 66.39 18.72 a, b, c Horizontal visibility at 6 m (HV1M, %) 48.89 26.75 35.91 0.25 42.43 26.76 29.76 21.73 a, b, c, d Canopy cover at 1 m (CC1M, %) 80.93 7.73 68.44 0.25 82.11 9.64 69.42 24.88 a, b, c Horizontal visibility at 1 m (HV1M, %) 19.47 16.41 0.16 0.16 27.50 22.21 23.47 22.50 a, b, c Shrub density (SHRD, individuals per m 2) 1.69 0.74 1.28 0.79 1.59 0.58 1.18 0.56 a, b, c Tree density (individuals per m 2) 0.19 0.13 0.18 0.15 0.18 0.09 0.17 0.1 a Slope (<°) 34.08 12.61 32.42 15.49 20.28 7.67 19.23 8.85 Aspect (<°) 248.5 79.3 201.2 95.4 152.1 76.6 169.4 87.5 Nest-patch (12.5 m radius) DBH of trees >5 cm (cm) 35.25 6.38 Tree height (m) 15.70 4.16 Tree density (TRED, individuals/ha) 0.18 0.08 Slope (<°) 32.94 7.99 Shrub density (individuals/m2) 3.48 1.5 Seedling density (SEED, individuals/m2) 1.53 0.46 Herbs density (individuals/m2) 4.39 1.42 Fern density (FERD, individuals/m2) 2.47 1.18 Adiantum density (individuals/m2) 1.48 0.88 Number of epiphytes on Trees 1.97 1.02 Density of ground epiphytes (individuals/m2) 0.72 0.38 Mushroom density (individuals/m2) 0.57 0.32 Licopodium density (individuals/m2) 0.05 0.03 Number of snags (individuals/ha) 9.24 5.37 Number of woody debris (individuals/ha) 3.70 2.01 Air temperature (°C) 15.69 2.51 Air humidity (%) 73.28 8.78 Mean canopy cover (CCPA, %) 84.62 19.59 Mean ground cover (%) 8.15 1.03 34.02 7.77 23.75 4.98 23.76 5.08 16.03 4.82 11.60 3.46 11.94 3.75 0.18 0.07 0.19 0.02 0.19 0.02 a, b 32.73 9.03 19.51 10.19 19.68 9.17 3.32 0.13 2.97 0.11 3.28 0.14 1.53 0.43 1.92 0.54 1.99 0.51 a, b 4.41 1.54 2.51 1.02 2.52 1.22 2.16 1.16 2.60 1.25 2.09 1.17 a, c 1.46 0.94 2.93 1.59 2.85 1.41 2.14 1.07 1.75 1.57 1.72 1.53 0.72 0.40 0.13 0.08 0.12 0.07 d 0.55 0.29 0.98 0.45 1.08 0.51 0.05 0.03 1.99 1.49 2.36 1.51 7.62 5.03 5.56 4.43 5.50 4.36 3.57 1.91 1.75 1.86 1.62 1.84 15.90 3.01 17.63 3.59 17.87 4.01 0.74 0.08 0.68 0.12 0.67 0.12 62.59 19.68 63.83 10.92 65.58 10.38 a, c 0.08 0.01 0.08 0.01 0.08 0.01 Table 2.3. Parameter es t imates for se lec ted habitat at t r ibutes that reflect the m a x i m u m effect on dai ly nest survival rates f rom a Mayf ie ld logistic regress ion mode l for Catharus frantzii (Ruddy-c a p p e d Night ingale Thrush) in the Hui tepec Biological Reserve , Centra l H igh lands of Ch iapas , dur ing 2 0 0 0 - 2 0 0 3 . Est imates are f rom a single g lobal mode l that inc luded fa te -known nests (n = 78) . Posi t ive es t imates indicate h igher nest surv iva l . Parameters Estimate S.E. Wald Chi-Square P-value Intercept 1.95 0.52 14.24 0.001 Dis tance to structure cover ing the nest (cm) 0.81 0.39 4.16 0.042 Hor izontal visibil i ty at 6 m (%) 1.84 0.79 5.45 0.019 Densi ty of g round ep iphytes ( ind iv iduals /m 2 ) 0.91 0.42 0.42 0.029 40 dl J J Figure 2 . 1 . (a) Number of territories, (b) pair density (pair per ha), and (c) nest density (nests per ha) from 2000 to 2003 of Catharus frantzii (Ruddy-capped Nightingale Thrush) in moist forest (77 ha, white bars) and dry-disturbed forest (70 ha, black bars) habitat types at the Huitepec Biological Reserve, Central Highlands of Chiapas. 41 a) Moist forest K> -6 r - n _ 2 2 1 — n 1 r—^ i 9 1 1 7 Apr Apr May Jun Jun 10 30 20 10 30 b) Dry-disturbed forest 1 6 2 2 2 — 1 1 4 1 1 -f-r1 J Apr Apr May Jun Jun Jul 10 30 20 10 30 20 Figure 2.2. Estimated onset of incuation from fate-known nests in (a) moist forest (n - 42 nests) and (b) dry-disturbed forest (n = 36 nests) of Catharus frantzii (Ruddy-capped Nightingale Thrush) at the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas, from 2000 to 2003. The vertical lines within the quantile boxes represent the sample median and boxes encompass the 25( to the 75 t h percentile. 42 • Abandoned • Depredated 12 Successful 20 V?. 26 m fed m 46 27 Moist forest | Dry-disturbed forest Known fate Moist forest | Dry-disturbed forest Fate inferred Figure 2.3. Fate of 176 nests of Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Huitepec Biological Reserve, Central Highlands of Chiapas, from 2000 to 2003. Fates were unknown for 12 nests in the moist forest and 5 in the dry-disturbed forest habitat types. 43 (a) Moist forest Figure 2.4. Biplot of canonical discriminant functions for the habitat attributes at nest-sites and random sites of Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Central Highlands of Chiapas, from 2000 to 2003. Selected variables for both models are listed in Table 2.2. (a) Moist forest (Wilk's X, u = 0.72, F5,2oi = 15.5, P< 0.001), and (b) Dry-disturbed forest (Wilk's X, u = 0.68, F 3,i4o= 21.8, P < 0.001). Internal circles correspond to 95% confidence limits for the group mean, and the external circles contain 50% of the normal contours. Biplot arrows from the grand mean (centroid) show those significant attributes selected by forward-stepwise procedure with P < 0.1. 44 0.98 -> 0.97 -0.96 -I 0.95 -S 0.94 -• Moist forest • Dry-disturbed forest 2000 2001 2002 2003 Year Figure 2.5. Daily nest survival rates (± Johnson's SE) for 78 fate-known nests from moist forest (white bars) and dry-disturbed forest (black bars) habitat types for Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Central Highlands of Chiapas, from 2000 to 2003. 45 (b) Female nest attentiveness I Incubating Nestling Breeding stage Figure 2.6. Parental activity (a, mean number of trips to the nest per hour ± 1SE) and parental nest attentiveness (b, ratio time per hour on-bouts ± 1SE) during incubating and nestling stages of successful (Moist forest - successful) and failed (Moist forest - failed) nests in the moist forest, and successful (Dry-disturbed forests - successful) and failed (Dry-disturbed forest - failed) nests in the dry-disturbed forest in the Central Highlands of Chiapas, for Catharus frantzii (Ruddy-capped Nightingale Thrush). Nest observations: 18 nests for moist forest and 15 for dry-disturbed forest habitat types. 46 R E F E R E N C E S Arcese, P., J. N. M. Smith, W. M. Hochachka, C. M. Rogers, and D. Ludwig. 1992. Stability, regulation, and the determination of abundance in an insular song sparrow population. Ecology 73:805-822. Artman, V. L , and J. F. Downhower. 2003. Wood Thrush (Hylocichla mustelina) nesting ecology in relation to prescribed burning of mixed-oak forest in Ohio. Auk 120:874-882. Badyaev, A. V., and C. K. Ghalambor. 2001. Evolution of life histories along elevational gradients: trade-off between parental care and fecundity. Ecology 82:2948-2960. Beltran, W., and G. H. Kattan. 2003. First record of the Slaty-backed Nightingale Thrush in the Central Andes of Colombia, with notes on its ecology and geographical variation. Wilson Bulletin 113:134-139. Boulton, R. L , P. Cassey, C. Schipper, and M. F. Clarke. 2003. Nest site selection by yellow-faced honeyeaters Lichenostomus chrysops. Journal of Avian Biology 34:267-274. Brawn, J. D., and S. K. Robinson. 1996. Source-sink population dynamics may complicate the interpretation of long-term census data. Ecology 77:3-12. Brown, W. P., and R. R. Roth. 2002. Temporal patterns of fitness and survival in the Wood Thrush. Ecology 83:958-969. Burke, D. M., and E. Nol. 2000. Landscape and fragment size effects on reproductive success of forest-breeding birds in Ontario. Ecological Applications 10:1749-1761. Carrete, M., J. A. Sanchez-Zapata, J. L. Telia, J. M. Gil-Sanchez, and M. Moleon. 2006. Components of breeding performance in two competing species: habitat heterogeneity, individual quality and density-dependence. Oikos 112:680-690. Chase, M. K., N. Nur, and G. R. Geupel. 2005. Effects of weather and population density on reproductive success and population dynamics in a Song Sparrow (Melospiza melodia) population: a long-term study. Auk 122:571-592. Clement, P., R. Hathway, C. Byers, and J. Wilczur. 2000. Thrushes. Christopher Helm Ltd., London, UK. Diario Oficial de la Federacion. 2002. Norma Oficial Mexicana NOM-059-ECOL-2001, Protection ambiental-Especies nativas de Mexico de flora y fauna silvestres-Categorias de riesgo y especificaciones para su inclusion, exclusion o cambio-Lista de especies en riesgo. Pages 1-85 in Diario Oficial de la Federacion. Secretaria de Medio Ambiente y Recursos Naturales, Mexico, DF. Donovan, T. M., and R. H. Lamberson. 2001. Area-sensitive distributions counteract negative effects of habitat fragmentation on breeding birds. Ecology 82:1170-1179. Easton, W. E., and K. Martin. 2002. Effects of thinning and herbicide treatment on nest-site selection by songbirds in young managed forests. Auk 119:685-694. 47 Farnsworth, G. L , K. C. Weeks, and T. R. Simons. 2000. Validating the assumptions of the Mayfield method. Journal of Field Ornithology 71:658-664. Fretwell, S. D., and H. L. Lucas. 1970. On the territorial behaviour and other factors influencing habitat distribution in birds. I. Theoretical development. Acta Biotheoretica 19:16-36. Ghalambor, C. K., and T. E. Martin. 2001. Fecundity-survival trade-offs and parental risk-taking in birds. Science 292:494-497. Gonzalez-Espinosa, M., S. Ochoa-Gaona, N. Rami'rez-Marcial, and P. Quintana-Ascencio. 1995. Current land use trends and conservation of old growth forest habitats in the highlands of Chiapas, Mexico. Pages 190-197 in M. H. Willson and S. A. Sader, editors. Conservation of Neotropical Migratory Birds in Mexico. Maine Agricultural and Forest Experimentation Station Misc., Maine. Gotelli, N. J., and A. M. Ellison. 2004. A Primer of Ecological Statistics. Sinuer Associates, Inc., Sunderland, MA. Grant, T. A., T. L. Shaffer, E. M. Madden, and P. J. Pietz. 2005. Time-specific variation in passerine nest survival: new insights into old questions. Auk 122:661-672. Gotmark, F., D. Blomqvist, O. Johansson, and J. Bergkvist. 1995. Nest site selection: a trade-off between concealment and view of the surroundings? Journal of Avian Biology 26:305-312. Harris, G. M., and S. L. Pimm. 2004. Bird species' tolerance of secondary forest habitats and its effects on extinction. Conservation Biology 18:1607-1616. Hazier, K. R. 2004. Mayfield logistic regression: a practical approach for analysis for nest survival. Auk 121:707-716. Hazier, K. R., A. J. Amacher, R. A. Lancia, and J. A. Gerwin. 2006. Factors influencing Acadian Flycatcher nestig success in an intensively managed forest landscape. Journal of Wildlife Management 70:353-360. Hensler, G. L., and J. D. Nichols. 1981. The Mayfield method of estimating nesting success: a model, estimators and simulation results. Wilson Bulletin 93:42-53. Hines, J. E., and J. R. Sauer. 1989. Program CONTRAST - a general program for the analysis of several survival or recovery rates estimates -. Fish. Wild. Tech. Rep. 24:1-7. Howlett, J. S., and B. J. Stutchbury. 1996. Nest concealment and predation in Hooded Warblers: experimental removal of nest cover. Auk 113:1-9. Johnson, D. H. 1979. Estimating nest success: The Mayfield method and an alternative. Auk 96:651-661. Laurance, W. F., and R. O. Bierregaard. 1997. Tropical Forest Remnants: Ecology, Management, and Conservation of Fragmented Communities. University of Chicago Press, Chicago. Londoho, G. A. 2005. A description of the nest and eggs of the Pale-eyed Thrush (Platycichla leucops), with notes on incubation behavior. Wilson Bulletin 117:394-399. 48 Long, A. J. 1995. The importance of tropical montane cloud forests for endemic and threatened birds. Pages 79-115 in L. S. Hamilton, J. O. Juvik, and F. N. Scatene, editors. Tropical Montane Cloud Forest. Ecological Studies 110. Spring-Verlag, New York, NY. Martin, J.-L., and M. Joron. 2003. Nest predation in forest birds: influence of predator type and predator's habitat quality. Oikos 102:641-653. Martin, T. E. 1995. Avian life-history evolution in relation to nest sites, nest predation and food. Ecological Monographs 65:101-127. Martin, T. E. 1996. Life-history evolution in tropical and south temperate birds: what do we really know? Journal of Avian Biology 27:263-272. Martin, T. E. 2001. Abiotic vs. biotic influences on habitat selection of coexisting species: climate change impacts? Ecology 82:175-188. Martin, T. E. 2004. Avian life-history evolution has an eminent past: does it have a bright future? Auk 121:289-301. Martin, T. E., and P. Li. 1992. Life history traits of open- vs. cavity-nesting birds. Ecology 73:579-592. Martin, T. E., J. Scott, and C. Menge. 2000. Nest predation increases with parental activity: separating nest site and parental activity effects. Proceedings of the Royal Society. London B 267:2287-2293. Mayfield, H. 1975. Suggestions for calculating nest success. Wilson Bulletin 87:456-466. Mazerolle, D. F., and K. A. Hobson. 2004. Territory size and overlap in male Ovenbirds: contrasting a fragmented and contiguous boreal forest. Canadian Journal of Zoology 82:1774-1781. McGarigal, K., S. Cushman, and S. Stafford. 2000. Multivariate Statistics for Wildlife and Ecology Research. Spring-Verlag, New York, NY. Morton, E. S., and B. J. Stutchbury. 2000. Demography and reproductive success in the Dusky Antbird, a sedentary tropical passerine. Journal of Field Ornithology 71:493-500. Muchai, M., and M. A. du Plessis. 2005. Nest predation of grassland bird species increases with parental activity at the nest. Journal of Avian Biology 36:110-116. Murphy, M. T. 2001. Habitat-specific demography of a long-distance, neotropical migrant bird, the Eastern Kingbird. Ecology 82:1304-1318. Naranjo, E. J., and E. Espinoza-Medinilla. 2001. Los mamfferos de la Reserva Ecologica Huitepec, Chiapas, Mexico. Revista Mexicana de Mastozoologia 5:58-67. Neter, J., M. H. Kutner, C. J. Nachtesheim, and W. Wasserman. 1996. Applied Linear Statistical Models. McGraw-Hill, Boston, MA. Newton, I. 2004. Population limitation in migrants. Ibis 146:197-226. Ochoa-Gaona, S., M. Gonzalez-Espinosa, A. A. Meave, and V. Sorani-DalBon. 2004. Effect of forest fragmentation on the woody flora of the highlands of Chiapas, Mexico. Biodiversity and Conservation 13:867-884. 49 Opdam, P., and J. A. Wiens. 2002. Fragmentation, habitat loss and landscape management. Pages 202-223 in K. Norris and D. J. Pain, editors. Conserving Bird Biodiversity: General Principles and their Application. Cambridge University Press, Cambridge, UK. Peterson, A. T., G. Escalona-Segura, and J. A. Griffith. 1998. Distribution and conservation of birds of Northern Central America. Wilson Bulletin 110:534-543. Raitt, R. J., and J. W. Hardy. 1970. Relationships between two partially sympatric species of thrushes (Catharus) in Mexico. Auk 87:20-57. Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, and D. F. DeSante. 1993. Handbook of Field Methods for Monitoring Landbirds. U.S. Forest Service Pacific Southewestern Research Station, General Technical Report PSW-GTR-144. Ramirez-Marcial, N., S. Ochoa-Gaona, M. Gonzalez-Espinosa, and P. Quintana-Ascencio. 1998. Analisis flori'stico y sucesional en la Estacion Biologica Cerro Huitepec, Chiapas, Mexico. Acta Botanica Mexicana 44:59-85. Remes, V. 2005. Birds and rodents destroy different nests: a study of Blackcap Sylvia atricapilla using the removal of nest concealment. Ibis 147:213-216. Ricklefs, R. E. 2000. Density-dependence, evolutionary optimization, and the diversification of avian life-histories. Condor 102:9-22. Ricklefs, R. E., and G. Bloom. 1977. Components of avian breeding productivity. Auk 94:86-96. Robinson, S. K., F. R. Thompson III, T. M. Donovan, and J. Faaborg. 1995. Regional forest fragmentation and the nesting success of birds. Science 267:1987-1990. Robinson, T. R., W. D. Robinson, and E. C. Edwards. 2000a. Breeding ecology and nest-site selection of song wrens in central Panama. Auk 117:345-354. Robinson, W. D., T. R. Robinson, S. K. Robinson, and J. D. Brawn. 2000b. Nesting success of understory forest birds in central Panama. Journal of Avian Biology 31:151-164. Rodenhouse, N., T. W. Sherry, and R. T. Holmes. 1997. Site-dependent regulation of population size: a new synthesis. Ecology 78:2025-2042. Rodenhouse, N. L , T. S. Sillett, P. J. Doran, and R. T. Holmes. 2003. Multiple density-dependence mechanisms regulate a migratory bird population during the breeding season. Proceedings of the Royal Society. London B 270:2105-2110. Rodriguez, A., A. M. Garcia, F. Cervera, and V. Palacios. 2006. Landscape and anti-predation determinants of nest-site selection, nest distribution and productivity in a Mediterranean population of Long-eared Owls Asio otus. Ibis 148:133-145. Roper, J. J. 1996. Nest predation and its importance for a Neotropical bird, the Western Slaty Antshrike (Thamnophilus atrinucha): The problem, experiments, and simulations. Ph. D. dissertation. University of Pennsylvania, Philadelphia, PA. 50 Rotella, J. J., S. J. Dinsmore, and T. L. Shaffer. 2004. Modeling nest-survival data: a comparison of recently developed methods that can be implemented in MARk and SAS. Animal Biodiversity and Conservation 27:1-19. Sail, J., L. Creighton, and A. Lehman. 2005. JMP Start Statistics: A Guide to Statistics and Data Analysis Using JMP and JMP IN Software, 3d edition. Thompson Learning, Belmont, Ca. SAS. 2004. Start Statistics, in. SAS Institute, Inc., Belmont, CA, USA. Schmidt, K. A., R. S. Ostfeld, and K. N. Smyth. 2006. Spatial heterogeneity in predator activity, nest survivorship, and nest-site selection in two forest thrushes. Oecologia 148:22-29. Sergio, F., and I. Newton. 2003. Occupancy as a measure of territory quality. Journal of Animal Ecology 72:857-865. Sherry, T. W., and R. T. Holmes. 1995. Summer versus winter limitation of populations: what are the issues and what the evidence? Pages 85-120 inl. E. Martin and D. M. Finch, editors. Ecology and Management of Neotropical Migratory Birds. Oxford University Press, New York, NY. Sieving, K. E., and J. R. Karr. 1997. Evian extinction and persistence mechanisms in lowland Panama. Pages 156-170 in R. O. Bierregaard and W. F. Laurance, editors. Tropical Forest Remnants: Their Ecology and Management. Chicago Press, Chicago, IL. Simberloff, D. 1994. Habitat fragmentation and population extinction of birds. Ibis 137:S105-S111. Sinclair, A. R. E. 1989. Population regulation in animals, in J. M. Cherret, editor. Ecological Concepts, The Distribution Ecology to An Undestanding of the Natural World. 29th Symposium of The British Ecological Society, Blackwell Scientific, Oxford, UK. Skutch, A. F. 1949. Do tropical birds rear as many young as they can nourish? Ibis 91:430-455. Skutch, A. F. 1960. Life histories of Central American birds. II. Pacific Coast Avifauna 34:1-593. Skutch, A. F. 1985. Clutch size, nesting success, and predation on nests of Neotropical birds, reviewed. Ornithological Monographs 36:575-594. Stanley, T. R. 2000. Modelling and estimation of stage-specific daily survival probabilities of nests. Ecology 81:2048-2053. Still, C. J., P. N. Foster, and S. H. Schneider. 1999. Simulating the effects of climate change on tropical montane cloud forests. Nature 398:608-610. Styrsky, J. N., J. D. Brawn, and S. K. Robinson. 2005. Juvenile mortality increases with clutch size in a Neotropical bird. Ecology 86:3238-3244. Saether, B.-E., and S. Engen. 2002. Pattern of variation in avian population growth rates. Phil. Trans. R. Soc. Lond B 357:1185-1195. Watson, D. M., and A. T. Peterson. 1999. Determinants of diversity in a naturally fragmented landscape: humid montane forest avifaunas of Mesoamerica. Ecography 22:582-589. Weidinger, K. 2002. Interactive effects of concealment, parental behaviour and predators on the survival of open passerine nests. Journal of Animal Ecology 71:424-437. 51 Weidinger, K. 2004. Relative effects of nest size and site on the risk of predation in open nesting passerines. Journal of Avian Biology 35:515-523. Wiens, J. A. 1976. Population responses to patchy environments. Ann. Rev. Ecol. Syst. 7:81-120. Willson, S. K. 2004. Obligate army-ant-following birds: a study of ecology, spatial movement patterns, and behaviour in Amazonian Peru. Ornithological Monographs 55:1-67. Zar, J. H. 1999. Biostatistical Analysis. Prentice Hall, New Jersey, NJ. 52 CHAPTER 3: POPULATION DYNAMICS OF THE MEXICAN ENDANGERED CATHARUS FRANTZII (RUDDY-CAPPED NIGHTINGALE THRUSH): HABITAT-SPECIFIC DENSITY, PRODUCTIVITY AND SURVIVAL INTRODUCTION Understanding changes in habitat heterogeneity and their influence on demographic parameters determining population dynamics (i.e., finite rate of population growth, X) is a key issue in conservation ecology. This is particularly important in ecosystems supporting high levels of biodiversity, and where human activities have extensively changed the environment (Myers et al. 2000, Manne and Pimm 2001). More than a third of the global avian biodiversity (36% of the total species) occurs in the Neotropics (Stotz et al. 1996), and montane habitats are key elements of this biodiversity as they contain numerous endemic species (Long 1995, Rejinfo et al. 1997, Young and MacDonald 2000). However, loss of forest habitats is modifying the distribution and abundance of many montane species (Watson 2003, Martinez-Morales 2005b, Tejeda-Cruz and Sutherland 2005). It is important to determine the current status of montane forest birds and to predict their response to habitat changes induced by human activities. The theoretical foundation for considering the relationship between habitat quality and demographic rates (i.e., per capita birth, death, emigration and immigration) is well developed for source-sink population dynamics (Pulliam 1988, Donovan and Thompson III 2001, Kawecki 2004). This approach has been applied to an increasing number of species living in landscapes with habitats that differ in quality and connected by dispersal (Norris and Pain 2002, Smith and Hellman 2002, Hanski and Gaggiotti 2004). The source-sink model distinguishes the capacity of habitat patchs to sustain stable or growing sub-populations in high-quality source patches (population growth, X > 1), or their dependency on immigrants, from source patches, to maintain patch occupancy in low-quality sink patches (X < 1; Pulliam 1996, Thomas and Kunin 1999). However, source-sink dynamics may be masked if patch quality and carrying capacity vary through time (Watkinson and Sutherland 1995, Kawecki 2004, Jonzen et al. 2005). Hence, estimating 53 demographic parameters to identify whether habitat patches act as either sources or sinks for target species is key for understanding the influence of human activities on the availability and suitability of habitats (With and King 2001, Smith and Hellman 2002). Populations that appear to be regulated by traditional density-dependent mechanisms may be the result of site-specific (i.e., territory) differences in demographic parameters (Rodenhouse et al. 1997). Both source-sink and site-dependent models of populations dynamics assume either ideal-despotic (Fretwell and Lucas 1970) or pre-emptive (Pulliam and Danielson 1991) models of individual settlement in which individuals will increase their fitness if they select the best sites. Individuals in a population differ in their capacity to acquire and hold high-quality sites, and many of them are forced to survive and breed in low-quality sites. Since natural habitats become modified through anthropogenic change, low-quality habitats become more prevalent in the landscape for many species (Donovan and Thompson III 2001). Studying variation of the population growth (X) in response to changes in demographic parameters will help us to better understand the consequences of reduction of habitat quality to population persistence (Donovan and Thompson III 2001, Holtet al. 2003). Little information exists on the demographic parameters of tropical forest birds (Ricklefs 1997, Morton and Stutchbury 2000, Willson 2004). Reports on nesting success varied from 8 to 57 percent for bird species in the lowland forests in Panama (Robinson et al. 2000, Styrsky et al. 2005), and to more than 60 percent in montane cloud forests (Skutch 1985). Estimates of survival rates of tropical forest birds also showed variation from lowland «>=0.52) to highland (cj)=0.85) forest habitats (Karr et al. 1990, Johnston et al. 1997, Ricklefs 1997, Sandercock et al. 2000). Such variation in nesting success and survival highlights the need to obtain more complete demographic data than nesting success and survival estimates (Saether and Engen 2003). In the Central Highlands of Chiapas, birds are exposed to highly seasonal rainfall patterns, cold temperatures, highly variable topography and vegetation composition and structure. Life-history theory predicts that birds breeding at high elevations are characterised by low reproductive rates and high longevity (Badyaev and Ghalambor 2001, Martin 2001a). A longer lifespan at the expense of low reproductive rates may represent a survivor life-history strategy (Saether and Bakke 2000, Norris 2004), but high reproductive rates have been hypothesized for bird populations in 54 isolated ecosystems, through re-nesting attempts after failure (Sieving and Karr 1997, Grzybowski and Pease 2005). For tropical montane forest birds, re-nesting to increase reproductive success may be a determinant of the finite rate of population growth (X). The breeding performance of Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Cerro Huitepec Biological Reserve was better in the moist forest habitat type with more understory vegetation (Chapter 2). Higher levels of nesting success and daily nest survival in the moist forest may suggest higher habitat-quality there than in the dry-disturbed forest. Human alterations likely affect population demographic rates by reducing habitat quality from local (Martin 1998, 2001b) to regional scales (Robinson etal. 1995, Brawn and Robinson 1996). Alteration in habitat quality may occur through changes in vegetation structure and composition (Brawn et al. 2001, Martin and Martin 2001, Seether and Engen 2003). However, information on the variation in demographic parameters determining population persistence of passerine, open-cup nesting species from montane cloud forests is limited. I studied C. frantzii in the Cerro Huitepec Biological Reserve, a montane cloud forest reserve in the Central Highlands of Chiapas, to evaluate habitat-specific variation in demographic parameters in two contiguous forest habitats (moist forest and dry-disturbed forest). I assessed the effects of population density, offspring production and adult annual survival on the finite rate of population growth (X) and persistence. Because of the long-term effects of selective vegetation harvesting in the region (Quintana-Ascencio et al. 2004), I predicted that the reduction of understory vegetation negatively affects bird productivity and survival. Under this prediction, the moist forest habitat type with more understory vegetation would support higher demographic rates than the dry-disturbed forest habitat type. Reducing understory vegetation may differentially affect demographic parameters, and these may result in different effects on the population growth rate (X). METHODS Study site. - The study site was the Cerro Huitepec Biological Reserve, a semi-isolated forest reserve in the Central Highlands of Chiapas, 4.5 km northwest of San Cristobal de las 55 Casas, Chiapas, Mexico (16°44'38"N; 92°40'15"W). The Huitepec reserve is 147 ha in size and contains a variety of natural and disturbed habitats ranging in elevation from 2230 m to 2710 m. The reserve includes six plant communities: grassland, shrub-land, second growth forest, dry oak forest, oak humid forest and evergreen cloud forest. For a more detailed description of the vegetation see (Ramfrez-Marcial et al. 1998). The mean annual temperature is 14.5°C and mean annual precipitation is 1300 mm. Rainfall is highly variable throughout the year, with a dry season (December - March), two transitional months (November and April) and a wet season (May -October). Firewood gathering was prevalent in the Huitepec Reserve, and the surrounding land is used for housing, agriculture and plantations of exotic pine. Open and secondary habitats around the reserve have recently increased. Fieldwork was conducted mainly from March to mid-September during 2000 to 2003, though annual survival and encounter rate data have been collected since 1995. C. frantzii, is a small (15-18 cm) forest dwelling passerine, monomorphic in size and colour, discontinuously distributed, highly sedentary and an inhabitant of the understory and dense edge undergrowth of humid montane forests from Central Mexico to Panama (Rowley and Orr 1964, Howell and Webb 1995, American Ornithologists' Union 1998, Clement et al. 2000). The species was listed as endangered on the Mexican Red List (Diario Oficial de la Federacion 2002) due to the loss of tropical montane forests in Mexico. Variation in abundance of C. frantzii throughout its range has been attributed to the distribution of remaining forest habitat and interactions with other Catharus species may limit its distribution to moist forest habitats (Raitt and Hardy 1970, Young et al. 1998, Watson 2003, Tejeda-Cruz and Sutherland 2004, Martinez-Morales 2005a). There is little information on the population structure and dynamics of C. frantzii. In the Northern and Central Highlands of Chiapas, C. frantzii occurs in a wide range of forest habitats (Raitt and Hardy 1970), and coexists with the migratory C. ustulatus (Swainson's Thrush) during autumn and spring migration (Chavez-Zichinelli 2002, Hiron 2005). The lack of year-round interactions with other nightingale Catharus thrushes may allow C. frantziiXo exploit a variety of habitat types in the Central Highlands of Chiapas. 56 I studied C. frantzii during two intervals: i) spring, summer, fall and winter of 1995-1999 at three main vegetation plots within the reserve; and ii) spring, summer and fall (mid March to early September) of 2000-2003 at the same sites plus two additional plots within the reserve. Moist forest included montane cloud forest (= 35 ha), wet-oak forest (~ 32 ha) and riparian forest (~ 10 ha), while dry-disturbed forest included dry-disturbed oak forest (= 40 ha) and second-growth (disturbed) forest (= 30 ha). These forest types varied in composition and structure, and show gradients in humidity, productivity, and disturbance history (Chapter 2). Intensive spot-mapping of singing males was used during mid-March to late May to establish territories and estimate the number of breeding pairs within the reserve (Ralph et al. 1993, Brawn and Robinson 1996). This method incorporates visual observations and songs to determine male territory boundaries (Brawn and Robinson 1996). I also broadcasted songs to verify territory ownership and reduce the chances of overestimating densities by including songs of unpaired males. Seasonal fecundity (i.e., female offspring per pair per breeding season) was estimated from the number of fledglings produced by known-fate nests (Chapter 2), and assuming a 1:1 sex ratio. The offspring fledged from a successful nest was the number of chicks recorded during the last visit before fledging. In most cases, I confirmed this number by observing them or their parents feeding them near the nest. However, reproductive output is also influenced by the number of nesting attempts within a breeding season (Anders and Marshall 2005). Thus, I combined data of female offspring per nest with an assumption of 1.5 nesting attempts per season to estimate seasonal productivity per habitat: # = ( H ) ( l - [ l - m r ) where (3 was the female offspring produced per female within a breeding season in /' habitat, n was the number of female offspring fledged per successful nest, m was the Mayfield estimate of overall daily nest survival (Chapter 2), and a was the mean number of nesting attempts per female during a breeding season (Anders and Marshall 2005). The number of nesting attempts (1.5) for C. frantzii at Huitepec was estimated considering the length of the breeding season (April 18 - August 18, = 110 d), the length of the nesting period (~ 33 d) and the period between nest-loss and re-nesting (~ 57 29 d), which suggested a limited number of re-nesting attempts (i.e., 0.5) after nest failures. Also, I assumed that individuals only had the potential to produce one successful brood per season. Capture-recapture-and-re-sight data collected over nine years were used to estimate the overall return rate. This rate is separated into apparent or local survival (())) and encounter (p) rates (Marshall et al. 1999, White and Burnham 1999). Individual birds were captured by a combination of passive and targeted mist netting. Three of the five passive mist-netting plots were the same through the nine-year study and an additional plot was added to the moist and dry-disturbed forests in 2000, respectively. Mist-netting plots were 500 - 900 m apart from each other. Birds were passively caught in a two-week period during spring, summer, fall and winter of 1995-1999, and spring and late summer of 2000-2003. In each plot, 10 to 12 mist-nets (2.5 x 12.5 m, 36-mm mesh) were placed in groups of one to three along trails and cut mist-netting lines. Mist-net lines were opened just before sunrise (06:15 - 06:30) and checked every 30-40 min for a five-hour period. Birds were weighed, colour banded and released at the site of capture. Apparent (local) survival probability (<)), hereafter survival, i.e. probability that a marked individual is still alive in the sampling area after a given unit of time) and encounter probability (p, i.e., percent chance that a marked individual is encountered, conditional of being alive within the study plot) were estimated using the Cormack-Jolly-Seber (CJS) mark-recapture-resighting models (Minta and Mengel 1989, Lebreton et al. 1992, Hastings 1997). Survival and encounter rate models included constant (<|)., p.) and time-dependent habitat- and sex- effects (§ h - s - u ph-s*<) (Sandercock ef al. 2000, Willson 2004). Estimates of sex and habitat interaction terms were based on models including only those individuals identified as either male or female (84 males and 48 females), while the model of overall adult survival and encounter probabilities included individuals captured as second year (SY) and after second year (ASY) (n = 196). Juvenile survival and encounter rates were calculated separately with a constant survival model (n = 25). Models for survival and encounter rates were selected for fit on the basis of the number of parameters and deviance, and the maximum likelihood approach with the lowest value for Aikaike's Information Criterion (Turchin 1998, Burnham and Anderson 2002). Over-dispersion of data (i.e., lack of independence) was evaluated with a goodness-of-fit test (GOF) for the completely 58 parameterized, time dependent model (tyh-s-t, ph-s-t) using the program RELEASE (Version 3.0, (Burnham etal. 1987). Encounter histories of individuals were analyzed using program MARK (White and Burnham 1999), following the approach outlined by (Lebreton et al. 1992). Normalized Akaike weights (Wj) were used to evaluate the relative support for different models in the candidate model set. These weights can be interpreted as the evidence that a particular model is most appropriate given the data and the set of models considered (Burnham and Anderson 2002). I considered model averaging when estimating time variation in survival and encounter rates (1995-2003) from the different models listed in Table 3.1. I assumed that models with AIC units differing by < 2 units similarly fit the data (Doncaster etal. 1997). Demographic parameters were related using Pulliam's model (Pulliam 1988, Pulliam and Danielson 1991, Pulliam 1996) to estimate population growth (X): where $ , d u K represents the estimated adult survival rate, $ U veni ie was the estimated juvenile survival rate and p represents seasonal fecundity. I used general linear models (GLM) two-way analysis of variance ANOVA: Type III sum of squares with a complete randomised design to test the effects of habitat (moist vs. dry-disturbed), year (2000-2003) and their interactions (Underwood 1997, Zar 1999) on breeding population density, seasonal fecundity and adult survival. Multiple regression analysis was performed to examine the relative effects of pair density, seasonal fecundity and adult survival on the population growth (X) across years. The original models containing two- and three-way interaction terms among pair density*seasonal fecundity*adult survival on X were reduced by sequentially discarding non-significant interaction terms ( P > 0.05). Prior to statistical analyses, variables were checked for statistical normality with Shapiro-Wilks W test and for equality of variance with Bartlet's test (Gotelli and Ellison 2004). When required, data were square root-transformed (distances) or arcsine-transformed (ratios) as appropriate to meet parametric assumptions. When factor interactions were significant, the role of each factor was evaluated independently with an univariate analysis. Statistical analyses were performed with SAS-JMP (Sail etal. 2005). All means are presented ± 1 SE, and considered significant at a = 0.05. 59 RESULTS The pair density (pairs/ha) of C. frantzii varied across habitats at Huitepec, being significantly higher in the moist forest than in the dry-disturbed forest (0.57 ±0.08 SE vs. 0.39 ± 0.07, ANOVA, F 1 i 6 = 12.2, P < 0.05). The mean number of fledglings was higher in the moist forest than the dry-disturbed forest (8.0 ± 1.37 vs. 3.25 ± 1.37; F 1 i 6 = 6.1, P < 0.05). However, the mean productivity (female offspring per pair per breeding season) did not differ between the moist forest and the dry-disturbed forest (0.13 ± 0.03 vs. 0.11 ± 0.03; F 1 6 = 0.2, P = 0.7). The pair density was not significantly associated with productivity in either the moist forest {r= 0.3, P= 0.5), or the dry-disturbed forest (r= -1.02, P-0.1), or both habitats combined (r =-0.02, P =0.7) . The negative association between pair density and productivity in the dry-disturbed forest may indicate possible density-dependent productivity in the dry-disturbed forest. The sampling effort (net/hours/season) was significantly lower during the first interval than the second interval (1995-1999 vs. 2000-2003; mean = 330 ± 22.7 SE vs. 575 ± 25.35), and higher in the moist habitat than in the dry-disturbed habitat (511 ± 24.05 vs. 367 ± 24.05; GLM-ANOVA, R 2 = 0.84, F 3 i l 4 = 23.38, P< 0.001). Between 1995-2003, 221 individual thrushes were permanently banded with uniquely numbered metal leg bands: 93 males, 53 females and 75 unknown (Fig. 3.1 .a). Twenty-five birds of unknown sex were either haching year (HY) or second year (SY: Fig. 3.1 .b). A combination of three coloured-plastic bands was added to individuals captured or recaptured from 2000 to 2003. The study population had a male-biased sex ratio and 35 % of the individuals showed no evidence of being breeders. Most banded birds were captured in the moist forest, particularly from 1995-1998 and 2002-2003. During 2000-2001 I captured more individuals in the dry-disturbed forest than in the moist forest (Fig. 3.1 .c). A high proportion of the individuals in the sampled population (66%), were captured once and never re-observed. The mean ratio of once-captured individuals (potentially non-breeders, floaters or transients) over the total captures did not vary annually between intervals (1995-1999 vs. 2000-2003; 0.34 ± 0.06 vs. 0.45 ± 0.07; F 1 | 1 6 = 1.36, P = 0.27), or between the moist forest and the dry-disturbed forest (0.45 ± 0.06 vs. 0.33 ± 0.06; F 1 i 1 6 = 1.9, P = 0.18). No interaction between 60 habitat and intervals was detected for the mean ratio of once-captured individuals (GLM, R2 = 0.19, F 3 i 1 4 = 0.01, P= 0.9). Of the re-captured or re-sighted individuals, 67 showed high levels of philopatry (usually re-captured in the same net line within plots), with just 9 individuals (13% of recaptures) recaptured in a different plot from where it was originally banded. Four of these latter individuals moved or dispersed within the same habitat type, and five moved to a different habitat. One individual that moved across habitats was originally banded in the moist forest and recaptured or sighted in the dry-disturbed forest (~ 700 m), while the remaining 4 birds moved in the opposite direction (400 -700 m). Variation in the sampling effort may have influenced the probabilities of inter habitat recapture, with higher probably of re-capturing individuals from the moist forest in the dry-disturbed forest, than vice versa. All completely parameterized models to estimate survival and encounter rates met the assumptions of mark-recapture methods based on goodness of fit tests (x 2 s 1.64, P > 0.19). The best models for adult survival had constant- and time-dependent survival and encounter rates (Table 3.1). Variation in adult survival and encounter rates suggested possible stability through time in both parameters, except for a large number of individuals captured in 2001 (Fig. 3.2.a). The adult survival in the reserve was (j> = 0.79. Neither adult survival, nor encounter rate varied from the moist forest to the dry disturbed forest (Table 3.2). Moreover, males and females exhibited similar survival across moist forest = 0.8 males and § = 0.77 females) and dry-disturbed forest = 0.83 males and <]> = 0.79 females; Fig. 3.2.b). Encounter rates did also not differ between sexes in the moist forest (p = 0.42 males and p = 0.35 females) and in the dry-disturbed forest (p = 0.31 males and p = 0.25 females). The survival for juvenile birds was = 0.67 and the encounter rate was p = 0.39, with model of constant survival and encounter rate. The breeding population at Huitepec averaged 71 pairs over the 4-year study period (Table 3.3), with a minimum value of 66 pairs in 2001, and a maximum value of 78 pairs in 2003. The estimated seasonal productivity varied from 0.74 to 0.99 (females offspring produced per successful pair) in the moist forest and from 0.82 to 0.99 in the dry-disturbed forests (Table 3.3). These numbers were low in 2001 and reached maximum values in 2000 in both habitats. 61 Estimation of the population growth indicated a stable population for C. frantzii at Huitepec (X = 1.09), when considering survival of <|> = 0.79 and a total productivity of 0.91 female offspring per successful pair per breeding season. The population growth showed slight annual variation within and across habitats (Table 3.4), with a range of 0.88 to 1.21 in the moist forests and 0.91 to 1.23 in the dry-disturbed forest. The mean population growth in the moist forest did not vary from the dry-disturbed forest (1.08 ±0.07 vs. 1.1 ± 0.07, F 1 i 6 = 0.04, P= 0.85). By performing multiple correlation analysis, I evaluated if annual variation in the population growth (X) was associated with pair density, seasonal fecundity and survival for the overall population at Huitepec (GLM ANOVA, Ft2 = 0.98, F 3, 4 = 103.6, P < 0.001). Pair density was not correlated with X (r- -0.05, F 1 4 = 0.79, P< 0.42, Fig. 3.3.a). Seasonal fecundity showed a significant negative correlation with X (r- -0.33, F 1 4 = 7.91, P = 0.04, Fig. 3.3.b). In contrast, survival had a positive and strong correlation with X (r= 0.93, F 1 4 = 310.7, P < 0.001, Fig. 3.3.c). Neither two- or three-way interaction terms were found among density, seasonal fecundity and survival in their association with X ( F 1 4 < 0.06, P s 0.82). D I S C U S S I O N The parameters determining the population growth of C. frantzii m a montane cloud forest reserve were likely similar across habitat types differing in understory vegetation. Although the moist forest supported higher pair densities than the dry-disturbed forest, productivity (female offspring/pair/year) and apparent adult survival did not vary between these habitat types, indicating similar habitat-quality for the C. frantzii at the Cerro Huitepec Ecological Reserve. Disturbance did not appear to affect negatively the processes determining the population dynamics of the C. frantzii. These results agree with previous reports on the unclear effects of habitat variation in demographic rates in bird populations of temperate forests (Holmes etal. 1996, Murphy 2001) and tropical lowland forests (Morton and Stutchbury 2000, Willson 2004). Despite higher nest success and daily nest survival rates in the moist forest compared to the dry-disturbed forest (Chapter 2), demographic rates such as productivity and adult survival did not differ between habitats. Pair density and the number of fledglings of C. frantzii varied across 62 habitats at Huitepec, being higher in the moist forest than the dry-disturbed forest. C. frantzii may select moist forest over the dry-disturbed forest because higher vegetation density provides more breeding opportunities and food (Gonzalez-Espinosa et al. 1995, Moron-Rios 2005). Differences in nest survival, nesting success and number of fledglings appeared to be important determinants of habitat selection for C. frantzii(Chapter 2). However, these components of reproductive success do not necessarily represent high-quality of habitats (Donovan and Thompson III 2001), since selection of moist forest habitats did not significantly increase per capita productivity and survival (i.e., individual fitness). A recent study has shown none significant association between nesting success and seasonal productivity in a Wood Thrush (Hylocychla mustelina) population from mixed deciduous upland forest in eastern North America (Underwood and Roth 2002). An alternative interpretation is that nests of C. frantzii in the moist forest are more difficult to find by observers (and potential predators) than in the dry-disturbed forest. This could influence productivity since species with cryptic nests in dense habitats, such as nightingale thrushes inhabiting montane cloud forests (Beltran and Kattan 2003), may be underestimated (Anders and Marshall 2005). Although, more total fledglings were produced in the moist forest, seasonal productivity did not indicate variation in breeding success between habitat types. One potential explanation is that some proportion of the pairs in the reserve may not breed. Indeed, pair density was not correlated with nest density (r = 0.09, F 1 6 = 0.05, P = 0.84). Pair density and nest density may result in over- and under-estimation of productivity, respectively. Thus, an intermediate value resulting from pair density and nest density that reflects a more feasible breeding population may increase the accuracy of productivity estimates (Anders and Marshall 2005). Since more females produced at least one fledgling in the moist forest than in the dry-disturbed forest, none significant differences in female fledglings per capita across habitats may suggest a negative density-dependent feedback on female productivity (Newton 1998). Adult survival and encounter rates did not vary across habitats and sexes, and were stable through the study period, except in 2001 (Fig. 3.1). In 2001, a high proportion of individuals were captured only once (66%), influencing survival estimates (Figure 3.2a). These may have been non-territorial individuals, floaters from the local population, or transient floaters being present in the 63 reserve only for short periods of time (Newton 1998). The presence of floaters may ensure recruitment, particularly of males, to the breeding population. Most recaptured individuals showed high levels of philopatry (88%) and were recaptured or re-sighted in the same site across the study. Movements or dispersal between habitats were infrequent (6% of recaptured individuals), with the primary direction of habitat shifts from dry-disturbed to moist forest (4/5 individuals). These individuals moved in this direction possibly in search of a better breeding opportunity (Chapter 2), when many bird species prospect for different or better places to breed after a nesting failure and they may explore sites that have an apparent higher quality (Murphy 2001). Also, because the shape of the study area and proximity of habitats, individuals may occasionally venture into neighbouring habitats. Thus, recorded movements may not necessarily represent dispersal at the observed spatial level (Holt et al. 2003, Schaefer 2006). Additional proximate factors and mechanisms explaining variation in adult survival across habitats may include conspecific and intraspecific competition in the moist forest that increases with increasing densities of competitors (Holmes et al. 1996). The moist forest appeared to support a richer predator community that may signify a higher risk of predation and compensate for seasonal productivity and adult survival. Other factors and mechanisms that may influence the estimated seasonal productivity and adult survival between habitats include chick development rates, juvenile survival and recruitment rates (Styrsky et al. 2005). My data suggest that individual fitness does not appear to vary from the moist forest to the dry-disturbed forest, when the former habitat is relatively undisturbed. My findings on the relationships between productivity and survival suggest that both habitat types are source habitats; thus, the entire reserve functions as a source. Individuals occur in source habitats due to environmental conditions for dispersal, knowledge, pre-emptive habitat selection, occupancy of better habitats, and stable population dynamics (Kruger and Lindstrom 2001, Holt et al. 2003, Jonzen et al. 2005). Variability in habitat conditions induced by changes in vegetation did not appear to affect population stability of C. frantzii in the Huitepec reserve (Opdam and Wiens 2002), despite the fact that declines in insect abundance have been recorded for the dry season (Chavez-Zichinelli 2002, Hiron 2005), and from moist to dry habitats (Moron-Rios 2005). Reduction in insect 64 abundance may force individuals to move further in search of better feeding areas. Thus, local movements by a portion of individuals within the population are likely. Productivity and apparent survival were important demographic parameters that explained fluctuations in population growth (X) of C. frantzii at Huitepec during the study period. However, this conclusion should be considered cautiously given the limited years of data for estimating productivity relative to adult survival. Adult survival showed a strong positive effect on X. Thus, habitat changes that negatively affect adult survival in the C. frantzii population at the Huitepec reserve will depress population densities. My results, that changes in X are explained more by changes in adult survival rather than productivity, support the findings from several other studies of tropical forest birds (Keast and Morton 1980, Stiles 1992, Morton and Stutchbury 2000, Willson 2004). C O N C L U S I O N Declines in montane cloud forest habitats in the Central Highlands of Chiapas have been shown to affect the distribution, abundance and habitat selection of resident and migratory birds (Gonzalez-Espinosa et al. 1995, Rappole et al. 2000). Basic demographic information needed for species conservation in the region is lacking for most resident species. I provided evidence that adult survival is the primary demographic trait driving X for the population of C. frantzii at the Huitepec reserve. Knowledge of the effects of factors determining population stability in the reserve, particularly the influence of dispersing or floater individuals, may provide better insights to understanding the relationships among demographic traits and their effects on population growth (X). It is important to note that the relationships between habitat quality and demographic traits that I presented here come from an apparent source population, breeding in a relatively heterogeneous, high quality habitat. 65 Table 3 .1 . Model selection results from program MARK for adult apparent survival (<\>) and encounter rates (p) for Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas, between 1995 and 2003. Only the five best models and the fully parameterized model are presented (f = year, n = habitat, s = sex, AlCc = Aikaike's Information Criterion, Deviance = model fit). Sample size = 196 individuals. Parameters Model AlCc AAICC AIC cWeight in Model Deviance (A) , p. 473.386 0.00 0.559 7 231.571 (B) <>. , p, 474.179 1.79 0.223 7 228.010 (C) <|>. , p. 475.965 2.58 0.154 2 244.66 (D) <|>f,p, 476.442 3.06 0.121 14 218.985 (E) h, p. 478.024 4.64 0.055 3 244.658 (F) §h+s+t i Ph+s 489.543 16.16 0.001 "•3 234.391 66 Table 3.2. Estimates of apparent adult apparent survival (<|>) and encounter rates (p) for Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas, from 1995 to 2003. Sample sizes are shown in parentheses. A. Adult survival in Moist forest (n = 132) Parameter Estimate SE Lower Upper 1:ct) 0.79 0.04 0.69 0.86 2 :p 0.38 0.05 0.28 0.49 B. Adult survival in Dry-disturbed forest (n = 64) Parameter Estimate SE Lower CI Upper CI 1:<|. 0.79 0.06 0.65 0.88 2 :p 0.32 0.07 0.20 0.46 C. Overall adult survival at Huitepec (n = 196) Parameter Estimate SE Lower CI Upper CI 1:<|> 0.79 0.04 0.71 0.86 2 :p 0.36 0.04 0.28 0.44 67 Table 3.3. Demographic parameter estimates over a four-year period (2000-2003) for Catharus frantzii (Ruddy-capped Nightingale Thrush) in two forest-habitat types (moist forest and dry-disturbed forest) at the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas. Seasonal fecundity represents the number of female offspring produced per breeding female within a breeding season. Seasonal productivity (p) represents the number of female offspring produced per successful female by the end of the breeding season. The intrinsic population growth rate (X) £ 1 may indicate a source population, while < 1 may indicate a sink population. Intrinsic Seasonal rate of Apparent Breeding Pair fecundity Seasonal populatic survival pairs density (female productivity growth Year (<|)) (number) (pairs*ha) offspring*female) (P) (X) A. Moist forest 2000 0.748 38 0.49 0.05 0.99 1.081 2001 0.918 46 0.59 0.16 0.74 1.167 2002 0.574 40 0.52 0.17 0.91 0.879 2003 0.902 51 0.66 0.15 0.91 1.207 B. Dry-disturbed forest 2000 0.748 31 0.44 0.03 0.99 1.080 2001 0.918 20 0.29 0.23 0.82 1.195 2002 0.574 30 0.43 0.12 0.99 0.907 2003 0.830 27 0.39 0.07 0.98 1.234 C. Total Populat ion 2000 0.736 69 0.47 0.04 0.99 1.069 2001 0.928 66 0.45 0.18 0.78 1.189 2002 0.565 70 0.48 0.15 0.93 0.877 2003 0.901 78 0.53 0.12 0.92 1.210 68 1995 1996 1997 1998 1999 Year 2000 2001 2002 2003 Figure 3 .1 . Total number of individuals (N = 221) of Catharus frantzii (Ruddy-capped Nightingale Thrush) banded in the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas each year from 1995 to 2003, by (a) sex, (b) age, and (c) habitat. 69 Figure 3.2. Estimates for Catharus frantzii (Ruddy-capped Nightingale Thrush) of (a) adult apparent survival (<(>) and (b) encounter rates (p) of birds initially banded as adults (males = 93, females = 53, and unknown = 75) at the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas, from 1995 to 2003. Point estimates were derived from the fully time-dependent Cormack-Jolly-Seber (CJS) Model (<t)habitaft,Phabitat*t)-70 (a) Pair density 1.3 1.2 1.1 1.0 H 0.! o.8 -r Pair density - I .7 (b) Seasonal productivity 0.8 -r . 0 0 .05 .10 .15 .20 Productivity (c) Apparent survival 0.8 -r i .25 Apparent survival Figure 3.3. Finite rate of population growth (X) vs. (a) pair density, (b) seasonal productivity (female offspring per pair per season), and (c) apparent survival of Catharus frantzii (Ruddy-capped Nightingale Thrush) at the Cerro Huitepec Ecological Reserve, Central Highlands of Chiapas, from 2000 to 2003. Finite rate of population growth (!) is plotted as a residual where the effect of one of the three independent variables in the overall model was removed through regression. Pair density: r= -0.05, F 1 | 4 = 0.79, P = 0.42; seasonal productivity: r= -0.33, F , 4 = 7.91, P = 0.05; apparent survival: r= 0.93, F 1 4 = 310.7, P < 0.001; Model R2 = 0.98, F 3 4 = 103.6, P < 0.001. 71 R E F E R E N C E S American Ornithologists' Union. 1998. Check-list of North American Birds, 7th. edition. American Ornithologists' Union, Washington, DC. Anders, A. D., and M. R. Marshall. 2005. Increasing the accuracy of productivity and survival estimates in assessing landbird population status. Conservation Biology 19:66-74. Badyaev, A. V., and C. K. Ghalambor. 2001. Evolution of life histories along elevational gradients: trade-off between parental care and fecundity. Ecology 82:2948-2960. Beltran, W., and G. H. Kattan. 2003. First record of the Slaty-backed Nightingale Thrush in the Central Andes of Colombia, with notes on its ecology and geographical variation. Wilson Bulletin 113:134-139. Brawn, J. D., and S. K. Robinson. 1996. Source-sink population dynamics may complicate the interpretation of long-term census data. Ecology 77:3-12. Brawn, J. D., S. K. Robinson, and F. R. Thompson III. 2001. The role of disturbance in the ecology and conservation of birds. Annu. Rev. Ecol. Syst. 32:251-276. Burnham, K. P., and D. R. Anderson. 2002. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. Springer-Verlag, New Yor, NY. Burnham, K. P., D. R. Anderson, G. C. White, C. Brownie, and K. H. Pollock. 1987. Design and analysis methods for fish survival experiments based on realese-recapture. American Fisheries Society, Monograph 5:1-437. Chavez-Zichinelli, C. A. 2002. Interacciones entre dos especies de zorzales (Catharus frantzii y C. ustulatus) que coexisten temporalmente en bosques de montana en los Altos de Chiapas. Tesis Licenciatura. Benemerita Universidad Autonoma de Puebla, Puebla, Mexico. Clement, P., R. Hathway, C. Byers, and J. Wilczur. 2000. Thrushes. Christopher Helm Ltd., London, UK. Diario Oficial de la Federacion. 2002- Norma Oficial Mexicana NOM-059-ECOL-2001, Protection ambiental-Especies nativas de Mexico de flora y fauna silvestres-Categorias de riesgo y especificaciones para su inclusion, exclusion o cambio-Lista de especies en riesgo. Pages 1-85 in Diario Oficial de la Federacion. Secretaria de Medio Ambiente y Recursos Naturales, Mexico, DF. Doncaster, C. P., J. Clobert, B. Doligez, G. L., and E. Danchin. 1997. Balanced dispersal between spatially varying local populations: an alternative to the source-sink model. American Naturalist 150:425-445. Donovan, T. M., and F. R. Thompson III. 2001. Modelling the ecological trap hypothesis: a habitat and demographic analysis for migrant songbirds. Ecological Applications 11:871-882. Fretwell, S. D., and H. L. Lucas. 1970. On the territorial behaviour and other factors influencing habitat distribution in birds. I. Theoretical development. Acta Biotheoretica 19:16-36. 72 Gonzalez-Espinosa, M., S. Ochoa-Gaona, N. Ramirez-Marcial, and P. Quintana-Ascencio. 1995. Current land use trends and conservation of old growth forest habitats in the highlands of Chiapas, Mexico. Pages 190-197 in M. H. Willson and S. A. Sader, editors. Conservation of Neotropical Migratory Birds in Mexico. Maine Agricultural and Forest Experimentation Station Misc., Maine. Gotelli, N. J., and A. M. Ellison. 2004. A Primer of Ecological Statistics. Sinuer Associates, Inc., Sunderland, MA. Grzybowski, J. A., and C. M. Pease. 2005. Renesting determines seasonal fecundity in songbirds: what do we know? what should we assume? Auk 122:280-291. Hanski, I., and O. E. Gaggiotti. 2004. Ecology, Genetics, and Evolution of Metapopulations. Elsevier Academic Press, Burlington, MA. Hastings, A. 1997. Population Biology: Concepts and Models. Springer, New York. Hiron, M. 2005. Coexistence of resident and migratory Catharus thrushes in the highlands of Chiapas, Mexico: on the potential of interspecific competition in mountain forests of the northern Neotropics. MSc. Thesis. Umea University, Umea, Sweden. Holmes, R. T., P. P. Marra, and T. W. Sherry. 1996. Habitat-specific demography of breeding Black-Throated Blue Warblers (Dendroica caerulescens): implications for population dynamics. Journal of Animal Ecology 65:183-195. Holt, R. D., M. Barfield, and A. Gonzalez. 2003. Impacts of environmental variability in open populations and communities: "inflation" in sink environments. Theoretical Population Biology 64:315-330. Howell, S., and S. Webb. 1995. A Guide to the Birds of Mexico and Northern Central America. Oxford University Press, Oxford, UK. Johnston, J. P., W. J. Peach, R. D. Gregory, and S. A. White. 1997. Survival rates of tropical and temperate passerines: a trinidadian perspective. The American Naturalist 150:771-789. Jonzen, N., J. R. Rhodes, and H. P. Possingham. 2005. Trend detection in source-sink systems: when should sink habitats be monitored? Ecological Applications 15:326-334. Karr, J. R., J. D. Nichols, M. Katheleen, Klimkiewicz, and J. D. Brawn. 1990. Survival rates of birds of tropical and temperate forests: will the dogma survive? The American Naturalist 136:277-291. Kawecki, T. J. 2004. Ecological and evolutionary consequences of source-sink population dynamics. Pages 387-411 in I. Hanski and O. E. Gaggiotti, editors. Ecology, Genetics, and Evolution of Metapopulations. Elsevier Academic Press, Burlington, MA. Keast, A., and E. S. Morton. 1980. Migrant birds in the neotropics: ecology, behavior, distribution, and conservation: proceedings of a symposium held at the Conservation and Research Center, National Zoological Park, Smithsonian Institution, October 27-29, 1977. Smithsonian Institution Press, Washington, D.C. 73 Kruger, O., and J. Lindstrom. 2001. Habitat heterogeneity affects population growth in goshawk Accipiter gentilis. Journal of Animal Ecology 70:173-181. Lebreton, J.-D., K. P. Burnham, J. Clobert, and D. R. Anderson. 1992. Modelling survival and testing biological hypothesis using marked animals: a unified approach with case studies, Ecological Monographs 62:67-118. Long, A. J. 1995. The importance of tropical montane cloud forests for endemic and threatened birds. Pages 79-115 in L. S. Hamilton, J. O. Juvik, and F. N. Scatene, editors. Tropical Montane Cloud Forest. Ecological Studies 110. Spring-Verlag, New York, NY. Manne, L. L., and S. L. Pimm. 2001. Beyond eight forms of rarity: which species are threatened and which will be the next? Animal Conservation 4:221-229. Marshall, M. R., R. R. Wilson, and R. J. Cooper. 1999. Estimating survival of Neotropical-Neartic migratory birds: are they dead or just dispersed? in R. Bonney, D. N. Pashley, R. J. Cooper, and L. Niles, editors. Strategies for Bird Conservation: The Partners in Flight Planning Process. Cornell Laboratory of Ornithology, Ithaca, NY. Martin, K. 2001a. Wildlife in alpine and sub-alpine habitats. Pages 285-310 in D. H. Johnson and T. A. O'Neil, editors. Wildlife-habitat relationships in Oregon and Washington. Oregon State University Press, Corvallis, OR. Martin, P. R., and T. E. Martin. 2001. Ecological and fitness consequences of species coexistence: a removal experiment with wood warblers. Ecology 82:189-206. Martin, T. E. 1998. Are microhabitat preferences of coexisting species under selection and adaptive? Ecology 79:656-670. Martin, T. E. 2001b. Abiotic vs. biotic influences on habitat selection of coexisting species: climate change impacts? Ecology 82:175-188. Martinez-Morales, M. A. 2005a. Landscape patterns influencing bird assemblage in fragmented Neotropical cloud forests. Biological Conservation 121:117-126. Martfnez-Morales, M. A. 2005b. Nested species assemblages as a tool to detect sensitivity to forest fragmentation: the case of cloud forest birds. Oikos 108:634-642. Minta, S., and M. Mengel. 1989. A simple population estimate based on simulation for capture-recapture and capture-resight data. Ecology 70:1738-1751. Morton, E. S., and B. J. Stutchbury. 2000. Demography and reproductive success in the Dusky Antbird, a sedentary tropical passerine. Journal of Field Ornithology 71:493-500. Moron-Rios, A. 2005. Los efectos del manejo forestal en la fauna de invertebrados del suelo. Ciencia y Tecnologia en la Frontera 3:38-40. Murphy, M. T. 2001. Habitat-specific demography of a long-distance, neotropical migrant bird, the Eastern Kingbird. Ecology 82:1304-1318. Myers, N., R. Mittermeier, C. Mittermeier, G. da Fonseca, and J. Kent. 2000. Biodiversity hot spots for conservation priorities. Nature 403:853-858. 74 Newton, 1.1998. Population Limitation in Birds. Academic Press, London, UK. Norris, K. 2004. Managing threatened species: the ecological toolbox, evolutionary theory and declining-population paradigm. Journal of Applied Ecology 41:413-426. Norris, K., and D. J. Pain, editors. 2002. Conserving Bird Biodiversity: General Principles and their Application. Cambridge University Press, Cambridge, UK. Opdam, P., and J. A. Wiens. 2002. Fragmentation, habitat loss and landscape management. Pages 202-223 in K. Norris and D. J. Pain, editors. Conserving Bird Biodiversity: General Principles and their Application. Cambridge University Press, Cambridge, UK. Pulliam, H. R. 1988. Sources, sinks, and population regulation. The American Naturalist 132:652-661. Pulliam, H. R. 1996. Sources and sinks: empirical evidence and population consequences. Pages 45-69 in O. E. Rhodes, R. K. Chesser, and M. H. Smith, editors. Population Dynamics in Ecological Space and Time. The University of Chicago Press, Chicago, IL. Pulliam, H. R., and D. J. Danielson. 1991. Sources, sinks, and habitat selection: a landscape perspective on population dynamics. The American Naturalist 137 (Supplement):S50-S66. Quintana-Ascencio, P. F., N. Ramirez-Marcial, M. Gonzalez-Espinosa, and M. Martinez-lco. 2004. Sapling survival and growth of coniferous and broad-leaved trees in successional highland habitats in Mexico. Applied Vegetation Science 7:81-88. Raitt, R. J., and J. W. Hardy. 1970. Relationships between two partially sympatric species of thrushes (Catharus) in Mexico. Auk 87:20-57. Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, and D. F. DeSante. 1993. Handbook of Field Methods for Monitoring Landbirds. U.S. Forest Service Pacific Southewestern Research Station, General Technical Report PSW-GTR-144. Ramirez-Marcial, N., S. Ochoa-Gaona, M. Gonzalez-Espinosa, and P. Quintana-Ascencio. 1998. Analisis flon'stico y sucesional en la Estacion Biologica Cerro Huitepec, Chiapas, Mexico. Acta Botanica Mexicana 44:59-85. Rappole, J. H., D. I. King, and P. Leimgruber. 2000. Winter habitat and distribution of the endangered golden-cheeked warbler (Dendroica chrysoparia). Animal Conservation 2:45-59. Rejinfo, L. M., G. P. Sevat, J. M. Goerk, B. A. Loiselle, and J. G. Blake. 1997. Patterns of species composition and endemism in the northen Neotropics: a case for conservation of montane avifaunas. Ornithological Monographs 48:577-594. Ricklefs, R. E. 1997. Comparative demography of the new world population of thrushes (Turdus spp.). Ecological Monographs 67:23-43. Robinson, S. K., F. R. Thompson III, T. M. Donovan, and J. Faaborg. 1995. Regional forest fragmentation and the nesting success of birds. Science 267:1987-1990. 75 Robinson, W. D., T. R. Robinson, S. K. Robinson, and J. D. Brawn. 2000. Nesting success of understory forest birds in central Panama. Journal of Avian Biology 31:151-164. Rodenhouse, N., T. W. Sherry, and R. T. Holmes. 1997. Site-dependent regulation of population size: a new synthesis. Ecology 78:2025-2042. Rowley, J. S., and R. T. Orr. 1964. The status of Frantzius1 Nightingale Thrush. Auk 81:308-314. Sail, J., L. Creighton, and A. Lehman. 2005. JMP Start Statistics: A Guide to Statistics and Data Analysis Using JMP and JMP IN Software, 3d edition. Thompson Learning, Belmont, Ca. Sandercock, B. K., S. R. Beissinger, S. H. Stoleson, R. R. Melland, and C. R. Hughes. 2000. Survival rates of a neotropical parrot: implications for latitudinal comparisons of avian demography. Ecology 81:1351-1370. Schaefer, J. A. 2006. Towards maturation of the population concept. Oikos 112:236-240. Sieving, K. E., and J. R. Karr. 1997. Evian extinction and persistence mechanisms in lowland Panama. Pages 156-170 in R. O. Bierregaard and W. F. Laurance, editors. Tropical Forest Remnants: Their Ecology and Management. Chicago Press, Chicago, IL. Skutch, A. F. 1985. Clutch size, nesting success, and predation on nests of Neotropical birds, reviewed. Ornithological Monographs 36:575-594. Smith, J. N. M., and J. J. Hellman. 2002. Population persistence in fragmented landscapes. Trends in Ecology and Evolution 17:397399. Stiles, G. F. 1992. Effects of a severe drought on the population biology of a tropical hummingbird. Ecology 73:1375-1390. Stotz, D. F., T. A. Parker III, and J. W. Fitzpatrick. 1996. Neotropical Birds: Ecology and Conservation. University of Chicago Press, Chicago. Styrsky, J. N., J. D. Brawn, and S. K. Robinson. 2005. Juvenile mortality increases with clutch size in a Neotropical bird. Ecology 86:3238-3244. Saether, B.-E., and 0 . Bakke. 2000. Avian life history variation and contribution of demographic traits to the population growth rate. Ecology 81:642-653. Saether, B.-E., and S. Engen. 2003. Variation in population growth rates. Pages 66-84 in R. M. Sibbly, J. Hone, and T. H. Cloutton-Brock, editors. Wildlife Population Growth Rates. Cambridge University Press, Cambridge, UK. Tejeda-Cruz, C , and W. J. Sutherland. 2004. Bird responses to shade coffee production. Animal Conservation 7:169-179. Tejeda-Cruz, C , and W. J. Sutherland. 2005. Cloud forest bird responses to unusually severe storm damage. Biotropica 37:88-95. Thomas, C. D., and W. E. Kunin. 1999. The spatial structure of populations. Journal of Animal Ecology 68:647-657. Turchin, P. 1998. Quantitative Analysis of Movements: Measuring and Modeling Population Redistribution in Animals and Plants. Sinauer Associates, Inc., Sunderland, MA. 76 Underwood, A. J. 1997. Experiments in Ecology. Cambridge University Press, Cambridge, UK. Underwood, T. J., and R. R. Roth. 2002. Demographic variables are poor indicators of Wood Thrush productivity. Condor 104:92-102. Watkinson, A. R., and W. J. Sutherland. 1995. Sources, sinks and pseudo-sinks. Journal of Animal Ecology 64:126-130. Watson, D. M. 2003. Long-term consequences of habitat fragmentation-highland birds in Oaxaca, Mexico. Biological Conservation 111:283-303. White, G. C , and K. P. Burnham. 1999. Program MARK: Survival estimation from populations of marked animals. Bird Study 46:120-138. Willson, S. K. 2004. Obligate army-ant-following birds: a study of ecology, spatial movement patterns, and behaviour in Amazonian Peru. Ornithological Monographs 55:1-67. With, K. A., and A. W. King. 2001. Analysis of landscape sources and sinks: the effect of spatial pattern on avian demography. Biological Conservation 100:75-88. Young, B. E., D. DeRosier, and G. V. N. Powell. 1998. Diversity and conservation of understory birds in the Tilaran Mountains, Costa Rica. Auk 115:998-1016. Young, B. E., and D. B. MacDonald. 2000. Birds. Pages 179-222 in N. M. Nadkarni and N. T. Wheelwright, editors. Monteverde: Ecology and Conservation of a Tropical Cloud Forest. Oxford University Press, New York, NY. Zar, J. H. 1999. Biostatistical Analysis. Prentice Hall, New Jersey, NJ. 77 CHAPTER 4: OCCUPANCY AND DETECTABILITY RATES OF CATHARUS FRANTZII (RUDDY-CAPPED NIGHTINGALE THRUSH) IN THE CENTRAL HIGHLANDS OF CHIAPAS, MEXICO INTRODUCTION Catharus frantzii (Ruddy-capped Nightingale Thrush) is a species restricted to cloud and humid pine-oak montane forests of the northern Neotropics. The species is listed as threatened by the Mexican Government because of the alarming decline of its habitat throughout its range (Diario Oficial de la Federacion 2002, Rangel-Salazar et al. In press). The species is fairly common to uncommon across montane forests in Mexico, (Howell and Webb 1995, Clement et al. 2000), and reported to prefer large patches of forest (> 3 000 ha) with large, tall trees and dense understory vegetation (Martinez-Morales 2001, Watson 2003, Tejeda-Cruz and Sutherland 2004). It has been reported to be sensitive and less detectable after a hurricane disturbance in the montane forests of western Chiapas (Tejeda-Cruz and Sutherland 2005). However, little is known about the causes of its patchy distribution and abundance (Watson 2003). The mountains of the Central Highlands of Chiapas may contain an important population for C. frantzii in the northern Neotropics, where the primary threats for the species are loss and fragmentation of montane forests (Rangel-Salazar et al. 2005). Recognition of the occupancy and abundance of species across suitable habitats is important for conservation (Krebs 2001, Newton 2003). Suitable habitat provides the environmental characteristics within the range of tolerance for organisms to survive or reproduce, or both (MacKenzie and Royle 2005). The loss and fragmentation of habitats caused by direct or indirect human actions is a major factor affecting distribution and abundance of species in many countries. In the Central Highlands of Chiapas, vegetation removal, which can alter the composition and configuration of habitat, is a current potential risk for understory songbirds. The alteration may be negative for species that depend on old-growth forest as habitat (Harris and Pimm 2004). Nonetheless, the importance of habitat in a region may depend on its availability and configuration (Fahrig and Nuttle 2005, Fortin and Dale 2005). 78 Anthropogenic disturbance in the Central Highlands of Chiapas has been suggested to influence the composition of old-growth forest in the region by reducing plant diversity (Gonzalez-Espinosa et al. 2004, Ochoa-Gaona et al. 2004), and potentially depleting resources for forest-dwelling songbirds (Gonzalez-Espinosa et al. 1995, Moron-Rfos 2005). Old-growth broadleaf forests have declined at a rate of 3.3% per year, resulting in degradation and loss of forested habitats (Ochoa-Gaona and Gonzalez-Espinosa 2000, Ochoa-Gaona 2001). Post-harvesting activities, including agriculture, grazing and housing, also reduce habitat supply for forest songbirds. A long period of time is required to regenerate these highly degraded forest sites (Galindo-Jaimes et al. 2002, Quintana-Ascencio et al. 2004). Some disturbed forests in the region may support stable lower density populations of C. frantzii (Chapter 3, Donovan and Thompson III 2001). Research is needed to identify the variation in distribution and abundance of bird populations in relation to the characteristics of habitat to evaluate the responses of individuals and populations to land management activities (Opdam and Wiens 2002). Currently, there are no published studies documenting the occupancy and detection probabilities of forest-dwelling songbirds in the Central Highlands of Chiapas as forest habitats are lost and fragmented in the region (Vidal et al. 1994, Rappole et al. 2000). It is possible that severe negative effects on populations of forest interior species have already occurred. Montane cloud forests in Chiapas host bird species of high conservation concern in the northern Neotropics (Watson and Peterson 1999, Watson 2003, Rangel-Salazar et al. 2005). The montane forest in the Central Highlands of Chiapas can be described as a mainland-archipelago system, with few large and numerous small patches of montane forests connected by various corridors in a predominantly agricultural and urban matrix. Because of the intensive land-use, most small patches of forest have been subjected to further habitat loss (Gonzalez-Espinosa et al. 1997, Alvarez-Moctezuma et al. 1999, Ochoa-Gaona and Gonzalez-Espinosa 2000). I selected C. frantzii as species to study because it is a typical inhabitant of cloud and pine-oak forests. It has also been identified as sensitive to the composition and configuration of montane forests (Tejeda-Cruz and Sutherland 2005). The species may be an indicator of processes if the 79 patch size is smaller than their daily movement distances, but shorter than the maximum dispersal range. Here, I report the results of a 2-year study at 5 montane forest sites in the Central Highlands of Chiapas, southern Mexico, including: (1) a description of occupancy and detection probabilities of C. frantzii; (2) a comparison of variation in abundance; and (3) a correlation of abundance with the critical habitat attributes identified in Chapter 2. METHODS Study site. - The study was conducted in the forests surrounding San Cristobal de las Casas, Central Highlands of Chiapas, Mexico (16° 44' N, 92° 38' W). This region of 6, 000 km 2 is a high elevation landmass (> 2000 m) with a main orientation NW-SE, and a temperate sub-humid climate. The original vegetation included 30% of about 9000 vascular plant species found in Chiapas and was comprised of associations of evergreen cloud forest, oak forest, pine forest, pine-oak forest and pine-oak-liquidambar forest (Breedlove 1981, Ochoa-Gaona and Gonzalez-Espinosa 2000). The native plant communities of these forest types have physiognomically similar floristic relationships to montane forests located in central and southern Mexico, and Central America and the Caribbean (Quintana-Ascencio and Gonzalez-Espinosa 1993). The Central Highlands of Chiapas are densely populated by first-nations people of Mayan origin who practice traditional shifting cultivation on steep slopes and who use firewood and other forest resources since pre-Columbian times (Gonzalez-Espinosa et al. 1997). I selected five forested sites for this study that were montane forests, are known to have C. frantzii present, are accessible by trail or unpaved roads, and are close to the Huitepec reserve (2 -7 km). Routes in each site were at least 2 km apart and 2-km in length. Surveys were conducted during the early breeding season (mid-March and mid-April) of 2002 and 2003, at 11 sampling points per site positioned 200 m apart (N- 55), using a fixed 50-m radius point-count method (Bibby et al. 2000, Sutherland et al. 2004). The selected routes had little or no traffic during surveys. Sampling routes started at the bottom of the central valley and continued upslope. Flags were placed 50 m before and after each sampling point to aid in estimating distances to vocalizing birds. 80 For species of high probability of detection (p > 0.5), three or more consecutive independent visits (K) per sampling point are required to conclude with certainty (a = 0.05) that the species is absent from sampling points with no detection (MacKenzie et al. 2004, MacKenzie and Royle 2005). Based on my experience at Cerro Huitepec Biological Reserve, C. frantzii has a high detectability in the region. We conducted four surveys per site per year (K= A). Surveys began at dawn (06:30 to 07:00 ET) and finished after 4 h. All observers covered all sites at least once through rotation. Two new points were added in 2003. Surveys were cancelled during windy (-15 km/hr or greater) or rainy conditions. Observers used playbacks of C. frantzii vocalizations from local recordings to increase detectability (Bibby et al. 2000). A sampling session at each point took seven minutes divided into a 3 minute period for listening, followed by 1 minute of playback, and a 3 minute listening period. Observers recorded both observations and vocalizations of C. frantzii. Only individuals detected within a 50 m radius were included in the analysis. I chose this distance after determining that observers could reliably detect the vocalizations of C. frantzii in all conditions and habitats. When there was uncertainty in detection (i.e., low vocalizations, variations in songs and calls), two surveys at the same point were used to confirm occupancy. Occupancy and detectability estimates.-1 recorded sampling-points where C. frantzii were detected as "occupied" and zero detections as "unoccupied" (MacKenzie et al. 2004). Using . encounter histories (i.e., contiguous series of encountered individuals) at each sampling point, I estimated occupancy percent chance a sampling point was occupied), and detection (p) probability (percent chance that an individual was recorded, if present at a sampling point). Because sampling points were along routes, occupancy estimates and detection probabilities may not be independent (Fortin and Dale 2005). However, I assumed that the sampling points were far enough apart to consider them independent during a single survey. Occupancy and detection probabilities modelled included a constant (y. , p.), year (y)- and visit (f)- effects / t , p / ( ; (MacKenzie et al. 2002, MacKenzie and Royle 2005). Model selection for occupancy and detection probabilities were based on the lowest value for Akaike's Information Criterion (AIC; (Burnham and Anderson 2002). Encounter histories were analyzed using the occupancy function of the program 81 MARK (White and Burnham 1999, Cooch and White 2004). Over-dispersion of data (i.e., lack of independence) was evaluated with a goodness-of-fit (GOF) test in program RELEASE (Burnham et al. 1987) for the completely parameterized model (\\ty,, p^,). Relative abundance (hereafter abundance) of singing males was estimated from the number of vocalizing individuals recorded within a 50 m radius at each point-survey. I assumed that recorded singing individuals were males. However, it is possible that females may also vocalize and influence abundance estimates. Playbacks did not significantly increase the number of singing males (Wilcoxon rank test Xs = 2.35, df- 431, P- 0.07), thus I used the mean abundance between broadcastings at each point-survey for the analysis. I collected data on habitat attributes within a 12.5 m-radius circular plot (i.e., 0.049 ha), centred at each sampling point (Bibby et al. 2000, Sutherland et al. 2004). Data collected included elevation (m), aspect (degrees), slope (%) and vegetation measurements: trees (estimated diameter at breast height (DBH) > 5 cm), tree height, and canopy cover. Tree, shrub and herb densities were estimated for the three most abundant species respectively in the plot. Additionally, I classified a vegetation type for each plot as old-growth broadleaf, second-growth broadleaf, and coniferous-broadleaf forests according to the most representative tree species. As well, each plot was characterized as "disturbed" or "non-disturbed" by anthropogenic activities. Cumulative and relative abundance among sites, years, sampling points and visits were analysed using general linear models of two- and three way analysis of variance (GLM-ANOVA; Zar 1999). The full initial models that contained two- and three-way interactions among site, year, point, and visit, were reduced by sequentially removing non-significant interaction terms ( P > 0.05). I modelled the correlation relationship between abundance (dependent variable) and habitat attributes (independent variables) using a multiple regression analysis (MRA) to test the null hypothesis that variation in abundance is not related to particular habitat attributes at site and regional levels. Candidate independent variables were selected for inclusion in the models using a forward-stepwise selection procedure. Prior to fitting the regression models, I tested each independent variable using univariate tests. All variables that were significant (a = 0.05) were then tested for pair-correlation before entering into the final model. When Pearson correlations between 82 any two variables were significant (a = 0.05) I selected the one with the higher univariate and lower probability values for inclusion in the final model. Prior to statistical analysis, variables were checked for normality and equal variance with Shapiro-Wilks Wand Bartllet's tests, respectively (Gotelli and Ellison 2004). Data were square root-transformed (linear) or arcsine-transformed (percentages) as appropriate to reach parametric assumptions. Non-parametric Wilcoxon/Kruskal-Wallis rank tests were used when data differed from parametric assumptions. When factor interactions were significant, independent univariate analysis were performed to evaluate the role of each factor independently. I performed all statistical analyses with SAS-JMP 5.01 (Sail etal. 2005) following Zar (1999) and Gotelli and Ellison (2004). All means are presented ± 1SE and tests were considered significant at a = 0.05. RESULTS In sites selected to have C. frantzii, birds were detected in 47 of 53 (89%) sampling points in 2002, and in 50 of 55 (91%) sampling points in 2003. Occupancy frequencies (i.e., number of visits with occupancy detection per sampling point) did not differ between years (mean occupancy frequencies, 2002: 1.08 ± 1.14, n = 53, vs. 2003: 1.17 ± 1.01, n = 55; Wilcoxon rank test x 2 = 0.04, P = 0.83). Two or three visits at each point were sufficient to determine occupancy / non-occupancy for C. frantzii as judged by the addition of new detections at a sampling point. One unoccupied point at the Callejon site in 2002 was detected as occupied at the very last visit in 2003. All sites but Huitepec had at least one (CVSJCH, Arcotete and Florecilla) or two (Callejon) unoccupied sampling points in the study. The top model indicated that C. frantzii showed high overall occupancy estimates (AlCc = 558.08 and AlCc weight = 0.29; occupancy = 0.91 ± 0.03; 95% confidence intervals [CI], lower = 0.64 and upper = 0.74), and detection probability (p = 0.69 ± 0.02; 95% CI, lower = 0.83 and upper = 0.95) within the suitable habitat in the region. Both occupancy (\|/2oo2 = 0.89 ± 0.05 vs. \ | / 2 0 0 3 = 0.92 ± 0.04) and detection probabilities (p 2 0 0 2 = 0.67 + 0.04 vs. p 2 0 0 3 = 0.71 + 0.03) did not vary between years (maximum likelihood ratio test x2 = 0.97, df = 2, P = 0.62). Sites, sampling points (s) 83 and visits (K) closely interacted, and they were strongly correlated to occupancy and abundance (GLM-ANOVA, R2 = 0.81, F 8 > 3 1 = 16.7, P< 0.001). The abundance of singing males of C. frantzii varied among sites and points, but not for year (GLM-ANOVA, F/2 = 0.91, F 5 4 , 5 3 = 10.3, P < 0.001). The site*point interaction term was significant (F4o, 67 = 10.2, P < 0.001). This significant interaction effect indicated that the abundance of singing males differed among sites but such variation was dependent on the point. When the point variable was fixed, differences in the variation of pair density were not explained by site alone. The average abundance of singing males did not differ among sites ( F 4 5 0 = 1.56, P= 0.19; Fig. 4.1). All habitat attributes differed by site (ANOVA; F s 2.94, P< 0.03; Table 4.2). The Huitepec reserve differed from all the other sites, since at Huitepec the dominant vegetation type was old-growth broadleaf forest with a low disturbance ratio. The dominant vegetation at the other sites was second-growth broadleaf or coniferous-broadleaf forest, and all sampling points showed evidence of anthropogenic disturbance. The anthropogenic disturbance varied from scattered harvested trees and understory reduction at Huitepec to the intensive harvesting practice of pollarding (a method of cutting broadleaf trees above levels reached by grazing animals) and an increase of coniferous trees as the dominant tree species with little understory in CVSJCH, Arcotete, Florecilla, and Callejon. Elevation, tree height and shrub and herb densities appeared to be the most important habitat attributes that differed among sites (Table 4.1). The habitat attributes that were correlated with detectability of singing males varied across sites {ft > 0.59; F> 4.3, P < 0.07; Table 4.2). The detectability of singing males correlated positively with slope, but negatively with increasing herb density at Huitepec. At CVSJCH, only tree height had a negative correlation with the detectability of singing males. Elevation and slope negatively correlated with the detectability of singing males at La Florecilla. At El Callejon, elevation, tree height, canopy cover and herb density negatively correlated with the detectability of singing males (Table 4.2). At Arcotete, no measured attribute correlated with the detectability of singing males. Habitat attributes of all sites were not correlated with the detectability of singing males (hf= 0.09; F 3, 5 1 = 1.63, P= 0.19; Table 4.2). 84 D I S C U S S I O N C. frantzii was recorded in most points in the sampled area (occupancy = 90% and probability of detection = 69%). Under the sampling conditions and design used, the number of repeated surveys per site may be reduced from 4 to 3 with only a small reduction in precision (MacKenzie et al. 2002, MacKenzie and Royle 2005). The number of sampling points necessary to achieve from 0.9 to 1 probability of detecting C. frantzii at least once during 4 surveys of an occupied sampling point (p*) varied from 88 to 195 sampling points (Fig. 4.2). Thus, sampling effort may be reduced in surveys (K= 3), while keeping the same number of sites and sampling points to achieve 0.98 of probability of detection, or in sampling points (124) with the same number of surveys to achieve a 0.95 of detection. Several studies have demonstrated the importance of species coexistence in determining occupancy and probabilities of detection of Catharus (Noon 1981, Rimmer et al. 1996, Nixon et al. 2001). C. frantzii coexists year round with other Catharus nightingale thrushes in most of its range. C. frantzii and C. occidentalis (Russet Nightingale Thrush) coexist in montane forests, and in these areas C. frantzii is restricted to dense, broad-leaf, riparian forests in deep canyons, while C. occidentalis occurs in a wider range of habitats from open oak woodlands to dense, closed fir-pine-oak forests in Oaxaca (Raitt and Hardy 1970, Watson 2003). In Oaxaca, C. frantzii occupied more cloud forest patches than C. occidentalis, suggesting an opposite relationship between range-size and patch occupancy in these two coexisting species (Watson 2003). In the western montane forests of Chiapas, where C. frantzii coexists with C. aurantirostris (Orange-billed Nightingale Thrush) and C. dryas (Spotted Nightingale Thrush), C. frantzii is restricted to montane cloud forest and recorded less frequently than C. aurantirostris and C. dryas (Tejeda-Cruz and Sutherland 2004, 2005). Thus, competition with congenerics may limit C. frantzii \o habitats of dense, broad-leaf moist forests in most of its range. In the Central Highlands of Chiapas, where C. frantzii occupies a wide range of montane forest habitats, they do not coexist with any other Catharus species, except for C. ustulatus (Swainson's thrushes) during migration (Outlaw et al. 2003, Hiron et al. 2006). The lack of year-round coexistence with other nightingale Catharus thrushes may allow C. frantzii \o exploit more habitat types in the Central Highlands of Chiapas. 85 Although they were distributed widely, the average detectability of C. frantzii was only 1.4 singing males per sampling point (Fig. 2.1). Detectability tended to be slightly higher in Huitepec and CVSJCH than in Arcotete, Callejon and La Florecilla. These lower values are comparable to those reported in the montane forest of Oaxaca (Watson 2003) and western Chiapas (Tejeda-Cruz and Sutherland 2004). In Oaxaca, C. frantzii had a relative detectability of 0.79 individuals across the four largest forest patches. In western Chiapas, the relative detectability of C. frantzii declined from 0.8 to 0.28 individuals/ha after a strong hurricane in 1998 (Tejeda-Cruz and Sutherland 2005). Variation in relative abundance of C. frantzii in the Central Highlands of Chiapas may be explained by the distribution of habitat across the region and whether habitats were intensively disturbed. The Huitepec reserve represents prime montane forest habitat, with medium to little disturbance (Table 4.1). The other sites presented signs of intense disturbance, which possibly reduced habitat suitability for C. frantzii. It is outstanding for the conservation of C. frantzii in the region that it occupies habitats that have experienced considerable disturbance. This probably indicates high levels of tolerance and resistance that allow C. frantzii Xo persist in the region. It is possible that I overestimated the relative abundance of singing males of C. frantzii \\ some singing birds were females. Females of other Catharus species have been reported to sing (e.g., C. bicknelli [Bicknell's Thrush] in Vermont; Rimmer et al. 1996). For C. bicknelli, recent research has shown that the home ranges of males tend to be large and overlap with other males (Strong et al. 2004). Thus, traditional survey methods based on broadcasting vocalizations may provide overestimates of abundance. Moreover, broadcasting vocalizations in more open habitats may result in less sound attenuation than closed habitats. Hence, individual thrushes in more open habitats would hear the broadcasting calls at greater distances, potentially more than 50 m, or observers may record individuals singing farther than 50 m. Despite the potential biases in data collection, the abundance of C. frantzii m Huitepec reserve is probably higher than the other sites, as well as higher than the low elevation habitats in western Chiapas and Oaxaca (Watson 2003, Tejeda-Cruz and Sutherland 2004). At higher elevations, C. frantzii may have advantages because they may be better adapted to colder environments than the other thrush species (Clement et al. 2000). At lower elevations, C. frantzii may compete with morphologically and behaviourally similar 86 species such as C. occidentalis, C. aurantirostris and C. dryas (Raitt and Hardy 1970, Outlaw et al. 2003). The abundance of C. frantzii could not be predicted from any habitat attributes considered in this study at the regional level. Typical suitable habitat for C. frantzii in the montane forests of central Mexico to Central America is "thick tangled undergrowth and ground vegetation in moist, humid or semi-humid pine-oak, and conifer cloud-forests, forest edges and clearings, woodlands and ravines, occasionally in bamboo thickets, vines and creepers of the subtropical and lower temperate zones between 1350 and 3500 m" (Clement e ta l . 2000). From this description, I expected to find a positive correlation between vegetation density, particularly tree, shrub and herb densities, and the detectability of C. frantzii. Variation in elevation, tree and shrub densities were larger in Huitepec, where more singing males were recorded. CVSJCH was the nearest site to Huitepec, and it had the next largest number of singing males. This site was highly disturbed and the only variable that related to the detectability of singing males was tree height. Although, at CVSJCH, tree and shrub densities were lower than in Huitepec, these variables did not differ from the rest of sites. Arcotete (a second growth broadleaf forest), and Florecilla and Callejon (coniferous broadleaf forests) also had high levels of anthropogenic disturbance. These sites had similar detectability of singing males. Understory tree species have been cited as an important habitat attribute for C. frantzii (Rowley and Orr 1964, Raitt and Hardy 1970, Clement et al. 2000), but understory tree density was not correlated with detectability of singing males at either all sites or individual sites. Contrary to my prediction, herb density was negatively correlated with the relative detectability of singing males at Huitepec and Callejon. Factors other than habitat attributes could also influence variation in abundance at the regional level. Individual competitive ability may affect settlement patterns, where better competitors occupy the more suitable habitats and less dominant individuals are forced to settle in less suitable habitats (Newton 1998, Sergio and Newton 2003). In the Huitepec Reserve, a large proportion (66%) of individuals captured in a nine-year study were never recaptured (Chapter 3). This may support the hypothesis that only a small proportion of individuals in the population are able to establish permanent territories. Similar findings have been reported for C. fuscater (Slaty-87 backed Nightingale Thrush) in Colombia's Central Cordillera (Beltran and Kattan 2003). Movements of temporary residents or floaters may obscure the relationship between habitat attributes and bird abundance at local and regional levels. Limited habitat availability results in a large floater population. Thus, both floaters and territorial individuals require resources from habitats of similar suitability to obtain sufficient resources for survival and reproduction (Fahrig and Nuttle 2005). C O N C L U S I O N My study showed that C. frantzii occupied a variety of forest patch types and conditions across the region. In the Central Highlands of Chiapas, C. frantzii makes extensive use of the region during the breeding season, occupying most of the suitable habitat from old growth broadleaf forests to relatively young, regenerating clear cuts of broadleaf and coniferous forests. Since the vegetation in the region is extensively harvested by local people (Gonzalez-Espinosa et al. 1995, Ochoa-Gaona 2001, Ochoa-Gaona et al. 2004), developing a plan for vegetation harvesting for firewood and construction material that ensures a continuous supply of regeneration of the appropriate tree species and understory density across the region is necessary to maintain suitable habitat for C. frantzii. More research is needed to answer basic questions about the use of second-growth broadleaf and coniferous regeneration habitats by C. frantzii in the region. First and most importantly, it is necessary to determine whether the species is successfully breeding in occupied sites, since lower foliage density and/or nest concealment can lead to increased rates of nest predation (Chapter 2). Then, the gradient of ages of regeneration of habitat suitability for C. frantzii must be determined. A larger area and greater number of sites with a gradient of habitat conditions would have to be surveyed to test this hypothesis. Finally, a study of colour-banded or radio-marked birds is needed to evaluate individual movements and its effects on the occupancy and detectability at a regional level, as well as possible differences in habitat selection, breeding performance and population dynamics, as well as their interaction with predators and competitiors. 88 Table 4 .1. Habitat attributes at 5 sites for surveys of Catharus frantzii (Ruddy-capped Nightingale Thrush) in 2002 and 2003 (n = 11 transects, 55 sampling points) in the Central Highlands of Chiapas. Habitat attributes were measured in 2003. Sites are: Huitepec (Cerro Huitepec Biological Reserve), CVSJCH (Camino Viejo a San Juan Chamula), Arcotete (Sendero Arcotete), Florecilla (Camino a La Florecilla), Callejon (Sendero Callejon). Dominant Forest Type: old-growth broadleaf (OGB), second-growth broadleaf (SGB), and coniferous - second-growth broadleaf (C-SGB). Disturbed Ratio: proportion of sampling points with disturbance by anthropogenic activities over the points in the site). Site Variables Huitepec CVSJCH Arcotete Florecilla Callejon Univariate Values Mean SE Mean SE Mean SE Mean SE Mean SE F P Elevation 2470.64 45.86 2319.18 13.08 2159.64 6.59 2330.73 10.49 2195 8.85 29.17 0.001 Aspect 148.55 46.82 194.55 9.66 222.00 47.32 207.27 19.54 96.73 39.48 2.84 0.034 Slope 24.27 3.61 17.36 2.63 22.91 1.94 33.45 7.06 12.27 2.37 2.94 0.027 Canopy cover 88.11 3.00 55.67 7.24 56.56 4.65 62.87 9.17 79.51 3.68 5.68 0.008 Tree height 14.18 1.48 8.89 0.52 7.46 0.49 10.35 1.31 16.08 1.27 11.68 0.001 DBH 29.77 3.91 16.71 1.04 15.93 1.65 21.94 2.55 23.25 1.97 4.91 0.002 Tree density 42.59 7.44 29.86 1.81 26.85 2.73 20.49 1.83 32.29 2.39 4.49 0.004 Shrub density 28.00 2.67 13.73 2.08 17.36 3.86 11.49 1.20 24.11 3.66 5.89 0.001 Herb density 15.18 3.04 42.73 5.19 29.73 4.89 15.91 2.32 17.55 4.07 6.78 0.001 Dominant Forest Type OGB SGB SGB C-SGB C-SGB Disturbed Ratio 0.36 1 1 1 1 Table 4.2. Multiple regression analysis examining the effect of habitat attributes on relative abundance of singing Catharus frantzii (Ruddy-capped Nightingale Thrush) at the site and regional levels in the Central Highlands of Chiapas, in 2002-2003. Columns give site and variable attributes, estimated coefficients, standard error (SE), ANOVA-F values, degrees of freedom (df) and probability (P). Generalized Linear Model (GLM-ANOVA) estimates are shown beside each site and for the overall region. Bolded values indicate significant values at a = 0.05. Variables Coefficient SE F df P Huitepec (F?= 0.92; F 4, 6 = 16.27, P = 0.002) Intercept 0.031 Slope 0.221 Tree Density -0.075 Shrub Cover 0.825 Herb Density -1.347 CVSJCH (F?= 0.59; F 2, 8 = 5.74, P= 0.03) Intercept 3.173 Tree Height -0.814 Tree Density -0.353 Arcotete (P?= 0.3; F,,9 = 3.3, P= 0.1) Intercept 0.597 Aspect -0.052 Florecilla (F?= 0.8; F 5, 5 = 4.3, P= 0.07) Intercept 33.445 Elevation -2.37 Aspect 0.218 Slope -0.442 Tree Height -0.911 Canopy Cover 0.402 Callejon (hf= 0.98; F 5 - 5 = 43.2, P= 0.004) Intercept 13.853 Elevation -0.914 Aspect -0.012 Tree Height -0.295 0.152 0.18 1 0.863 0.039 31.285 1 0.001 0.037 2.785 1 0.146 0.261 10.113 1 0.019 0.223 35.95 1 0.001 0.833 3.81 1 0.005 0.265 9.429 1 0.015 0.163 4.689 1 0.062 0. 165 3.61 1 0.005 0.028 3.334 1 0.101 12.167 2.75 1 0.04 0.894 7.015 1 0.045 0.087 6.189 1 0.055 0.121 13.284 1 0.014 0.362 6.298 1 0.053 0.201 3.992 1 0.102 4.356 3.18 1 0.024 0.339 7.24 1 0.043 0.009 1.839 1 0.233 0.094 9.773 1 0.021 90 Canopy Cover -0.836 0.101 69.183 1 0.001 Herb Density -0.545 0.181 9.075 1 0.029 Overall locality (fl*= 0.09; F 4 , 5 0 = 1.63, P = 0.19) Intercept -1.891 1.449 -1.3 1 0.197 Elevation 0.201 0.114 3.119 1 0.083 Slope -0.054 0.037 2.104 1 0.153 Tree Height -0.119 0.095 1.598 1 0.211 91 03 0) ro E c n g ' c n c 'to .Q E 3 3.5 - | 3 2.5 2 1.5 1 0.5 -0 --0.5 SGB SGB C-SGB C-SGB OGB 1 1 1 1 Arcotete CVSJCH Callejon Florecilla Huitepec Site Figure 4 .1 . Box-plots of average relative abundance of singing male Catharus frantzii (Ruddy-capped Nightingale Thrush) across five mixed forest sites during 2002 and 2003 in the Central Highlands of Chiapas. The horizontal line within each box represents the sample median and boxes encompass the 25 t h to the 75 t h percentile. Bars represent the 10 t h and 90 t h percentiles, and the grand horizontal line represents the average abundance of singing males per sampling point = 1.43 ± 0.79. Dominant Forest Type: old-growth broadleaf (OGB), second-growth broadleaf (SGB), and coniferous - second-growth broadleaf (C-SGB). 92 250 i CD E -z. 50 -0 -J 1 1 1 1 1 1 1 1 1 1 1 0.99 0.98 0.97 0.96 0.95 0.94 0.93 0.92 0.91 0.9 Level of precision (p*) Figure 4.2. Estimates of the total number of survey points (TSP) required for achieving a specific level of precision (p*) for estimating occupancy (HJ) and detection probability (p) parameters for Catharus frantzii (Ruddy-capped Nightingale Thrush) in the Central Highlands of Chiapas. Initial parameter values were estimated in MARK. Estimates were based on a standard design of occupancy (MacKenzie and Royle 2005). 93 R E F E R E N C E S Alvarez-Moctezuma, J. G., S. Ochoa-Gaona, B. H. J. de Jong, and M. L. Soto-Pinto. 1999. Habitat y distribution de cinco especies de Quercus (Fagacea) en la Meseta Central de Chiapas, Mexico. Revista de Biologia Tropical 47:351-358. Beltran, W., and G. H. Kattan. 2003. First record of the Slaty-backed Nightingale Thrush in the Central Andes of Colombia, with notes on its ecology and geographical variation. Wilson Bulletin 113:134-139. Bibby, C. J., N. D. Burgess, D. A. Hill, and S. Mustoe. 2000. Bird Census Techniques. Academic Press, London, UK. Breedlove, D. E. 1981. Introduction to the flora of Chiapas. Flora de Chiapas 1:1 -35. Burnham, K. P., and D. R. Anderson. 2002. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. Springer-Verlag, New Yor, NY. Burnham, K. P., D. R. Anderson, G. C. White, C. Brownie, and K. H. Pollock. 1987. Desing and anlysis methods for fish survival experiments based on realese-recapture. American Fisheries Association Monograph 5. Clement, P., R. Hathway, C. Byers, and J. Wilczur. 2000. Thrushes. Christopher Helm Ltd., London, UK. Cooch, E., and G. White. 2004. Program MARK: A Gentle Introduction, in http://www.phidot.org/software/mark/docs/book/ Diario Oficial de la Federacion. 2002. Norma Oficial Mexicana NOM-059-ECOL-2001, Proteccion ambiental-Especies nativas de Mexico de flora y fauna silvestres-Categorias de riesgo y especificaciones para su inclusion, exclusion o cambio-Lista de especies en riesgo. Pages 1-85 in Diario Oficial de la Federacion. Secretaria de Medio Ambiente y Recursos Naturales, Mexico, DF. Donovan, T. M., and F. R. Thompson III. 2001. Modelling the ecological trap hypothesis: a habitat and demographic analysis for migrant songbirds. Ecological Applications 11:871-882. Fahrig, L., and W. K. Nuttle. 2005. Population ecology in spatially heteregenous environments. Pages 95-118 in G. M. Lovett, C. G. Jones, M. G. Turner, and K. C. Weathers, editors. Ecosystem Function in Heterogeneous Landscapes. Springer-Science, New York, NY. Fortin, M.-J., and M. Dale. 2005. Spatial Analysis. A Guide for Ecologists. Cambridge University Press, Cambridge, UK. Galindo-Jaimes, L., M. Gonzalez-Espinosa, P. Quintana-Ascencio, and L. Garcia-Barrios. 2002. Tree composition and structure in disturbed stands with varying dominance by Pinus spp. in the highlands of Chiapas, Mexico. Plant Ecology 162:259-272. Gonzalez-Espinosa, M., S. Ochoa-Gaona, N. Ramfrez-Marcial, and P. Quintana-Ascencio. 1995. Current land use trends and conservation of old growth forest habitats in the highlands of Chiapas, Mexico. Pages 190-197 in M. H. Willson and S. A. Sader, editors. Conservation 94 of Neotropical Migratory Birds in Mexico. Maine Agricultural and Forest Experimentation Station Misc., Maine. Gonzalez-Espinosa, M., S. Ochoa-Gaona, N. Ramirez-Marcial, and P. F. Quintana-Ascencio. 1997. Contexto vegetacional y florfstico de la agricultura en los Altos de Chiapas. Pages 91-124 in M. R. Parra-Vazquez and B. M. Diaz-Hernandez, editors. Los Altos de Chiapas: agricultura y crisis rural. El Colegio de la Frontera Sur, San Cristobal de Las Casas, Chiapas, Mexico. Gonzalez-Espinosa, M., J. M. Rey-Benayas, N. Ramirez-Marcial, M. A. Huston, and D. Golicher. 2004. Tree diversity in the northern Neotropics: regional patterns in highly diverse Chiapas. Ecography 27:741-756. Gotelli, N. J., and A. M. Ellison. 2004. A Primer of Ecological Statistics. Sinuer Associates, Inc., Sunderland, MA. Harris, G. M., and S. L. Pimm. 2004. Bird species' tolerance of secondary forest habitats and its effects on extinction. Conservation Biology 18:1607-1616. Hiron, M., J. L. Rangel-Salazar, and B. Christensen. 2006. Coexistence of resident and migratory Catharus thrushes in a montane forest reserve of the northern Neotropics: on habitat use and morphology. Ornitologia Neotropical 17:1-12. Howell, S., and S. Webb. 1995. A Guide to the Birds of Mexico and Northern Central America. Oxford University Press, Oxford, UK. Krebs, C. J. 2001. Ecology: The Experimental Analysis of Distribution and Abundance, 5th edition. Benajmin Cummings, New York. MacKenzie, D. I., J. D. Nichols, G. B. Lachman, S. Droege, J. A. Royle, and C. A. Langtimm. 2002. Estmating site occupancy rates when detection probabilities are less than one. Ecology 83:2248-2255. MacKenzie, D. I., and J. A. Royle. 2005. Designing occupancy studies: general advice and a l locat ing survey effort. Journal of Applied Ecology 42:1105-1114. MacKenzie, D. I., J. A. Royle, J. A. Brown, and J. D. Nichols. 2004. Occupancy estimation and modeling for rare and elusive populations. Pages 149-172 in W. L. Thompson, editor. Sampling Rare and Elusive Species. Island Press, Washington, D.C. Martinez-Morales, M. A. 2001. Forest fragmentation effects on bird communities of montane cloud forest in eastern Mexico. University of Cambridge, Cambridge, UK. Moron-Rios, A. 2005. Los efectos del manejo forestal en la fauna de invertebrados del suelo. Ciencia y Tecnologia en la Frontera 3:38-40. Newton, I. 1998. Population Limitation in Birds. Academic Press, London, UK. Newton, I. 2003. The Speciation and Biogeography of Birds. Academic Press, London, UK. 95 Nixon, E. A., S. B. Holmes, and A. W. Diamond. 2001. Bicknell's thrushes (Catharus bicknelli) in New Brunswick clear cuts: their habitat associations and co-ocurrence with Swainson's thrushes (Catharus ustulatus). Wilson Bulletin 113:33-40. Noon, B. R. 1981. The distribution of an avian guild along a temperate elevational gradient: the importance and expression of competition. Ecological Monographs 51:105-124. Ochoa-Gaona, S. 2001. Traditional land-use systems and patterns of forest fragmentation in the highlands of Chiapas, Mexico. Environmental Management 27:571-586. Ochoa-Gaona, S., and M. Gonzalez-Espinosa. 2000. Land use and deforestation in the highlands of Chiapas, Mexico. Applied Geography 20:17-42. Ochoa-Gaona, S., M. Gonzalez-Espinosa, A. A. Meave, and V. Sorani-DalBon. 2004. Effect of forest fragmentation on the woody flora of the highlands of Chiapas, Mexico. Biodiversity and Conservation 13:867-884. Opdam, P., and J. A. Wiens. 2002. Fragmentation, habitat loss and landscape management. Pages 202-223 in K. Norris and D. J. Pain, editors. Conserving Bird Biodiversity: General Principles and their Application. Cambridge University Press, Cambridge, UK. Outlaw, D. C , G. Voelker, B. Mila, and D. J. Girman. 2003. Evolution of long-distance migration and historical biogeography of Catharus thrushes: a molecular phylogenetic approach. Auk 120:299-310. Quintana-Ascencio, P. F., and M. Gonzalez-Espinosa. 1993. Afinidad fitogeografica y papel sucesional de la flora lehosa de los bosques de pino-encino de Los Altos de Chiapas, Mexico. Acta Botanica de Mexico 21:43-57. Quintana-Ascencio, P. F., N. Ramfrez-Marcial, M. Gonzalez-Espinosa, and M. Martfnez-lco. 2004. Sapling survival and growth of coniferous and broad-leaved trees in successional highland habitats in Mexico. Applied Vegetation Science 7:81-88. Raitt, R. J., and J. W. Hardy. 1970. Relationships between two partially sympatric species of thrushes (Catharus) in Mexico. Auk 87:20-57. Rangel-Salazar, J. L., P. L. Enrfquez-Rocha, and T. Will. 2005. Diversidad de aves en Chiapas: prioridades de investigacion para su conservacion. Pages 265-323 in M. Gonzalez-Espinosa, N. Ramfrez-Marcial, and L. Ruiz-Montoya, editors. Diversidad Biologica en Chiapas. Plaza y Valdes, Mexico City, Mexico. Rangel-Salazar, J. L., E. Santana-Castellon, J. Carrillo, and P. L. Enrfquez-Rocha. In press. Zorzal de Frantzius (Catharus frantzii). in P. Escalante-Pliego, D. Ayala, and V. Nequiz, editors. Libra Rojo de las Aves de Mexico, Mexico D.F. Rappole, J. H., D. I. King, and P. Leimgruber. 2000. Winter habitat and distribution of the endangered golden-cheeked warbler (Dendroica chrysoparia). Animal Conservation 2:45-59. 96 Rimmer, C. C , J. L. Atwood, K. P. McFarland, and L. R. Nagy. 1996. Population density, vocal behavior, and recommended survey methods for Bicknell's Thrush. Wilson Bulletin 108:639-649. Rowley, J. S., and R. T. Orr. 1964. The status of Frantzius' Nightingale Thrush. Auk 81:308-314. Sail, J., L. Creighton, and A. Lehman. 2005. JMP Start Statistics: A Guide to Statistics and Data Analysis Using JMP and JMP IN Software, 3d edition. Thompson Learning, Belmont, Ca. Sergio, F., and I. Newton. 2003. Occupancy as a measure of territory quality. Journal of Animal Ecology 72:857-865. Strong, A. M., C. C. Rimmer, and K. P. McFarland. 2004. Effect of prey biomass on reproductive success and mating strategy of Bicknell's Thrush (Catharus bicknelli), a polygynandrous songbird. The Auk 121:446-451. Sutherland, W. J., I. Newton, and R. E. Green, editors. 2004. Bird Ecology and Conservation: A Handbook of Techniques. Oxford University Press, Oxford, UK. Tejeda-Cruz, C , and W. J. Sutherland. 2004. Bird responses to shade coffee production. Animal Conservation 7:169-179. Tejeda-Cruz, C , and W. J. Sutherland. 2005. Cloud forest bird responses to unusually severe storm damage. Biotropica 37:88-95. Vidal, R. M., C. Macias-Caballero, and C. D. Duncan. 1994. The ocurrence and ecology of the Golden-cheecked Warbler in the Highlands of northern Chiapas, Mexico. Condor 96:684-691. Watson, D. M. 2003. Long-term consequences of habitat fragmentation-highland birds in Oaxaca, Mexico. Biological Conservation 111:283-303. Watson, D. M., and A. T. Peterson. 1999. Determinants of diversity in a naturally fragmented landscape: humid montane forest avifaunas of Mesoamerica. Ecography 22:582-589. White, G. C , and K. P. Burnham. 1999. Program MARK: Survival estimation from populations of marked animals. Bird Study 46:120-138. Zar, J. H. 1999. Biostatistical Analysis. Prentice Hall, New Jersey, NJ. 97 CHAPTER 5: CONCLUSIONS POPULATION ECOLOGY OF A TROPICAL MONTANE FOREST BIRD I found that the breeding performance of Catharus frantzii alticola (Ruddy-capped Nightingale Thrush) varied across different conditions of understory vegetation structure and composition in the Cerro Huitepec Biological Reserve, Central Highlands of Chiapas. In the moist forest habitat type containing undisturbed understory vegetation, indicators of breeding success (i.e., pair density, nest density, nesting success and daily nest survival rates) were higher than in the dry-disturbed forest habitat type with removed understory vegetation. Habitat attributes related to nest concealment at the site and patch levels explained the selection of sites for breeding, as well as variation in daily nest survival. Furthermore, nesting success increased with decreasing parental visitation rates, as well with increasing female nest attentiveness. These findings indicate that the breeding performance of C. frantzii at the Huitepec reserve is negatively affected by removal of the understory vegetation. I observed a better breeding performance per nesting attempt in the moist forest than in the dry-disturbed forest, however, productivity and adult survival, important demographic parameters determining the finite rate of population growth (A) of C. frantzii in the Huitepec Reserve, did not differ between habitat types. Thus, no variation in habitat-specific vital traits (productivity and survival) was observed across habitats indicating that the dry-disturbed forest contributes substantively to the population stability of C. frantzii at the Huitepec Reserve. At the regional level, C. frantzii was found occupying most of the sites surveyed, with high rates of detectability across suitable, forested habitats in the region. However, no habitat attributes were correlated consistently with bird abundance. The incorporation of habitat variation in models of population dynamics has changed markedly over the last few decades. Recently, the evolutionary and ecological consequences of source-sink population dynamics for biodiversity conservation was recognized (Hanski and Gaggiotti 2004). The concept of source-sink population structure and dynamics emphasizes the balance among births and deaths, with a dispersal propensity from source- to sink-habitats. 98 Dif ferences in habitat qual i ty p roduce a net of d ispersa l of individuals f rom high qual i ty, expor ter habi tats to low qual i ty, importer habi tats (With and King 2001) . Even though birth rates do not equal death rates wi th in a popula t ion, populat ion stabil i ty can be possib le wi th a net of emigrat ion to, or immigrat ion f rom, ne ighbour ing popula t ions. However , at the Hui tepec Reserve , d i f ferences in breed ing success indicators (pair densi ty, nest ing success and dai ly nest survival) be tween the mois t forest a n d the dry-d is turbed forest do not suppor t a source-s ink st ructure for C. frantzii at a local level , s ince es t imated vital rates in both habi tats cont r ibuted equivalent ly to the populat ion stabil i ty. Nagy and Ho lmes (2004) have raised the quest ion of the reliabil i ty of est imat ing nest ing success and dai ly nest survival as sur rogates of product iv i ty. For instance, product iv i ty in a marked populat ion of Hylocychla mustelina (Wood Thrush) w a s poor ly assoc ia ted wi th nest ing success and dai ly nest surv iva l , a m o n g severa l o ther reproduct ive indicators ( U n d e r w o o d and Roth 2002) . Nest ing success var ies spat ial ly across av ian spec ies f rom tropical lowland forests (Rob inson et a l . 2000 ) , and m a y not necessar i ly reflect product iv i ty or reproduct ive rate. Habi tat-speci f ic reproduct ive rates m a y vary accord ing to whe ther a spec ies has recrui tment- or survival- l i fe history s t rategies (Saether and Bakke 2000) . T h u s , for a survivor- l i fe history s t rategy spec ies such as C. frantziifrom tropical mon tane forests , habi tat-speci f ic survival rate represented the most impor tant demograph ic rate for the spec ies at the Hui tepec reserve. Adul t survival posit ively corre la ted wi th the populat ion g rowth (A) of C. frantzii a\ the Hui tepec reserve. My es t imates of populat ion g rowth (A) sugges ted that an adult survival rate of lower than 7 0 % , the populat ion cou ld decl ine wi th the current pair densi ty and product iv i ty rates (Fig. 3.3c). Adul t survival es t imates were in f luenced in part by individuals cap tu red once . Hence , m y est imates of adul t survival for C. frantzii wi th in the Hui tepec reserve, appeared to be in f luenced by a net d ispersa l at the regional scale (Kaweck i 2004 , Schaefer 2006) . T h e predict ive power of es t imated vital rates for popu la t ion , dec l ines wi th decreas ing spat ial scale (Schaefer 2006) . T h u s , individual m o v e m e n t at a regional level m a y be an impor tant c o m p o n e n t of the populat ion dynamics of C. frantzii &\ the Hui tepec Reserve . A third of the indiv iduals in the populat ion s h o w e d high levels of phi lopatry wi th very few individuals swi tch ing habitat types . T h e s e indiv iduals 99 represented the core individuals within the population and were the individuals establishing territories in the Huitepec reserve. A small proportion of C. fuscater (Slaty-backed Nightingale Thrush) individuals in a population in the central highlands of Colombia established breeding territories, while a large proportion of individuals were considered floaters or dispersers (Beltran and Kattan 2003). Thus, the local populations of nightingale thrushes inhabiting tropical montane forests may be structured by a small proportion of territorial individuals and a large proportion of dispersing individuals; an open, dispersed dominated population (Schaefer 2006), suggesting that habitats are saturated. Population "openness" of nightingale thrushes from tropical montane forests may promote equilibrium properties for habitats maintaining a set of populations connected by a net of dispersal. Asymmetric dispersal rates or individual movements at the regional level can create source-sink dynamics (Kawecki 2004), with the Huitepec Reserve being a stable, source population. Regionally, C. frantzii showed high levels of occupancy with high rates of detectability in a variety of habitat types and conditions. However, I found no consistent correlation between habitat attributes and singing male abundance at site level. In the Central Highlands of Chiapas, human disturbance appeared to modify both vegetation composition and configuration. These changes in vegetation varied from site to site, influenced in part by environmental characteristics. As long as changes in vegetation composition and configuration allow tree stands with understory vegetation, the stability and persistence of C. frantzii populations in the region should be possible. A similar abundance was recorded in both old growth broadleaf forest and coniferous-broadleaf disturbed forests. These findings contrast with previous reports that C. frantzii was being restricted to the montane cloud forest north of Chiapas (Martinez-Morales 2001, Watson 2003), and in western Chiapas (Tejeda-Cruz and Sutherland 2004). To my knowledge, this study is the first to examine how habitat variation influences population dynamics, productivity and survival, and the distribution of C. frantzii at the regional level (Rangel-Salazar et al. In press). Future research should involve examining ecological processes (e.g., predator-prey interactions, habitat selection, competition, and natal and breeding dispersion) that enable C. frantzii \o use a wider variety of habitats and conditions in the Central Highlands of 100 Chiapas compared to other regions. Particularly, research needs to examine the role of floater individuals to the population stability of C. frantzii at the Huitepec Reserve, and to determine the importance of the montane cloud forest in the reserve to the persistence of C. frantzii in the region. There has been little research on the effects of heterogeneity of habitat on population ecology of tropical montane forest birds. In this study, I presented estimates of habitat-specific demographic traits in a system of two contiguous, but distinct, habitats that represent original and modified montane forests in the region. The inclusion of habitat variation in quality into studies of populations of C. frantzii in other areas of tropical montane forests may provide new insights into the role of life history traits, habitat-specific demography, and behaviour in the population ecology of tropical montane forest birds. CONSERVATION ECOLOGY OF CATHARUS FRANTZII IN THE MONTANE FORESTS OF CHIAPAS Catharus frantzii is listed as threatened by the Mexican Government (Diario Oficial de la Federacion 2002) because of the loss and fragmentation of tropical montane forests in the country. Previous to this study, there was no evaluation of population status and how habitat variation affects population dynamics (Rangel-Salazar et al. In press), and my study of C. frantzii population at the Huitepec Reserve and at nearby sites is the only current investigation. Thus, current information for the species is insufficient to suggest any change in its status at the state and national levels. I showed that human-induced disturbance in forest habitats can result in reductions in pair density for C. frantzii. Current land-use patterns in the Central Highlands of Chiapas are affecting the availability of resources used by birds from old-growth humid forests, particularly for those species depending on understory plants (Gonzalez-Espinosa et al. 1995). Reduction in understory plant cover may represent a decrease in food resources, nest substrates and cover from predators. In particular, nest predation was the most important factor of nest failure in this study. Habitat characteristics and nest predation have emerged as the main interacting factors driving fluctuations in avian populations (Heske et al. 2001, Martin and Joron 2003). Habitat variation in time and space often has consequences for predator-prey interactions, and human-induced disturbance 101 promotes changes in these interactions, such as the numbers and function of predators in forest habitats (Mahon and Martin 2006). Different predator types and abundances may account for variable predation rates among different habitat types or neighbouring sites. It is important to identify conditions under which predators change in function and numbers, such that they could change nest predation conditions for montane cloud forest birds. The primary causes of bird species decline in the tropics are loss and fragmentation of forested habitats (Harris and Pimm 2004). Here, I showed that C. frantzii used a variety of habitats, and seemed tolerant to secondary and disturbed habitats. Habitat loss and fragmentation in the region can lead to severe consequences, such as increased edge effects, while decreasing connectivity and reducing habitat heterogeneity (Opdam and Wiens 2002). Whether the set of C. frantzii populations in the region will persist may be a function of its response to habitat disturbance and habitat-based variation in demography (Murphy 2001). I presented how the breeding performance of C. frantzii at the Huitepec Reserve varied across different conditions of structure and composition of understory vegetation. In the moist forest the breeding performance was higher only at the nesting attempt level than in the dry-disturbed forest. This variation was attributable to nest concealment and parent behaviour. However, I found no variation in habitat-specific vital rates (productivity and survival) and both forest types contributed to population stability of C. frantzii at the Huitepec Reserve. C. frantzii was found widely distributed, but in low densities, across suitable habitats and abundance did not differ from old growth broadleaf forest to coniferous-broadleaf disturbed forest. Adult survival was found to be the most important vital rate determining population growth (A), and thus strongly influenced the population dynamics of C. frantzii at the Huitepec Reserve. Overall, the population of C. frantzii aX the Huitepec Reserve appeared stable by a compensatory variation in demographic rates such as productivity and survival. Long-term population stability and persistence is possible through the habitat-specific net reproductive and dispersal rates. Conservation efforts must focus at both the local and regional levels since human disturbance appeared to differentially modify both vegetation composition and configuration from site to site (Rangel-Salazar et al. 2005, Wolf 2005). My data suggest that C. frantzii showed good 102 resistance to disturbance, given the ability of the species to persist in both pristine and disturbed forests. These conditions are important for the conservation of the species since it is likely that conservation efforts can improve their status in the region. The future stability and persistence of C. frantzii populations should be promising for the region if habitat changes in vegetation are managed to maintain suitable habitats. It is fortunate that C. frantzii can tolerate considerable disturbance of the understory, and may represent a good system to elucidate how species respond to habitat disturbance in montane forests. The challenge is determining the limits of tolerance and the relevant spatial scales necessary for elucidating the ultimate causes of population trends. 103 R E F E R E N C E S Beltran, W., and G. H. Kattan. 2003. First record of the Slaty-backed Nightingale Thrush in the Central Andes of Colombia, with notes on its ecology and geographical variation. Wilson Bulletin 113:134-139. Diario Oficial de la Federacion. 2002. Norma Oficial Mexicana NOM-059-ECOL-2001, Protection ambiental-Especies nativas de Mexico de flora y fauna silvestres-Categorias de riesgo y especificaciones para su inclusion, exclusion o cambio-Lista de especies en riesgo. Pages 1-85 in Diario Oficial de la Federacion. Secretaria de Medio Ambiente y Recursos Naturales, Mexico, DF. Gonzalez-Espinosa, M., S. Ochoa-Gaona, N. Ramirez-Marcial, and P. Quintana-Ascencio. 1995. Current land use trends and conservation of old growth forest habitats in the highlands of Chiapas, Mexico. Pages 190-197 in M. H. Willson and S. A. Sader, editors. Conservation of Neotropical Migratory Birds in Mexico. Maine Agricultural and Forest Experimentation Station Misc., Maine. Hanski, I., and O. E. Gaggiotti. 2004. Ecology, Genetics, and Evolution of Metapopulations. Elsevier Academic Press, Burlington, MA. Harris, G. M., and S. L. Pimm. 2004. Bird species' tolerance of secondary forest habitats and its effects on extinction. Conservation Biology 18:1607-1616. Heske, E. J., S. K. Robinson, and J. D. Brawn. 2001. Nest predation and Neotropical migrant songbirds: piecing together the fragments. Wildlife Society Bulletin 29:52-61. Kawecki, T. J. 2004. Ecological and evolutionary consequences of source-sink population dynamics. Pages 387-411 in I. Hanski and O. E. Gaggiotti, editors. Ecology, Genetics, and Evolution of Metapopulations. Elsevier Academic Press, Burlington, MA. Mahon, L. C , and K. Martin. 2006. Nest survival of chickadees in managed forests: habitat, predator, and year effect. Journal of Wildlife Management In press. Martin, J.-L, and M. Joron. 2003. Nest predation in forest birds: influence of predator type and predator's habitat quality. Oikos 102:641-653. Martinez-Morales, M. A. 2001. Forest fragmentation effects on bird communities of montane cloud forest in eastern Mexico. University of Cambridge, Cambridge, UK. Murphy, M. T. 2001. Habitat-specific demography of a long-distance, neotropical migrant bird, the Eastern Kingbird. Ecology 82:1304-1318. Nagy, L. R., and R. T. Holmes. 2004. Factors influencing fecundity in migratory songbirds: is nest predation the most important? Journal of Avian Biology 35:487-491. Opdam, P., and J. A. Wiens. 2002. Fragmentation, habitat loss and landscape management. Pages 202-223 in K. Norris and D. J. Pain, editors. Conserving Bird Biodiversity: General Principles and their Application. Cambridge University Press, Cambridge, UK. 104 Rangel-Salazar, J. L , P. L. Enrfquez-Rocha, and T. Will. 2005. Diversidad de aves en Chiapas: prioridades de investigacion para su conservacion. Pages 265-323 in M. Gonzalez-Espinosa, N. Ramfrez-Marcial, and L. Ruiz-Montoya, editors. Diversidad Biologica en Chiapas. Plaza y Valdes, Mexico City, Mexico. Rangel-Salazar, J. L , E. Santana-Castellon, J. Carrillo, and P. L. Enriquez-Rocha. In press. Zorzal de Frantzius (Catharus frantzii). in P. Escalante-Pliego, D. Ayala, and V. Nequiz, editors. Libro Rojo de las Aves de Mexico, Mexico D.F. Robinson, W. D., T. R. Robinson, S. K. Robinson, and J. D. Brawn. 2000. Nesting success of understory forest birds in central Panama. Journal of Avian Biology 31:151-164. Schaefer, J. A. 2006. Towards maturation of the population concept. Oikos 112:236-240. Saether, B.-E., and 0 . Bakke. 2000. Avian life history variation and contribution of demographic traits to the population growth rate. Ecology 81:642-653. Tejeda-Cruz, C , and W. J. Sutherland. 2004. Bird responses to shade coffee production. Animal Conservation 7:169-179. Underwood, T. J., and R. R. Roth. 2002. Demographic variables are poor indicators of Wood Thrush productivity. Condor 104:92-102. Watson, D. M. 2003. Long-term consequences of habitat fragmentation-highland birds in Oaxaca, Mexico. Biological Conservation 111:283-303. With, K. A., and A. W. King. 2001. Analysis of landscape sources and sinks: the effect of spatial pattern on avian demography. Biological Conservation 100:75-88. Wolf, J. H. D. 2005. The response of epiphytes to anthropogenic disturbance of pine-oak forests in the highlands of Chiapas, Mexico. Forest Ecology and Management 212:376-393. 105 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0075067/manifest

Comment

Related Items