|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 AVIAN LIFE HISTORY DIVERSIFICATION ALONG AN ANDEAN ELEVATION GRADIENT : A NESTING PERSPECTIVE By GUSTAVO ADOLFO LONDONO GUERRERO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFIL LMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 G ustavo A dolfo L ondoo G uerrero
3 To my m om who always invest ed i n my education and support ed my dreams and adve ntures to study birds in Colombia and other countries support desp ite my long trips to the field
4 ACKNOWLEDGEMENTS I thank my parents for their unconditional suppo rt to become a biologist and explore different jungles search ing for many species of birds and learning about their natural history My wife and eternal love Rosario have been an important part of my career and life. Since we first meet and got to know each other in 1999, when I became deeply in love with her, she e ncourage me to pursue my goals. This support, love and believing in my projects and ideas were stronger since we started dating again in 2006 and got married in 2010, she have been an important motivation to continue during the difficult times. I want to t hank her for understanding and been patience during my long absents due to the long field season during four years with little communication. This dissertation project could not be possible without the invaluable intellectual support and guidance of my two advisers Scott Robinson and Douglas Levey, thanks to them I was able to have many successful proposals that support ed my field expenses and develope d a succes s ful project. I also want to thank them for always have time to talk and provide invaluable advis e that mad e me a better scientist and person. I also want to thank my committee member s Jamie Gillooly, Jane Brockmann, and Daniel Hahn for long and productive conversations and discussions. All the data collected during t his dissertation was possible thanks to the priceless help of volunteers f r om many countries that join me every fall between 2007 and 2011 and spent long hours searching for tropical bird nest in remote areas, with little communication and limited variety of food. I specially want to thank R. Aracil, A. Chinome, S. David, C. Flrez, J. Garizbal, M. Loaiza, A. Miller D. Ocampo S. Prez, C. Smith, G. Valencia for participating in the project during two or more field seasons.
5 T he contribution of every volunteer in the field was very im portant to collect novel information about tropical bird nesting strategies for this reason I want to thank A. Adams, J. Bailey, K. Becraft, J. Bermdez, S. Borchert, C. Bota, J. Brennan, J. Brown, R. Brunner, A. Campuzano, A. Carter, J. Echeverri, K. Eli ot, M. Estrada, S. Flake, E. Fishel, T. Forrester, M. Fuss, D. Grayson, D. Giordano, D. Guevara, R. Grey, R. Hanauer, J. Hazlehurst, E. Johnston, M. Kelley, C. Kingwell, P. Marting, B. McLaughlin, J. Miller, W. Minehart III, A. Montoya, J. Muoz, N. Peralt a, A. Quispe, D. Rincn, M. Sanchez, C. Shake, L. Silver, L. Southcott, E. Tenorio, S. Thornton, J. Ungvari Martin S. Valdez, S. Van Remoortere, J. Vooz, C Zurita. All th e field work was also possible thank s to the logistic support of owners and work ers in Pantiacolla llodge, Tono station, Cock of the Rock lodge and Wayqecha reserach station, for this reason I want to thank D. Blanco, W. Borda, R. Catpo, L. Ccoyo, D. Hermoza, F. Llacma, A. Moscoso, G. Moscoso, E. Ochoa, A. Tejedor, N. Torrez, M. Van Vlaar dingen I thank the INRENA and SERNAP for providing the permits and allowing me to work in the buffer zone and inside the Manu National Park during all these years. Finally I want to thank my lab mates, professor and friends that made my experience at the univer s i ty of florida an u nforgateble one J. Allen, J. Gi llooly, A. Martinez, J. Jankowski, J. Hoover, J. Oswald, K. Paterson, C. Pizano, W. Sh elsky, D. Steadman, and J. Ungvari Martin a n d K. Villet. Funding was received from Pamela and Alexander F. Skutc h Research Award, PSTC Compton International Fellowship Alexander Wetmore Award, TCD Amazon Conservation Leadership Initiative, Gordon and Betty Moore Graduate Scholarships at the University of Florida for Tropical Forest Conservation in the Andes, Asocia cin para
6 la Conservacin de la Cuenca Amaznica Louis Agassiz Fuertes Award, and Dexter Fellowship.
7 TABLE OF CONTENTS Page ACKNOWLEDGEMENTS ................................ ................................ ............................... 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 INTRODUCTORY REMARKS ................................ ................................ ................ 13 2 CHANGES IN NEST PREDATION, PREDATOR IDENTITY AND BIRD COMMUNITY COMPOSITION ALONG AN ELEVATION GRADIENT IN THE ANDES ................................ ................................ ................................ ................... 18 Introduct ion ................................ ................................ ................................ ............. 18 Methods ................................ ................................ ................................ .................. 20 Results ................................ ................................ ................................ .................... 24 Discussion ................................ ................................ ................................ .............. 27 3 EFFECTS OF BIOTIC AND ABIOTIC FACTORS ON AVIAN CLUTCH INVESTMENT AND INCUBATION BEHAVIOR ALONG A TROPICAL ELEVATION GRADIENT ................................ ................................ ........................ 39 Introduction ................................ ................................ ................................ ............. 39 Methods ................................ ................................ ................................ .................. 41 Results ................................ ................................ ................................ .................... 44 Discussion ................................ ................................ ................................ .............. 47 4 TANAGERS OF THE HIGH ANDES HAVE AN ULTRA SLOW PACE OF LIFE .... 60 Introduction ................................ ................................ ................................ ............. 60 Methods ................................ ................................ ................................ .................. 61 Results ................................ ................................ ................................ .................... 65 Discussion ................................ ................................ ................................ .............. 66 5 CONCLUSION ................................ ................................ ................................ ........ 75 APPENDIX A NUMBER OF NESTS DEPREDATED BY INDICATED SPECIES OF TAXONOMIC GROUPS. ................................ ................................ ......................... 78
8 B NUMBER OF NESTS ENCOUNTERED FOR THE 84 BIRD SPECIES IN THE MANU GRADIENT AND THE SPECIES ASSOCIATED WITH EACH COMMUNITY. FROM THE 84 SPECIES, TEN WHERE SHARED AMONG COMMUNITIES. ................................ ................................ ................................ ..... 82 C VOLUNTEERS THAT WORKED DURING LONG HOURS AND MONTHS SEARCHING AND MONITORING BIRD NESTS BEWTEEN AUGUST AND DECEMBER TO MAKE THIS DISSERTATION POSSIBLE. ................................ .. 85 LIST OF REFERENCES ................................ ................................ ............................... 86 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 96
9 LIST OF TABLES Table P age 2 1 Site characteristics for stations found along the Manu elevational gradient, including geographic coordinates, elevational range, average mean monthly temperature and average canopy height of the forest, SD. .............................. 34 3 1 Avian incubation investment predictions in environments with different nest predation risk and ambient temperatures. ................................ .......................... 57 3 2 Full model for general and community analysis of covariance (ANCOVA). ........ 58 3 3 Output from the Principal Component Analysis (PCA) realized in the 56 species and 62 individuals ................................ ................................ .................. 59 4 1 Association between environmental and nesting variables in the tanager clade species. Principal Component Analysis output. ................................ ........ 70 4 2 Association between environmental and nesting variables in the tanager clade species. Principal Component Analysis output. ................................ ........ 71 A 1 Number of nests depredated by indicated species of taxonomic groups. ........... 78 A 2 Tanager species list. Environmental and nesting variables associated with the tanager clade ................................ ................................ ................................ 79 B 1 Number of nests encountered for the 84 bird species in the manu gradient and the species associated with each community ................................ .............. 82
10 LIST OF FIGURES Figure Page 2 1 Daily nest predation rates along an Andean elevational gradient between 350 3100m ................................ ................................ ................................ ......... 35 2 2 Percentage of nests depredated by different predators at four stations distributed along the 2750 m elevational gradient. ................................ ............. 36 2 3 Percent of diurnal and nocturnal nest predation events distributed along the elevational gradient ................................ ................................ ........................... 37 2 4 Dissimilarity values for bird species composition and differences in daily nest predation along the elevational gradient between 850 and 3100m .................... 38 3 1 Associations between the main factors that aff ect bird embryonic development (egg mass and egg temperature) ................................ .................. 53 3 2 Variation in selective factors among bird communities on the elevation gradient ................................ ................................ ................................ .............. 54 3 3 Principal Component Analysis among nesting variables ................................ .... 55 3 4 Change s in nesting variables ................................ ................................ .............. 56 4 1 Principal component axes that explain 65% of the variation for nesting variables associated with tanagers nesting in the Manu elevation gradient ....... 72 4 2 Clutch size phylogenetic signal in the tanager clade ( = 89.98, P <0.0001) ..... 73 4 3 Principal component axes that explain 68% of the variation associated with the nesting variables in the tanager clade 74 C 1 Volunteers that worked during long hours and months searching and monitoring bird nests bewteen August and Decembe r to make this dissertation possible. ................................ ................................ .......................... 85
11 A bstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy AVIAN LIFE HISTORY DIVERSIFICATION ALONG AN ANDEAN ELEVATION GRADIENT: A NESTING PERSPECTIVE By Gustavo Adolfo Londoo Guerrero May 2013 Chair: Scott K. Robinson Cochair: Douglas J. Levey Major: Zoology Many tropical birds have life history traits such as small clutches and long fitness in the face of a relatively benign climate and intense biotic constraints on reproduction -namely, nest predation and competition. Nest mortality plays an important selective role in shaping avian life h istory traits, ecology and the composition of communities. Nest predation is the most common type of nest failure; thus, nest predators directly influence avian reproductive success. To evaluate these ideas I examine three perspectives: 1) if variation in nest predation rates and predator frequency could act as ecological factors influencing species turnover patterns along an Andean elevational gradient. 2) If interspecific shifts in life history traits of tanagers along a tropical elevation gradient are pr omote by changes in abiotic factors with elevation. 3) If tropical bi r d species evolving under relatively stable climates make similar investment decisions within elevation bands, but to differ considerably among elevation bands. To evaluate these ideas I study nesting behavior, success and predation on an elevation gradient (350 3100m) in the Manu National Park, Cusco and
12 Madre de Dios, Peru between 2007 and 2011 I found that: 1) Nest predation rates decreased non linear with an abrupt decrease between 1 100 1350m, the zone in which predation by snakes and primates declined steeply. In general, community turnover was highest between zones in which nest predation rates and composition of the predator community changed the most. 2) Six of the 13 tanager spec ies had a clutch size of one egg that was associated with longer incubation and nestling periods, lower nestling growth, and colder climates. 3) Incubation period increase with elevation and incubation strategies varied considerable among species within el evation band. These results suggest that nesting birds decision in the elevation gradient vary accordin g ly with changes in abiotic and biotic factors, which relative importance change with elevation. Dramatic shifts in nest predator communities and nest p redation rates can explain variation in nesting success of species, and the slow pace of life in tanagers in the high Andes results from cold temperatures.
13 CHAPTER 1 INTRODUCT ORY REMARKS One of the key goals of life history studies is to understand h ow biotic (e. g., predation) and abiotic (e. g., temperature) factors affect behavior and reproductive effort of an organism throughout its life ( Reznick & Endler 1982 ; Stearns 1992 ; Roff 2002 ) Variation in the number of eggs among populations of oviparous species, for example, has been linked to differences i n parental care ( Stearns 1992 ; Roff 2002 ). Oviparous species such as birds, which invest greatly in parental care, tend to produce fewer eggs than similar taxa with less parental care. Although increasing investment in parental care typically accelerates embryonic development, it is costly for the parents and may constrain future reproduction. In particular, parental care entails trade offs between self ma intenance behaviors (e. g., foraging) and embryonic development. These trade offs are exacerbated when nesting oviparous species encounter environments with harsh biotic and abiotic factors such as high predation and cold temperatures that increase reprodu ctive and parental care costs ( Ghalambor & Martin 2002 ; Lima 2002 2009 ) Although it is generally accepted that investment decisions vary in contrasting environments ( Berven 1982a b ; Baur & Raboud 1988 ; Badyaev 1997 ; Rohr 1997 ; Hancock et al. 1998 ; Badyaev & Ghalambor 2001 ) there is controversy about the proximate mechanisms that promote this variation ( Badyaev & Ghalambor 2001 ; Ricklefs & Wikelski 2002 ; Martin 2004 ) One such controversy concerns the hypothesized mechanism s underlying the small clutch sizes and long incubation periods observed in many tropical birds. These traits have been generalized as latitudinal with a low pace o f life: lower feeding rates and slower nestling growth ( Skutch 1949 ;
14 Martin et al. 2011 ) lower adult mortality ( Gill & Haggerty 2012 ) and lower metabolic rates ( Wikelski et al. 2003 ; Wiersma et al. 2007 ) Alexander Skutch, for example, hypothesized that high biotic constraints, in particular high rates of nest predation, would select for smaller clutches and the reduced parental investment per nesting attempt associated w ith the slow pace of life (see also, ( Martin et al. 2006 ; Martin et al. 2007 ; Martin et al. 2011 ) Yet, high rates of nest predation do not explain the longer incubation periods in species with small clutch sizes ( Skutch 1 985 ; Ricklefs & Wikelski 2002 ; Robinson et al. 2010 ; Roper et al. 2010 ) To date, most studies of these traits (e. g., small clutches, long incubation and nestling period) have involved latitudinal comparisons that might be confounded by multiple other factors that vary with latitude (e.g., productivity, day length, phylogenetic structure, seasonal ity; ( Ricklefs & Wikelski 2002 ; Martin 2004 ; Robinson et al. 2010 ) One possible way to control for many of these variables (e. g., climate, nest predation pressure, productivity, phylogenetic structure) is to focus on variation in life history traits in contrasting environments within the tropics, especially in environments that vary predictably in rates of nest predation and in abiotic variables such as temperature. Therefore, in my thesis I decided to exploit the variation in these selective pressures along an elevation gradient within the Neotropics by expl oring how nesting strategies, particularly incubation, responds to these contrasting environments. Tropical mountains have great promise as a place to uncover the mechanisms responsible for the slow pace of life observed in the reproductive behavior of t ropical birds ( Ricklefs & Wikelski 2002 ) because nest predation and ambient temperature change in a predictable way. The proportion of nests depredated ( Skutch 1985 ; Boyle
15 2008 ) and ambient temperature ( Janzen 1967 ; Ghalambor et al. 2006 ) decrease linearly with elevation, which makes i t possible to test if changes in investment in eggs and incubation are associated with nest predation, ambient temperature, or both. Overall, bird species that nest in areas with high nest predation have smaller clutches and eggs ( Martin et al. 2006 ) increase the proportion of the day they spend on the nest incubating the eggs ( Ghalambor & Martin 2002 ) and have shorter incubation periods ( Martin et al. 2 007 ) In contrast, species that nest in colder climates have smaller clutches ( Badyaev & Ghalambor 2001 ; Lu et al. 2010 ) increase their foraging trips ( Conway & Martin 2000a ) and have longer incubation periods ( Lu et al. 2010 ). Therefore I predicted that bird species should change incubation investment decisions throughout tropical elevation gradients. I predicted that bird species nesting at high and low elevation would have smaller clutches and eggs compared with birds nesting at middle elevations, but that incubation constancy should be similar acros s the gradient. To test these predictions I used data from 84 species for which I had incubation period data along this 2750 m Andean elevation gradient in southeastern Peru. Before I could test these predictions, however, I needed to know if indeed nes t predation decreases linearly with elevation, as reported in previous studies ( Skutch 1985 ; Boyle 2008 ) Those s tudies were based either on data from artificial nests baited with eggs ( Boyle 2008 ) or on data from many widely scattered sites in Central America ( Skutch 1985 ) A better test of the hypothesis that nest predation changes systematically with elevation would be based on observations of real nests along one elevation gradient. In the first chapter of my thesis, I estimated the daily nest predation rate from data on 1584 nests of 245 species along the Manu Road in Peru. To help
16 understand the mechanisms underlying any changes in nest predation with elevatio n, I used camera traps to document the identity of nest predators at four different elevation ranges (350 500 m, 700 1100 m, 1200 1600 m, and 2300 3100 m). I found that nest predation decreases with increased elevation, but it does not do so linearly, as previously suggested ( Skutch 1985 ; Boyle 2008 ) I found that nest predation decreases dramatically between foothill and cloud forest bird communities (110 0 1350 m). Furthermore, I found that this decrease in nest predation was associated with a dramatic decrease in nest predation by snakes -from 46.4% in the lowlands to 20.8% in the cloud forest and 0%; forests from 2300 3100m. The majority of bird nests at all elevations were depredated during the day, therefore, selection for behaviors around the nest (e.g., number of trips to and from the nest) should be similar along the gradient. In the second chapter I explored how the abrupt change in nest predati on in the foothills and the linear decrease in ambient temperature over the entire gradient affected incubation time across 84 species. First, I evaluated whether daily nest predation rates among the 84 species followed the same general pattern that I docu mented for the sample from the entire community of 245 species. I found that nest predation decreased significantly in the cloud forest (1200 1600 m), but increased at higher elevations. In general, I found that changes in investment decisions were mainly associated with ambient temperature. Although the clutch size (the number of eggs produced per nesting attempt) was relatively constant along the gradient, egg mass increased with elevation. Similarly, egg temperature decreased with elevation, which suppo rted the hypothesis that low temperatures constrain embryonic development
17 ( Badyaev & Ghalambor 2001 ; Olson et al. 2006 ) Although incubation period did not vary significantly among bird communities, it had a tendency to increase with elevation. In the third and final chapter I show that for one clade (tanagers) the "p ace of life" varies substantially even within the tropics and document further clutch reduction than other tropical species ( Martin 2004 ; Robinson et al. 2010 ) Of particular interest, I discovered that six species of mountain tanagers had one egg clutches, thus decreasing their potential fitness and reproductive investment by 50% relati ve to species with the typical clutch size of two eggs, all else being equal. Based on a literature review for the monophyletic tanager clade, this is the first time one egg clutches have been reported. Clutch size has a strong phylogenetic signal in this clade, but it is variable among species in the mountain tanager sub clade. Additionally, I found that tanager species with one egg clutches inhabit higher elevations and had longer incubation and nestling periods than tanager species with two and three e gg clutches. I also collected more detailed nesting data for 12 tanager species that nest in the elevation gradient. These data also showed that species with one egg clutches inhabited high elevations and had longer nestling periods and slower nestling gro wth rates. Taken together, these results suggest that the combination of constant severe abiotic constraints (e. g., cold temperature) and high nest predation severely constrain tanager nesting and lead to an extremely slow pace of life; tropical tanagers living at high elevations are likely "making the best of a bad situation."
18 CHAPTER 2 CHANGES IN NEST PREDATION PREDATOR IDENTITY AND BIRD COMMUNITY COMPOSITION ALONG AN ELEVATION GRADIENT IN THE ANDES Introduction Nest mortality is proposed to play an i mportant role in avian ecology and community composition ( Cresswell 1997 ; Martin et al. 2000 ; Lima 2009 ; Cresswell 2010 ; McKinnon et al. 2010 ) Predation is the most important type of nesting failure an d, hence, the most important type of predator prey interaction is likely with nest predators ( Ricklefs 1969 ; Martin & Roper 1988 ) Experimental studies have demonstrated that birds use nest predator cues to choose among habitats, such that individuals exposed to habitats with different levels of predation risk prefer those habitats with lower risk ( Fontaine & Martin 2006 ; M orosinotto et al. 2010 ; Parejo & Aviles 2011 ) Spatial variation in nest predation has also been used to explain the evolution of bird migration along large scale (e.g., latitudinal) and small scale (e. g., elevational) gradients ( Boyle 2008 ; McKinnon et al. 2010 ) Taken together, these studies suggest that reg ional differences in avian life histories and reproductive success can be caused by spatial variation in nest predation ( Matysiokov et al. 2011 ) Although nest predation rates vary across landscapes ( Chalfoun et al. 2002 ; Martin et al. 2006 ; Matysiokov et a l. 2011 ) our understanding of the factors responsible for this variation is limited. Nest concealment, nest type, and parental behavior have been used to explain variation in nest predation, but results are rarely consistent among studies ( Cresswell 1997 ; Fontaine et al. 2007 ; Matysiokov et al. 20 11 ) Variation in the identity and relative importance of nest predators is increasingly considered to be an important predictor of nesting success and an important explanatory factor in patterns of nest predation ( Benson et al. 2010 ; Reidy & Thomson
19 2012 ) For example, one possible source of variation in predation rates is spatial variation in nest predator co mposition ( Rodewald & Kearns 2011 ; Cox et al. 2012 ) Studies of forest bird communities, however, have not yet linked geographical variation in avian nest predation rates with variation in predator community composition. Understanding the relative importance of different nest predators and their activity patterns should improve our understanding of which life history traits may allow bird species to th rive in different habitats or at particular locations along environmental gradients ( Lima 2002 ) Habitats dominated by diurnal predators, for example, will impose very different selective regimes on a vian life histories than those dominated by nocturnal predators ( Carter et al. 2007 ; Reidy & Thomson 2012 ) While the high avian community turnover found along tropical elevational gradients has been partially linked to biotic factors such as competit ion and vegetation structure ( Jankowski et al. 2012 ) other biotic factors such as nest predation may also influence species elevational distributions and could have far reaching effects across montane species. Evidence from both artificial nests ( Boyle 2008 ) and from studies of actual nests from widely scattered sit es at different elevations ( Skutch 1985 ) shows that nest predation varies w ith elevation in the Neotropics. If there is substantial variation in actual nest predation rates with elevation, as these studies have suggested, then I may also expect nest predation to be an important driver of changing species composition with elevatio n, potentially reinforcing the elevational range limits of multiple species. In addition to changes in the rate of nest predation with elevation, shifts in the predator community may also be critical determinants of nesting success of species at different elevations and may also underlie patterns of species turnover.
20 Here I report patterns of nest predation along a large scale elevational gradient (350 3100m) in the eastern Andes of Peru ( Walker et al. 2006 ) a diverse region characterized by h igh avian species turnover ( Patternson et al. 1998 ; Jankowski et al. 2012 ) Specifically, I sought to answer four questions: (1) Do overall rates of nest predation vary with elevation? (2) Does taxonomic composition of nest predators vary with elevation? (3) Do daily activity patterns of predators and the timing of nest predation (nocturnal vs. diurnal) vary with elevation? (4) Do abrupt shifts in nest predation along the gradient corres pond with abrupt shifts in bird species composition? If variation in nest predation rate is one of the mechanisms underlying shifts in the composition of bird communities along this gradient, then I predicted that there will be substantial bird community t urnover in zones where there are dramatic changes in nest predation rates (e. g., between 1000 1500 m) ( Terborgh 1971 ; Rahbek 1977 ; Patternson et al. 1998 ; Herzog et al. 2005 ; Jankowski et al. 201 2 ) Methods Study site : This study was conducted along an elevational gradient in the southeastern Andes of Peru in the Manu Biosphere Reserve between 2008 and 2011, during the height of the breeding season each year (August to December). Data collection took place at four stations distributed along this gradient (Table 1). The low elevation station, Pantiacolla, is located in the buffer zone of Manu National Park between the Pantiacolla ridge and the Alto Madre de Dios River. The site features lowland tr opical rainforest, with a combination of high canopy floodplain and terra firme habitat. The second station, Tono, is found in the Andean foothills near the Tono River inside Manu National Park. This site contains typical premontane forest with rolling hil ls and outlying ridges, where forest canopy is slightly shorter than lowland forest. The mid
21 elevation station, San Pedro, is found within the Kcosipata Valley near the intersection of the San Pedro and Kcosipata Rivers. This site contains lower montane cloud forest, with a high density of mosses and epiphytes and an average canopy height half that of foothill forest. The high elevation station is at the Centro de Investigacin Wayqecha near forest treeline and harbors the last low canopy cloud forest mov ing upslope before puna grassland becomes dominant at higher elevations. The San Pedro and Wayqecha stations are found along the Manu road, which descends along the eastern slope of the Andes down to the foothills ( Patternson et al. 1998 ; Jankowski et al. 2012 ) Nest search: To address whether nest predation varies with elevation, I searched for and monitored nests along the gradient following protocols approved by the University of Flor ida IACUC. Nest searching teams were assembled at each station, where an average of nine (Pantiacolla), seven (Tono), seven (San Pedro) and five (Wayqecha) people searched for nests for ten hours a day, six days a week from mid August until mid December. At all stations, each team member was assigned to different 5 10 ha plots, where they searched for and monitored nests throughout the season. To cover additional elevations, I conducted weekly nest searches between 1500 and 3000 m along the Manu Road. I used two techniques to monitor nests, depending on nesting stage: (1) During the incubation period I positioned a thermal sensor inside the nest connected to a U12 4 channel Hobo data logger (Onset Computer Corporation, MA, USA http://www.onsetcomp.com) th at recorded nest temperature every minute, allowing us to continuously monitor incubation activity. Thermistors registered arrivals and departures of birds, consistently recording abrupt changes in temperature (<1.5 C). (2) For nests with eggs or young, I deployed motion
22 cameras (Rapidfire professional; Reconyx, Inc. WI, USA; between 10 and 27 per station) placed on tripods 0.5 to 1m from the nest and covered with vegetation to reduce disturbance and conspicuousness. Cameras were programmed to take ten ph otos when camera sensors detected movement at the nest. Once a camera or data logger was set, it was removed either after a predation event, when eggs hatched (only nests with data loggers), or when the nestlings left the nest (only nests with cameras). Th ese techniques allowed us to continuously monitor more than 90% of encountered nests. Nests with nestlings were visited daily and some were continuously monitored with motion cameras based on camera availability. Nests with data loggers and cameras were vi sited every six days to download data and replace batteries, and nests without data loggers or cameras were visited every three days, except those along the road, which were visited weekly. Daily nest survival probability did not vary among nets monitorin g techniques: nest with camera traps (0.95; CI 95% = 0.94 0.96, n = 122), data loggers (0.96; CI 95% = 0.95 0.96, n = 558), camera trap and data logger (0.96; CI 95% = 0.95 0.97, n = 182) and traditional three to four day monitoring (0.93; CI 95% = 0.92 0. 93, n = 721). To reduce disturbance and decrease the impact created by unfamiliar objects around the nest, I covered the data loggers, cables, and camera traps with vegetation (e.g., dry leaves, moss). Also, to reduce disturbance and avoid nest abandonment I avoided revisiting nests during the laying period, when birds are more prone to abandon their nest; eggs were not measured and monitoring equipment was not put in place until 2 3 days after clutch completion in most nests.
23 Data analyses : To estimate d aily nest predation rate, I used a nest survival model constructed using the program MARK ( Rotella 2007 ) This model allowed us to estimate daily nest survival probability using the number of exposure days to calculate probability of su rvival. I used all nests with known fate (depredated or successful), regardless of monitoring technique (n = 1584). Variables in the model included: (1) the first date an active nest was encountered; (2) the last date the nest was checked and eggs or nest lings were present; (3) the last date the nest was checked; (4) the fate of the nest (0 = successful, 1 = depredated); and (5) the nest location within continuous elevational bands of 250m, between 350 and 3100 m (i.e., 350 600m, 600 850m, etc.). The gradi ent was divided into 250m bands because it was the minimum range with adequate sample size that would not affect nest daily survival probabilities. I examined the change in taxonomic composition of nest predators with elevation using photos provided by cam eras to identify predators. I tested for significant variation with elevation in the proportion of these broad taxonomic groups (i.e., birds, mammals, and reptiles) using a Chi squared test. To examine whether the time at which nest predation occurred (d ay vs. night) varied with elevation, I noted the time of nest predation (n = 505 nests). Daytime was defined as hours between 06:00 and 18:00 and nighttime between 18:01 and 05:59. Time of nest predation was usually obvious for nests with camera traps. F or nests with data loggers, a predation event was marked by prolonged departures of incubating adults that broke the normal cycle of incubation. Although less direct than the use of cameras, the accuracy of this technique has been independently validated ( Libsch et al. 2008 )
24 To address whether changes in nest predation corresponded with areas of high turnover in bird species along the elevational gradient, I compared daily nest predation risk w ith data on species turnover from elevational zones between 850 3100m (as sufficient lowland species distribution data are not yet available). For daily nest predation risk, I estimated daily nest predation probability for the nine 250 m elevational zones between 850 and 3100 m. Species turnover data come from a large number of point count surveys (n=940) conducted during the early avian breeding seasons (July November) of 2006 2009 at 235 survey sites (see Jankowski et al. 2012 for more details on survey m ethods). Species detected during any visit to the survey site were considered present. Survey sites within 250 m elevational zones were combined to create a site species presence absence matrix for 367 bird species and nine 250 m elevational zones. Differ ences in species composition between elevational segments ( Lennon et al. 2001 ) (to create a dissimilarity matrix. I used a generalized dissimilarity model (GDM; which allows for nonlinear relationships between predictor and response matrices ( Ferrier et al. 2007 ) to determine the variation in bird dissimilarity among zones explained by differences in daily nest predation risk. Results Daily nest predation was estimated for 1584 nest s from 245 bird species and predator identity was documented in 136 cases. Additionally, the time of day when nest predation occurred was recorded for 505 nests. Daily n est predation rates : Community wide nest predation rates decreased no nlinearly with ele vation ( Figure 2 1 ). Daily nest predation rates were high at low elevations (7.1%; 350 600m), reached a peak at foothill elevations (9.8%; 850 1100m),
25 abruptly decreased to 3.8% above 1350m, and remained fairly constant at higher elevations (1.3 4.7%). Hen ce, daily nest predation in the foothills was 1.4 times greater than in the lowlands and 2 3 times greater than elevations above the foothills. Taxonomic composition of predator communities : All taxonomic groups were recorded as nest predators at all eleva tions, with the exception of reptiles and invertebrates, which were never recorded predating nests at high elevations ( Figure 2 2 ). Overall, reptiles and mammals were the most frequent nest predators (37.8% and 38.5% of all predation events, respectively) followed by birds (16.9%) and invertebrates (6.7%). The relative importance of each predator type changed with elevation (X2 = 58.3, df = 6, P < 0.0001, n = 138; Figure 2). Reptiles were the most frequent predators recorded at the two lower stations, co mprising 39.4% of predation events in the lowlands (350 600m) and 46.4% of predation events in the foothills (750 1100m). The frequency of reptile predation events decreased abruptly to 20.8% above 1250m where only one nocturnal species of snake was record ed between 1250 and 1600m (Oxyrhopus petola; Appendix 1). Within the reptile category, snakes were responsible for 95.7% of predation events, and among snakes, one species, Pseustes sulphureus, was responsible for 78.3% of predation events. The frequency of nest predation by mammals decreased by 83.3% from the lowlands to the foothills (750 1100m) and then increased by 76.5% at higher elevations ( Figure 2 2 ). Among mammalian predators, marsupials and primates were detected with the same frequency and toget her accounted for 72% of all mammalian predation events. Primates accounted for 65% of mammalian nest predation events between 350 600m and 35% between 1250 1600m, but I did not document any primate nest predation event in the foothills between 750
26 1100m. In contrast, marsupials accounted for 30% of the mammalian predation events between 350 550m, 20% between 750 1100m, and 45% between 1250 1600m. Nest predation frequency by birds was low below 1600m, but birds were the most frequent nest predators, along with mammals, at higher elevations ( Figure 2 2 ). Among birds, most (92%) predation events between 350 1100m were by toucans. At high elevations (2500 3000m), 66% of bird predation events were by one species, Turdus fuscater. Invertebrate nest predation was the rarest type of nest loss at all elevations, except between 1250 1600m; ants accounted for 70% of all invertebrate predation events. Timing of nest predation : Overall, 331 predation events occurred during the day and 174 occurred at night ( Figure 2 3a). Nocturnal and diurnal predation events varied among predator taxa. Birds and reptiles depredated nests mainly during the day: 91.7% and 73.3% of bird and reptile predation events, respectively, occurred during daylight hours ( Figure 2 3a,c). In con trast, mammals tended to depredate nests during the night (56.4%; Figure 2 3b). Diurnal nest predation did not change significantly with elevation among birds (R2 = 0.09, P = 0.17) or among snakes (R2 = 0.001, P = 0.80; Figure 2 3a,c), but mammalian nest predation was exclusively nocturnal at high elevations (R2 = 0.12, P =0.009; Figure 2 3b). As elevation increased, nest predator communities became dominated by a few nocturnal mammals and diurnal birds (e.g., Turdus fuscater). Nest predation and bird spec ies turnover : Along the elevational gradient, I found that abrupt increases in rates of nest predation occurred at the same two elevational zones where the composition of bird species changed most dramatically ( Figure 2 4). When comparing the bird species composition of adjacent elevational
27 zones, I found the highest species turnover between the elevational zones of 850 1100m and 1100 1350m where approximately 60% of bird species were replaced. This shift in bird species composition was associated with a 2 .2% increase in daily nest predation compared with the lowlands and a 6.0% increase compared to most elevations above the foothills ( Figure 2 4). Similarly, the second largest shift in bird species composition occurred between the elevational zones of 1850 2100 and 2100 2350m, where roughly 50% of bird species were replaced. At this elevation, I observed a 3.4% increase and 1.6% decrease in daily nest predation compared to elevations immediately below and above this elevation, respectively ( Figure 2 4). A cross the entire gradient, differences in nest predation rates alone explained 29.8% of the variation in compositional dissimilarity of species among elevational zones, and were positive (R2 = 0.33, P = 0.08). Discussion Although nesting success increased with elevation, it did not do so in a linear fashion, as previously suggested ( Skutch 1985 ; Boyle 2008 ) The dail y rate of nest predation I recorded at low elevations, 7%, is near the upper end of the range of daily nest predation values reported for tropical forest birds (0.038 0.083 ) ( Robinson et al. 2000 ; Martin et al. 2006 ) In contrast, the rate of daily nest predat ion of 0.098 in the foothills is among the highest rate s recorded for a tropical bird community. Furthermore, this nest predation rate in the foothills is even higher than those recorded from small forest fragments thought to be population sinks for nesti ng songbirds in North America ( Robinson et al. 1995 ) Above 13 50m, nest predation rates abruptly declined from 9. 8% to 3.8 %, beginning to approach those more typical of temperate and cloud forest habitats (~0.038; Martin et al. 2006).
28 I found support for my prediction that nest predation rates and frequency of nest p redators change most abruptly in zones of high bird species turnover. My analysis of species turnover found that the abrupt increase in nest predation between 850 1350m coincided with the largest peak in bird species turnover along the gradient. Likewise an increase in nest predation between 1850 2350 coincided with a smaller peak in species turnover along the gradient High species turnover in foothill elevations (i.e., between 1000 1500m) has been reported along other tropical mountains ( Rahbek 1977 ; Herzog et al. 2005 ) ; however, the abiotic (e.g., temperature, precipitation) and biotic (e.g., competition, vegetation structure) f actors considered in these studies have not provided a mechanism for why so many lowland bird species disappear above this elevational band, or alternatively, why so many highland bird species do not extend their ranges downslope into a thermally benign f oothill environment. Physiological limitations have been proposed as a mechanism to explain the narrow distribution of birds on elevational gradients ( Janzen 1967 ; Ghalambor et al. 2006 ) and competition has been proposed for setting the lower and upper distr ibution s of species ( Terborgh 1971 ; Remsen & Graves 1995 ) There are no data available, however, on avian physiology for species on tropical mountains to test the former hypot hesis, and interspecific competition may explain some congener replacements but not community wide turnover patterns ( Terborgh 1971 ; Jankowski et al. 2010 ) I propose that high nest predation pressure below cloud forest elevations at least partially explains why many montane tropical species have not extended their ranges downslope to warmer and more productive site s. The extraordinarily high nest predation found in the foothills in combination with lower temperatures compared to lowland habitats may also truncate the upper
29 distribution of lowland birds Rahbek (1997) demonstrated that the high peak in bird species richness observed in the foothills of tropical mountains is mainly driven by the overlap of elevational ranges limits between highland and lowland communities (see also Herzog et al. 2005) and proposed that foothill tropical bird communities may be margin al populations that are poorly adapted to local conditions. My results support this hypothesis and provide a plausible mechanism high rates of nest predation may make foothills suboptimal habitat for most birds. Thus, despite the high species richness obs erved in the foothills of tropical mountains, this elevational zone may represent a sink habitat, where lowland and highland nesting birds may not be able to cope with local conditions, specifically, a combination of cooler temperatures (relative to lowlan d thermal regimes) and high nest predation. The abrupt decrease in rates of nest predation above the Andean foothills was coincident with a shift in the relative frequency of common nest predators at lower elevations. The loss of diurnal snakes and mammals likely played a role in the community wide decrease in predation rates from the foothills to cloud forest. Snake predation frequency increased in the foothills where m ammalian nest predation frequency decreased, which made snakes by far the most importan t predators in this zone. Among snakes the species most frequently observed depredating nests in the lowlands was Pseustes poecilonotus (83%), which has been reported as a common nest predator in other lowland tropical forests ( Robinson et al. 2005 ; Tarwater 2008 ) Given how frequently P. sulphureus preyed on nests below 1100 m and its complete absence as a nest predator above 1200 m the elevational distribution of this single
30 species may be largely responsible for the overall drop in nest predation above 1200 m the Andea n foothills. Both diurnal and nocturnal predation events were recorded at all elevations and were observed in all taxonomic groups of predators, but the frequency of diurnal predation decreased with elevation, primarily because snakes and diurnal mammals became increasingly rare at higher elevations. Because many of the life history and behavioral traits associated with reducing nest predation rates (e.g., number of foraging trips, active mobbing of nest predators) are only affected by diurnal nest predat ors, this reduction of diurnal nest predation at high elevations may relax selection on these traits. Nocturnal nest predation, however, is not trivial anywhere on the gradient T he combination of nocturnal and diurnal predation could favor diverse nestin g strategies across bird species, depending on the suite of common predators and differences in nest searching strategies among those predators. For example, bird species whose nests are more likely to be depredated by common diurnal predators (i.e., snak es, primates and birds) may show a reduced number of foraging trips ( Fontaine et al. 2007 ) but I would not expect similar traits for species that are depredated mainly by noctu rnal mammals. It is likely that species nesting at low elevations and in the foothills require a suite of adaptations that enable them to cope with extreme levels of nest predation; such characteristics could include short nesting cycles ( Badyaev & Ghalambor 2001 ; Lu et al. 2010 ) and minimizing the number of trips to the nest (with correspondingly long incubation bouts ; Badyaev and Ghalambor 2001 ). Such high selection pressures imposed by predation may be strongest for common species if nest predation is density dependent ( Martin & Roper 1988 ) with the result that rare species, with low population
31 densities, have an advantage in these environments ( Martin & Roper 1988 ) Species nesting at higher elevations, which encounter cooler environments, should be expected to have longer nesting cycles and make more frequent trips to and from the nest ( Badyaev 1997 ; Conway & Martin 2000a ; Lu et al 2010 ) but these are traits that would make them less able to cope with high levels of nest predation at lower elevations, especially those in the foothills. A comparison of life history traits of species nesting above and below this transition zone w ould shed light on the potential for this abrupt shift in nest predation rates to act as a habitat filter ( Reznick & Endler 1982 ) Although the increment al increase in nest predation observed in the second zone of species turnover from 1850 2350m seems small relative to the changes in the foothill region it may impose a strong selection pres sure in nesting traits above this elevation compared to the foothills. Birds nesting in cold temperatures can increase energy expenditure up to three times basal metabolic rate ( Williams 1996 ) and the low productivity of these en vironments ( Salinas et al. 2011 ) may not supply enough food to compensate for harsh weather conditions ( Moreno 1989 ) Thus, I propose that middle elevations ( cloud f orest ) can serv e as a refuge for nesting birds, because nesting birds encounter moderate temperatures, high productivity and low nest predation at this elevation. In contrast, birds nesting at high elevation encounter an increase in nest predation, low tem perature and productivity, and birds nesting at lower elevations encounter high productivity but also high nest predation. Additionally, Boyle (2008) hypothesized that species inhabiting low elevations with high nest predation rates could increase fitness by moving upslope to breed, seeking refuge from predation. A previous study o f elevation migration at this site,
32 however, did not find frequent migration from elevations with high nest predation to elevations with low nest predation ( Merkord 2010 ) I hypothesize that the extra time and energy of nesting in colder climates at high elevations would require more time for development, resulting in longer nesting periods with increasingly colder thermal regimes ( Badyaev 1997 ; Lu et al. 2010 ) and this could counteract presumed benefits of having lower daily predatio n rates at high elevations. While I agree with Boyle (2008) that birds could benefit from moving upslope to breed, for elevational gradients with strong nonlinear change s in daily predation rates, such as Manu, I would expect that this benefit would only be realized for species moving from the lowlands, especially from the foothills, to lower cloud forest (1500 2000m) where thermal regimes are moderate and predation rates are low. These movements, however, are uncommon at my study site (Merkord 2010). Co nclusion : My study revealed that changes in the relative importance of different nest predators and their activity patterns likely contribute to variation in nesting success along an elevational gradient. Furthermore, my work suggests that this variation may limit the elevational distributions of some tropical bird species, with community wide implications. Nest predation seems especially likely to contribute to changes in bird species composition in the foothills of tropical mountains, and possibly withi n upper cloud forest. I view this study as a first step towards evaluating the hypothesis that nest predation contributes to species turnover and life history evolution along Neotropical montane gradients, and propose that nest predation is a critical fac tor both promoting and maintaining the high beta diversity and extraordinary regional diversity found in the Andes ( Rahbek & Graves 2001 ) The next steps fo r testing this
33 hypothesis will require data on life history traits as potential adaptations to nest predation, such as egg and clutch mass, incubation and nestling periods and behaviors during nesting, consideration of abiotic factors that change along the gradient (i.e., temperature, rainfall and productivity) and biotic factors such as interspecific competition. Understanding the interactions and relative importance of these factors will provide a more complete picture of the ecological mechanisms that p romote species turnover along tropical elevational gradients.
34 Table 2 1 Site characteristics for stations found along the Manu elevational gradient, including geographic coordinates, elevational range, average mean monthly t emperature and average canopy height of the forest, SD.
35 Figure 2 1 Daily nest predation rates along an Andean elevational gradient between 350 3100m. Boxes represent 75% of data for each band, and bars show 95% confident intervals. The numbers above boxes show the number of nests included in the analysis for each elevational band.
36 Figure 2 2 Percentage of nests depredated by different predators at four stations distributed a long the 2750 m elevational gradient.
37 Figure 2 3 Percent of diurnal (gray bars) and nocturnal (black bars) nest predation events distributed along the elevational gradient for a) all the nests for which the time of predatio n is known, obtained from data logger and camera trap records (n= 505) and percent of diurnal and nocturnal nest predation events where predator identity, recorded exclusively with camera traps, is known for b) birds, c) mammals, and d) reptiles.
38 Figu re 2 4 Index; black line) and differences in daily nest predation (gray line) along the rity Index compares bird species composition of two adjacent 250 m elevational zones so that peaks in species turnover with elevation can be identified. Daily nest predation is estimated for each 250 m elevational zone, and differences in daily nest predat ion between adjacent elevational zones allow peaks in daily nest predation change to be identified. The x axis (Elevational Zone Comparisons) indicates the elevation of the lower of the two adjacent zones compared.
39 CHAPTER 3 EFFECTS OF BIOTIC AND ABIO TIC FACTORS ON AVIAN CLUTCH INVESTMENT AND INCUBATION BEHAVIOR ALONG A TROPICAL ELEVATION GRADIENT Introduction Food resources and parental care provided to the embryo will affect the survival of progeny ( Stearns 1992 ; Reznick et al. 1996 ; Roff 2002 ) ; therefore, d ecisions made during the reproductive period affect other life history components Most oviparous organisms can produce hundreds or even millions of eggs because their offspring require minimal parental care. In most birds, however, eggs require extensive parental care and they would not hatch without precise temperature regulation provided by the attendant adults ( Turner 2002 ; Williams 2012 ) Egg care, therefore, involves a tradeoff between self maintenance and reproductive investment, which ultimately could limit clutch size ( Lack 1968 ; Clutton Brock 1991 ) T radeoffs in investment decisions are usually related to variation in nest predation risk and ambient temperature ( Conway & Martin 2000b ) Birds living in the tropi cs, for example, have been hypothesized to reduce investment in each clutch in favor of increased adult survival in the face of severe nest predation rates and a benign climate ( Ricklefs & Wikelski 2002 ; Roper et al. 2010 ) Many t ropical birds, however, also have relatively long incubation periods, a trait that would increase vulnera bility to nest predation ( Tieleman et al. 2004 ; Martin et al. 2007 ) Before I can resolve this paradox, I ne ed studies of birds from the same region in which nest predation and other potential selective factors such as ambient temperature and nest predation vary so that I can associate investment decisions with these factors Tropical elevation gradients offer an opportunity to test the relative contributions of nest predation rates and ambient temperature to investment decisions. Along elevation gradients, bird species with similar evolutionary histories have been evolving
40 under different biotic and abiotic co nditions, but have experienced relatively stable climates ( Brumfield & Edwards 2007 ) Ambient temper ature decreases steadily with elevation, whereas nest predation rates decrease abruptly at the border between lower elevations and cloud forest ( Jankowski et al. 2013 ) ; Chapter 1 ). Under these conditions I m ight expect species to make similar reproductive investment decisions within elevation bands, but to differ considerably among elevation bands if predation is an important factor in the evolution of nesting behavior Neotropical elevation gradients, at lea communities ( Stotz et al. 1996 ) which provides an opportunity to search for repeated patterns of responses to similar and different biotic and abi otic conditions. Investment decisions during nesting periods can be partitioned into egg and nestling stages The egg is a closed system in which all the food resources are fixed from the beginning and the transformation rate of these resources to form the embryo depends largely on the amount of heat energy transferred by the incubating parent ( Brua 2002 ; Deeming 2002b ) In contrast, nestling growth relies mainly on quantity and quality of food provided by the parents ( Skutch 1949 ; Lack 1968 ; Ricklefs 1984 ) In this paper I focus only on the egg stage, because the embryonic development rate depends exclusively o n egg temperature, which is regulated by the incubating parent ( Turner 2002 ) and parental behavior or decisions will be affected by e xternal factors such as predation risk and ambient temperature ( Lack 1968 ; Skutch 1985 ; Conway & Martin 2000b ; Ghalambor & Martin 2000 ) Specifically, I evaluate how life history traits involving the egg stage vary along an elevation gradient in response to the decrease in nest predation and ambient temperature with elevation ( Janzen 1967 ; Skutch 1985 ;
41 Boyle 2008 ; Jankowski et al. 2013 ) I hypothesize d that the observed variation in incubation length among bird species w ould be associated with differences in egg si ze and egg temperature ( Turner 2002 ; Gillooly et al. 2008 ) and the remaining variation w ould be associated with differences in nest predation risk and ambient temperature. Based on allometric models, I predicted that species with larger eggs w ould have longer incubation periods and eggs maintained at warmer temperatures w ould have shorter incubation periods. S econd, I predicted that birds nesting in elevation bands characterized by high rates of nest predation would have smaller clutches and eggs ( Skutch 1985 ; Martin et al. 2006 ; Zanette et al. 2011 ) longer incubation bouts ( Ghalambor & Martin 2000 ; Ghalambor & Martin 2002 ; Fontaine & Martin 2006 ) which would result in higher egg temperatures and shorter incuba tion periods ( Martin et al. 2007 ) (Table 1). In contrast, birds living in colder elevation bands should have smaller clutches ( Badyaev 1997 ; Badyaev & Ghalambor 2001 ; Lu et al. 2010 ) cooler egg temperatures as a result of hi gher egg cooling rate s during foraging trips as a result of lower ambient temperatures, and, therefore, longer incubation periods ( Lu et al. 2010 ) (Table 3 1) In this study, I use d nesting data that I gather ed from 84 tropical bird species nesting in contrasting environments (high and low predation and ambient temperature) in a diverse community ( Walker et al. 2006 ) located in the eastern slope of the Andes in southeastern Peru between 350 and 31 00 m. Methods Avian nesting data were collected from 2008 to 2011 from an elevation range of 350 to 3100 m in the southeastern Peruvian Andes. Daily nest searches where conducted by 6 8 people at each of the four field stations distributed along the Manu R oad gradient. The Pantiacolla L odge station located in a lowland rainforest adjacent to
42 the Alto Madre De Dios r iver included an elevation range from 350 to 550 m (12 39' 21.4" S; 71 13' 54.6" W) The Tono station located in the foothills of the Andes n ext to the Tono river included an elevation range between 800 and 1000m ( 12 57' 58.2" S, 71 34' 05.3" W) The Cock of the Rock Lodge station was located in cloud forest at the intersection of the San Pedro and Kcosipata rivers between 1200 and 1500 m ( 1 3 03' 19.4" S; 71 32' 48.5" W) The fourth station, Wayqecha, was in elfin forest between 2700 and 3000 m (13 10' 30.1" S; 71 35' 14.6" W) When a nest was found with eggs I set up two thermistors, one inside a fresh egg and a second one outside the nest; both were attached to U12 4 channel hobo data logger (Onset Computer Corporation, MA, USA http://www.onsetcomp.com ) that was program med to store and record temperatures every minute. I only placed thermisto r s inside fresh egg s (no embryonic development) to measure egg temperature, by drilling a hole in the wider end of the egg and the hole was sealed with super glue. The thermistor outside the nest that provided nest microclimate information was attached to a branch near the nest (~ 15 cm). Because inserting the thermistor in the egg stopped embryonic development, I did not drill more than 4 eggs per species to reduce population impacts. The overall fitness impact produced by egg drilling was very small beca use nest predation rates were very high (see below) and the overall number of nests encountered with fresh eggs was small. Therefore, the likelihood that any of the eggs sacrificed for this study would fledge was very small and the proportion of eggs for a ny one species was also very small. All the animal manipulations were approved by the University of Florida IACUC # 201004590.
43 Incubation period was measured as the number of days between the dates on which the last egg was laid and the last egg hatched. A ll of my egg and clutch mass analyses were conducted with fresh eggs (no embryonic development) to avoid biases caused by egg water loss ( Rahn & Ar 1974 ) Nest monitoring and nest predation estimation : Data loggers allowed us to continuously monitor nest survival and to reduce nest visitation rate to one visit every six days. Nests without data loggers were visit ed every 3 4 days. To estimate nest predation I used bird nests with known fate s in a nest survival model constructed within the program MARK ( Dinsmore et al. 2002 ; Rotella 2007 ) This model allowed us to estimate daily nest predation rate by using the number of exposure days to calculate the probability of predation. I used all n ests with known fate (depredated or successful) regardless of monitoring technique (n = 1134); on average I had 13.5 nests per species (17.2 standard deviation; min max = 1 86). Data Analysis : As expected, egg (r 2 = 0.89, p <0.0001) and clutch (r 2 = 0.72, p <0.0001) mass were highly correlated with body mass; therefore in my analyses I corrected for body mass using ln (egg mass/body mass) and ln (clutch mass/body mass). In all linear models I checked for normality using the Shapiro test ( Gotelli & Ellison 2004 ) ; if the variables were not normal I transformed th em. All but nest predation variables were normalized using logarithmic transformation and the nest predation data were Arcsine transformed. Allometry: To evaluate the effect of metabolic rate combined with egg mass and temperature on incubation period (o ntogenetic growth rate), I used the equation proposed by ( Gillooly et al. 2002 ) and teste d by ( Gillooly et al. 2008 ) on a particular
44 embryonic d evelopment stage. I used my data to test the main prediction that the logarithm of the body size corrected time to egg hatching, ln( T i m i 1/4 ), should be a linear function of the inverse absolute temperature, 1/ k T, with a slope of reflecting t he exponential effects of temperature on metabolic rate. Explaining variation in incubation period : To explain what factors affect incubation period among my 84 bird species I conducted an Analysis of Covariance (ANCOVA) with bird species as a random effe ct, and, bird body mass, egg temperature, egg and clutch mass, nest microclimate, and daily ne st predation as response variables. Community comparison : To evaluate if the nesting (incubation length, egg mass, clutch mass, egg temperature) and selective (ne st microclimate and nest predation) variables occurred randomly throughout the gradient or if they were associated with particular bird communities along the gradient, I used two different approaches. My first approach was to search for general association s among all variables and the bird communities of different elevation bands. To do this I conducted a Principal Component Analys i s (PCAs) that included the 62 individuals and 56 species for which I had complete nesting information. In my second approach, I quantified how my different nesting variables and selective factors varied among the four bird communities by conducting an Analysis of Variance (ANOVA). In this analysis I used incubation period, egg temperature, egg mass, clutch mass, nest microclimat e, and daily ne st predation as response variables, and bird community as a factor. I s s as post hoc tests of what? Results Among the 84 bird species with known incubation periods from my study sites, the average incubation period was 1 8 days ( 4.2 SD), clutch size 2.2 eggs ( 0.63),
45 egg mass 5.76g ( 8.7) and egg temperature 34.80 C ( 1.5). Average proportion of nests depredated daily was 0.04 ( 0.03), and average nest microclimate was 18.7 C ( 3.9). Incubation length : The obser ved variation in incubation period was mainly associated with changes in egg temperature (F 1,62 = 5.33 p = 0.03) and marginally explained by changes in daily nest predation (F 1,62 = 3.21 p = 0.09) and bird body mass (F 3,62 = 3.33 p = 0.08) ( Table 3 2 ). In general bird eggs that were incubated at higher temperatures had shorter incubation periods (Fig ure 3 1a). Thus, my data suppor t the metabolic prediction (Figure 3 1a): a plot of the logarithm of mass corrected time to egg hatching, ln ( t i m i 1/4 )), versus the inverse absolute temperature, 1/ k T, yielded a significant relationship with a fitted slope of 0.62 (eV; CI 95% 0.52 0.71). However, the metabolic prediction only explained 20% of the variation ( r 2 = 0.20, p <0.0001) observed among my 84 birds spe cies. The remaining variation was weakly associated with nest daily predation rate ( r 2 = 0.08, p = 0.02; Figure 1b), but not with nest microclimate ( r 2 = 0.02, p = 0.91; Fig ure 3 1c). Elevation gradient and community comparisons : Daily nest predation varied among bird communities (F 3,91 = 6.08 p = 0.0008 ; Fig ure 3 2 a), in which the highest daily nest predation rates were recorded in the foothill forest community (6.9% 4.2%), follow ed by lowland (5.3% 2.9), elfin (3.9% 2.0) and cloud (3.3% 2.1 %) forest communities. Nest predation rates differed significantly between the cloud forest and the foothill (Tukey: p = 0.008) and lowland (Tukey: p = 0.002) bird communities, and elfin forest nest predation rates were nearly significantly different from the foothill
46 community (Tukey: p = 0.07; Figure 2 a). N est microclimate dec reased linearly with elevation ( r 2 = 0.86, p <0.0001 ; Figure 3 2 b). Among the 56 bird species and 62 individuals for which I had complete information, I found different incubation str ategies and selective factors among bird communities along the gradient ( Figure 3 3 ; Table 3 3 ). Most elfin forest species had similar incubation period s wh ereas most lowland species shared similarities i n egg temperature and nest microclimate ( Figure 3 3 ). Among communities, e gg mass had a strong tendency to increase in high elevation bird communities (F 3,91 = 2.64 ; p = 0.05), in which birds in elfin forest communities laid between 9.0 and 20.5% larger eggs (Tukey: p = 0.07; Figure 3 3a). In contrast, cl utch mass did not change significantly among bird communities (F 3,91 = 1.64 p = 0.19; Figure 3 3b). Average egg temperature varied among bird communities (F 3,91 = 3.19 p = 0.03), in which egg temperatures decrease d at higher elevations ( Figure 3 3c). However, the only communities that differed significantly in egg temperature were the lowland and elfin forest communities, in which birds nesting in the elfin forest communities maintained their eggs on average 1.6 C cooler (Tukey: p = 0.02; Figure 3 3c) Although average incubation period increased with elevation (F 3,91 = 2.58, p = 0.05), nearly significant differences only occurred between cloud and lowland forest (Tukey: p = 0.08; Figure 3 3 d). Overall, however, the observed variation i n incubation per iod among bird communities in my full model was explained mainly by differences in daily ne st predation (F 3,62 = 4.51, p = 0.01) and clutch mass (F 3,92 = 6.91, p = 0.001) ( Table 3 2 ).
47 Discussion My results partially supported my predictions. Although in g eneral there was a significant negative correlation between daily nest predation rate and incubation period, incubation period, clutch and egg mass, and egg temperature did not change significantly among elevations with contrasting nets predation risk. In contrast, ambient temperature was not significant associated with general variation in bird incubation period. Thus, my data suggest that nest predation rates but not ambient temperature, contribute to the suite of life history traits characteristic of ma ny tropical bird species. However, the observed variation in clutch size, egg mass, egg temperature and incubation period were concordant with elevation prediction (Table 3 1). Throughout the rest of the discussion I will discuss how my results advance our understanding of how nest predation and ambient temperature could shape incubation strategies in tropical bird communities. Even though egg and clutch mass were not associated with variation in incubation period, I observed significant interactions betwee n communities and clutch and egg mass that were significantly associated with incubation periods. Bird species inhabiting higher elevations had longer incubation periods and invested more per egg despite the lack of a linear decrease in daily nest predatio n rate, which is puzzling as nest predation has been linked to the decrease in nesting investment observed in tropical areas ( Skutch 1985 ; Martin et al. 2006 ) More confusing, yet, is the observed trend of clutch size reduction with elevation, which could be the reason w hy I did not observe an increase in clutch mass with elevation despite the incremental gain in egg mass. Because egg mass, clutch mass, and clutch size did not change significantly between high ( l owland and foothill) and low predation areas (Cloud forest) my data
48 suggest that daily nest predation rate is a poor predictor of clutch mass and egg mass in the Neotropics, as suggested by Martin et al. ( 2006). Instead egg mass increases a pattern reported previously ( Berven 1982a b ; Howard & Wallace 1985 ; Blanck enhorn 1997 ; Rohr 1997 ; Hancock et al. 1998 ; Lu et al. 2010 ) and clutch size decrease s, another patt ern that has been shown in other areas ( Zang 1980 ; Krementz & Handford 1984 ; Badyaev 1997 ; Badyaev & Ghalambor 2001 ; Lu et al. 2010 ) In addition to these changes, nest predation at high elevations was just as hig h as in the lowlands, but ambient temperature and productivity ( Salinas et al. 2011 ) are lower Because incubating large clutches ( Visser & Lessells 2001 ; Nord & Nilsson 2012 ) at colder temperature ( Haftorn & Reinertsen 1985 ) is costly, I propose that increasing egg mass and decreasing clutch size is a nesting strategy t hat increase s nesting success in harsh environment s such as the high Andes. Providing extra resources for t he embryo may reduce embryonic mortality at high elevations. E mbryos at high elevation s encounter pronounced cooling periods (lower ambient temperatures) during incubation recess, which increases embryonic stress reflected in smaller embryo nic mass and low er yolk reserves ( Olson et al. 2006 ) Under these conditions, extra resources within eggs could increase embryon ic survival probabilities by providing the energy necessary for increased metabolic rates. In contrast, clutch size reduction may be a dvantageous because it reduces the high energetic cost of egg formation and incubation in this harsh environment ( Biebach 1981 ; Haftorn & Reinertsen 1985 ; Tinberger & Dietz 1994 ; Williams 1996 ; Bryant & Bryant 1999 ) Similarity, and not mutually exclusively, low primary productivity at high elevations ( Salinas et al. 2011 ) could reduce the amount of food resources available during the nesting period, which would increase the cost of
49 egg production and potentially increase the amo unt of foraging time required during incubation recesses. Hence, c lutch size reduction s at high elevations could reduce incubation and egg production cost s allowing birds to increase investment in the remaining eggs and decrease intervals between nesting attempts as a result of reduced stress on the adults ( Roper et al. 2010 ) Because of the high energet ic costs encounter ed by birds nesting at high elevation s, it is unlikely that birds can simply increase their time on the nest incubating their eggs to reduce egg heat loss. However, birds can modify their incubation behavior when incubating at colder temp eratures ( Conway & Martin 2000a ) by increasing the number of foraging trips, which could allow them to frequently refuel their energetic demands and simultaneously reduce egg cooling time (short trips). But, increasing the number of foraging trips could increase nest predation probability ( Fontaine & Martin 2006 ; Ibez lamo & Soler 2012 ; Smith et al. 2012 ) e specially at this elevation in which all diurnal nest predation is conducted by birds that use visual cues (Chapter 1). This possible change in the frequency and duration of off bouts could be the reason why I observed an increase in daily nest predation rate between the cloud and elfin forest. Thus, the high cost of nesting in colder ambient temperatures at higher elevations could limit the options available to reduce negative fitness effect s such as nest predation. At high elevation birds are investing more per egg but reducing their clutch size, which is the opposite strategy to that used by lowland birds under similar nest predation but warmer climate a nd higher primary productivity ( Salinas et al. 2011 ) The harsh conditions encountered by nesting birds at high elevations may limit their ability to employ behaviors that could reduce nest p redation (e. g., a reduced number of
50 trips to and from the nest), which forces them to use other strategies (e. g., decrease clutch size). However, until I have quantified how and if bird incubation behaviors change with elevation however, I cannot draw s trong conclusions. Daily nest predation rate s w ere negatively associated with overall incubation period ( Bosque & Bosque 1995 ; Martin et al. 2007 ) which was supported by the nearly significant differences i n incubation period between lowland (high nest predation and short incubation period) and cloud forest (low nest predation and long incubation period) bird c ommunities. Overall, however, daily nest predation explained only a small percentage of the variation observed in the incubation period throughout the gradient. Birds incubating at elevations with high nest predation rates but colder temperatures (foothill and elfin forest communities) did not reduce their incubation periods ; in stead average incubation period increased at th ese elevation s Thus, lower ambient temperatures could constrain birds from adopting life history strategies such as reducing trips to and from the nest that might reduce nest predation rates Additionally I did not find differences i n egg and clutch mass among communities with contrasting daily nest predation rate s (foothill lowland and cloud forest; Figure 3 3a,b), suggesting that at le ast in my tropical gradient egg and clutch mass investment are not associated with differences in nest predation. Instead, it appears that these variables were more strongly associated with ambient temperature. Another surprising result was that nest micr oclimate did not contribute to the overall incubation period variation, in spite of the strong association observed between low temperature nest microclimate and low egg temperature. Although frequent egg cooling periods slow ed embryonic development ( Olson et al. 2006 ; Nord & Nilsson
51 2012 ) I know little about how the length and frequency of these cooling periods affect embryonic development and, thus, the length of the incubation period. T he relationship between egg mass ( m ) and incubation period ( t) ln ( t i m i 1/4 )), and nest microclimate (1/ k T), had a slope (0.62) close to the predicted value based on allometric models of metabolism (0.65 ) ( Gillooly et al. 2008 ) Nevertheless, this relationship was relatively weak and only accounted for 20% of the variation, suggesting that other factors may determine inve stment in clutches and incubation. The residual variation was negatively correlated with n est predation rates but not ambient temperature. W hen I compare d if clutch and incubation investment were associated with different bird communities along the gradie nt, I found that communities had distinctive nesting traits associated with their unique combination of abiotic (temperature) and biotic (nest predation) conditions Although allometric hypotheses are intended for comparisons among organisms that differ in mass by several order of magnitude ( Brown et al. 2004 ) it is interesting that I found that bird incubation period is fundamentally related to metabolic properties, despite the limited range of body masses represented in my sample ( Gillooly et al. 2002 ; Gillooly et al. 2008 ) In general bird species that maintained their eggs at warmer temperatures had shorter incubation periods ; however, incubation period corrected for egg mass varied g reatly among species that had similar average egg temperatures ( Martin 2004 ; Tieleman et al. 2004 ; Martin et al. 2007 ) This variation could result from differences in egg investment among bird species, in which those with proportionally larger eggs and higher nutrient content with in eggs have longer incubation periods ( Martin 2008 ) ( Figure 1a ). Variation in the length of incubation bouts, which tend to be shorter in small species that lay large eggs,
52 could also account for differences in the length of the incubation period because of differences in the length of time when eggs are warm enough to allow embryonic development. In general, my results only partially supported my hypothesis that nest predation is a primary dr iver of the suite of life history traits observed in tropical birds. The different incubation strategies and investment decisions documented in the Manu Andean elevation gradient reflect the complexity and diversity of life history strategies concealed in this hyper diverse tropical area. Overall bird species that encounter higher daily nest predation rate s and maintained their egg at warmer temperatures had shorter incubation periods ( Martin et al. 2007 ) These characteristics were mainly associated with lowland bird communities. But, generally, daily nest predation and egg temperature alone explained only a small percentage of the variation observed on this tropic al elevation gradient ( Martin 2004 ; Tieleman et al. 2004 ; Robinson et al. 2010 ) The large variation in incubation strategies observed along the elevation gradient and within bird communities suggest that previously documented contradictory results may be resolved by studying two different tropical forest bir d communities, lowlands ( Tieleman et al. 2004 ) and cloud forest ( Martin et al. 2007 ) which have contrasting incubation strategies. Therefore I need to explore additional factors that could explain the observed variation within tropical communities ( Martin 2004 ) for example, by comparing nesting strategies among tropical bird species wi thin a community ( Ricklefs & Bra wn 2012 )
53 Figure 3 1 Association s between the main factors that affect bird embryonic development (egg mass and egg temperature). A) Graphical test of the metabolic prediction (Gil l ooly et al. 2008), in which the fitted slope of 0.62 is very close to the predicted slope 0.65. Association of the residual variation observed in the metabolic prediction with the B) daily nest predation rate and C) nest microclimate.
54 Figure 3 2 Variation in selective factors among bird communities on the elevation gr adient, A) daily nest predation rate (Proportion of nests depredated daily) and B) Nest microclimate.
55 Figure 3 3 Principal Component Analysis among nesting variables (egg and clutch mass, incubation period, egg temperature ) and selective factors (nest microclimate and daily nest predation rate) in four bird communities along an Andean elevation gradient, Cusco, Peru. Data are from 54 species and 62 individuals
56 Figure 3 4 Chang es i n nesting variables (A: egg mass; B: clutch mass; C: Egg temperature ; and D: incubation period) among bird communities of different elevation bands along the Manu gradient. Communities where nesting variables are close to be significant.
57 Table 3 1 Avian incubation investment predictions in environments with diff erent nest predation risk and ambient temperatures. Nest Predation Ambient Temperature Life History Trait High Low High Low Egg mass Low High Low High Clutch size Low High High Low Egg Temperature High Low High Low Incubation period Short Long Shor t Long
58 Table 3 2 Full model for general and commun ity a nalysis of covariance (ANCOVA). Variables df F value Probability I)Variables that explain incubation period (error df = 62) Egg mass 1 1.5 0.23 Clutch mass 1 1.83 0.18 B ody mass 1 3.33 0.08 Daily nest predation rate 1 3.21 0.09 Egg temperature 1 5.33 0.03 Nest microclimate 1 1.24 0.27 II) Incubation period and bird communities (error df = 62) Community 3 0.16 0.93 Egg mass 1 1.14 0.71 Clutch mass 1 0 .87 0.36 Daily nest predation rate 1 6.45 0.02 Egg temperature 1 0.85 0.36 Nest microclimate 1 0.8 0.38 Body mass 1 0.09 0.76 Community*Egg mass 3 1.37 0.27 Community*Clutch mass 3 6.91 0.001 Community*Daily nest predation rate 3 4.51 0.01 C ommunity*Egg temperature 3 1.26 0.31 Community*Nest microclimate 3 0.26 0.86 Community*Body mass 3 0.33 0.81
59 Table 3 3 Output from the Principal Component Analysis (PCA) realized in the 56 species and 62 individuals from the four bir d communities (lowland, foothill, cloud and elfin forest) distributed along the Manu elevation gradient, Cusco, Peru. Variables Component1 Component2 Component3 Egg mass 0.55 0.81 0.3 Clutch mass 0.46 0.49 0.21 Incubation period 0.36 0.22 0.57 D aily nest predation 0.36 0.12 0.63 Ambient temperature 0.21 0.74 0.11 Egg temperature 0.41 0.33 0.37 standard deviation 1.57 1.07 1.05 Proportion of variance 0.41 0.19 0.18
60 CHAPTER 4 TANAGER S OF THE H IGH ANDES HAVE AN ULTRA SLOW PACE OF LIFE Introduction Life history traits related to reproductive effort have been used to characterize ( Stearns 1992 ; Roff 2002 ) Many lines of evidence suggest for example, that tropical birds have a suite of life history traits that can be characte pace of life ( Ricklefs & Wikelski 2002 ) This suite of traits includes small clutches, long incubation and nestling periods, low nest attentiveness, a nd low feeding rates of young ( Moreau 1944 ; Lack 1947 ; Skutc h 1949 ; Martin 2004 ; Jetz et al. 2008 ; Robinson et al. 2010 ; Martin et al. 2011 ; Gill & Haggerty 2012 ) More recently, studies of metabolic rates have shown that tropical birds also have lower basal metabolic rates than their temperate counterparts ( Wikelski et al. 2003 ; Wiersma et al. 2007 ) further supporting the hypothesis that tropical birds have a slow pace of life. Typically, traits associated with a slow pace are considered adaptations to a climatically benign, but biotically challenging environment in which adult survival is favored at the expense of each nesting attempt ( Martin 2004 ; Robinson et al. 2010 ) Not all tropical environments however, are climatically benign. High elevations of tropical mountains, for example, are much colder and less productive year round than nearby lowlands ( Janzen 1967 ; Ghalambor et al. 2006 ; Salinas et al. 2011 ) Given this contrast, one might expect tropical birds to vary in their nesting strategies along elevation gradients in ways that reflect the greater abiotic constraints to nesting at higher elevations. Speci fically, I predict ed that species nesting at higher elevations will have lower fecundity (e. g., reduced clutch size ) ( Badyaev 1997 ; Badyaev & Ghalambor
61 2001 ; Lu et al. 2010 ) compared with lowland counterparts. I tested this prediction by examining how egg and clutch mass, nest attentiveness, egg temperature, nest pre dation risk, nestling growth, and development times of eggs and nestlings vary among n eotropical tanagers ( Sedano & Burns 2010 ; Jetz et al. 2012 ) inhabiting different elevations. Additionally I conducted a literature review and contrasted my results with other species in the tanager clade to evaluate if my findings for high elevation tanagers were present in othe r species of tanagers and if they were randomly distributed in the clade. Specifically I compared variation in elevation distribution, clutch size, and incubation and nestling development times in the tanager clade. I focused on tanagers because they oc cur at all elevations at densities sufficient to obtain data necessary to quantify life history traits, and they do not show the pattern of decreasing nest predation rates with elevation documented for the community as a whole ( Skutch 1985 ; Boyle 2008 ) see chapter 1 Therefore, tanagers provide an opportunity to study how life histories of tropical birds respond to i ncreasing abiotic constraints without the additional complication of associated changes in nest predation rates, which are well known to be strong determinants of life histories ( Martin 1995 2004 ) I show that tanagers nesting at high elevations have an especially slow paced li fe history strategy, which reflects an environment with harsh abiotic factors. Methods Study site : This study has two components, one based on original data and one based on published data. Original data came from a field study conducted between 350 an d 3100 m elevation in the Manu National Park, in the adjacent departments of Cusco and Madre de Dios, which are in the Andes of southeastern Peru. The goal of this study was to provide highly detailed information on nesting behavior and nest predation for as
62 many tanager species as possible along the elevational gradient. Annually from 2008 to 2011, I worked at four stations distributed along the gradient: Pantiacolla (350 900 m; 12 39' 21.4" S; 71 13' 54.6" W), Tono (800 1100 m; 12 57' 58.2" S, 71 34' 05.3" W), San Pedro (1100 1600 m; 13 03' 19.4" S; 71 32' 48.5" W) and Wayqecha (2300 3100 m; 13 10' 30.1" S; 71 35' 14.6" W). Nest searching : was conducted between August and December, six days per week, ten hours per day between 6:00 and 18:00. Seven ty two field assistants found and monitored 206 nests from 13 tanager species (Thraupidae; Appendix 1 ). All eggs and nestlings were weighed to the nearest 0.05g. Eggs and nestlings were weighed the day they were found, and nestlings were weighed daily unti l they fledged or died. Data loggers and motion cameras allowed continuous monitoring of nests; those without these monitoring systems were checked every 2 4 days. The incubation period was calculated as the number of days between clutch completion and wh en the last egg hatched. Similarly, the nestling period was estimated from the day the last nestling hatched to the day the last nestling left the nest. When eggs or young nestlings disappeared between nest checks, I assumed they had been depredated. If nestlings with primary (flight) feathers that were more than half grown disappeared between nest checks, I assumed they had fledged. Incubation behavior : I was able to monitor incubation behavior in 81 nests of nine species. To determine incubation behavio r and nest survival, I placed one t hermo couple (2x3 mm) inside the nest under the eggs and a second t hermo couple outside the nest; both were attached to data loggers (Onset Computer Corporation, http:// www.onsetcomp.com) that recorded and stored temperatur e every minute. The
63 thermocouple inside the nest allowed us to quantify length of incubation bouts (time on the nest incubating the egg) and foraging bouts (when the bird was absent from the nest). Bouts were obviously demarcated by rapid changes in temper ature of at least 1C. The thermocouple outside the nest provided data on ambient microclimate. I placed a third thermocouple inside the fresh egg (undeveloped embryo) on a subset of 17 nests (5 species) to monitor egg temperature, which can be used to un derstand differences in incubation period among bird species ( Martin et al. 2007 ) To reduce nesting mortality occasioned by inserting the thermocouple inside the fresh egg, particularly in species with one egg clutches, I did not drill more than three fresh eggs per species. In analyses I included only days in which I collected incubation data for 24 continuous hours (from 00:00 to 23:59), a sample consisting of 52 0 incubation days and 9 4 temperature days. All the animal manipulations were approved by the university of Florida IACUC # 201004590. Data from Literature : To augment the data from my field study, I obtained data from published studies on clutch size, incu bation and nestling period, and elevation range for additional species of tanagers. I used these data to evaluate if the similar associations that I found among clutch size, elevation, and incubation and nestling periods on the Manu elevation gradient app lied to the tanager clade. The literature review mainly came from two review books ( Isler & Isler 1999 ; Hilty 2011 ) ; additionally I reviewed large databases (ISI Web of Science 1864 2013), SORA ( http://sora.unm.edu/node ), and google scholar for small publications not indexed in the Web of Science.
64 Data Analysis: For species along my elevat ion gradient, I used PCA to evaluate the relationship s among the following variables: body mass, number of foraging tr ips from the nest, nest attentiveness, nest microclimate incubation period, egg temperature, daily nest predation rate nestling growth r ate nestling period, and clutch size. I only included in my analyses eggs for which I had initial mass (n = 187) to avoid mass variation produced by water loss after the onset of incubation ( Rahn & Ar 1974 ) Nestling growth rate was calculated using a log istic equation transformation K ( Ricklefs 1967 ) E gg and clutch mass comparison were made using logarithmic transformed data to correct for body mass. Nest predation was estimated using a nest surviva l model constructed within the program MARK ( Dinsmore et al. 2002 ; Rotella 2007 ) This model allowed us to estimate daily nest predation rate by using the number of exposure days to calcul ate the probability of predation. I used all nests with known fate (depredated or successful), regardless of monitoring technique (n = 198). Variables included in the model were: 1) the first day that an active nest was encountered, 2) the last day the nes t was checked and eggs or nestlings were present, 3) the last day the nest was checked, 4) the fate of the nest (0 = successful, 1 = depredated). To estimate nesting success, I raised the probability of nesting success to the length of the average nesting period observed in the Manu gradient, which was 0.935 39 and 0.946 31 for species with one and two egg clutches, respectively. For the general literature review, I calculated averages for each species (including those from the Manu elevation gradient) and c onducted PCA to explore associations among clutch size, elevation distribution, incubation and nestling period.
65 To evaluate if clutch size was randomly distributed in the tanager clade I used a recently published bird phylogeny ( Jetz et al. 2012 ) that included all the Manu gradient tanagers for the analysis. I used two methods to estimate phylogenic signal: K statistic ( Blomberg et al. 2003 ) and and I compared the two measures between the observed tanager tree and a tanager tree without phylogen et ic structure, using a Chi square test ( Pagel 1999 ) Results I found that six of the 13 tanager species distributed along the Manu elevation gradient for which I obtained reliable nesting information had one egg clutches (n= 101 nests). In the ele vation gradient clutches from tanagers with single eggs weighed on average 56% less than those from species with two egg clutches; incubation periods in tanagers with one egg clutches were 22 % (5 days) longer than those with two egg clutches and nestling periods averaged 22% longer (4 days) in tanagers with one egg clutches. When I explored nesting characteristics associated with the one egg clutch in the tanagers along the Manu elevation gradient, I found that the first three PCA axes explained 77.4% of t he variation (Table 3 1). The one egg clutch species were associated with longer incubation and nestling periods, lower nestling growth and colder climates (PC1 44.4%; Figure 3 1). The second PC2 axis explained 20.7% and the main variables were body mass nest attentiveness and egg temperature, but they varied largely among species with the same clutch sizes (PC2 20.7%; Figure 3 1). The one egg clutch size tanager species in the Manu gradient were not randomly 89.98, P<0.0001; Figure 3 2). I found that the three PCA axes associated with nesting and elevation characteristics explained
66 82.9% variance among tanager species with different clutch sizes (Table 3 1). In general species with smaller clutches had longer incubation and nestling periods and inhabited higher elevations (PC1 48.7%; Figure 3 3). The second axis (PC2) explained 20.7% of the variation and was largely influenced by adult body mass, but body mass varied largely among species with the same clutch size ( Figure 3 3). Discussion Several life history traits of the tanager clade changed with elevation in ways that generally conformed to the expectation that increased abiotic stressors select for a slower pace of life. Clutch size decreased from t wo or three eggs in most species at low elevations to one egg in the majority of higher elevation tanagers. Although one egg species partially compensate d for clutch reduction by increasing egg mass on average by 15.7 %, their clutch mass was still much (44 .9%) less than those with two eggs Species with one egg clutches also had longer incubation and nestling periods, and lower nestling growth rates than species with two egg clutches. Therefore, even though tanagers nesting at high elevations substantially reduced their investment in each clutch, development rate and parental care did not decrease. In the rest of the discussion I develop the argument that one egg clutches and long incubation and nestling periods which reflect an ultra slow pace of life f or a Neotropical passerine bird ( Deeming 2002a ; Martin et al. 2006 ; Auer et al. 2007 ; Jetz et al. 2008 ) result from the addition of harsh abiotic factors at high elevations to the already documented harsh biotic constraints of the tropics The one egg clutch in six sp ecies of high elevation tanagers illustrates a new extreme in life history trade offs of tropical passerines. This dramatic reduction in reproductive investment by nearly 50% compared to most other tropical passerines
67 that have two egg clutches ( Jetz et al. 2008 ) is very rare in passerines. Altho ugh bird clutch size reduction has been documented previously along elevation gradients ( Badyaev & Ghalambor 2001 ; Lu et al. 2010 ) these studies reported a small clutch reducti on, between 10.9 and 15.6% (from 5.5 to 4.9, and 3.2 to 2.7 eggs), from low to high elevation s, compared to the 50% (from two to one egg) reduction observed in tanagers in my gradient Although tanagers significantly reduced their clutch size and mass with elevation, species with one egg clutches at least partially compensated by laying larger eggs. Proportionally larger eggs in tropical passerine species are associated with lower nest attentiveness, in which species compensate for their reduced parental ca re by laying larger eggs ( Martin 2008 ) Although, average nest attentiveness was lower in one egg (60.212.8) compared to two egg (66.110.6) clutches, it varied greatly among species. Thus, other selective pressures may influence egg mass increases and clutch reduction in high elevation tanagers (see below). Nest predation is commonly hypothesized to be an important selective pressure affecting bird nesting invest ment decisions ( Lima 2009 ) ; high nest predation rates have been associated with small er clutches and eggs ( Skutch 1985 ; Martin et al. 2006 ) Nest predation rates in my study, however, did not vary with elevation in th e tan ager clade and did not vary with clutch size suggest ing that predation is unlikely to be a causal mechanism for the observed reduction in clutch size with elevation in my study system Overall nesting success, however, did decrease because of the observed longer incubation and nestling periods at higher elevations. Another important selective factor that is commonly proposed to affect bird nesting strategies and investment decisions is ambient temperature ( Conway & Martin 2000b ) The c osts of producing eggs,
68 incubating clutches, and fe e ding nestlings are all higher in colder climates ( Biebach 1981 ; Haftorn & Reinertsen 1985 ; Tinberger & Dietz 1994 ; Williams 1996 ; Bryant & Bryant 1999 ) all of which would select for clutch reduction in high elevation Andean tanagers. Because nest predation rate did not vary with elevation among tanagers, the ~12C reduction in ambient temperature observed could well be the dominant selective factor favoring clutch reduction in my system. Lower temperatures may also be responsible for the lo ng er incubation and ne stling periods of high elevation tanagers. Previous studies on elevation gradients also found that species inhabiting higher elevations had longer incubation and nestling periods ( Badyaev & Ghalambor 2001 ; Lu et al. 2010 ) The lower nestling growth rates observed in tanagers with one egg clutches suggest that the quantity and/or quality of food delivered to nestlings may be reduced at high elevations. Because d aily nest predation was similar between one and two egg clutch tanagers it is unlikely that differences in tanager nestling growth are promoted by changes in nest predation. Alternatively, colder nest microclimate s may impose higher thermoregulatory costs o n developing nes tlings, which might make it necessary for nestlings from one egg clutch es to invest less in growth and more in maintenance ( Mertens 1977 ) Similarly, lower egg temperatures may be responsible for longer incubation periods. In support of this possibility tanagers with one egg clutches have lower egg temperatures (32.01.4) compared to species with two egg clutches (34.20.4) clutches, althou gh there is a great deal of variation and my sample size is small (5 species). Because nest attentiveness did not vary among species, lower egg temperatures may result from the rapid cooling of eggs during foraging trips at higher elevations, which would s low
69 embryonic development and increase stress and metabolic rates ( Olson et al. 2006 ) Thus, the incremental inc rease in egg mass documented in tanager species inhabiting high elevations could reflect selection to provide extra resources for the embryo to survive during frequent cooling periods. In conclusion, my results suggest that the extreme clutch size reducti on and associated slow pace of life in high productivity ( Salinas et al. 2011 ) and low nesting success ( Robinson et al. 2000 ; Martin et al. 2006 ; Robinson et al. 2010 ) Tanagers with single egg clutches had longer incubation and nestling periods than those at lower elevations with two egg clutches. Given the constant high nest pr edation rates along the entire elevation gradient in tanagers there should be intense select ion for rapid incubation and development ; therefore, the long nesting period at high elevations suggests that shorter periods are not an option. Indeed, correcting for length of nesting cycle, tanagers of high elevation actually have lower nesting success than those with two egg clutches nesting at lower elevations. Thus, maintaining viable populations at high elevations might only be possible with frequent re nesti ng ( Roper et al. 2010 ) and high survival of adults ( Gill & Haggerty 2012 ) and fledglings ( Tarwater et al. 2011 ; Gill & Haggerty 2012 ) Given that most high elevation tanagers are relatively large, 51% larger than those in the lowlands, they may live longer ( Peterson & Wroblewski 1984 ; Brown et al. 2004 ) Until I have data on these additional life history variables, however, I will n ot be able to assess the whole suite of life history traits that characterize tanagers living at these high elevations.
70 Table 4 1. Association between environmental and nesting variables in the tanager clade species. Principal Component Analysis output. Variables PC1 PC2 PC3 PC4 Body mass (g) 0.31 0.47 0.03 0.20 Number of foraging trips 0.27 0.39 0.50 0.11 Incubation Atentiveness (%) 0.17 0.43 0.32 0.45 Nest microclimate (C) 0.35 0.01 0.32 0.41 Incubation period 0.42 0.11 0.22 0.17 Egg te mperature 0.13 0.45 0.28 0.30 Nest mortality (daily nest predation) 0.26 0.14 0.54 0.41 Nestling growth (k) 0.36 0.31 0.12 0.27 Nestling period 0.30 0.34 0.28 0.41 Clutch size 0.44 0.01 0.16 0.22 Standard deviation 2.11 1.44 1.11 0.95 Prop ortion of Variance (%) 44.44 20.65 12.31 9.12
71 Table 4 2 Association between environmental and nesting variables in the tanager clade species. Principal Component Analysis output. Variables PC1 PC2 PC3 PC4 Incubation period 0.49 0.14 0.43 0.75 Nestling period 0.43 0.47 0.43 0.62 Elevation 0.48 0.02 0.65 0.06 Body mass (g) 0.25 0.86 0.33 0.19 Clutch size 0.53 0.13 0.32 0.15 Standard deviation 1.5611 0.9937 0.85 0.68799 Proportion of Variance (%) 48.74 19.75 1 4.45 9.467
72 Figure 4 1 Principal component axes that explain 65% of the variation for nesting variables associated with tanagers nesting in the Manu elevation gradient. The first axis explain 44% of the variation, ambient temperature, nestling grow th (K), incubation period and clutch size are the main variables associated with this axis variation. The second axis explained 21% of the variation and the main variables associated with this variation are egg temperature, nest attentiveness and body mass
73 Figure 4 2. Clutch size phylogenetic signal in the tanager clade ( = 89.98, P <0.0001). This tree is base d on publish ed information by Jetz et al. 2012, and clutch information was obtained from literature (see appendix B).
74 Figure 4 3. Principal com ponent axes that explain 68% of the variation associated with the nesting variables in the tanager clade. The first axis explains 49% of the variation and incubation, nestling period, elevation and clutch size are the main variables associated with this va riation. The second axis explains 20% of the variation and body mass and nestling period are the main variables associated with this variation.
75 CHAPT ER 5 CONCLUSION Tropical mountains offer an exceptional opportunity to explore the mechanism that may be r esponsible for the slow pace of life observed in neotropical bird species (Ricklefs and Wikelski 2002). Not only do important selective factors such as nest predation and temperature change in a predictable way along the gradient, but there is also enough variation in nesting strategies that I can link variation in investment decisions with contrasting selective pressures. Overall the harsh environment (e. g., colder temperatures) encountered by nesting bird at high elevation seems to have a stronger impac t on investment decision made by nesting tropical birds than variation in nest predation. However, both nest predation and ambient temperature seem to shape the slow pace of life observed in tropical nesting birds, in areas where both abiotic and biotic co nditions are harsh such as in elfin forest, there is an even slower pace of life available to birds. My study revealed that changes in the relative importance of different nest predators and their activity patterns likely contribute to variation in nesting success along an elevational gradient. Furthermore, my work suggests that this variation may limit the elevational distributions of some tropical bird species, with community wide implications. Nest predation seems especially likely to contribute to cha nges in bird species composition in the foothills of tropical mountains, and possibly within upper cloud forest. I view this study as a first step towards evaluating the hypothesis that nest predation contributes to species turnover and life history evolu tion along Neotropical montane gradients, and propose that nest predation is a critical factor both promoting and maintaining the high beta diversity and extraordinary regional diversity found in the
76 Andes ( Rahbek & Graves 2001 ) The next steps for testing this hypothesis will require data on life history traits as potential adaptations to nest predation, such as egg and clutch mass, incubation and nestling p eriods and behaviors during nesting, consideration of abiotic factors that change along the gradient (i.e., temperature, rainfall and productivity) and biotic factors such as interspecific competition. Understanding the interactions and relative importanc e of these factors will provide a more complete picture of the ecological mechanisms that promote species turnover along tropical elevational gradients. In general, my results partially supported the hypothesis that nest predation is a primary driver of t he suite of life history traits observed in tropical birds. The different incubation strategies and investment decisions documented in the Manu Andean elevation gradient reflect the complexity and diversity of life history strategies concealed in this hype r diverse tropical area. Overall bird species that encounter higher daily nest predation rates and maintained their egg at warmer temperatures had shorter incubation periods ( Martin et al. 2007 ) These characteristics were mainly associated with lowland bird communities. But, generally, daily nest predation and egg temperature alone explained only a small percentage of the variation observed on this tropical elevat ion gradient ( Martin 2004 ; Tieleman et al. 2004 ; Robinso n et al. 2010 ) The large variation in incubation strategies observed along the elevation gradient and within bird communities suggest that previously documented contradictory results may be resolved by studying two different tropical forest bird communi ties, lowlands ( Tieleman et al. 2004 ) and cloud forest ( Martin et al. 2007 ) which have contrasting incubatio n strategies. Therefore I need to explore additional factors that could explain
77 the observed variation within tropical communities ( Martin 2004 ) for example, by comparing nesting strategies among tropical bird species within a co mmunity ( Ricklefs & Brawn 2012 ) Finally, the extreme clutch size reduction and associated slow pace of life in high perpetually cold environment characterized by low productivity ( Salinas et al. 2011 ) and low nesting success ( Robinson et al. 2000 ; Martin et al. 2006 ; Robinson et al. 2010 ) Tanagers with single egg clutches had longer incubation and nestling periods than those at lower elevations with two egg clutches. Given the constant high nest predation rates along the entire elevation gradient in tanagers, there should be intense selection for rapid incubation and development; therefore, the long nesting period at high elevations suggests that shorter pe riods are not an option. Indeed, correcting for length of nesting cycle, tanagers of high elevation actually have lower nesting success than those with two egg clutches nesting at lower elevations. Thus, maintaining viable populations at high elevations mi ght only be possible with frequent re nesting ( Roper et al. 2010 ) and high survival of adults ( Gill & Haggerty 2012 ) and fledglings ( Tarwater et al. 20 11 ; Gill & Haggerty 2012 ) Given that most high elevation tanagers are relatively large, 51% larger than those in the lowlands, they may live longer ( Peterson & Wroblewski 1984 ; Brown et al. 2004 ) Until I have data on these additiona l life history variables, however, I will not be able to assess the whole suite of life history traits that characterize tanagers living at these high elevations.
78 APPENDIX A NUMBER OF NESTS DEPREDATED BY INDICATED SPECIES OF TAXONOMIC GROUPS. Table A 1. Number of nests depredated by i ndicated species of taxonom ic gr oups. Predator Taxonomic Group Species Number of Nest Predated Birds 25 Emerald Toucanet ( Aulacorhynchus prasinus ) 4 Chestnut tipped Toucanet ( Aulacorhynchus derbianus ) 1 Curl crested Aracari ( Pteroglossus beauharnaesii ) 1 White throated Toucan ( Ramphastos tucanus ) 1 Golden collared Toucanet ( Selenidera reinwardtii ) 3 Barred Forest Falcon ( Micrastur ruficollis ) 2 White Collared Jay ( Cyanolyca viridicyanus ) 1 Andean Pygmy Owl ( Glaucidium jardinii ) 1 Barn Owl ( Tyto alba ) 1 Great Thrush( Turdus fuscater ) 8 House Wren ( Troglodytes aedon ) 1 Mammals 56 White lipped Peccary ( Tayassu pecari ) 1 Ocelot ( Leopardus pardalis ) 1 Common Opossum ( Dedelphis marsupialis ) 2 Whi te fronted capuchin ( Cebus albifrons ) 3 Tufted capuchin ( Sapajus apella ) 6 Saddle back Tamarin ( Saguinus fuscicollis ) 9 Tayra ( Eira barbara ) 1 Andean Skunk ( Conepatus sp ) 1 Marmosa ( Marmosa lepida ) 4 Marmosa ( Marmosa spp .) 11 Unknown Marsupia l 3 Oryzomys spp. 5 Unknown Rodents 9 Reptiles 46 Gold Tegu ( Tupinambis teguixin ) 2 Puffing Snake ( Pseustes poecilonotus ) 36 Amazon Tree Boa ( Corallus hortulanus ) 1 Rainbow Boa ( Epichrates cenchria ) 2 Forest Flame Snake ( Oxyrhopus petol a ) 4 Mussurana ( Clelia clelia ) 1 Invertebrates 9 Bullet Ant ( Paraponera clavata ) 1 Unknown Ants 5 Bees 2 Snail 1 Total 136
79 Table A 2. Tanager species list. Environmental and nesting variables associated with the tanager clade S pecies Clutch size Incubation period (days) Nestling Period (days) Elevation (m) Refrences Anisognathus igniventris 1 21 21 3050 This study Bangsia edwardsi 2 1050 Isler and Isler 1999, Hilty 2011 Buthraupis montana 1 20 2700 This study Camarhynchus heliobates 3.5 12 10 500 Isler and Isler 1999, Hilty 2011 Camarhynchus pallidus 3.5 500 Isler and Isler 1999, Hilty 2011 Camarhynchus parvulus 3.5 500 Isler and Isler 1999, Hilty 2011 Camarhynchus pauper 3.5 14.5 500 Isler and Isler 1999, Hilty 20 11 Camarhynchus psittacula 3.5 12 11 500 Isler and Isler 1999, Hilty 2011 Certhidae olivacea 3.5 500 Isler and Isler 1999, Hilty 2011 Chlorochrysa calliparea 1 21 1450 This study Chlorornis riefferii 1 22 2500 This study Cissopis leverianus 2.5 12 .5 15 1150 Isler and Isler 1999, Hilty 2011 Coereba flaveola 3 12.5 17 1200 Isler and Isler 1999, Hilty 2011 Diuca diuca 3 1250 Isler and Isler 1999, Hilty 2011 Diuca speculifera 2.5 4750 Isler and Isler 1999, Hilty 2011 Euneornis campestri 3 750 Isler and Isler 1999, Hilty 2011 Geospiza conirostris 3.5 12 14 500 Isler and Isler 1999, Hilty 2011 Geospiza difficilis 3.5 12 14 500 Isler and Isler 1999, Hilty 2011 Geospiza fortis 3.5 12 12 500 Isler and Isler 1999, Hilty 2011 Geospiza fuliginosa 3.5 500 Isler and Isler 1999, Hilty 2011 Geospiza magnirostris 3.5 12 14 500 Isler and Isler 1999, Hilty 2011 Geospiza scandens 3.5 12 12.5 500 Isler and Isler 1999, Hilty 2011 Gubernatrix cristata 2.5 350 Isler and Isler 1999, Hilty 2011 Iridosorn is jelskii 1 2950 Isler and Isler 1999, Hilty 2011 Lophospingus pusillus 3 650 Isler and Isler 1999, Hilty 2011 Loxigilla barbadensis 4 170 Isler and Isler 1999, Hilty 2011
80 Table A 2. Continued Species Clutch size Incubation period (days) Nestlin g Period (days) Elevation (m) Refrences Loxigilla noctis 3 495 Isler and Isler 1999, Hilty 2011 Loxigilla portoricensis 3 500 Isler and Isler 1999, Hilty 2011 Loxigilla violacea 3.5 1000 Isler and Isler 1999, Hilty 2011 Loxipasser anoxanthus 3.5 900 Isler and Isler 1999, Hilty 2011 Melanospiza richardsoni 2 475 Isler and Isler 1999, Hilty 2011 Melopyrrha nigra 4 450 Isler and Isler 1999, Hilty 2011 Neothraupis fasciata 2.5 13 11.7 825 Duca and Marini 2011, Isler and Isler 1999 Orchesticu s abeillei 2 1175 Isler and Isler 1999, Hilty 2011 Paroaria capitata 3.5 250 Isler and Isler 1999, Hilty 2011 Paroaria coronata 3 250 Isler and Isler 1999, Hilty 2011 Paroaria gularis 3 200 Isler and Isler 1999, Hilty 2011 Pinaroloxias inornata 3.5 500 Isler and Isler 1999, Hilty 2011 Pipraeidea melanonota 2.5 14 17 1900 Alquezar et al. 2010, This study Saltator rufiventris 3 3000 Isler and Isler 1999, Hilty 2011 Schistochlamys melanopis 2 1175 Isler and Isler 1999, Hilty 2011 Schistoc hlamys ruficapillus 2 600 Isler and Isler 1999, Hilty 2011 Stephanophorus diadematus 4 1300 Isler and Isler 1999, Hilty 2011 Tanagar cyanocephala 3 12.5 15 600 Isler and Isler 1999, Hilty 2011 Tangara arthus 2 14.5 1550 Isler and Isler 1999, Hilty 2011 Tangara cayana 2 850 Isler and Isler 1999, Hilty 2011 Tangara chilensis 2.5 16 850 Isler and Isler 1999, Hilty 2011 Tangara cucullata 2 500 Isler and Isler 1999, Hilty 2011 Tangara cyanicollis 2 17 15 1350 Isler and Isler 1999, Hilty 2011 Ta ngara dowii 2 1900 Isler and Isler 1999, Hilty 2011 Tangara guttata 2 13 15 1150 Skutch 1989 Tangara gyrola 2 13.5 15.5 1300 Skutch 1989
81 Table A 2. Continued Species Clutch size Incubation period (days) Nestling Period (days) Elevation (m) Refrences Tangara heinei 2 14 1900 Freeman and Greeney 2009 Tangara icterocephala 2 13.2 15 1150 Skutch 1989 Tangara inornata 2 600 Isler and Isler 1999, Hilty 2011 Tangara larvata 2 14.5 15 475 Skutch 1989 Tangara lavinia 3 600 Isler and Isler 1999, Hilt y 2011 Tangara mexicana 2.5 13 600 Isler and Isler 1999, Hilty 2011 Tangara nigrocincta 2 13 17 800 Isler and Isler 1999, Hilty 2011 Tangara schrankii 2 15 13.5 700 Isler and Isler 1999, Hilty 2011 Tangara seledon 3 17 550 Isler and Isler 1999, Hilty 2011 Tangara vassorii 2 2500 Isler and Isler 1999, Hilty 2011 Tangara vitriolina 2 1350 Isler and Isler 1999, Hilty 2011 Tangara xanthocephala 2 1800 This study Thraupis abbas 2.5 1050 Isler and Isler 1999, Hilty 2011 Thraupis bonariensis 3 1275 Isler and Isler 1999, Hilty 2011 Thraupis cyanocephala 1 22 21 2250 Isler and Isler 1999, Hilty 2011 Thraupis cyanoptera 2 750 Isler and Isler 1999, Hilty 2011 Thraupis episcopus 2 13 17.5 900 Isler and Isler 1999, Hilty 2011 Thraupis ornata 3 1025 Isler and Isler 1999, Hilty 2011 Thraupis palmarum 2 14 19 950 Isler and Isler 1999, Hilty 2011 Tiaris bicolor 3 430 Isler and Isler 1999, Hilty 2011 Tiaris canorus 2.5 750 Isler and Isler 1999, Hilty 2011 Tiaris fuliginosus 2.5 1550 Isler and Isler 1999, Hilty 2011 Tiaris obscurus 3.5 1300 Isler and Isler 1999, Hilty 2011 Tiaris olivaceus 3 1150 Isler and Isler 1999, Hilty 2011 Tangara chrysotis 2 1700 Isler and Isler 1999, Hilty 2011
82 APPENDIX B NUMBER OF NESTS ENCOUNTERED FOR THE 84 BIRD SPECIES IN THE MANU GRADIENT AND THE SPECIES ASSOCIATED WITH EACH COMMUNITY. FROM THE 84 SPECIES, TEN WHERE SHARED AMONG COMMUNITIES. Table B 1 Number of nests encountered for the 84 bird species in the manu gradi ent and the species associated with each community. From the 84 species, ten where shared among communities. Species Number of nest Forest c ommunity Anisognatus igniventris 23 Elfin Arremon torquatus 2 Elfin Atlapetes melanolaemus 25 Cloud/Elfin Automolus infuscatus 3 Lowland Basil eutrus chrysogaster 4 Foothill Buthraupis montana 6 Elfin Catharus dryas 28 Cloud Chlorornis riefferii 3 Elfin Chlorospingus flavigularis 20 Cloud Chlorothraupis carmioli 23 Lowland/Foothill Conopophaga ardesiaca 26 Cloud Doryfera johannae 1 Lowlan d Doryfera ludoviciae 16 Cloud Dysithamnus mentalis 13 Cloud Elaenia albiceps 3 Cloud Elaenia pallatangae 20 Elfin Euphonia mesochrysa 1 Cloud Euphonia xantogaster 51 Foothill/Cloud Formicarius analis 7 Lowland/Foothill Geotrygon montana 72 Lowland /Foothill Glaucis hirsutus 3 Lowland Glyphorynchus spirurus 8 Lowland Grallaria guatimalensis 4 Cloud Heliodoxa leadbeateri 5 Cloud Hemitriccus granadensis 2 Elfin Hydropsalis climacocerca 12 Cloud Hylophylax naevius 70 Lowland/Foothill Lanio vers icolor 7 Lowland Leptopogon amaurocephalus 8 Lowland Leptotila rufaxilla 7 Lowland Lochmias nematura 2 Cloud
83 Table B 1 Continued Species Number of nest Forest c ommunity Manacus manacus 2 Lowland Micromonacha lanceolata 6 Foothill Mionectes ma cconnelli 9 Lowland/Foothill/Cloud Mionectes olivaceus 4 Foothill/Cloud Myiadestes ralloides 15 Cloud Myioborus miniatus 28 Cloud Myiodinastes chrysocephalus 16 Cloud Myiotheretes fuscorufus 7 Cloud Myrmeciza fortis 1 Elfin Nyctiphrynus ocellatus 2 Lowland Ochthoeca rufipectoralis 9 Lowland Octhornis litoralis 54 Elfin Odontophorus stellatus 2 Lowland Onycorhynchus coronatus 2 Foothill Ortalis guttata 2 Lowland Peryglina leuconota 4 Lowland/Cloud Phaeothlypis fulvicauda 2 Lowland Phaethornis guy 5 Lowland Phlegopsis nigromaculata 4 Foothill Pipra chloromeros 3 Lowland Pipraeidea melanonota 11 Lowland Platyrinchus mystaceus 2 Elfin Premnoplex brunnescens 1 Cloud Psarocolius atrovirens 24 Cloud Pyrrhomyias cinnamomea 18 Cloud Ramphocelus carbo 86 Elfin Rhynchocyclus fulvipectus 44 Lowland/Cloud Rupicola peruviana 6 Cloud Saltator maximus 8 Cloud Sayornis nigricans 4 Cloud Schistocicla leucostigma 20 Cloud Schizoeaca helleri 26 Lowland Sepophaga cinerea 5 Elfin Streptoprocne rutil a 19 Cloud Tangara cyanicollis 11 Elfin Tangara schrankii 3 Cloud Terenotricus erythrurus 59 Lowland
84 Table B 1 Continued Species Number of nest Forest c ommunity Tersina viridis 2 Lowland Thraupis cyanocephala 2 Lowland Tinamus guttatus 21 Elfin Tinamus major 4 Foothill Troglodytes aedon 2 Lowland Trogon melanurus 3 Elfin Turdus albicollis 2 Lowland Turdus fuscater 13 Lowland Turdus haxwelling 33 Elfin Turdus leucops 8 Cloud Uropsalis segmentata 3 Cloud Willisornis poecilonotus 3 Elfin X iphorhynchus guttatus 5 Lowland Zonotricha capensis 11 Lowland Total 1134
85 APPENDIX C VOLUNTEERS THAT WORKED DURING LONG HOURS AND MONTHS SEARCHING AND MONITORING BIRD NESTS BEWTEEN AUGUST AND DECEMBER TO MAKE THIS DISSER TATION POSSIBLE. Figure C 1. Volunteers that worked during long hours and months searching and monitoring bird nests bewteen August and December to make this dissertation possible.
86 LIST OF REFERENCES Auer S.K., Bassar R.D., Fontaine J.J. & Martin T.E. (2007). Breeding biology of passerines in a subtropical montane forest in northwestern Argentina. Condor 109, 321 333. Badyaev V.A. (1997). Avian Life History Variation along Altitudinal Gradients: An Example with Cardueline Finches. Oe cologia 111, 365 374. Badyaev V.A. & Ghalambor K.C. (2001). Evolution of Life Histories along Elevational Gradients: Trade Off between Parental Care and Fecundity. Ecology 82, 2948 2960. Baur B. & Raboud C. (1988). Life History of the Land Snail Arianta arbustorum Along an Altitudinal Gradient. Journal of Animal Ecology 57, 71 87. Benson T.J., Brown J.D. & Bednarz J.C. (2010). Identifying predators clarifies predictors of nest success in a temperate passerine. The Journal of animal ecology 79, 225 34. B erven K.A. (1982a). The Genetic Basis of Altitudinal Variation in the Wood Frog Rana sylvatica. I. An Experimental Analysis of Life History Traits. Evolution; international journal of organic evolution 36, 962 983. Berven K.A. (1982b). The Genetic Basis o f Altitudinal Variation in the Wood Frog Rana sylvatica. II. An Experimental Analysis of Larval Development. Oecologia 52, 360 369. Biebach H. (1981). Energetic costs of incubation on different clutch sizes in Starlings (Sturnus vulgaris). Ardea 69, 141 142. Blanckenhorn W.U. (1997). Altitudinal life history variation in the dung flies Scathophaga stercoraria and Sepsis cynipsea. Oecologia 109, 342 352. Blomberg S.P., Garland T. & Ives A.R. (2003). Testing for phylogenetic signal in comparative data: beh avioral traits are more labile. Evolution; international journal of organic evolution 57, 717 745. Bosque C. & Bosque M.T. (1995). Nest Predation as a Selective Factor in the Evolution of Developmental Rates in Altricial Birds. American Naturalist 145, 2 34 260. Boyle W.A. (2008). Can variation in risk of nest predation explain altitudinal migration in tropical birds? Oecologia 155, 397 403.
87 Brown J.H., Gillooly J.F., Allen A.P., Savage V.M. & West G.B. (2004). Toward a Metabolic Theory of Ecology Ecolog y 85, 1771 1789. Brua R.B. (2002). Parent embryo interactions. In: Avian Incubation Behaviour, Environment, and Evolution (ed. Deeming DC). Oxford University Press Oxford, New York. Brumfield R.T. & Edwards S.V. (2007). Evolution into and out of the Andes : a Bayesian analysis of historical diversification in Thamnophilus antshrikes. Evolution; international journal of organic evolution 61, 346 67. Bryant S.M. & Bryant D.M. (1999). Heating Nest Boxes Reveals an Energetic Constraint on Incubation Behaviour in Great Tits, Parus major. Proceedings. Biological sciences / The Royal Society 266, 157 162. Carter G.M., Legare M.L., Breininger D.R. & Oddy D.M. (2007). Nocturnal nest predation: a potential obstacle to recovery of a Florida Scrub Jay population. Jour nal of Field Ornithology 78, 390 394. Chalfoun A.D., Thomson F.R. & Ratnaswamy M.J. (2002). Nest predators and fragmentation: a review and meta analysis. Conservation Biology 16, 306 318. Clutton Brock T.H. (1991). The evolution of parental care Princet on University Press, Princeton, New Jersey. Conway C.J. & Martin T.E. (2000a). Effects of ambient temperature on avian incubation behavior. Behavioral Ecology 11, 178 188. Conway C.J. & Martin T.E. (2000b). Evolution of passerine incubation behavior: Infl uence of food, temperature, and nest predation. Evolution; international journal of organic evolution 54, 670 685. Cox W.A., Thompson F.R. & Faaborg J. (2012). Species and temporal factors affect predator specific rates of nest predation for forest songbi rds in the Midwest. The Auk 129, 147 155. Cresswell W. (1997). Nest predation: the relative effects of nest characteristics, clutch size and parental behaviour. Animal behaviour 53, 93 102. Cresswell W. (2010). Predation in bird populations. Journal of O rnithology 152, 251 263. Deeming D.C. (2002a). Behaviour patterns during incubation. In: Avian incubation: Behaviour, Environment and Evolution (ed. Deeming DC). Oxford University press Oxford, New York, pp. 63 87.
88 Deeming D.C. (2002b). Embryonic developm ent and utilisation of egg components. In: Avian Incubation Behaviour, Environment, and Evolution (ed. Deeming DC). Oxford univeristy Press Oxford, New York, pp. 43 53. Dinsmore S.J., White G.C. & Knopf F.L. (2002). Advanced techniques for modeling avian n est survival. Ecology 83, 3476 3488. Ferrier S., Manion G., Elith J. & Richardson K. (2007). Using generalized dissimilarity modelling to analyse and predict patterns of beta diversity in regional biodiversity assessment. Diversity and Distributions 13, 252 264. Fontaine J.J., Martel M., Markland H.M., Niklison A.M., Decker K.L. & Martin T.E. (2007). Testing ecological and behavioral correlates of nest predation. Oikos 116, 1887 1894. Fontaine J.J. & Martin T.E. (2006). Parent birds assess nest predation risk and adjust their reproductive strategies. Ecology letters 9, 428 34. Ghalambor C.K. & Martin T.E. (2000). Parental investment strategies in two species of nuthatch vary with stage specific predation risk and reproductive effort. Animal behaviour 60 263 267. Ghalambor K.C., Huey B.R., Martin R.P. & Tewksbury J.J. (2006). Are Mountain Passes Higher in the Tropics? Janzen's Hypothesis Revisited. Integrative and Comparative Biology 46, 5 17. Ghalambor K.C. & Martin T.E. (2002). Comparative manipulatio n of predation risk in incubating birds reveals variability in the plasticity of responses. Behavioral Ecology 13, 101 108. Gill A.S. & Haggerty M.T. (2012). A comparison of life history and parental care in temperate and tropical wrens. Journal of Avian Biology 43, 461 471. Gillooly J.F., Charnov E.L., West G.B., Savage V.M. & Brown J.H. (2002). Effects of size and temperature on developmental time. Nature 417, 70 73. Gillooly J.F., Londoo G.A. & Allen A.P. (2008). Energetic constraints on an early dev elopmental stage: a comparative view. Biology letters 4, 123 6. Gotelli N.J. & Ellison A.M. (2004). A primer of Ecological Statistics Sinauer Associates, Inc., Sunderland, Massachusetts. Haftorn S. & Reinertsen R.E. (1985). The Effect of Temperature and Clutch Size on the Energetic Cost of Incubation in a Free Living Blue Tit (Parus caeruleus). Auk 102, 470 478. Hancock A.M., Hughes J.M. & S. B.E. (1998). Influence of Genetic and Environmental Factors on Egg and Clutch Sizes among Populations
89 of Paratya australiensis Kemp (Decapoda: Atyidae) in Upland Rainforest Streams, South East Queensland. Oecologia 115, 483 491. Herzog S.K., Kessler M. & Bach K. (2005). The elevational gradient in Andean bird species richness at the local scale: a foothill peak and a high elevation plateau. Ecography 28, 209 222. Hilty S.L. (2011). Family Thraupidae (Tanagers). In: Handbook of the birds of the world (ed. Del Hoyo J, Elliott, A. and Christie, D. A.). Lynx Edicions Barcelona, Spain. Howard J.H. & Wallace R.L. (19 85). Life History Characteristics of Populations of the Long toed Salamander (Ambystoma macrodactylum) from Different Altitudes. American Naturalist 113, 361 373. Ibez lamo J.D. & Soler M. (2012). Predator induced female behavior in the absence of male incubation feeding: an experimental study. Behavioral Ecology and Sociobiology 66, 1067 1073. Isler M.L. & Isler P.R. (1999). The Tanagers Smithsonian Institution Press, Washington, D.C., USA. Jankowski J.E., Londoo G.A., Robinson S.K. & Chappell M.A. (2013). Exploring the role of physiology and biotic interactions in determining elevational ranges of tropical animals. Ecography 36, 001 012. Jankowski J.E., Merkord C.L., Rios W.F., Cabrera K.G., Revilla N.S., Silman M.R. & Gillman L.N. (2012). The rela tionship of tropical bird communities to tree species composition and vegetation structure along an Andean elevational gradient. Journal of Biogeography n/a n/a. Jankowski J.E., Robinson S.K. & Levey D.J. (2010). Squeezed at the top: Interspecific aggress ion may constrain elevational ranges in tropical birds. Ecology 91, 1877 1884. Janzen H.D. (1967). Why Mountain Passes are Higher in the Tropics. American Naturalist 101, 233 249. Jetz W., H. S.C. & K. B. G. (2008). The Worldwide Variation in Avian Clutc h Size across Species and Space. Plos 6, 2650 2657. Jetz W., Thomas G.H., Joy J.B., Hartmann K. & Mooers A.O. (2012). The global diversity of birds in space and time. Nature 491, 444 8. Krementz D.G. & Handford P. (1984). Does Avian Clutch Size Increase with Altitude? Oikos 43, 256 259.
90 Lack D. (1947). The significance of clutch size. I. Intraspecific variations. Ibis 89, 302 352. Lack D. (1968). Ecological adaptations for breeding in birds Methuen & Co, London, UK. Lennon J.J., Koleff P., Greenwood J. J.D. & Gaston K.J. (2001). The geographical structure of British bird distributions: diversity, spatial turnover and scale. Journal of Animal Ecology 70, 966 979. Libsch M.M., Batista C., Buehler D., Ochoa I., Brawn J. & Ricklefs R.E. (2008). Nest Predati on in a Neotropical Forest Occurs during Daytime. The Condor 110, 166 170. Lima S.L. (2002). Putting predators back into behavioral predator prey interactions. Trends in Ecology and Evolution 17, 70 75. Lima S.L. (2009). Predators and the breeding bird: behavioral and reproductive flexibility under the risk of predation. Biological Reviews 84, 485 513. Lu X., Yu T., Liang W. & Yang C. (2010). Comparative breeding ecology of two White bellied Redstart populations at different altitudes. Journal of Field O rnithology 81, 167 175. Martin T.E. (1995). Processes organizing open nesting bird assemblages: competition or nest predation? Evolutionary Ecology 2, 37 50. Martin T.E. (2004). Avian life history evolution has an eminent past: does it have a bright futu re? Auk 121, 289 301. Martin T.E. (2008). Egg size variation among tropical and temperate songbirds: an embryonic temperature hypothesis. Proceedings of the National Academy of Sciences of the United States of America 105, 9268 71. Martin T.E., Auer S. K., Bassar R.D., Niklison A.M. & Lloyd P. (2007). Geographic variation in avian incubation periods and parental influences on embryonic temperature. Evolution; international journal of organic evolution 61, 2558 69. Martin T.E., Bassar R.D., Bassar S.K., Fontaine J.J., LLoyd P., Matewson H.A., Niklison A.M. & Chalfoun A. (2006). Life history and ecological correlates of geographic variation in egg and clutch mass among passerine species. Evolution; international journal of organic evolution 60, 390 398. Martin T.E., Lloyd P., Bosque C., Barton C.D., Biancucci L.A., Cheng Y.R. & Ton R. (2011). Growth rate variation among passerine species in tropical and temperate sites: an antagonistic interaction between parental food provisioning and nest predation ris k. Evolution; international journal of organic evolution 65, 1607 1522.
91 Martin T.E., Martin R.P., Olson C.R., Heidinger B.J. & Fontaine J.J. (2000). Parental Care and Clutch Sizes in North and South American Birds. Science 287, 1482 1485. Martin T.E. & R oper J.J. (1988). Nest predation and nest site selection of a western population of the Hermit Thrush. Condor 90, 51 57. responds differently to nest predation risk across hemispheres. Animal behaviour 82, 1347 1356. McKinnon L., Smith P.A., Nol E., Martin J.L., Doyle F.I., Abraham K.F., Gilchrist H.G., Morrison R.I. & Bety J. (2010). Lower predation risk for migratory birds at high latitudes. Science 327, 326 7. Merkord C.L. (2010). Seasonality and elevationla migration in an Andean bird community. In. Univeristy of Missouri Columbia. Mertens J.A.L. (1977). Thermal Conditions for Successful Breeding in Great Tits (Parus major L.). I. Relation of Growth and Development o f Temperature Regulation in Nestling Great Tits. Oecologia 28, 1 29. Moreau R.E. (1944). Clutch size: a comparative study, with special reference to African. Ibis 86, 286 347. Moreno J. (1989). Energetic constraints on uniparental incubation in the wheat ear Oenanthe oenanthe L. Ardea 77, 107 115. Morosinotto C., Thomson R.L. & Korpimaki E. (2010). Habitat selection as an antipredator behaviour in a multi predator landscape: all enemies are not equal. The Journal of animal ecology 79, 327 33. Nord A. & N ilsson J. (2012). Context dependent costs of incubation in the pied flycatcher. Animal behaviour 84, 427 436. Olson C.R., Vleck C.M. & Vleck D. (2006). Periodic Cooling of Bird Eggs Reduces Embryonic Growth Efficiency. Physiological and Biochemical Zoo logy 79, 927 936. Pagel M. (1999). Inferring the historical patterns of biological evolution. Nature 401, 877 884. Parejo D. & Aviles J.M. (2011). Predation risk determines breeding territory choice in a Mediterranean cavity nesting bird community. Oecol ogia 165, 185 91. Patternson B.D., Stotz D.F., Solari S., Fitzpatrick J.W. & Pacheco V. (1998). Contrasting Patterns of Elevational Zonation for Birds and Mammals in the Andes of Southeastern Peru. Journal of Biogeography 25, 593 607.
92 Peterson J. & Wroble wski J.S. (1984). Mortality rates of fishes in the pelagic ecosyste. Canadian Journal of Fisheries and Aquatic Sciences 41, 1117 1120. Rahbek C. (1977). The relationship among area, elevation, and regional species richness in neotropical birds. American Na turalist 149, 875 902. Rahbek C. & Graves G.R. (2001). Multiscale Assessment of Patterns of Avian Species Richness. Proceedings of the National Academy of Sciences 98, 4534 4539. Rahn H. & Ar A. (1974). The avian egg: Incubation time and water loss. Cond or 76, 147 152. Reidy J.L. & Thomson F.R. (2012). Predatory identity can explain nest predation patterns. In: Video surveillance of nesting birds (eds. Ribic CA, Thompson FR & Pietz PJ). University of California Press Berkeley, CA. Remsen J.V. & Graves W .S. (1995). Distribution patterns of Buarremon Brush finches (Emberizinae) and interspecific competition in Andean bird. Auk 112, 225 236. Reznick D. & Endler J.A. (1982). The Impact of Predation on Life History Evolution in Trinidadian Guppies (Poecilia reticulata). Evolution; international journal of organic evolution 36, 160 177. Reznick N.D., Butler J.M., Rodd H.F. & Ross P. (1996). Life history evolution in guppies (Poecilia reticulata) 6. diferential mortality as a mechanism for natural selection. E volution; international journal of organic evolution 50, 1651 1660. Ricklefs R.E. (1967). A graphical method of fitting equtions to growth curves Ecology 48, 978 983. Ricklefs R.E. (1969). An analysis of nesting mortality in birds Smithsonian instituti on press, Washington, D.C. Ricklefs R.E. (1984). The Optimization of Growth Rate in Altricial Birds. Ecology 65, 1602 1616. Ricklefs R.E. & Brawn J. (2012). Nest attentiveness in several Neotropical suboscine passerine birds with long incubation periods. Journal of Ornithology 154, 145 154. Ricklefs R.E. & Wikelski M. (2002). The physiology/life history nexus. Trends in Ecology and Evolution 17, 462 468. Robinson S.K., Thompson F.R., Donovan T.M., Whitedead D.R. & Faaborg J. (1995). Regional forest frag mentation and the nesting success of migratory birds. Science 267, 1987 1990.
93 Robinson W.D., Hau M., Klasing K.C., Wikelski M., Brawn J.D., Austin S.H., Tarwater C. & Ricklefs R.E. (2010). Diversification of Life Histories in New World Birds. The Auk 1 27, 253 262. Robinson W.D., Robinson T.R., Robinson S.K. & Brawn J.D. (2000). Nesting Success of Understory Forest Birds in Central Panama. Journal of Avian Biology 31, 151 164. Robinson W.D., Rompr G. & Robinson T.R. (2005). Videography of Panama nests shows snakes are the principal predator. Ornitologia Neotropical 16, 187 195. Rodewald A.D. & Kearns L.J. (2011). Shifts in Dominant Nest Predators along a Rural to Urban Landscape Gradient. The Condor 113, 899 906. Roff A.D. (2002). Life history evolutio n Sinauer Associates, Inc., Sunderland, Massachusetts. Rohr D.H. (1997). Demographic and Life History Variation in Two Proximate Populations of a Viviparous Skink Separated by a Steep Altitudinal Gradient. Journal of Animal Ecology 66, 567 578. Roper J .J., Sullivan A.K. & Ricklefs R.E. (2010). Avoid nest predation when predation rates are low, and other lessons: testing the tropical temperate nest predation paradigm. Oikos 119, 719 729. Rotella J. (2007). Modeling nest survival data: recent improvement s and future directions. Studies in Avian Biology 34, 145 148. Salinas N., Malhi Y., Meir P., Silman M., Roman Cuesta R., Huaman J., Salinas D., Huaman V., Gibaja A., Mamani M. & Farfan F. (2011). The sensitivity of tropical leaf litter decomposition to t emperature: results from a large scale leaf translocation experiment along an elevation gradient in Peruvian forests. The New phytologist 189, 967 77. Sedano R.E. & Burns K.J. (2010). Are the Northern Andes a species pump for Neotropical birds? Phylogenet ics and biogeography of a clade of Neotropical tanagers (Aves: Thraupini). Journal of Biogeography 37, 325 343. Skutch A.F. (1949). Do tropical birds rear as many young as they can nourish? Ibis 91, 430 Skutch A.F. (1985). Clutch size, nesting succe ss, and predation on nests of neotropical birds, reviewed. In: Ornithological Monographs (ed. Foster MS). American ornithologists' union Washington, D.C., pp. 575 594.
94 Smith P.A., Tulp I., Schekkerman H., Gilchrist H.G. & Forbes M.R. (2012). Shorebird incu bation behaviour and its influence on the risk of nest predation. Animal behaviour 84, 835 842. Stearns S.C. (1992). The evolution of life histories Oxford University press, New York. Stotz D.F., Fitzpatrick J.W., Parker T.A. & Moskovits D.K. (1996). Neo tropical Birds ecology and conservation Univeristy of Chicago press, Chicago, Illinois. Tarwater C.E. (2008). Predators at Nests of the Western Slaty Antshrike (Thamnophilus atrinucha). The Wilson Journal of Ornithology 120, 620 624. Tarwater C.E., Rickl efs R.E., Maddox J.D. & Brawn J.D. (2011). Pre reproductive survival in a tropical bird ad its implications for avian life historie. Ecology 92, 1271 1281. Terborgh J. (1971). Distribution on Environmental Gradients: Theory and a Preliminary Interpretatio n of Distributional Patterns in the Avifauna of the Cordillera Vilcabamba, Peru. Ecology 52, 23 40. Tieleman B.I., Williams J.B. & Ricklefs R.E. (2004). Nest attentivenessandegg temperaturedo not explainthe variationin incubationperiodsin tropicalbirds. F unctional Ecology 18, 571 577. Tinberger J.M. & Dietz M.W. (1994). Parental Energy Expenditure During Brood Rearing in the Great Tit (Parus major) in Relation to Body Mass, Temperature, Food Availability and Clutch Size. Functional Ecology 8, 563 572. Tu rner S.J. (2002). Maintenance of egg temperature. In: Avian Incubation: Behaviour, Environment, and Evolution (ed. Deeming DC). Oxford University press Oxford, New York, pp. 119 142. Visser M.E. & Lessells C.M. (2001). The costs of egg production and incub ation in great tits (Parus major). Proceedings. Biological sciences / The Royal Society 268, 1271 7. Walker B., Stotz D.F., Pequeo T. & Fitzpatrick J.W. (2006). Birds of the Manu Biosphere Reserve. Fieldiana Zoology 110, 23 49. Wiersma P., Muoz Garcia A., Walker A. & Williams J.B. (2007). Tropical birds have a slow pace of life. Proceedings of the national academy of science 104, 9340 9345. Wikelski M., Spinney L., Schelsky W., Scheuerlein A. & Gwinner E. (2003). Slow pace of life in tropical sedentary birds: a common garden experiment on four stonechat populations from different latitudes. Proceedings of the royal society B 270, 2383 2388.
95 Williams D.T. (2012). Physiological adaptations for breeding in birds Princeton University press, Princeton, New Jersey. Williams J.B. (1996). Energetics of avian incubation. In: Avian energetics and nutritional ecology (ed. Carey C). Chapman and Hall New York. Zanette L.Y., White A.F., Allen M.C. & Clinchy M. (2011). Perceived predation risk reduces the number of o ffspring songbirds produce per year. Science 334, 1398 401. Zang V.H. (1980). The altitudinal influence on the population density and on the breeding biology of Ficedula hypoleuca, Parus palustris, Parus major and Parus ater in the Harz Mountains (Germany ). Journal fr Ornithologie 121, 371 386.
96 BIOGRAPHICAL SKETCH Gustavo Adolfo Londoo was born in July of 1976 and grew up in Cali, Colombia. During his childhood hood he want it to be a soccer player, but also spent most of his summer vacation in a fam ily farm, swimming in rivers, hiking and riding horses in the mountains and valley next to the Cauca river. After finishing high school at Gimnasio la Colina in 1993 he spent one year as an exchange student learning English in Morton, Illinois USA (the Pum pkin capital of the world). After a year, when he returned to the country, he moved to Bogota to start his undergraduate studies in Biology at los Andes University, in August of 1994. He became interested in birds after participating in field techniques co urse at the Tinigua National Park, Meta, Colombia, in December of 1997. The following summer he participate in a three months study exploring dietaries Mitu salvini ). In January of 1999 he and four other friends retur ned to the same place to collect data for their undergraduate thesis on Mitu salvini ) and Black Curassows ( Crax alector ) during seven months. After finishing his undergraduate in March of 2000, he spent one month on Costa Rica counting Great Crassows ( Crax rubra ) and three months in a Marbled Murrelet ( Brachyramphus marmoratus ) project in the Queen Charlotte Islands and Desolation Sound in British Columbia, Canada. In February of 2011 he spent one year at Cos ha Cashu biological station where he not only study the Razor billed Curassows ( Mitu tuberosum ), but he became interested in the nesting biology of tropical birds that year Gustavo monitored more than 130 nest from 57 bird species and published the first n esting records for three neotropical species. When he returned to Colombia he was planning to go back to Peru but due to the lack of funding the study was cancel and Gustavo started work at a local NGO EcoAndian in June of
97 2002. During two years at EcoAndi na Gustavo studied the biology of three Andean Guans at Santuario de Fauna y Flora Otn Quimbaya, Risaralda, Colombia. In the fall of 2004 Gustavo began his graduate studies at the University of Florida, under the supervision of Scott Robinson and Douglas Levey, where he studied nesting strategies of the Northern Mockingbird ( Mimus polyglottus ) und er different temperature regimes he traveled to the Manu National park during the fall to explore and collect preliminary data for his doctoral study. This dissertat ion is the result of 23 moths of fieldwork in an elevation gradient in Southeastern Peru between 350 and 3100 m. Outside his field regularly to Washing DC, to visit Rosario Casta eda his fianc e and now current wife