Northeastern Naturalist Vol. 21, No. 3 J.A. Cava, J.D. Riddle, and R.P. Thiel 2014 495 2014 NORTHEASTERN NATURALIST 21(3):495–505 Apparent Survival of Woodpeckers and Nuthatches in Wisconsin Jenna A. Cava1,*, Jason D. Riddle2, and Richard P. Thiel3 Abstract - Few annual survival and capture-probability estimates exist for sittid and picid species common in North America. We used a mark–recapture study and robust design analysis in Program MARK to estimate annual survival rates based on a sample of 51 Sitta carolinensis (White-breasted Nuthatch), 12 Picoides pubescens (Downy Woodpecker), and 15 Picoides villosus (Hairy Woodpecker) wintering in central Wisconsin, 2006–2013. Apparent survival probability was similar between the two woodpecker species (Downy Woodpecker: p = 0.51, 95% CI = 0.34–0.68; Hairy Woodpecker: p = 0.52, 95% CI = 0.36– 0.68). Apparent annual survival modeled as constant across time was low for both sexes of White-breasted Nuthatch (0.25 [CI = 0.12–0.44] and 0.28 [CI = 0.14–0.49] for males and females, respectively), but there was some evidence for annual variation in survival. All three species showed evidence for a trap-happy response in which recapture probability was higher than original capture probability, but it was stronger in the White-breasted Nuthatches than the two woodpecker species. There is little evidence of temporary emigration for any of the woodpecker taxa we studied. Our results provide baseline demographic data for these species in Wisconsin and will be useful in planning future trapping studies. Introduction Basic demographic data often are lacking for species considered common, widespread, or not threatened. However, it is important to obtain demographic data as a baseline for future comparisons and expand our knowledge of these species. Picoides pubescens L. (Downy Woodpecker), Picoides villosus L. (Hairy Woodpecker), and Sitta carolinensis (Latham) (White-breasted Nuthatch) are common species in North America and are currently designated as species of least concern by the IUCN (Grubb and Pravosudov 2008, Jackson and Ouellet 2002, Jackson et al. 2002). Research focused on the demographics of these species is scarce (Grubb and Pravosudov 2008, Jackson and Ouellet 2002, Jackson et al. 2002), and information is often embedded within articles investigating several species and general topics (e.g., Karr et al. 1990). Most demographic studies, while valuable, are geographically limited and provide an incomplete view of survivorship in these widely distributed species (e.g., Doherty and Grubb 2002, Karr et al. 1990). The Monitoring Avian Productivity and Survivorship (MAPS) Program provides low-resolution estimates from regions across North America; however, these estimates may not be directly comparable to other studies (Michel et al. 2006). 1Department of Biology, University of Wisconsin-Stevens Point, Stevens Point, WI 54481. 2Wildlife Ecology and Management, College of Natural Resources, University of Wisconsin- Stevens Point, WI 54481. 37167 Deuce Road, Tomah, WI 54660. *Corresponding author - jennaacava@gmail.com. Manuscript Editor: Jeremy Kirchman Northeastern Naturalist 496 J.A. Cava, J.D. Riddle, and R.P. Thiel 2014 Vol. 21, No. 3 Thus far, all studies reporting survivorship for any of these species have utilized Cormack-Jolly-Seber (CJS)-type models, which allow estimation of apparent annual survival and capture probabilities (Cormack 1964, DeSante et al. 1998, Doherty and Grubb 2002, Jolly 1965, Karr et al. 1990, Seber 1965). Another method for estimating these parameters is Pollock’s robust design, which uses additional closed-capture information (Kendall 2001, Kendall et al. 1997, Pollock 1982). Modified versions of these models allow temporary movement of individuals from the study site, which accounts for an additional source of variation in capture probabilities (Kendall and Nichols 1995, Kendall et al. 1997, Pollock 1982). Neither of these methods discerns probabilities of permanent emigration and mortality, resulting in an estimate of apparent survival that concedes some degree of uncertainty, as opposed to true survival (Gilroy et al. 2012, Lebreton et al. 1992). Some researchers have developed methods to reduce bias in survival estimates by using predicted dispersal rates combined with survival data (Gilroy et al. 2012), including emigration and mortality data in capture histories (Horton and Letcher 2008), or increasing the area covered for resighting individuals (Marshall et al. 2004) to provide more accurate survival estimates. These methods for obtaining estimates of survivorship require prior knowledge of a population’s movement patterns that is not available for many species or collection of additional data that may be impractical given a study’s objectives or resources (Horton and Letcher 2008). Estimation of temporary movements, capture probabilities, recapture probabilities, and survival rates via model selection potentially provide improved demographic and ecological information for understudied species that scientists may use to identify future research priorities. Our objectives were (1) to provide apparent survival and capture probabilities for the Downy Woodpecker, Hairy Woodpecker, and White-breasted Nuthatch in Wisconsin using the robust design method, and (2) determine the extent of temporary movement and trap response. Field-site Description Our study site was North Bluff, an approximately 259-ha Quercus spp. (oak) and Populus spp. (aspen) forest located within the southwest corner of Sandhill Wildlife Area, WI (N44°19'6.1", W90°10'50.6"). The surrounding area is a mix of marsh and forest patches. The trapping scheme comprised 2 concentric rings of suet-baited live traps with 13 traps surrounding the base of the bluff and 7 around the top of the bluff during 2006–2008, and 14 surrounding the base and 9 around the top during 2009–2013. We affixed traps to live tree trunks 1.2 m–1.4 m above the ground and spaced them about 160 m apart along the arc of the concentric rings (Fig. 1). Methods We made our traps based on a design by Fiske (1968). Trapping commenced during late January and ended with the snowmelt in early March each year. We trapped Saturdays and Sundays depending on weather restrictions and time of snowmelt, Northeastern Naturalist Vol. 21, No. 3 J.A. Cava, J.D. Riddle, and R.P. Thiel 2014 497 for an average of 6 trapping sessions per year (range = 5–9 sessions). We opened traps at 06:00 and checked them at 09:00, 12:00, and 15:00. Traps were wired shut between Saturday at 15:00 and Sunday at 06:00 to prevent loss of bait and wired open during the week so birds could consume excess bait without capture. We banded birds with sequentially numbered aluminum US Geological Survey bands in all years. We determined sex by plumage (Pyle et al. 1987). We captured 51 White-breasted Nuthatches (29 males, 22 females), 12 Downy Woodpeckers (6 males, 6 females), and 15 Hairy Woodpeckers (8 males, 7 females) during the 8 years of this study (Table 1). All recaptures occurred at traps as opposed to resighting birds, so that we estimated capture probabilities, not detection probabilities in our models. Figure 1. Aerial view of our study site, located within Sandhill Wildlife Area, WI. Traps 21–23 were not used during 2006–2008. Northeastern Naturalist 498 J.A. Cava, J.D. Riddle, and R.P. Thiel 2014 Vol. 21, No. 3 All statistical analyses were conducted in Program MARK (White and Burnham 1999). We assessed apparent annual survival (ϕ), within-year initial capture (p) and recapture (c) probabilities, and between-season temporary emigration probability (γ' and γ'') using the robust design (Kendall et al. 1997, Pollock 1982). The robust design assumes a closed population during the secondary sampling period, which in our study was between late January and early March. None of the species we studied reproduce during this time period, and the limited evidence available on these and related species suggests little dispersal occurs then (Grubb and Pravosudov 2008, Jackson and Ouellet 2002, Jackson et al. 2002). We did not expect dispersal of White-breasted Nuthatches to occur during this time period because winter territories are established during the fall (Butts 1931)), and in Sitta europaea L. (European Nuthatch) dispersal is rare after an individual’s first summer (Mattysen and Schmidt 1987). Downy Woodpeckers are known to inhabit their breeding home range during January–March, which suggests it is unlikely many disperse during late winter (Kellam 2003). We compared models that predicted a trap response, which allowed p and c to differ, to models that predicted no trap response, which held c equal to p. Because there was no indication that capture or recapture probabilities varied between years, we held p and c constant over time in all models. Temporary emigration can be modeled as Markovian or random using gamma- prime (γ', the probability that an individual will be absent from the study area if it was absent during the previous primary sampling session) and gammadouble- prime (γ'', the probability that an individual will be absent from the study area if it was present during the previous primary sampling session) (Kendall et al. 1997). For Markovian movement, the probability of an individual being on the study site depends on whether it was on or off the study site during the previous sampling session, and both γ'and γ'' may vary in the models. For random movement, the probability of an individual being on the study site is independent of its prior location, and γ' and γ'' are set equal to each other in the models. An absence of emigration can also be modeled by setting both gammas equal to zero. Results of preliminary analyses showed that we were unable to estimate Table 1. Hairy Woodpecker, Downy Woodpecker, and White-breasted Nuthatch captures and recaptures for each species per year; all data were collected at Sandhill Wildlife Area, WI, 2006–2013. Hairy Woodpecker Downy Woodpecker White-breasted Nuthatch Captures Recaptures Captures Recaptures Captures Recaptures 2006 0 0 4 0 15 6 2007 7 4 12 9 34 27 2008 7 6 4 4 12 7 2009 15 9 15 12 31 27 2010 17 15 9 9 39 25 2011 11 10 0 0 0 0 2012 6 5 6 4 39 29 2013 12 11 19 19 19 17 Total 75 60 69 57 189 138 Northeastern Naturalist Vol. 21, No. 3 J.A. Cava, J.D. Riddle, and R.P. Thiel 2014 499 temporary emigration as Markovian, so we only considered classical random emigration and an absence of emigration in our models (Kendall et al. 1997). Thus, in our models, we either held both γ' and γ'' constant over time (constant random emigration), or fixed γ at zero (no emigration). We modeled Downy and Hairy Woodpeckers together because of the similarity between their ecologies; White-breasted Nuthatches had their own model set. We constructed the initial woodpecker model set to investigate species and year effects on annual survival: ϕ was either held constant over time and between species, allowed to vary over time but constant between species, held constant over time but allowed to vary between species, or allowed to vary over time and between species. We then constructed an initial White-breasted Nuthatch model set to test sex and year effects on annual survival: ϕ was either held constant over time and between sexes, allowed to vary over time but constant between sexes, held constant over time but allowed to vary between sexes, or allowed to vary over time and between sexes. Models in both sets also investigated presence versus absence of a trap response with p and c always held constant over time and between species or sexes for woodpeckers or nuthatches, respectively. Finally, we used our models to assess random emigration versus an absence of temporary emigration, with both γ' and γ'' always held constant over time and between species or sexes for woodpeckers or nuthatches, respectively. All possible combinations of the parameters listed for each model set were included. Results Several woodpecker models were unable to estimate all parameters, including those that allowed γ' and γ'' to vary from zero or allowed survival to vary by year and species. We removed these models from our analysis to produce a final reduced set containing 6 models (Table 2). The top model (5W) accounted for double the AICc weight of the second-best model (13W), and together these had the majority of support (73%). Both models showed no difference in survival between Downy Table 2. Woodpecker reduced model set, ordered by AICc weight; all data were collected at Sandhill Wildlife Area, WI, 2006–2013. ϕ = apparent annual survival, p = initial capture probability, c = recapture probability, γ' = probability of the individual being off the study area given it was also off the study area during the previous primary sampling session, and γ'' = probability of the individual being off the study area given it was on the study area during the previous primary sampling session. # = number of parameters. AICc Model ϕ p c γ' γ'' AICc weight # Deviance 5W Constant Constant Constant 0 0 539.21 0.51 3 475.04 13W Constant * * 0 0 540.85 0.22 2 478.77 6W Species Constant Constant 0 0 541.22 0.19 4 474.92 14W Species * * 0 0 542.83 0.08 3 478.66 7W Year Constant Constant 0 0 549.93 0.00 9 472.50 15W Year * * 0 0 551.55 0.00 8 476.42 *p (initial capture probability) and c (recapture probability) were set equal to each other. Northeastern Naturalist 500 J.A. Cava, J.D. Riddle, and R.P. Thiel 2014 Vol. 21, No. 3 and Hairy Woodpeckers. Models 6W and 14W estimated each species’ survival separately and collectively represented the rest of support, suggesting there may be some difference between the species. Models 7W and 15W allowed survival to vary by year but held almost none of the support (less than 1%), indicating constant survival over time. Model-averaged apparent annual survival probability was 0.51 (95% CI = 0.34–0.68) and 0.52 (95% CI = 0.36–0.68) for Downy and Hairy Woodpeckers, respectively. Models 5W, 6W, and 7W indicated a trap response, while 13W, 14W, and 15W did not. Model-averaged recapture probability was higher than initial capture probability when estimated separately, but the confidence interval of p overlapped the point estimate of c (p = 0.31, 95% CI = 0.21–0.43; c = 0.39, 95% CI = 0.32– 0.46). All models in the woodpecker set indicated no temporary movement from the study site (γ' = 0 and γ'' = 0, but see below). All White-breasted Nuthatch models that allowed survival to vary by year or by year and sex were unable to estimate all survival parameters and we removed them from further analyses. However, models 3N and 7N, which allowed survival to vary by year, held the majority of AIC weight (67%) in the initial model set. This result provided evidence of annual variation in survival, with estimates ranging from 0.12 (95% CI = 0.02–0.53) to 0.60 (95% CI = 0.25–0.87) for both sexes. We did not include these models in the reduced model set used for averaging because the inestimable survival parameters for 2011 and 2012 would have interfered with the averaging procedure. The final reduced model set contained 8 models (Table 3). Similar apparent annual survival between the sexes held most of the support, but there was some evidence of a sex difference (Table 3). Model-averaged apparent annual survival was 0.25 (95% CI = 0.12–0.44) for males and 0.28 (95% CI = 0.14–0.49) for females. The top four models allowed for trap response while the Table 3. White-breasted Nuthatch reduced model set, ordered by AICc weight; all data were collected at Sandhill Wildlife Area, WI, 2006–2013. ϕ = apparent annual survival, p = initial capture probability, c = recapture probability, γ' = probability of the individual being off the study area given it was also off the study area during the previous primary sampling session, and γ'' = probability of the individual being off the study area given it was on the study area during the previous primary sampling session. # = number of parameters. AICc Model ϕ p c γ' γ'' AICc weight # Deviance 5N Constant Constant Constant Zero Zero 667.95 0.40 3 466.41 1N Constant Constant Constant ** ** 668.70 0.28 4 465.07 6N Sex Constant Constant Zero Zero 669.42 0.19 4 465.80 2N Sex Constant Constant ** ** 670.25 0.13 5 464.50 9N Constant * * ** ** 678.78 0.00 3 477.25 10N Sex * * ** ** 680.32 0.00 4 476.69 13N Constant * * Zero Zero 681.91 0.00 2 482.44 14N Sex * * Zero Zero 683.35 0.00 3 481.82 *p (initial capture probability) and c (recapture probability) were set equal to each other. **Probability of the individual being off the study area if it was off the study area during the previous primary sampling session set equal to the probability of the individual being off the study area if it was on the study area during the previous primary sampling session. Northeastern Naturalist Vol. 21, No. 3 J.A. Cava, J.D. Riddle, and R.P. Thiel 2014 501 rest did not; collectively these top 4 held >99% of the support, indicating that a trap response occurred. Nuthatches showed a trap-happy response with higher recapture probability than initial capture probability: p = 0.23, 95% CI = 0.14–0.35; c = 0.44, 95% CI = 0.39–0.50. Models that indicated an absence of emigration from the study site held slightly more support than models that indicated some temporary random emigration (Table 3). This finding suggests that there is a small rate of temporary emigration (model averaged: γ' = 0.10, 95% CI = 0.002–0.898, and γ'' = 0.10, 95% CI = 0.002–0.898). Discussion We found low to moderate annual survival for three species of permanent resident birds living in central Wisconsin: White-breasted Nuthatch, Downy Woodpecker, and Hairy Woodpecker. Average annual survival for White-breasted Nuthatches was only half that of Downy or Hairy Woodpecker survival, resulting in higher individual turnover. Although we were unable to provide estimates for each year, the model set indicated that there were yearly differences. High annual variation and low survival makes this study population susceptible to large fluctuations. Our model-averaged White-breasted Nuthatch apparent survival estimates are most similar to those obtained in a fragmented forest landscape in Ohio (0.27 ± 0.06; Doherty and Grubb 2002) and slightly lower than those obtained in Maryland (0.35 ± 0.01; Karr et al. 1990). We found similar apparent annual survival between Downy and Hairy Woodpeckers. Our small sample size may have prevented us from observing a noteworthy difference in survival between these species, but the current lack of data in the literature prevents us from substantiating whether or not we are missing a true difference. Apparent annual survival for these two species was moderate and consistent. Existing Downy Woodpecker apparent annual survival estimates differ, with the lowest estimate obtained in Ohio (0.26 ± 0.11; Doherty and Grubb 2002) and a higher one obtained in Maryland (0.64 ± 0.07; Karr et al. 1990). However, Doherty and Grubb (2002) observed a positive correlation between annual survival probability and woodlot size in Ohio, with their model-averaged estimate (0.26 ± 0.11) on the low end of the range of estimates. Apparent survival estimates from larger woodlots in Ohio were similar to our estimates from Wisconsin, where our study site was in a section of a large forest. We found multiple apparent survival estimates for Downy Woodpeckers and White-breasted Nuthatches, but the MAPS program was the only source we found for Hairy Woodpeckers. The MAPS program provides the following apparent annual survival estimates for the north-central region during 1989–2006: White-breasted Nuthatch survival = 0.526 ± 0.138, Downy Woodpecker survival = 0.393 ± 0.054, and Hairy Woodpecker survival = 0.552 ± 0.114 (Michel et al. 2006). These estimates are higher than ours for the White-breasted Nuthatch, slightly lower for the Downy Woodpecker, and similar for the Hairy Woodpecker. There are several potential explanations for these differences: the MAPS program has a larger spatial scale over which data were collected, it uses models that have been modified to reduce bias caused by the Northeastern Naturalist 502 J.A. Cava, J.D. Riddle, and R.P. Thiel 2014 Vol. 21, No. 3 inclusion of transient individuals in the dataset, and it did not consider temporary movement of individuals (Michel et al. 2006, Nott and DeSante 2000). Wisconsin is near the eastern edge of the MAPS north-central region, which extends west to central Montana, south to Nebraska, and north into Alberta and Saskatchewan (Michel et al. 2006). Annual survival is calculated from data collected across this entire region (Michel et al. 2006) and probably varies among areas and habitat types and quality when studied at a finer scale. Inclusion of transient individuals depresses survival estimates from unadjusted CJS and robust design analyses because these individuals are uncatchable after they leave the study area (Nott and DeSante 2000, Pradel at al. 1997). Our dataset may have included transient individuals of each species but their inclusion does not explain all the variation because not all of our estimates were lower than the MAPS estimates. The MAPS estimates also may have been impacted by the exclusion of temporary emigration in their models (Pollock et al. 1990). Our estimates may have been affected by small sample size, but our results are broadly consistent with results of studies with larger sample sizes (Doherty and Grubb 2002, Karr et al. 1990). Recapture-sample sizes for these species are also small in the MAPS north-central region study because, in order to exclude transients, they calculated survival inference based only on individuals that were recaptured within their initial year of capture (White-breasted Nuthatch: 8 recaptures of 150 captures, Downy Woodpecker: 58 of 583, Hairy Woodpecker: 13 of 107; Michel et al. 2006). In our study, recapture probability was higher than initial capture probability for all 3 species, demonstrating a trap-happy response when attracted with bait. This is the first study to demonstrate a trap response for any of these species. Failure to account for a trap-happy response when it exists can impact the estimates of other important demographic parameters. Studies that provide a single-capture probability are not directly comparable to our capture parameters (Doherty and Grubb 2002, Michel et al. 2006). The majority of individuals we captured consistently returned to traps, although there were several exceptions which may have been transients, individuals that became trap-shy, or birds whose home-range edge coincided with the edge of our trapping grid. Despite these exceptions, all capture probabilities per trapping session were high enough to suggest that we caught most of the birds available for capture each year. Downy Woodpeckers, Hairy Woodpeckers, and White-breasted Nuthatches are permanent residents across their ranges, but detailed movement information is currently lacking (Grubb and Pravosudov 2008, Jackson and Ouellet 2002, Jackson et al. 2002). White-breasted Nuthatches are known to engage in irruptive movements (Grubb and Pravosudov 2008), which may explain their absence on our study site in the winter of 2011; however, we cannot rule out complete mortality because none of the individuals on the study site in 2010 returned. Our data and models suggest that there is low to no temporary emigration occurring, although low sample sizes may account for a complete lack of temporary emigration observed in the woodpeckers and the difficulty in estimating it in the White-breasted Nuthatches. Indeed, the majority of individuals on our study site either (1) stayed on site for several seasons Northeastern Naturalist Vol. 21, No. 3 J.A. Cava, J.D. Riddle, and R.P. Thiel 2014 503 before disappearance, (2) were present for a single season, but were captured multiple times within the season, or (3) disappeared from the site after a single capture. These observations are not the result of temporary movement, but reflect mortality and permanent emigration. Estimating true survival requires distinction between the rates of mortality and permanent emigration in a population (Horton and Letcher 2008, Marshall et al. 2004). Methods that can distinguish these two processes often require collection of additional data on dispersal and mortality (Gilroy et al. 2012, Horton and Letcher 2008, Marshall et al. 2004). Prior knowledge of permanent emigration rates also can increase certainty in apparent survival estimates. Apparent survival is the product of true survival and site fidelity (Lebreton et al. 1992). If the site fidelity rate is 100% (i.e., permanent emigration = 0%), then true survival is equal to the apparent survival estimate. Therefore apparent survival estimates for populations displaying little to no permanent emigration should be closer to true survival than apparent survival estimates for populations with high rates of permanent emigration. We were unable to find information on the rate of permanent emigration for any of our study species. Telemetry or more extensive mark–recapture studies will be necessary to determine the extent of site fidelity in these species. Our results may be useful in planning future trapping studies for these birds. The capture probabilities we recorded indicate that baited, mesh-wire traps are effective for capture of wintering woodpeckers and White-breasted Nuthatches. We also found that relatively few, short trapping sessions were required to capture nearly all resident individuals in our study area. This method can be especially helpful in studies that require unique marks on as many birds as possible, including research concerning home ranges, behavioral interactions, and dispersal. However, obtaining large sample sizes will require significant investment in materials and time if this method is used. Acknowledgments We thank Dr. Jeremy Kirchman and two anonymous reviewers for their helpful comments on this manuscript. We thank Dr. Kenneth Pollock for a helpful review of a previous version of this manuscript. We also thank past project leaders A. Purdy, B. Sadler, B. Winter, K. Witkowski, R. Sheldon, E.E. Scherer, and all other student volunteers who assisted in data collection. We thank the staff at Sandhill Wildlife Area, Babcock, WI, and the Wisconsin Department of Natural Resources, Madison, WI, for use of their facilities and field site. The University of Wisconsin-Stevens Point Student Chapter of the Wildlife Society and Student Government Association provided logistical and financial support. Banding was conducted under permit number 21040 issued to R.P. Thiel. The University of Wisconsin-Stevens Point Institutional Animal Care and Use Committee approved trapping, handling, and marking protocols (protocol number 20011.11.12). We have no conflict of interest to declare. Literature Cited Butts, W.K. 1931. A study of the Chickadee and White-breasted Nuthatch by means of marked individuals. Part III: The White-breasted Nuthatch (Sitta carolinensis cookei). Bird-Banding 2:59–76. Northeastern Naturalist 504 J.A. Cava, J.D. Riddle, and R.P. Thiel 2014 Vol. 21, No. 3 Cormack, R.M. 1964. Estimates of survival from the sighting of marked animals. Biometrika 51:429–438. DeSante, D.F, D.R. O’Grady, K.M. Burton, P. Velez, D. Froehlich, E.E. Feuss, H. Smith, and E.D. Ruhlen. 1998. The Monitoring Avian Productivity And Survivorship (MAPS) program sixth and seventh annual report (1995 and 1996). Bird Populations 4:69–122. Doherty, P.F., and T.C. Grubb. 2002. Survivorship of permanent-resident birds in a fragmented forested landscape. Ecology 83:844–857. Fiske, J. 1968. Woodpecker trap. Eastern Bird Banding Association 31:154–155. Gilroy, J.J., T. Virzi, R.L. Boulton, and J.L. Lockwood. 2012. A new approach to the apparent survival problem: Estimating true survival from mark–recapture studies. Ecology 93(7):1509–1516. Grubb, T.C., Jr., and V.V. Pravosudov. 2008. White-breasted Nuthatch (Sitta carolinensis), In A.Poole (Ed.). The Birds of North America Online. Cornell Lab of Ornithology, Ithaca, NY. Available online at http://bna.birds.cornell.edu/bna/species/054. Accessed Accessed 10 March 2013. Horton, G.E., and B.H. Letcher. 2008. Movement patterns and study area boundaries: Influences on survival estimation in capture–mark–recapture studies. Oikos 117:1131–1142. Jackson, J.A., and H.R. Ouellet. 2002. Downy Woodpecker (Picoides pubescens), In A. Poole (Ed.). The Birds of North America Online. Cornell Lab of Ornithology, Ithaca, NY. Available online at http://bna.birds.cornell.edu/bna/species/613. Accessed 10 March 2013 Jackson, J.A., H.R. Ouellet, and B.J. Jackson. 2002. Hairy Woodpecker (Picoides villosus). In A. Poole (Ed.). The Birds of North America Online. Cornell Lab of Ornithology, Ithaca, NY. Available online at http://bna.birds.cornell.edu/bna/species/702. Accessed 10 March 2013. Jolly, G.M. 1965. Explicit estimates from capture–recapture data with both death and immigration– stochastic model. Biometrika 52:225–247. Karr, J.R., J.D. Nichols, M.K. Klimkiewicz, and J.D. Brawn. 1990. Survival rates of birds of tropical and temperate forests: Will the dogma survive? American Naturalist 136:277–291. Kellam, J.S. 2003. Downy Woodpecker pair-bond maintenance in winter: Proximate and ultimate mechanisms. Ph.D. Thesis. Purdue University, West Lafayette, IN. 196 pp. Kendall, W.L. 2001. The robust design for capture–recapture studies: Analysis using program MARK. Pp. 357–360, In R. Field, R.J. Warren, H. Okarma, and P.R. Sievert (Eds.). Wildlife, Land, and People: Priorities for the 21st Century. The Wildlife Society, Bethesda, MD. 437 pp. Kendall, W.L., and J.D. Nichols. 1995. On the use of secondary capture–recapture samples to estimate temporary emigration and breeding proportions. Journal of Applied Statistics 22:751–762. Kendall, W.L., J.D. Nichols, and J.E. Hines. 1997. Estimating temporary emigration using capture–recapture data with Pollock’s robust design. Ecology 78:563–578. Lebreton, J.D., K.P. Burnham, J. Clobert, and D.R. Anderson. 1992. Modeling survival and testing biological hypotheses using marked animals: A unified approach with case studies. Ecological Monographs 62:67–118. Marshall, M.R., D.R. Diefenbach, L.A. Wood, and R.J. Cooper. 2004. Annual survival estimation of migratory songbirds confounded by incomplete breeding-site fidelity: Study designs that may help. Animal Biodiversity and Conservation 27:59–72. Matthysen, E., and K. Schmidt. 1987. Natal dispersal in the Nuthatch. Ornis Scandinavica 18:313–316. Northeastern Naturalist Vol. 21, No. 3 J.A. Cava, J.D. Riddle, and R.P. Thiel 2014 505 Michel, N., D.F. DeSante, D.R. Kaschube, and M.P. Nott. 2006. The Monitoring Avian Productivity And Survivorship (MAPS) Program annual reports, 1989–2003. NBII/MAPS avian demographics query interface. Available online at http://www.birdpop.org/nbii/ NBIIHome.asp. Accessed December 2006. Nott, M.P., and D.F. DeSante. 2002. Demographic monitoring and the identification of transients in mark–recapture models. Pp. 727–736, In J.M. Scott, P.J. Heglund, M. Morrison, M. Raphael, J. Haufler, and B. Wall (Eds.). Predicting Species Occurrences: Issues of Scale and Accuracy. Island Press, Covello, CA. Pollock, K.H. 1982. A capture–recapture design robust to unequal probability of capture. Journal of Wildlife Management 46:757–760. Pradel, R., J. Hines, J.D. Lebreton, and J.D. Nichols. 1997. Estimating survival probabilities and proportions of transients using capture–recapture data. Biometrics 53:60–72. Pyle, P., S.N.G. Howard, R.P. Yunick, and D.F. DeSante. 1987. Identification Guide to North American Passerines. State Creek Press, Bolinas, CA. 732 pp. Seber, G.A.F. 1965. A note on the multiple recapture census. Biometrika 52:249–259. White, G.C., and K.P. Burnham. 1999. Program MARK: Survival estimation from populations of marked animals. Bird Study Supplement 46:120–138.