2009 SOUTHEASTERN NATURALIST 8(1):83–92 Recovery of Nesting Bald Eagles in Texas Sarah T. Saalfeld1,*, Warren C. Conway1, Ricky Maxey2, Chris Gregory3, and Brent Ortego4 Abstract - One of the most successful conservation stories in United States’ history resulted in Haliaeetus leucocephalus (Bald Eagle) being removed recently from the federal endangered species list. Few studies, however, have documented regional long-term recovery trends for Bald Eagles. We quantified Bald Eagle nesting density, distribution, and productivity trends by using aerial surveys of nests located in 69 counties in eastern Texas from 1971–2005. The total number of occupied nests, productive nests, and offspring produced increased exponentially during that time, with the most dramatic increases occurring from 1995–2005. Since 1971, the total number of occupied nests increased 13% per year, from 5 in 1971 to 157 in 2005, and the total number of young produced also increased 13% per year, from 6 in 1971 to 195 in 2005. Apparent nest success estimates (50–100%) and mean brood size (1–2 young/nest), however, remained relatively consistent from 1971 to 2005. By 1989, Bald Eagles in Texas exceeded recovery goals set by the Southeastern States Bald Eagle Recovery Team (i.e., >0.9 young produced/occupied nest, >1.5 young produced/successful nest, and >50% of nests successful in raising at least 1 young). Continued regional increases in Bald Eagle nesting activity may warrant specific attention, particularly as related to anthropogenic pressures and interactions in increasingly fragmented and urbanized habitats. Introduction Haliaeetus leucocephalus Linnaeus (Bald Eagle) populations have increased dramatically after being placed on the endangered species list in 1973 (Buehler 2000). Populations have increased and expanded throughout the United States, with nesting attempts documented in all of the lower 48 states (US Fish and Wildlife Service 2007). In Texas, nesting Bald Eagles have increased since the early 1970s, when only 8 nests were documented from the lower reaches of rivers within coastal areas (Ortego 2005). Bald Eagles in Texas have recently reoccupied portions of their assumed historical range, now nesting in high densities on tall trees near rivers within the coastal region and near inland reservoirs (Ortego et al. 2006) where nesting had not been previously documented (Campbell 1995). Although the federal government removed the Bald Eagle from the endangered species list (US Fish and Wildlife Service 2007), it remains on the threatened list of Texas, and still receives federal protection from the 1940 Bald Eagle and Golden Eagle Protection Act and the Migratory Bird Treaty Act of 1918. Therefore, continued monitoring of long-term changes in density, 1Arthur Temple College of Forestry and Agriculture, Stephen F. Austin State University, Nacogdoches, TX 75962. 2Texas Parks and Wildlife Department, Karnack, TX 75661. 3Texas Parks and Wildlife Department, Livingston, TX 77351. 4Texas Parks and Wildlife Department, Victoria, TX 77901. *Corresponding author - saalfeldst@ titan.sfasu.edu. 84 Southeastern Naturalist Vol. 8, No. 1 distribution, and productivity is important to evaluate current status as related to recovery plans. According to the Southeastern States Bald Eagle Recovery Team (US Fish and Wildlife Service 1989), along with documentation of population vigor and supporting habitats, recovery within southeastern states, including Texas, was defined to occur when the most recent 3 years of data document >600 occupied breeding areas distributed in >75% of the historical range within southeastern United States, ≥0.9 young produced per occupied nest, ≥1.5 young produced per successful nest, and ≥50% of nests successfully fl edged ≥1 young. By the late 1980s and early 1990s, it appeared that Bald Eagles attained recovery in eastern Texas, but no comprehensive nor quantitative formal documentation of this recovery exists. Therefore, our objectives were to quantify and examine recovery trends of nesting Bald Eagles throughout Texas and to determine how current nesting activity and productivity correspond to regional recovery goals. Methods Personnel of Texas Parks and Wildlife Department collected aerial survey data for Bald Eagle nests throughout eastern Texas from 1971–2005 (Ortego et al. 2006). We conducted surveys annually from February–April to monitor known nest locations and to locate new nests. Additionally, we located nests reported from private individuals to verify existence and incorporate into future surveys. We classified nests as occupied/unoccupied by presence/ absence of adult eagles near nests and/or presence of eggs and/or young in nest. We revisited occupied nests observed prior to hatching to determine productivity (i.e., number of eaglets observed within each occupied nest) in April and May. We also aerially surveyed territories in which an active nest was not located during the February–April survey to locate missing nests in April and May. We defined a productive nest as one in which at least 1 eaglet was observed during either survey. We determined nesting success using the ratio of productive nests to occupied nests, where productivity was defined as the mean number of young produced per occupied nest, and mean brood size was defined as the mean number of young produced per productive nest. Calculations of nesting success, productivity, and mean brood size included only nests in which productivity was determined. We used simple linear regression (PROC REG; SAS Institute 1999) to examine relationship(s) between year and total number of (1) occupied nests, (2) productive nests, and (3) young produced, as well as (4) nest success, (5) productivity, and (6) mean brood size. Once we determined that the total number of (1) occupied nests, (2) productive nests, and (3) young produced was exponentially related to year (i.e., highest R2 value), we transformed these metrics using a natural logarithmic (ln) transformation prior to performing linear regression. However, all untransformed values are presented. We also used correlation analysis (PROC CORR; SAS Institute 1999) to examine the relationship between total number of young produced and productive nests observed each year, as well as between total number of occupied nests, productive nests and young produced. Finally, to quantify 2009 S.T. Saalfeld, W.C. Conway, R. Maxey, C. Gregory, and B. Ortego 85 change in rates of increase in total number of (1) occupied nests, (2) productive nests, and (3) young produced per year, we used simple linear regression (PROC REG; SAS Institute 1999) to fit a linear model to untransformed data during 3 predefined time periods: 1971–1982 (early), 1983–1994 (middle), and 1995–2005 (late). We then used analysis of variance (ANOVA; SAS Institute 1999) to compare regression slope coefficients between early, middle, and late periods for each nesting parameter. Some biases could occur when using aerial survey data (Nesbitt et al. 1998). Specifically, underestimation of occupancy and production may occur if nests could not be relocated or productivity determined, either within or among years. To estimate this potential bias, we determined percentage of (1) known nests not relocated during aerial surveys in subsequent years and (2) known occupied nests in which productivity could not be determined within the same year. Although difficult to quantify, underestimates of nesting activity and productivity may also occur if eagles abandoned nests or nestlings left nests prior to observation(s). However, these estimates are assumed to be consistent between and among years. Additionally, increases in nesting activity observed could be a result of increased knowledge of known territories among years and increased reporting of nests from the public. Results From 1971–2005, we made 2211 observations of 482 Bald Eagle nests located in 248 individual breeding territories. Bald Eagle nests were found in 69 counties within the Coastal Prairies, Edwards Plateau, Oaks and Prairies, Osage Plains, Pecos and Staked Plains, Rolling Red Plains, and West Gulf Coastal Plain physiographic regions (modified from Texas Parks and Wildlife Department [2007]) of Texas (Fig. 1). Bald Eagle nesting activity (i.e., total number of occupied nests: R2 = 0.996, productive nests: R2 = 0.993, and total number of young produced: R2 = 0.990) increased exponentially from 1971–2005 (Appendix 1, Fig. 2). Year was positively associated with and accounted for significant variation in the total number of occupied nests, productive nests, and young produced (Table 1). However, year was not associated with nest success, productivity, nor mean brood size (Table 1), indicating that individual nest metrics have remained fairly constant in time. As expected, the total number of young produced each year from all Bald Eagle nests throughout Texas was positively correlated with the total number of productive nests observed each year (R = 0.998, P < 0.001). Additionally, the total number of occupied nests throughout Texas was positively correlated with total number of productive nests (R = 0.997, P < 0.001) and total number of young produced (R = 0.994, P < 0.001) each year. Specifically, the total number of occupied nests (annual rate = 0.127 ± 0.004), productive nests (annual rate = 0.127 ± 0.004), and young produced (annual rate = 0.130 ± 0.005) increased each year from 1971–2005 (Table 1 and Appendix 1). For all metrics tested, regression slopes differed (P < 0.001) among time periods (early: 1971–1982, middle: 1983–1994, and late: 1995–2005). Specifically, regression slopes were greater (P < 0.001) for 86 Southeastern Naturalist Vol. 8, No. 1 Table 1. Simple linear regression models comparing year to measures of Bald Eagle nesting activity and productivity collected from aerial surveys of nests in Texas, 1971–2005. Intercept Year Dependent variable R2 P-value Estimate SE P-value EstimateA SE P-value Ln (occupied nests) 0.96 <0.01 -250.06 8.59 <0.01 0.13 0.00 <0.01 Ln (productive nests) 0.96 <0.01 -250.26 8.49 <0.01 0.13 0.00 <0.01 Ln (young produced) 0.96 <0.01 -255.14 9.16 <0.01 0.13 0.01 <0.01 Nesting successB 0.01 0.55 3.66 4.71 0.44 0.00 0.00 0.55 ProductivityC <0.01 0.95 0.61 9.76 0.95 0.00 0.01 0.95 Mean brood sizeD 0.02 0.44 -3.76 6.86 0.59 0.00 0.00 0.44 AAnnual percent increase in dependent variable. BNesting success was defined as the ratio of productive nests to occupied nests. CProductivity was defined as the mean number of young produced per occupied nest. DMean brood size was defined as the mean number of young produced per productive nest. Figure 1. Location and number of Bald Eagle nesting territories per Texas physiographic region (modified from Texas Parks and Wildlife [2007]) located from aerial surveys of nests, 1971–2005. Shaded areas correspond to counties in which ≥1 nest was found. 2009 S.T. Saalfeld, W.C. Conway, R. Maxey, C. Gregory, and B. Ortego 87 the late period than the early and middle periods and for the middle period when compared to the early period (total number of occupied nests: early rate = 0.269, middle rate = 3.210, late rate = 11.809, F2,29 = 223.99; productive nests: early rate = 0.224, middle rate = 2.587, late rate = 9.064, F2,29 = 194.59; and young produced: early rate = 0.413, middle rate = 4.066, late rate = 15.945, F2,29 = 111.05; Fig. 2), indicating that rates of these parameters have increased more dramatically during the last 11 years. The percentage of known nests not relocated during aerial surveys in subsequent years for all years combined was 4.4%. Similarly, the percentage of known occupied nests in which productivity could not be determined within the same year for all years combined was 9.5%. Figure 2. Total number of occupied nests (A), productive nests (B), and young produced (C) for Bald Eagles located from aerial surveys of nests in Texas, 1971–2005, as well as fitted exponential models and rates of increase from 1971–1982, 1983– 1994, and 1995–2005. 88 Southeastern Naturalist Vol. 8, No. 1 Discussion Nesting activity (i.e., number of occupied and productive nests and young produced) of Bald Eagles in Texas increased exponentially between 1971 and 2005 with the most dramatic increases occurring within the last 11 years (i.e., 1995–2005; see Fig. 2). Individual productivity rates (i.e., nesting success, productivity, and mean brood size) remained consistent throughout this study. One of the main causes attributed to Bald Eagle declines in the mid-1900s was pesticide use, mainly DDT, causing egg shell thinning and ultimately decreasing productivity (Buehler 2000). Since the ban on use of DDT in 1972, recovery of Bald Eagle hatching rates and productivity has occurred throughout most of North America (Buehler 2000, Campbell 1995, Grier 1982), although at differing rates regionally (see Buehler 2000, Dykstra et al. 2005). In Texas, Bald Eagles may (1) have recovered from the effects of DDT prior to the time this study was initiated (i.e., 1971) or (2) not have been as severely impacted by DDT as in other regions as indicated by high productivity rates observed in this study (i.e., range 1.20–1.45 young/occupied nest from 1971–2005; mean for all years = 1.27 young/occupied nest). As suggested by Buehler et al. (1991), productivity rates >0.74 young/occupied nest are indicative of an expanding population. Therefore, we hypothesize that the main cause for Bald Eagle recovery post-DDT in Texas can be attributed to increased population growth from consistently, high productivity rather than increased individual nesting success and/or productivity. Moreover, high fl edgling survival rates and recruitment, strong fidelity to natal nesting areas for breeding, and creation of suitable nesting habitat (i.e., man-made inland reservoirs) may have also facilitated increased population growth and post-DDT recovery of Bald Eagles in Texas. In a previous study on Bald Eagles in Texas conducted from 1985–1991, Mabie et al. (1994) estimated a fl edgling survival rate of 97%, a recruitment rate of 33.3%, and a fidelity rate to natal nesting areas for breeding of 20%, all of which may positively affect population growth. In addition, creation of additional suitable nesting habitat in the form of inland reservoirs (see below) also occurred during the same time period (i.e., 1960s–1970s) as the ban on use of DDT. By increasing suitable nesting habitat, the Bald Eagle population of Texas was able to expand and grow without habitat limitations. Although not directly comparable because of regional area dissimilarities (Coastal Prairies = 4,658,340 ha; Oaks and Prairies = 10,738,000 ha; West Gulf Coastal Plain = 6,481,950 ha), we found differing recovery trends throughout the 3 physiographic regions where we located the majority of eagle nests (i.e., Coastal Prairies, Oaks and Prairies, West Gulf Coastal Plain; see Fig. 1). Within the West Gulf Coastal Plain region, we found a greater increase in the number of occupied Bald Eagle nests (1 to 75 from 1986–2004) than the other 2 physiographic regions (Coastal Prairies = 5 to 40 from 1971–2005, Oaks and Prairies = 1 to 40 from 1980–2005; Fig. 3), which ultimately resulted in more Bald Eagle nests within the West Gulf Coastal Plain region by 2005, despite a longer history of nesting within the Coastal Prairies and Oaks and Prairies regions. One possible explanation may be that these differences are caused by habitat availability for nesting Bald Eagles within each physiographic region. Bald Eagles 2009 S.T. Saalfeld, W.C. Conway, R. Maxey, C. Gregory, and B. Ortego 89 within all 3 physiographic regions nest within suitable riparian habitat adjacent to rivers and creeks; however, man-made inland reservoirs host the majority of nesting eagles within the Oaks and Prairies and West Gulf Coastal Plain regions (B. Ortego, unpubl. data). Essentially, these reservoirs, along with the maturation of forested areas surrounding them, create additional nesting habitat within these regions where nesting has not been previously documented (Campbell 1995). Additionally, the West Gulf Coastal Plain region is the only region with large tracts of public land (i.e., National Forest Lands) managed for nesting Bald Eagles. Essentially, these areas provide the structural habitat necessary for nest placement near quality foraging grounds and protection from human disturbance. As Bald Eagle populations increased from 1990–2005, nesting expanded throughout the Coastal Prairies region and into the Oaks and Prairies and West Gulf Coastal Plain regions where suitable habitat was found in forested areas near rivers, creeks, and man-made inland reservoirs. Figure 3. Total number of occupied nests (A), productive nests (B), and young produced (C) for Bald Eagles for 3 physiographic regions (Coastal Prairies, Oaks and Prairies, and West Gulf Coastal Plain) located from aerial surveys of nests in Texas, 1971–2005. 90 Southeastern Naturalist Vol. 8, No. 1 Nesting Bald Eagles in Texas have exceeded productivity goals (i.e., >0.9 young/occupied nest, >1.5 young/productive nest, and >50% of nests successful in raising ≥1 young) set by the Southeastern States Bald Eagle Recovery Team (US Fish and Wildlife Service 1989) since 1989 (Appendix 1). Similar results were also observed in Florida (Nesbitt et al. 1998) and Oklahoma (Jenkins and Sherrod 2005), where increasing nesting numbers of Bald Eagles have occurred since the mid-1980s and early 1990s. In Oklahoma, it was estimated that productivity has reached and/or exceeded recovery goals set by the Northern States Bald Eagle Recovery Plan (>1 young fl edged per productive nest) since 1991 (Jenkins and Sherrod 2005). Despite their rapid population increase in eastern Texas, specific threats to current populations still exist. Although additional suitable nesting habitat has been created in the form of inland reservoirs and maturation of forested lands surrounding a few of these reservoirs, disease, habitat loss, disturbance, and human tolerance are still major concerns for nesting populations (see Buehler 2000, US Fish and Wildlife Service 1989). Currently, nesting Bald Eagles in Texas appear to have an adequate amount of habitat for nesting as indicated by the exponential growth in the total number of occupied nests and consistent productivity. However, if sufficient nesting locations become limiting, the number of nesting pairs and young produced may plateau at some point in the near future. Two major habitat requirements necessary for nesting by Bald Eagles are suitable nest trees near productive foraging areas (i.e., large bodies of water and/or permanent fl owing creeks and rivers; Garrett et al. 1993, Livingston et al. 1990, Murphy 1965) and freedom from human disturbance (Andrew and Mosher 1982, Garrett et al. 1993, Murphy 1965, Thompson and McGarigal 2002). However, because of increased recreation and development pressure, these habitat characteristics are becoming increasingly rare not only in Texas, but throughout the lower 48 states (McGarigal et al. 1991). To continue protecting this vulnerable species and allow nesting populations to increase further, specific measures should be taken to identify, conserve, and limit human disturbance around potential nesting habitat throughout Texas. Some potential nesting habitat is likely to be located on private land; therefore, it will also be important to continue public education about the needs of eagles and the importance of conserving nesting habitat regardless of ownership. Acknowledgments We thank all of the individuals who assisted with data collection from Texas Parks and Wildlife Department and US Forest Service during the time this research was conducted. We also thank John Steele for initial efforts on this research, as well as Tawnya Brummett, Ron Mize, Patricia Johnson, Todd Johnson, and Bill Bartush from the US Forest Service, Jeffrey Reid from the US Fish Wildlife Service, and technicians Thad Choate and Kellon Harris, who assisted with this project. We also thank the landowners who contributed information on nesting eagles and who continue to protect and manage their land for nesting eagles. Fred LeBlanc of The Woodlands Operating Corporation, the Arthur Temple College of Forestry and Agriculture at Stephen F. Austin State University, the Texas Parks and Wildlife Department, the US Forest Service, and the US Fish and Wildlife Service provided financial and logistical support for this research. 2009 S.T. Saalfeld, W.C. Conway, R. Maxey, C. Gregory, and B. Ortego 91 Literature Cited Andrew, J.M., and J.A. Mosher. 1982. Bald Eagle nest-site selection and nesting habitat in Maryland. Journal of Wildlife Management 46:382–390. Buehler, D.A. 2000. Bald Eagle (Haliaeetus leucocephalus). In A. Poole and F. Gill (Eds.). The Birds of North America. Philadelphia, No. 506. The Birds of North America, Inc., Philadelphia, PA. Buehler, D.A., J.D. Fraser, J.K.D. Seegar, G.D. Therres, and M.A. Byrd. 1991. Survival rates and population dynamics of Bald Eagles on Chesapeake Bay. Journal of Wildlife Management 55:608–613. Campbell, L. 1995. Endangered and Threatened Animals of Texas: Their Life History and Management. Texas Parks and Wildlife Department, Austin, TX. 130 pp. Dykstra, C.R., M.W. Meyer, S. Postupalsky, K.L. Stromborg, D.K. Warnke, and R.G. Eckstein. 2005. Bald Eagles of Lake Michigan: Ecology and contaminants. Pp. 365–392, In T. Edsall and M. Munawar (Eds.). State of Lake Michigan: Ecology, Health, and Management. Ecovision World Monograph Series. Backhuys Publishers, Leiden, The Netherlands. Garrett, M.G., J.W. Watson, and R.G. Anthony. 1993. Bald Eagle home range and habitat use in the Columbia River Estuary. Journal of Wildlife Management 57:19–27. Grier, J.W. 1982. Ban of DDT and subsequent recovery of reproduction in Bald Eagles. Science 218:1232–1235. Jenkins, M.A., and S.K. Sherrod. 2005. Growth and recovery of the Bald Eagle population in Oklahoma. Wildlife Society Bulletin 33:810–813. Livingston, S.A., C.S. Todd, W.B. Krohn, and R.B. Owen, Jr. 1990. Habitat models for nesting Bald Eagles in Maine. Journal of Wildlife Management 54:644–653. Mabie, D.W., M.T. Merendino, and D.H. Reid. 1994. Dispersal of Bald Eagles fl edged in Texas. Journal of Raptor Research 28:213–219. McGarigal, K., R.G. Anthony, and F.B. Issacs. 1991. Interactions of humans and Bald Eagles on the Columbia River estuary. Wildlife Monographs 115:1–47. Murphy, J.R. 1965. Nest-site selection by the Bald Eagle in Yellowstone National Park. Proceedings of the Utah Academy of Science 42:261–264. Nesbitt, S.A., M.A. Jenkins, S.K. Sherrod, D.A. Wood, A. Beske, J.H. White, P.A. Schulz, and S.T. Schwikert. 1998. Recent status of Florida’s Bald Eagle population and its role in eagle reestablishment efforts in the southeastern United States. Proceedings of the Annual Conference Southeastern Association of Fish and Wildlife Agencies 52:377–383. Ortego, B. 2005. Bald Eagle nest survey and management. Performance report, Federal Aid Grant No. T-1, Texas Parks and Wildlife Department, Austin, TX. Ortego, B., C. Gregory, and D. Schmidt. 2006. Bald Eagles in Texas. Texas Birds Annual 2:11–16. SAS Institute. 1999. SAS/STAT Software, Version 9. SAS Institute Inc., Cary, NC. Texas Parks and Wildlife. 2007. Texas partners in fl ight: Ecoregional bird checklist. Available online at http://www.tpwd.state.tx.us/huntwild/wild/birding/pif/ecoregions/. Accessed 1 January 2008. Thompson, C.M., and K. McGarigal. 2002. The infl uence of research scale on Bald Eagle habitat selection along the lower Hudson River, New York (USA). Landscape Ecology 17:569–586. US Fish and Wildlife Service. 1989. Southeastern states Bald Eagle recovery plan. Southeastern States Bald Eagle Recovery Team, US Fish and Wildlife Service, Atlanta, GA. US Fish and Wildlife Service. 2007. Endangered and threatened wildlife and plants: Removing the Bald Eagle in the lower 48 states from the list of endangered and threatened wildlife. Federal Register 72:37346–37372. 92 Southeastern Naturalist Vol. 8, No. 1 Appendix 1. Year, nesting activity, and productivity of Bald Eagles collected from aerial surveys of nests in Texas, 1971–2005. Mean Occupied Productive Young Nesting brood Year nests nests produced successA ProductivityB sizeC 1971 5 3 6 0.60 1.20 2.00 1972 3 3 3 1.00 1.00 1.00 1973 D 4 2 3 0.67 1.00 1.50 1974 3 3 2 0.67 0.67 1.00 1975 3 2 5 1.00 1.67 1.67 1976 2 3 3 1.00 1.50 1.50 1977 D 4 2 5 1.00 1.67 1.67 1978 D 4 3 5 1.00 1.67 1.67 1979 D 4 3 5 1.00 1.67 1.67 1980 D 5 3 6 1.00 2.00 2.00 1981 6 5 7 0.83 1.17 1.40 1982 D 8 6 10 0.86 1.43 1.67 Mean, 1971–1982 4.25 3.17 5.00 0.89 1.39 1.56 1983 8 4 7 0.50 0.88 1.75 1984 11 9 15 0.82 1.36 1.67 1985 15 8 10 0.53 0.67 1.25 1986 17 15 24 0.88 1.41 1.60 1987 D 19 12 22 0.67 1.22 1.83 1988 D 21 15 22 0.75 1.10 1.47 1989 D 26 19 29 0.76 1.16 1.53 1990 27 19 29 0.70 1.07 1.53 1991 36 24 38 0.67 1.06 1.58 1992 D 39 27 41 0.71 1.08 1.52 1993 D 39 32 53 0.84 1.39 1.66 1994 D 42 34 53 0.85 1.33 1.56 Mean, 1983–1994 25.00 18.17 28.58 0.72 1.14 1.58 1995 42 30 50 0.71 1.19 1.67 1996 49 33 51 0.67 1.04 1.55 1997 D 55 36 55 0.75 1.15 1.53 1998 D 64 47 71 0.81 1.22 1.51 1999 D 66 48 73 0.75 1.14 1.52 2000 D 81 63 99 0.82 1.29 1.57 2001 D 104 70 107 0.72 1.10 1.53 2002D 115 92 151 0.89 1.47 1.64 2003 D 123 93 148 0.96 1.53 1.59 2004 D 144 109 190 0.89 1.56 1.74 2005 D 157 112 195 0.82 1.45 1.74 Mean, 1995–2005 90.91 66.64 108.18 0.80 1.28 1.60 Mean, all years 38.60 28.26 45.51 0.80 1.27 1.58 ANesting success was defined as the ratio of productive nests to occupied nests. BProductivity was defined as the mean number of young produced per occupied nest. CMean brood size was defined as the mean number of young produced per productive nest. DNot all productivity outcomes were determined for this year. Nesting success, productivity, and mean brood size include only nests with known productivity.