Differential physiological responses to prey availability by the great egret and white ibis

The Journal of Wildlife Management; DOI: 10.1002/jwmg.445 Differential Physiological Responses to PreyAvailability by the Great Egret and White Ibis GARTH HERRING,1,2 Department of Biological Sciences, Florida Atlantic University, 777 Glades Road, Boca Raton, FL 33431, USA DALE E. GAWLIK, Department of Biological Sciences, Florida Atlantic University, 777 Glades Road, Boca Raton, FL 33431, USA ABSTRACT In long-lived species, the balance between the benefits of reproduction and the costs fromreduced survival or productivity is particularly challenging in dynamic environments like wetlands, wherefood levels vary greatly year to year. Some wetland species exhibit changes in reproductive strategies inresponse to food availability but whether physiological responses function in a similar manner is unclear. Wecompared the pre-breeding physiological responses (fecal corticosterone [FCORT], heat shock protein 60[HSP60], and mass) of 2 species of wading birds with contrasting foraging strategies (great egret [Ardea alba],an exploiter, and white ibis [Eudocimus albus], a searcher) during years with contrasting levels of preyavailability. Both species were in good physiological condition, with low levels of HSP60 and FCORT,during a year with high prey availability (2006). In a contrasting year with lesser prey availability (2007),HSP60 and FCORT concentrations indicated that ibis physiological condition was reduced, whereas egretsshowed little change. Egrets and male ibis increased body mass, whereas female ibis decreased mass, in theyear with low prey availability. Although poorly understood, we hypothesize that the differential responsebetween female ibis and the others is associated with differential investment strategies based on long-termcosts of reproduction. Model results identified prey availability and the 2-week water recession rate as theprimary habitat variables that were associated with the physiological condition of white ibises, whereas greategret physiological condition was influenced mostly by 2-week water recession rate. Our results support thehypothesis that prey availability and hydrological factors play crucial roles in regulating populations of wadingbirds in the Florida Everglades. The results of this study show a more complete pathway by which hydrologicpatterns affect wading birds, and it suggests that ibis are more sensitive to habitat conditions than are egrets.
This information can be used to refine species models designed to evaluate water management scenarios andwill improve our ability to manage and restore wetland ecosystems ß 2012 The Wildlife Society.
KEY WORDS Ardea alba, corticosterone, Eudocimus albus, Florida Everglades, heat shock proteins, nesting ecology,pre-breeding.
Reproduction is a costly undertaking for most species, with birds to their local environment should reflect local condi- those individuals that reproduce more than once having to tions and their inherent ability to forage within those balance the cost of reproduction with surviving to future reproduction attempts (Ricklefs 1977, Partridge and Harvey One way birds respond to unpredictable food availability 1988, Roff 1992, Golet and Irons 1999). One mechanism in during the reproductive period is to use different forag- birds that increases productivity while minimizing risk of ing strategies that allow them to maximize food intake mortality, is improving pre-breeding physiological condition while potentially minimizing competition. Wading birds prior to egg laying, thereby increasing the likelihood of have been categorized relative to a foraging strategy contin- successful breeding, (i.e., capital breeders; Drent and Daan uum bounded by searchers and exploiters (Gawlik 2002), 1980). The value of this mechanism depends on the amount analogous to the cream-skimmer–crumb-picker continuum of food that is available in the environment, the ability of the (Brown et al. 1997). Searcher species (e.g., white ibis bird to access it, and the demands of the pre-breeding period.
[Eudocimus albus], wood stork [Mycteria americana]) select If birds are able to recognize changes in habitat conditions, high quality foraging patches and abandon those sites when then their life-history traits are predicted to track an optimal patch quality declines (e.g., prey densities reach a critical reaction norm with regard to the particular environment giving up density; Gawlik 2002). Exploiter species (e.g., encountered (Stearns 1992, Kawecki and Stearns 1993, great egret [Ardea alba], great blue heron [A. herodias]) follow Kisdi et al. 1998). Accordingly, the observed responses of a different rule in regards to when they should leave a patch,leaving when patch quality is lower and after the searchers Received: 12 December 2011; Accepted: 30 May 2012 have abandoned the patch (Gawlik 2002). These foragingstrategies evolved over long periods of time under natural wetland conditions; the ability of birds to predict conditions Present address: U.S. Geological Survey, Forest and Rangeland may be more difficult when wetlands become highly altered Ecosystem Science, 25 Center, 3200 SW Jefferson Way, Corvallis,OR 97331, USA.
Herring and Gawlik  Ibis and Egret Pre-Breeding Physiology One method for understanding the role that food avail- iological condition might play an influential role in setting up ability plays in avian ecology is to examine the physiological these responses observed later in the reproductive cycle.
condition of birds. Physiological biomarkers have the poten- Our objectives were to compare the physiological responses tial to respond directly and predictably to ecological con- of pre-breeding great egrets and white ibises to landscape- ditions, making them useful for quantifying how wading level prey availability and the hydrologic variables that birds respond to fluctuating resource levels, particularly in influence foraging conditions in the Florida Everglades.
highly stochastic ecosystems (Herring et al. 2011). Studies We focused on the great egret and white ibis because they of physiological responses can identify the mechanisms have different foraging strategies and have shown different by which 1 foraging strategy becomes more beneficial trends in nest numbers; great egret increased steadily during than another under particular environmental conditions.
the 1990s, whereas the white ibis declined or remained stable Further, understanding these physiological responses during (Crozier and Gawlik 2003a). White ibises are more restrict- the pre-breeding period may provide a basis for understand- ed in their use of habitats than are great egrets (Gawlik 2002, ing subsequent species-specific differences in reproductive Beerens et al. 2011); therefore we expected that the physio- logical condition of white ibises would also be more restricted One physiological biomarker is the hormone corticoste- by habitat conditions than would that of great egrets. We rone, which serves as a physiological signal to modify behav- hypothesized that during good habitat conditions (e.g., hy- ior and metabolism in response to potentially adverse change.
drologic conditions that produce predictable patches of con- Corticosterone is released into the blood stream via the centrated prey), great egrets and white ibises would be in adrenocortical tissue when birds become stressed (e.g., low good pre-breeding physiological condition. However, during food availability) inducing a response, and allowing them to poor habitat conditions (hydrologic conditions that fail to overcome the short-term deficiency (Astheimer et al. 1992, produce predictable patches of concentrated prey), we Wingfield 1994). Increased corticosterone levels in birds expected white ibises to be in poorer physiological condition have been correlated with food shortages (Kitaysky et al.
than great egrets. We used measures of physiological condi- 2003, Herring et al. 2011). Corticosterone can be measured tion that we expected would represent a temporal continuum either directly from the blood stream or excreta, although of responses from the short and medium term (FCORT fecal corticosterone (FCORT) metabolite levels are less than metabolites; Wasser et al. 2000, Herring et al. 2011) to long circulating levels (Wasser et al. 2000) because of rapid and term (heat shock protein 60 [HSP60]; Sørensen et al. 2003, extensive metabolization before excretion.
Toma´s et al. 2004, Herring et al. 2011). We also examined An alternative but less utilized group of physiological changes in body mass (a common measure of body condition; parameters for measuring nutritional condition are the stress Ankney and MacInnes 1978, Afton and Ankney 1991) in proteins, which function as molecular chaperones for pro- egrets and ibises throughout the pre-breeding stage.
teins within cells (Linquist 1986, Bierkens 2000). Duringperiods of increased stress, the molecular chaperone role isamplified to minimize cell protein damage (Locke and Noble1995, Ra˚berg et al. 1998). Their delayed response relative toplasma corticosterone (Burel et al. 1992) suggests that theybetter indicate chronic long-term stress (Martı´nez-Padilla et al. 2004, Toma´s et al. 2004) and should be independent ofcapture stress. In many species and taxa, stress proteins are The Florida Everglades is a large subtropical oligotrophic induced in response to a variety of stressors including heavy wetland in southern Florida with pronounced annual wet and metals (Werner and Nagel 1997, Martı´nez et al. 2001), dry seasons (Obeysekera et al. 1999). The dry season, when nutritional stress (Merino et al. 2002), food limitation most wading birds reproduce, is typically from November (Herring et al. 2011), and others (see Herring and Gawlik until May but can vary slightly between years. Over half of the spatial extent of the Everglades has been lost to In the Florida Everglades, environmental conditions, such agriculture and urban development. The hydrologic as water depth, recession rate, and landscape-level prey patterns of remnant areas were greatly altered, thus changing availability are known to influence breeding success and the timing, magnitude, and predictability of seasonal wading bird foraging (Frederick and Collopy 1989, prey concentration events that are critical to wading bird Gawlik 2002, Herring et al. 2010). Gawlik (2002) proposed reproduction (Ogden 1994). We conducted our study in that species-specific differences in foraging strategies could the 3 Water Conservations Areas (WCAs 1–3: 57,951, account for species-specific population trends under the same 54,390, and 236,984 ha, respectively) that comprise the environmental conditions, although he did not identify a northern Everglades. Because we captured wading birds mechanistic pathway of the effects. Most recently, Herring early in the dry season, our capture sites were restricted to et al. (2010, 2011) found that differences in nesting success the short hydroperiod portions of the WCAs, where prey and physiological condition observed during the breeding first become concentrated and available to wading birds.
season were linked in part to species-specific foraging strate- The study area was bounded to the north by 26.68408, gies (searchers vs. exploiters) and prey availability across the the east by À80.22178, the west by À80.85608, and the landscape. As a result, we expected adult pre-breeding phys- We recorded tarsus length (middle of midtarsal joint to the end of tarso-metatarsus), wing chord, wing flattened, exposed culmen length, bill depth, and mass for both Hydrologic conditions in the Florida Everglades during the species, and curved bill length for white ibis only. We 2006 dry season were near optimal for wading bird nesting as recorded all measurements to the nearest 1 mm using suggested by Gawlik (2002). Water depths were above aver- calipers or a wing ruler, except mass, which we measured age at the start of the dry season and then receded unhin- to the nearest 5 g using a spring scale. We captured all dered by significant water level reversals, which result in the birds during the morning (0600 hours to 1000 hours). We redistribution of concentrated prey and diminish accessible banded all birds with uniquely numbered United States foraging patches because of increased water depths (Herring Geological Survey bands and attached radio transmitters et al. 2011). This stable, protracted recession in 2006 fostered to all birds for a related study examining habitat selection comparatively elevated prey densities (Herring et al. 2010, and nest survival of egrets and ibises (Herring et al. 2010, 2011) and a large number of nesting wading birds (Cook and Beerens et al. 2011). We sexed all birds later using Call 2006). The widespread recession and succeeding abnor- DNA analysis (Zoogen Services Inc., Davis, CA; Herring mally delayed wet season of 2006 produced unusually low water levels during the 2006 wet season, and culminated inan official drought during the 2007 dry season (Herring et al.
2010). The 2007 dry season was parallel to that of 2006, with We homogenized thawed fecal samples and divided them few hydrological reversals and a moderately continuous re- into 2 equal 1-ml wet portions, which we dried using a cession; however, water depths were lower, hydroperiods Labconco CentriVap Concentro (Labconco, Kansas City, were shorter, and mean prey density and biomass were MO). We mixed dried samples (approx. 0.25 g) with 5 ml of relatively low (Herring et al. 2010, 2011). Concurrent studies 95% ethanol and vortexed for 30 minutes. After centrifuga- on the physiological conditions of egret and ibis nestlings tion (15 min, 2,500g) we transferred the supernatant to a during these same 2 years found that both species responded new vial, and then evaporated it under a stream of nitrogen negatively to the decreased habitat conditions in 2007, but gas. We resuspended CORT metabolites in diluted extrac- the effect was greater for ibis (Herring 2008, Herring et al.
tion buffer and measured them using the Correlate-EIATM 2010). Based on the fact that ibis were found to be food Corticosterone Enzyme Immunoassay Kit (EIA; Rothschild limited during 2007 (Herring et al. 2011), we classified 2006 et al. 2008, Herring and Gawlik 2009) following the as a year with good habitat conditions and 2007 as a year with manufacturer’s instructions (Assay Designs, Inc., Ann Arbor, MI). We determined inter- and intra-assay coeffi- cients of variation for FCORT internal standards to be 7% We captured great egrets and white ibises during the pre- and 11%, respectively, for egrets, and 8% and 9%, respec- breeding season using either a net-gun or modified flip trap tively, for ibises. We validated EIAs for FCORT metabolites (Herring et al. 2008a) and decoys (Crozier and Gawlik (Herring et al. 2011). We also validated the assumption that 2003b, Heath and Frederick 2003) between 10 January FCORT levels did not change after freezing (Herring and and 23 March. Capture dates corresponded with the seasonal pattern of the 2 species arriving in the Everglades in great We washed red blood cells 3 times using phosphate- numbers prior to the upcoming breeding season, which buffered saline, centrifuged the red blood cells, and removed began in late March to early April. Both species generally the pellet after the final wash. We mixed the red blood cell spend the non-breeding season throughout the southeastern pellet with 1Â extraction reagent and a protease inhibitor United States and move slowly back to the Everglades as the cocktail (Sigma, St. Louis, MO), placed it in a vortexer for breeding season approaches (Heath et al. 2009, McCrimmon 5 minutes, and then sonicated it for 1 minute. We centri- et al. 2011). During 2006 and 2007, we captured and sam- fuged samples again (15 min, 2,500g) and removed the pled 209 adult birds (79 great egrets [49 F, 30 M] and 130 supernatant. We measured HSP60 (HSPD1) in the super- white ibises [67 F, 63 M]) across the entire pre-breeding natant using EIA kits specific to just those stress proteins and not all other HSP60 family members. We determined inter- Upon capture of a bird, we immediately placed a hood on its and intra-assay coefficients of variation for HSP60 internal head to minimize movement during subsequent sampling standards to be 5% and 7%, respectively. We ran all samples and measurements. We collected up to 1 ml of blood from in duplicate, and used the means of duplicates in subsequent the brachial vein using a 27.5-gauge needle and stored analyses. We validated all EIA kits using serial dilutions and samples in heparinized vacuutainers, placed on ice until spike tests to determine percent recovery (Herring et al.
transport to the lab. We then extracted up to 2 ml of fecal material directly from the cloaca of the adult using a micro-pipette. We stored fecal samples in micro centrifuge tubes and placed them on ice. In the lab, we centrifuged (15 min, We used the Everglades Depth Estimation Network 10,000g) blood samples, to separate plasma and red blood (EDEN; USGS 2006) to estimate water depth and water cells. We subsequently froze blood and fecal samples at level recession rate at foraging sites where adults were captured. The EDEN used a network of water level gauges Herring and Gawlik  Ibis and Egret Pre-Breeding Physiology to produce a water surface model that, when combined influenced the response of each physiological parameter for with a ground elevation model, provided an estimate of each species. We used an information-theoretic approach water depth for the entire freshwater portion of the (Akaike 1974, Burnham and Anderson 2002) that ranked Greater Everglades. The EDEN calculated water level stage competing models developed from a biological understand- in 400-m by 400-m grid cells at daily time steps accounting ing of wading foraging ecology and evidence from previous for evapotranspiration, rainfall, and sheet flow. The estimat- ed water depths were accurate to within 5 cm (Liu et al.
We ran separate models for each species to understand 2009). A depth of 0 cm indicated that the water surface was their individual physiological responses (FCORT, HSP60, at the average ground elevation. In such cases, standing mass) to differing habitat conditions (i.e., hydrology and water was present in small areas within a cell.
prey availability and date within years) with sex as a covari- To estimate recession rate and water depth at foraging ate. We included a variable for mean prey availability for sites, we first used ArcGIS 9.1 (Environmental Systems each year because food abundance is considered one of the Research Institute, Inc., Redlands, CA) to delineate the most influential determinants of nesting effort (Lack 1954, EDEN grid cells within 3 km of each capture site. We Ricklefs 1968) and it may be linked to the size of wading assumed this area contained the habitat that birds used bird populations in the Everglades (Kahl 1964, Kushlan within 2 weeks prior to their capture because radio-tagged 1977, Gawlik 2002, Herring et al. 2010). We included birds regularly returned to within about 3 km of the standardized date to better understand the physiological capture site for the subsequent several days. We extracted condition changes over the weeks leading up to reproduc- daily water depth values for the 7 and 14 days preceding tion. We included measurements of the 1-week and 2-week a bird’s capture for all cells within the foraging area, and mean water depth (depth) and the quadratic forms of water used the mean depth and change in depth over each period depth (depth þ depth2) 1 week and 2 weeks before we made as the measures of water depth and recession rate, respec- an estimate of physiological condition because water depth tively. Positive recession rates indicate decreasing water can limit foraging (Kushlan 1976, Gawlik 2002). We also included models with the 1-week and 2-week mean reces-sion rate (recession) and the quadratic form of recession rate (recession þ recession2) 1 week and 2 weeks before we made We measured the biomass of wading bird prey that was an estimate of physiological condition because recession rate available to wading birds across the ecosystem in 2006 is a determinant of wading bird nesting success (Frederick and 2007 in a concurrent study (see Pierce and Gawlik and Collopy 1989, Herring et al. 2010) and distributions in [2010] for details on sampling design). We estimated prey the Everglades (Bancroft et al. 2002, Russell et al. 2002, availability as the mean biomass of all prey (fishes and macro- Beerens et al. 2011). We included the sex of a bird in the invertebrates) captured during the period of our adult trap- models as a covariate to account for potential differences in ping and sampling. Our measure of prey biomass was an acceptable surrogate for prey availability because we obtained In the models of mass, we also included an index of our samples from shallow water with an open habitat struc- body size as a covariate to account for some of the variance ture, 2 factors that make prey highly vulnerable to capture by in the analysis associated with structural size differences.
birds (Gawlik 2002). Thus, density was the primary deter- The body-size index consisted of the standardized scores minant of availability in our case. We determined mean prey of the first principal component PC1 obtained from a biomass during 2006 and 2007 to be 41.82 g/m2 (Æ18.78 principal component analysis on 5 morphometric measure- SE, n ¼ 91) and 6.55 g/m2 (Æ0.72 SE, n ¼ 168), respec- ments taken from each bird (Afton and Ankney 1991, tively. Although we averaged these estimates across the Esler et al. 2001). The first principal component explained landscape, they did demonstrate large differences between 59%, 53%, 50%, and 78% of the overall variation among years, as indicated by non-overlapping 95% confidence inter- morphometric measurements for female and male egrets vals (Gawlik et al. 2008), and we found them to be similar to and ibises, respectively. Global models included all main differences from an independent measure of prey density at effects, excluding the body size covariate, and their more localized sites during the same years (Herring et al.
2010, Beerens et al. 2011). Florida Atlantic University We used Akaike’s Information Criterion values adjusted Institutional Animal Care and Use Committee (Protocol for small samples sizes (AICc) in all models (Burnham and A0534) approved the research techniques, and we conducted Anderson 2002). We calculated differences in AICc (Di) and the research under United States Fish and Wildlife Service Akaike weights (wi). We considered competing models with Research Permit 23354 and Florida Fish and Wildlife 2 to be equally plausible and models with Conservation Commission Scientific Research Permit 4 to have less support (Burnham and Anderson 2002). To assess the relative influence of each parameter, wealso calculated their parameter weights by summing Akaike weights across all models that included each variable. We We used Proc Mixed in SAS (SAS Institute, Inc., Cary, calculated the model-averaged parameter estimates and their NC), specifying the maximum likelihood variance estimator standard errors using the full set of models (Burnham and (Littell et al. 1996) to determine which habitat variables most contained the variables for the 2-week recession rate and2-week depth (w ¼ only the 2-week recession rate was 1.6 and 1.8 times more Three plausible models explained FCORT metabolite levels likely than the next 2 models (Table 1). Models containing in pre-breeding great egrets. The first model contained only the 2-week recession rate had a combined AICc weight of 0.67, with little evidence for effects of 2-week depth (0.26), second model contained the variables for 2-week recession sex (0.22), prey availability (0.15), date (0.06), 1-week reces- sion rate (0.03), or 1-week water depth (0.03). All of the Table 1. Akaike’s Information Criterion (adjusted for small sample sizes; AICc) model selection for adult great egret and white ibis fecal corticosterone(FCORT), heat shock protein 60 (HSP60), and mass. Samples were collected in Water Conservation Areas 2A, 3A, and the Arthur R. Marshall LoxahatcheeNational Wildlife Refuge between 10 January and 23 March in both 2006 and 2007. Models presented only include those that were within 3 AICc units of the topmodels (DAIC ¼ Sex, date, 2-week recession, prey availability Sex, date, 2-week depth, prey availability, PC1e Sex, date, 2-week depth, prey availability, sex  prey availability, PC1 Sex, 2-week recession, prey availability, PC1 Sex, date, 2-week depth, 2-week recession, prey availability, PC1 Sex, date, 2-week recession, prey availability, PC1 Sex, date, 1-week depth, prey availability, PC1 2-Week depth, 2-week recession, prey availability Date, 2-week recession, prey availability Sex, 2-week depth, 2-week recession, prey availability Sex, date, 2-week recession, prey availability Sex, date, 2-week depth, prey availability Sex, 2-week depth, 2-week recession, prey availability Sex, date, 2-week recession, prey availability, sex  prey availability, PC1 Sex, date, 2-week depth, prey availability, sex  prey availability, PC1 Sex, date, 1-week recession, prey availability, sex  prey availability, PC1 Sex, date, 1-week depth, prey availability, sex  prey availability, PC1 Sex, 1-week depth, prey availability, sex  prey availability, PC1 Sex, 2-week recession, prey availability, sex  prey availability, PC1 Sex, 2-week depth, prey availability, sex  prey availability, PC1 Sex, 1-week recession, prey availability, sex  prey availability, PC1 Sex, date, 2-week depth, 2-week recession, prey availability, sex  prey availability, PC1 a Number of estimated parameters in the model including the variance plus the intercept.
b Second-order Akaike’s Information Criterion (AICc).
c The difference in the value between AICc of the current model and the value of the most parsimonious model.
d Likelihood of the model given the data, relative to models in the candidate set.
e First principal component for body measurements.
Herring and Gawlik  Ibis and Egret Pre-Breeding Physiology variables of influence for great egret FCORT metabolitelevels had beta coefficient estimates with 95% confidenceintervals that overlapped zero, suggesting none of them had alarge effect (Table 2). Great egrets had similar FCORTconcentrations between the year with high prey availabilityand the year with low prey availability (Fig. 1).
The most parsimonious model explaining differences in HSP60 concentrations in pre-breeding great egrets includedsex, date, and the 2-week recession rate (w ¼ No other models of great egret HSP60 concentrations werecompetitive (Table 1). The weight of the evidence suggestedthat the top model was 3.4 and 7.7 times more likely toexplain HSP60 levels than the next 2 models (Table 1).
Parameter likelihoods showed that the most influential var-iables were 2-week recession (0.88), sex (0.77), and date Table 2. Variable weights and weighted parameter estimates Æ standarderror (SE) from general linear mixed models evaluating the response of greategret and white ibis fecal corticosterone metabolites (FCORT), heat shockprotein 60 (HSP60), and mass in the Everglades, 2006–2007. Explanatoryvariables were included if variable weight exceeded 0.10; variables with thegreatest relative support have variable weights close to 1.0.
Figure 1. Model-averaged least-square mean estimates of great egret and white ibis corticosterone metabolites (FCORT), heat shock protein 60 (HSP60), and mass controlling for structural size during a year with elevated (2006; light bar) and low (2007; dark bar) prey availability in the Everglades.
(0.70), with less influence from prey availability (0.31) and 2-week water depth (0.24), and little effect from 1-week recession rate (0.02) or 1-week water depth (0.00). Great egret HSP60 concentrations decreased by 4.6 ng/ml with each 1-cm increase in recession rate, were on average 1.3 ng/ ml greater in females, and decreased by 0.11 ng/ml with each advancing day of the pre-breeding season (Table 2). Great egrets had similar HSP60 concentrations between the year with high prey availability and the year with low prey avail- The most parsimonious of 5 competitive models predicting mass contained the variables sex, date, 2-week depth, and receiving support (<2.0 DAICc; Table 1). Of those compet- sex, date, 2-week depth, and prey availability was 2.3 times itive models, the best model (1.7 and 1.8 times more likely more likely than the next competitive model (Table 1). The than the next 2 models) contained the variables for sex, date, other plausible models also included a term for 2-week prey availability, 2-week recession rate, and the interaction of recession rate. Parameter likelihood values suggested great sex  prey availability (Table 1). Parameter likelihood egret mass was most influenced by sex (0.97), prey availability values suggested white ibis mass was most influenced (0.84), date (0.63), 2-week water depth (0.57), and 2-week by sex (0.98), prey availability (0.98), the interaction of recession rate (0.34). However, the beta coefficient 95% sex  prey availability (0.98), and date (0.60), and to a lesser confidence intervals for the 2-week water depth and 2- extent by the 2-week recession rate (0.43). We found even week recession rate overlapped zero indicating a weak effect less support for 2-week water depth (0.28), 1-week water from these variables, and suggesting that mass was mostly depth (0.26), and 1-week recession rate (0.17; Table 2). The influenced by sex and prey availability. We found little beta coefficient 95% confidence intervals for date overlapped support for the 1-week recession rate (0.11), and 1-week zero indicating a weak effect and suggesting that mass was water depth (0.06; Table 2). Great egret mass was greater in most influenced by sex and prey availability. The supported males than females on average by 48 g, increased by 1.4 g/ interaction of sex  prey availability showed that in response day during the pre-breeding stage, and was greater during to decreasing prey availability, females decreased their mass the year with low prey availability on average by 69 g by 51 g, whereas males increased their mass by 72 g (Table 2, Fig. 1). On average males were 63 g heavier than females(Table 2, Fig. 1).
White IbisThe model selection process identified 3 competitive models to explain FCORT metabolites levels in pre-breeding white The physiological condition of great egrets was influenced ibises; the best model included the 2-week recession rate and less consistently by low prey availability than was the physi- ological condition of white ibises. We found strong support and 2.5 times more likely than the next 2 competitive models for the effects of sex on differences in mass and HSP60 in (Table 1). The other 2 competitive models contained vari- ibises. Of the landscape variables, we also found strong ables similar to those in the top model with the addition of support for the influence of prey availability on white ibis sex and 1-week recession rate (Table 1). Parameter likeli- FCORT and HSP60 concentrations (low prey availability hood values suggested that white ibis FCORT metabolite increased concentrations), and mass. We also found moder- levels were most influenced by prey availability (0.77) and ate support for the influence of the 2-week recession rate on 2-week recession rate (0.49), with moderate support for sex ibis FCORT and SP60 concentrations, and mass.
(0.39), and little support for the 2-week depth (0.21), date In the case of great egrets, sex and date influenced both (0.19), 1-week recession rate (0.18), and the 1-week depth mass and HSP60, whereas prey availability influenced egret (0.09; Table 2). However, the beta coefficient 95% confi- mass. The 2-week recession rate influenced HSP60 concen- dence intervals for the 2-week recession rate and sex both trations (high recession rates decreased HSP60 concentra- overlapped zero indicating a weak effect from these variables tions) and FCORT concentrations (although the beta and suggesting that FCORT levels were mostly influenced coefficient suggested this effect was small). Positive recession by prey availability. White ibis FCORT levels increased rates produce a high density of prey within local patches and during the year with decreasing prey availability on average move the location of available patches across the landscape because the ecosystem has a gradual slope southward.
The model selection process identified 4 competitive mod- Although we saw no evidence of negative effects of high els to explain HSP60 levels in pre-breeding white ibises recession rate over the range we observed, we would expect (Table 1). The best model contained the variables sex, the negative effects at a greater recession rate because the loca- 2-week recession rate, and prey availability (w ¼ tion of available patches can move too quickly and be beyond 1.5 and 1.8 times the support of the 2 next best models an effective foraging distance from a colony, thus causing (Table 1). Parameter likelihood values suggested white ibis nest abandonment (Bancroft et al. 1994). Also, a rapid adult HSP60 levels were most influenced by prey availability recession rate may reduce prey availability by drying patches (0.95), sex (0.68), and 2-week recession rate (0.56), with less before birds have a chance to access their prey, thereby support for date (0.34), 2-week water depth (0.27), and 1- reducing prey availability at a landscape level rather than week recession rate (0.23), and no support for 1-week mean water depth (0.08; Table 2). Also, the beta coefficient 95% Prey availability influenced the physiological condition of confidence intervals for date overlapped zero indicating a pre-breeding great egrets less than did recession rate based weak effect. White ibis HSP60 level increased during the on the model selection results. The response of adult egrets year with low prey availability on average 6.7 ng/ml and was to greater levels of prey availability was unexpected in the on average 1.4 ng/ml greater in females than males (Table 2, case of mass; lesser prey availability resulted in an increase in mass. The greater influence of recession rate than actual prey As with great egrets, we found considerable uncertainty in availability suggests that recession may affect foraging the models of white ibis mass, with 7 competitive models patches in more ways than simply concentrating prey. We Herring and Gawlik  Ibis and Egret Pre-Breeding Physiology speculate that an additional way in which positive recession prey availability, so this strategy might have some long-term rates may affect foraging birds is by improving their ability to find predictable feeding patches because new patches are Great egrets have several characteristics that may allow always formed just down the elevation gradient from existing them to increase their body masses during years when prey availability is low. Their long legs, broad diet, and exploiter Great egret FCORT metabolite concentrations in both foraging strategy allow great egrets to exploit a wide variety sexes were similar between years, as were HSP60 levels.
of habitat conditions (e.g., water depths; Gawlik 2002, These results suggested that egrets may not respond as Beerens et al. 2011), thus reducing the frequency at which acutely as do ibises to variability in prey availability.
they need to search for new foraging patches as water levels Indeed, the physiological condition of great egrets was change. Indeed, both species flew shorter distances to forag- most influenced by the 2-week recession rate, whereas ing patches in the year with low prey availability (Beerens prey availability and the 2-week recession rate were more 2008) and in doing so may have reduced the costs of foraging.
influential for white ibis physiological condition. We also Further, although not widely reported in the literature found differential responses between species to changes in (McCrimmon et al. 2011), great egrets foraged nocturnally landscape prey availability throughout the remainder of the on numerous occasions (Herring 2008) and this foraging breeding season for other parameters. During the low prey strategy could improve their overall daily intake of prey availability year, white ibises laid smaller clutches and fledged fewer chicks (Herring et al. 2010), and the physiological This study demonstrated consistently different physiologi- condition of those chicks was poorer (Herring 2008). Great cal results between great egrets and white ibises to reduced egrets maintained their clutch size, but fledged fewer chicks prey availability; ibises had greater short-term and long-term than in a good year (Herring et al. 2010), although the stress levels whereas egret stress levels remained similar physiological condition of those chicks was similar to that relative to the year with high prey availability. The extent of chicks hatched into the good year (Herring 2008).
to which our results can be extrapolated to other species Our data do not explain why the pattern of body mass in along the searcher and exploiter continuum will not be clear female ibis differed from that of male ibis and egrets, nor are until the responses of other species to variability in land- comparable findings available in the literature to clearly scape-level prey availability have been examined. Comparing understand these results. However, we believe that the species along the searcher and exploiter continuum will most plausible hypothesis is that the costs and benefits of improve our understanding of the generalized patterns different reproductive approaches differ for female ibis.
within these 2 groups and potentially yield new insights During good years, both egrets and ibises may not benefit into how sympatric species can respond differently to the from an increase in body mass leading up to breeding, same environmental conditions, with implications for sub- because prey availability is high and predictable, and mini- mizing mass reduces energetic costs. Models of foragingstrategies (McNamara and Houston 1987, Anholt and Werner 1998) predict that animals in rich environments The rate that water rises or falls is a parameter of high will spend less time feeding than those in poor environments.
management importance because it can often be manipulated Most research in this field has focused on the response of by managers and it has broad ecologically significance. Water birds to condition of low food availability (see Gosler et al.
recession rate and other variables that determine prey avail- 1995, Rogers and Reed 2003) or unpredictable food (see ability were influential in almost every egret and ibis physio- Pravosudov and Grubb 1997, Cuthill et al. 2000), demon- logical model, demonstrating the fundamental importance of strating an increase in mass or lipid reserves storage. A key maintaining good foraging habitat throughout the period of point that all of these studies demonstrated inadvertently time leading up to, and through, the breeding season.
is that during good years, birds do not need to have high Although shallow water conditions are needed to attract mass levels. Abreu and Kacelnik (1999) experimentally dem- wading birds to the Everglades during the early dry season, onstrated this response in European starlings (Sturnus vul- steady recession rates, and suitable foraging depths may act garis), when starlings had predictable food sources they together as a switch to initiate the physiological changes that lost mass. However, this strategy of preparing for unpredict- precede nesting. Recent research, consistent with our find- able food availability would not explain the response of ings, demonstrated that a recession rate of 0.5 cm/day was female ibis during the year with low prey availability nor optimal for maximizing nest success in great egrets (Herring why male ibis would have a different strategy under the et al. 2010), but that lesser recession rates were acceptable same conditions. Perhaps female ibises are minimizing management targets for foraging wading birds (Beerens long-term costs of reproduction to increase their likelihood et al. 2011) if the landscape had a high biomass of prey.
of potential future reproductive success as suggested by If prey biomass is unknown, then maintaining recession rates theoretical models (Drent and Daan 1980). This finding around 0.5 cm/day throughout the pre-breeding and breed- is supported by the fact that female ibises lay fewer eggs ing seasons is a reasonable management target in our study during years with lesser prey availability (Herring et al.
site. Nevertheless, the greater sensitivity of white ibis than 2010). However, female ibises were more stressed as evi- great egrets to habitat conditions may lead to different denced by greater levels of HSP60 during the year with low population responses, as we believe has occurred in the Everglades. Incorporating these species-specific responses Crozier, G. E., and D. E. Gawlik. 2003a. Wading bird nesting effort as an into models used to evaluate water management scenarios index of wetland ecosystem integrity. Waterbirds 26:303–324.
Crozier, G. E., and D. E. Gawlik. 2003b. The use of decoys in attracting will greatly improve our ability to effectively manage and wading birds. Journal of Field Ornithology 69:276–287.
Cuthill, I. C., S. A. Maddocks, C. V. Weall, and E. K. M. Jones.
2000. Body mass regulation in response to changes in feeding predict- ability and overnight energy expenditure. Behavioural Ecology 11:189–195.
Funding for research was provided by the United States Fish Drent, R. H., and S. Daan. 1980. The prudent parent: energetic adjustments in avian breeding. Ardea 68:225–252.
and Wildlife Service. We thank T. Dean for his support in Esler, D., J. B. Grand, and A. D. Afton. 2001. Intraspecific variation in implementing this study. We appreciate the enormous nutrient reserve use during clutch formation by lesser scaup. Condor efforts of J. Beerens in helping coordinate field research and capture of all the wading birds in this study. We appre- Frederick, P. C., and M. W. Collopy. 1989. Nesting success of Ciconiiform ciate the support and cooperation of the Arthur R. Marshall species in relation to water conditions in the Florida Everglades. Auk106:625–634.
Loxahatchee National Wildlife Refuge staff. M. Cook, P.
Gawlik, D. E. 2002. The effects of prey availability on the numerical Dixon, H. Herring, S. Milton, J. Volin, and 2 anonymous response of wading birds. Ecological Monographs 72:329–346.
reviewers provided valuable comments on previous drafts of Gawlik, D. E., G. Herring, B. Botson, and J. M. Beerens. 2008. The this manuscript. We thank our field research crews and relationship between Everglades landscape prey availability and wadingbird habitat selection and reproductive measures in M. I. Cook, editor.
fellow researchers that assisted in collection of field data: South Florida Wading Bird Report, Volume 14. South Florida Water T. Anderson, T. Beck, J. Beerens, B. Botson, E. Call, H.
Management District: West Palm Beach.
Herring, N. Hill, A. Horton, B. Imdieke, M. Kobza, and S.
Golet, G. H., and D. B. Irons. 1999. Raising young reduces body condition and fat scores in black-legged kittiwakes. Oecologia 120:530–538.
Gosler, A. G., J. J. Greenwood, and C. Perrins. 1995. Predation risk and the cost of being fat. Nature 337:621–623.
Heath, J. A., and P. C. Frederick. 2003. Trapping white ibis with rocket nets Abreu, F. B., and A. Kacelnik. 1999. Energy budgets and risk-sensitive and mist nets in the Florida Everglades. Journal of Field Ornithology foraging in starlings. Behavioral Ecology 10:338–345.
Afton, A. D., and C. D. Ankney. 1991. Nutrient-reserve dynamics of Heath, J. A., P. C. Frederick, J. A. Kushlan, and K. L. Bildstein. 2009.
breeding lesser scaup: a test of competing hypothesis. Condor 93:89–97.
White ibis (Eudocimus albus). Account 009 in A. Poole, editor. The birds Akaike, H. 1974. A new look at the statistical model identification. IEEE of North America online. Cornell Lab of Ornithology, Ithaca, New York, Transactions on Automatic Control AC 19:716–723.
Anholt, B. R., and E. E. Werner. 1998. Predictable changes in predation Herring, G. 2008. Constraints of landscape level prey availability on physi- mortality as a consequence of changes if food availability and predation ological condition and productivity of great egrets and white ibises in the risk. Evolutionary Ecology 12:729–738.
Florida Everglades. Dissertation, Florida Atlantic University, Boca Raton, Ankney, C. D., and C. D. MacInnes. 1978. Nutrient reserves and repro- ductive performance of female lesser snow geese. Auk 95:459–471.
Herring, G., and D. E. Gawlik. 2007. The role of stress proteins in the study Astheimer, L. B., W. A. Buttemer, and J. C. Wingfield. 1992. Interactions of allostatic overload in birds, use and applicability to current studies in of corticosterone with feeding activity and metabolism in passerine birds.
avian ecology. Scientific World Journal 7:1596–1602.
Herring, G., and D. E. Gawlik. 2009. Stability of avian fecal corticosterone Bancroft, G. T., A. M. Strong, R. J. Sawicki, W. Hoffman, and S. D. Jewell.
metabolite levels in frozen avian feces. Journal of Wildlife Management 1994. Relationships among wading bird foraging patterns, colony locations, and hydrology in the Everglades. Pages 615–657 in S. M.
Herring, G., D. E. Gawlik, and J. M. Beerens. 2008a. Evaluating two new Davis and J. C. Ogden, editors. Everglades: The Ecosystem and Its methods for capturing large wetland birds. Journal of Field Ornithology Restoration. St. Lucie Press, Delray Beach, FL.
Bancroft, G. T., D. E. Gawlik, and K. Rutchey. 2002. Distribution of Herring, G., D. E. Gawlik, and J. M. Beerens. 2008b. Predicting the sex of wading birds relative to vegetation and water depths in the northern great egrets and white ibises. Waterbirds 31:306–311.
Everglades of Florida, USA. Waterbirds 25:265–277.
Herring, G., D. E. Gawlik, M. I. Cook, and J. M. Beerens. 2010. Sensitivity Beerens, J. M. 2008. Hierarchical resource selection and movements of two of great egret and white ibis nesting to landscape prey availability. Auk wading bird species with divergent foraging strategies in the Florida Everglades. Thesis, Florida Atlantic University, Boca Raton, USA.
Herring, G., M. I. Cook, D. E. Gawlik, and E. M. Call. 2011. Food Beerens, J. M., D. E. Gawlik, G. Herring, and M. I. Cook. 2011. Dynamic availability for nestling white ibis is expressed through some physiological habitat selection by two wading bird species with divergent foraging stress indicators: a food supplementation experiment. Functional Ecology strategies in a seasonally fluctuating wetland. Auk 128:651–662.
Bierkens, J. G. E. A. 2000. Applications and pitfalls of stress-proteins in Kahl, M. P. 1964. Food ecology of the wood stork (Mycteriaa mericana).
biomonitoring. Toxicology 153:61–72.
Brown, J. S., B. P. Kotler, and W. A. Mitchell. 1997. Competition between Kawecki, T. J., and S. C. Stearns. 1993. The evolution of life histories in birds and mammals: a comparison of giving-up densities between crested spatially heterogeneous environments: optimal reaction norms revisited.
larks and gerbils. Evolutionary Ecology 11:757–771.
Burel, C., V. Mezger, M. Pinto, M. Rallu, S. Trigon, and M. Morange.
Kisdi, E., G. Meszena, and L. Pasztor. 1998. Individual optimization: 1992. Mammalian heat shock protein families. Expression and functions.
mechanisms shaping the optimal reaction norm. Evolution and Burnham, K. P., and D. R. Anderson. 2002. Model selection and inference: a Kitaysky, A. S., E. V. Kitaiskaia, J. F. Piatt, and J. C. Wingfield. 2003.
practical information-theoretic approach. Springer-Verlag, New York, Benefits and costs of increased levels of corticosterone in seabird chicks.
Cook, M. I., and E. M. Call. 2006. System-wide summary. Pages 1–2 in M.
Kushlan, J. A. 1976. Site selection for nesting colonies by the American Cook and E. Call, editors. South Florida wading bird report, Volume 12.
white ibis Eudocimus albus in Florida. IBIS 118:590–593.
South Florida Water Management District, West Palm Beach, Florida, Kushlan, J. A. 1977. Population energetics of the American white ibis. Auk Herring and Gawlik  Ibis and Egret Pre-Breeding Physiology Lack, D. 1954. The Natural Regulation of Animal Numbers. Oxford Pravosudov, V. V., and T. C. Grubb. Jr. 1997. Management of fat reserves University Press, London, United Kingdom.
and food caches in tufted titmice (Parus bicolor) in relation to unpredictable Lantz, S. M., D. E. Gawlik, and M. I. Cook. 2010. The effects of water food supply. Behavioural Ecology 8:332–339.
depth and submerged aquatic vegetation on the selection of foraging Ra˚berg, L., M. Grahn, D. Hasselquist, and E. Svensson. 1998. On the habitat and foraging success of wading birds. Condor 112:160–169.
adaptive significance of stress-induced immunosupression. Proceedings of Linquist, S. 1986. The heat shock response. Annual Review of Biochemistry the Royal Society of London Series B 265:1637–1641.
Ricklefs, R. E. 1968. Patterns of growth in birds. IBIS 110:419–451.
Littell, R. C., G. A. Milliken, W. W. Stroup, and R. D. Wolfinger. 1996.
Ricklefs, R. E. 1977. On the evolution of reproductive strategies in birds: SAS system for mixed models. SAS Institute, Cary, North Carolina, USA.
reproductive effort. American Naturalist 111:453–478.
Liu, Z., J. C. Volin, V. D. Owen, L. G. Pearlstine, J. R. Allen, F. J. Mazzotti, Roff, D. A. 1992. The evolution of life histories: theory and analysis.
and A. L. Higer. 2009. Validation and ecosystem applications of the Chapman and Hall, London, United Kingdom.
EDEN water-surface model for the Florida Everglades. Ecohydrology Rogers, C. M., and A. K. Reed. 2003. Does avian winter fat storage integrate temperature and resource conditions? A long-term study. Journal of Avian Locke M., and E. G. Noble. 1995. Stress proteins: the exercise response.
Canadian Journal of Applied Physiology 20:155–167.
Rothschild, D. M., T. L. Serfass, W. I. Seddon, L. Hegde, and R. S. Fritz.
Martı´nez, J., J. Pe´rez-Serrano, W. E. Bernadina, and F. Rodrı´guez- 2008. Using fecal glucocorticoids to assess stress levels in captive river Caabeiro. 2001. In vitro stress response to elevated temperature, hydrogen otters. Journal of Wildlife Management 72:138–142.
peroxide and mebendazole in Trichinella spiralis muscle larvae.
Russell, G. J., O. L. Bass, Jr., and S. L. Pimm. 2002. The effect of International Journal of Parasitology 29:1457–1464.
hydrological patterns and breeding-season flooding on the numbers and Martı´nez-Padilla, J., J. Martı´nez, J. A. Da´villa, S. Merino, J. Moreno, and J.
distribution of wading birds in Everglades National Park. Animal Milla´n. 2004. Within-brood size differences, sex and parasites determine blood stress protein levels in Eurasian kestrel nestlings. Functional Sørensen, J. G., T. N. Kristensen, and V. Loeschke. 2003. The evolutionary and ecological role of heat shock proteins. Ecology Letters 6:1025– McCrimmon, D. A., Jr., J. C. Ogden, and G. T. Bancroft. 2011. Great egret (Ardea alba). Account 570 in A. Poole and F. Gill, editors. The birds of Stearns, S. C. 1992. The evolution of life histories. Oxford University Press, North America. Academy of Natural Sciences, Philadelphia, Pennsylvania and American Ornithologists’ Union, Washington, D.C., USA.
Toma´s, G., J. Martı´nez, and S. Merino. 2004. Collection and analysis of McNamara, J. M., and A. I. Houston. 1987. Starvation and predation as blood samples to detect stress proteins in wild birds. Journal of Field factors limiting population size. Ecology 68:1515–1519.
McNeil, R., P. Drapeau, and R. Pierotti. 1993. Noctunality in colonial U.S. Geological Survey. 2006. The Everglades Depth Estimation Network waterbirds: occurrence, special adaptations, and suspected benefits.
(EDEN) for support of ecological and biological assessments. Fact Sheet 2006-3087, U.S. Department of the Interior, Washington, D.C., USA.
Merino, S., J. Martı´nez, A. P. Møller, A. Barbosa, F. de Lope, and F.
Wasser, S. K., K. E. Hunt, J. L. Brown, K. Cooper, C. M. Crockett, U.
Rodrı´guez-Caabeiro. 2002. Blood stress protein levels in relation to sex Bechert, J. J. Millspaugh, S. Larson, and S. L. Monfort. 2000. A general- and parasitism of barn swallows (Hirundo rustica). Ecoscience 9:300–305.
ized fecal corticosterone assay for use in a diverse array of nondomestic Obeysekera, J., J. Browder, L. Hornung, and M. A. Harwell. 1999. The mammalian and avian species. General and Comparative Endocrinology natural South Florida system I: climate, geology, and hydrology. Urban Werner I., and R. Nagel. 1997. Stress proteins HSP60 and HSP70 in three Ogden, J. C. 1994. A comparison of wading bird nesting dynamics, 1931– species of amphipods exposed to cadmium, diazinon, dieldrin and flour- 1946 and 1974–1989 as an indication of changes in ecosystem conditions anthane. Environmental Toxicology Chemistry 16:2393–2403.
in the southern Everglades. Pages 533–570 in S. Davis and J. Ogden, Wingfield, J. C. 1994. Modulation of the adrenocortical response to stress in editors. Everglades: the ecosystem and its restoration. St. Lucie Press, birds. Pages 520–528 in K. G. Davey, R. E. Peter, and S. S. Tobe, editors.
Perspectives in comparative endocrinology. National Research Council of Partridge, L., and P. H. Harvey. 1988. The ecological context of life history Pierce, R. L., and D. E. Gawlik. 2010. Wading bird foraging habitat selection in the Florida Everglades. Waterbirds 33:494–503.

Source: http://www.science.fau.edu/biology/gawliklab/papers/HerringGandDEGawlik2012.pdf