Northeastern Naturalist 20 A.E. Stupik, T. Sayers, M. Huang, T.A.G. Rittenhouse, and C.D. Rittenhouse 22001155 NORTHEASTERN NATURALIST V2o2l.( 12)2:,2 N0–o3. 11 Survival and Movements of Post-fledging American Kestrels Hatched from Nest Boxes Annie E. Stupik1, Tom Sayers2, Min Huang3, Tracy A.G. Rittenhouse1, and Chadwick D. Rittenhouse1,* Abstract - Population declines of cavity-nesting birds are being recorded worldwide with degraded habitat, reduced prey availability, and limited nest-site availability implicated as potential causal factors for declines. A possible aid for avian species limited by breedingsite availability is the construction of nest boxes. Although nest boxes are commonly used for Falco sparverius (American Kestrels), little is known about the survival of young from fledging to the onset of migration. Using radiotelemetry, we recorded post-fledging survival and movement of 11 juvenile American Kestrels in northeastern Connecticut from June to September 2013. We used the Kaplan-Meier procedure, adjusted for staggered entry of individuals over time, to estimate daily survival at 0.270 (95% CI 0.01, 0.53) at the onset of migration. Causes of mortality included predation (n = 4 American Kestrels), exposure (n = 2), and unknown (n = 1). During the post-fledging period, the farthest net distance that we recorded for an American Kestrel from its natal box was 16.1 km. The 3 American Kestrels that we tracked from the nest box to the onset of migration demonstrated different patterns: one made a sudden, long-distance movement to a site 8 km away from the nest box; one undertook a series of 1–5-km movements away from the nest box and eventually settled in an area 2 km from the nest box; and one made consistent movements between the nest-box area and sites 1–5 km away. Our results indicate that although many young American Kestrels die within the first month of fledging, those that survive make pre-migratory movements up to 16 km from their nest box. Extending conservation and management efforts beyond the nest-box area may be an important step towards maintaining American Kestrel populations. Introduction Unexplained population declines of cavity-nesting birds are being recorded worldwide (Catry et al. 2013, Hipkiss et al. 2013). Potential causes of population declines include habitat degradation, reduced prey availability, and limited accessibility to nesting sites (Newton 1994). A possible solution to the problem of species population-loss and limited breeding-site availability is the construction of nest boxes. Nest boxes play an important role in the conservation of bird species worldwide (Lambrechts et al. 2011). However, some species continue to exhibit population declines even when nest boxes are provided and occupied (Hipkiss et al. 2013, Robillard et al. 2013). Why are cavity-nesting species declining even when their reproductive output is increased? Understanding what happens to 1Wildlife and Fisheries Conservation Center, Department of Natural Resources and the Environment, University of Connecticut, 1376 Storrs Road, Unit 4087, Storrs, CT 06269-4087. 229 Mihaliak Road, Tolland, CT 06084. 3DEEP–Wildlife Division, 391 Route 32, North Franklin, CT 06254. *Corresponding author - chadwick.rittenhouse@uconn.edu. Manuscript Editor: Jean-Pierre Savard Northeastern Naturalist Vol. 22, No. 1 A.E. Stupik, T. Sayers, M. Huang, T.A.G. Rittenhouse, and C.D. Rittenhouse 2015 21 cavity-nesting birds once they fledge from a nest box is important for developing a greater understanding of population declines. Falco sparverius L. (American Kestrel), a small grassland falcon native to North America, is a cavity-nesting species exhibiting population declines, especially in the northeastern US (Farmer and Smith 2009, Smallwood et al. 2009). Reproducing populations still exist in Connecticut, but the species is listed as threatened due to lack of knowledge and perceived population declines (CT DEP 2010). Adult American Kestrels breed within the state, and then migrate to winter grounds in the southern US and Central America. American Kestrel populations benefit from the construction of nest boxes (Katzner et al. 2005, Smallwood and Collopy 2009) because their availability improves nestling survival (90%) to fledging (Smallwood and Bird 2002). However, the nestling stage represents a small portion of an American Kestrel’s life, and understanding factors affecting juvenile and adult life-stages is also necessary for developing effective conservation and management actions. In comparison to high nestling and adult survival, juvenile survival may be limiting in American Kestrel populations. Estimates of adult survival from banding returns include 57% (Roest 1957), 45% (Henny 1972), and 75% (Hinnebusch et al. 2010). Estimates of juvenile survival to dispersal for American Kestrels are highly variable (e.g., 40% and 92% annually in the same study; Varland et al. 1993) due in part to defining natal dispersal as the first instance of leaving the natal area. A radiotelemetry study in Pennsylvania reported that winter survival was 61% (Farmer et al. 2006). Apparent low juvenile survival may be due to low survival during the post-fledging, fall migration, over-wintering, or spring migration periods, or may be due to dispersal outside of monitored areas. The movements of American Kestrels that have fledged from their nest box but not yet migrated have not been well studied, primarily due to the difficulty in documenting initial dispersal movements by fledglings. Re-sighting of juveniles in the following year is one method of estimating natal dispersal distance, defined as the net movement of an individual from its place of birth to a prospective breeding site the following season (Miller and Smallwood 1997). We also know from re-sightings that adult American Kestrels exhibit low nest-box fidelity. One study found that only 23% of males and 13% of females occupied the same nest box in consecutive years (Steenhof and Peterson 2009). However, although they may not often use the same box, non-migratory populations of American Kestrels breed within 8 km of their natal box (Miller and Smallwood 1997). The objectives of our study were to quantify survival and movement patterns of juvenile American Kestrels during the post-fledging period. Based on existing literature, we hypothesized that post-fledging American Kestrel survival would be lower than adult survival (Henny 1972, Hinnebusch et al. 2010, Roest 1957, Varland et al. 1993). Results from previous radio-tracking efforts (T. Sayers and M. Huang, unpubl. data) led us to hypothesize that post-fledging American Kestrels would remain within 1 km of their natal box for approximately 14 days before making a single large movement out of the natal area. Northeastern Naturalist 22 A.E. Stupik, T. Sayers, M. Huang, T.A.G. Rittenhouse, and C.D. Rittenhouse 2015 Vol. 22, No. 1 Field Site Description Our study area was Tolland and Hartford counties in Connecticut (Fig. 1). All American Kestrel nest boxes, except one were located on active farms, including Christmas tree farms, cornfields, hayfields, cow pasture, and goat pasture. During the study, a solar-panel array was installed in one of the hayfields. One of our nest boxes was on a grassland that was used for recreation. All nest-box locations included perches—Christmas trees, snags, or fence posts—and a field area characterized by a wide variety of grass and flower species inhabited by insects. We used wooden nest boxes with exterior dimensions of 30.5 × 45.7 cm and interior dimensions of 22.9 × 22.9 cm. The entrance hole was 7.6 cm in diameter. We added pine shavings to each next box to a depth of ~10 cm and mounted boxes 3.6–4 m off the ground, with the entrance hole oriented east–southeast to buffer against spring northerly and westerly winds. We monitored each nest box approximately once per week to determine number of eggs, number of hatchlings, and age of the nestlings. We studied only birds that nested in nest boxes. We do not know if the adult American Kestrels in our study were the same individuals that inhabited the nest box during the previous year. We banded nestling birds at ~14–18 days old, and deployed the radio-transmitters on 1 nestling per nest box at ~28–30 days old. Methods Materials We conducted the radio-telemetry portion of this study in the summer of 2013, beginning in June and ending in September. We attached BD2 transmitters (Holohil Industries, ON, Canada) to 11 nestling American Kestrels. The transmitters were 1.9 cm × 0.95 cm wide, and weighed 1.9 g. The average mass of a nestling was 125 g (T. Sayers, unpubl. data), thus the transmitters were approximately 1.52% of the American Kestrel’s mass. We used a figure-eight backpack-design harness system (Buehler et al. 1995) to secure the transmitter on the bird’s back, but with Figure 1. Map of study area in Tolland and Hartford counties, CT. Northeastern Naturalist Vol. 22, No. 1 A.E. Stupik, T. Sayers, M. Huang, T.A.G. Rittenhouse, and C.D. Rittenhouse 2015 23 the harness anchored to the legs rather than the wings. We used 0.1-in-wide Teflon tape as the harness material. The width of the loop depended on the size of the bird, and ranged from 7.0–8.0 cm long. We used homing (Mech 1983), triangulation (Cochran and Lords 1963, Verts 1963), and systematic search methods to locate American Kestrels daily from the time of radio-transmitter deployment until either death or the initiation of migration. For homing and triangulation methods, we used a handheld Yagi antenna paired with a R-1000 Telemetry Receiver (Communications Specialists, Inc., Orange, CA). For systematic searches, we used a vehicle-mounted whip antenna and drove 1-minute latitude by 1-minute longitude grids centered on the last known American Kestrel location. Survival analysis We estimated survival using the Kaplan-Meier procedure (Kaplan and Meier 1958), adjusted for staggered entry because we deployed transmitters over time (Pollock et al. 1989). We did an analysis based on Julian days and an analysis based on American Kestrel age and found no difference between these analyses. The staggered-entry, daily-survival estimate required daily totals of the number of American Kestrels at risk and the number of American Kestrel deaths. We assigned death events to the first day after a confirmed sighting of a live bird (n = 4 American Kestrels) or the last day of known movement when we were unable to visually observe the bird (n = 3). We considered all live American Kestrels at risk until death occurred or the birds moved from the study and search areas. We censored one American Kestrel for which the fate was not known. Assignment as censored had no impact on the survival estimate because censoring a bird from the data is equivalent to a death (Pollock et al. 1989). We calculated survival as: St = 1 – (Nd / Nr), where St is survival at time t, Nd is number of deaths at time t, and Nr is number at risk at time t. Once we had this initial survival value, we calculated subsequent daily-survival rates factoring in survival on the previous day as: St = St-1 * (1 – Nd / Nr) Using the above method, we estimated survival for all American Kestrels, individuals that fledged early in the season (prior to 7 July), and those that fledged late in the season. We tested for differences in survival among early vs. late-fledging American Kestrels using a log-rank test, which provided a chi-square statistic with 1 degree of freedom (Cox and Oakes 1984). Movement analysis We calculated 3 metrics to describe American Kestrel movements from the natal nest box. We defined the minimum distance moved (dist_move) as the Euclidean distance between 2 consecutive relocations for an individual American Kestrel, the total distance (total_dist) as the sum of all dist_moves(s) for an individual, and the net distance (net_dist) as the straight-line distance between the nest box Northeastern Naturalist 24 A.E. Stupik, T. Sayers, M. Huang, T.A.G. Rittenhouse, and C.D. Rittenhouse 2015 Vol. 22, No. 1 and last telemetry location for each bird (Turchin 1998). We used LOAS version 4.0 (Ecological Software Solutions, LLC 2013) to estimate point locations from triangulated locations and the Geospatial Modeling Environment version 7.2 (Beyer 2012) to calculate movement metrics. We log-transformed the movement metrics of minimum distance moved, total distance, and net distance moved to meet the assumption of normality, and performed an ANOVA using the R language and environment for statistical analyses (version 2.15.2; R Core Development Team 2012). We also used daily distance moved as the response variable in mixed-effects models to account for fixed effects of fledgling age (in days) and a random effect for each American Kestrel. We structured the random effect such that each bird had a unique intercept and unique slope. We fitted all models with the lmer function in the lme4 package for the R language and environment for statistical analyses (version 2.15.2; R Core Development Team 2012). Results We tracked 11 American Kestrels (8 females and 3 males) until they flew out of range, the transmitter battery died, or they died. Seven tracked American Kestrels (6 females and 1 male) died during the course of the study (Table 1). We censured one additional American Kestrel from the data set at its last known location on 22 July 2014 because its signal vanished only 12 days after leaving the nest box or ~2 weeks before we expected migratory movements to begin. This individual was alive and making small-scale movements the day before the radio-signal was lost. We obtained 316 total telemetry relocations over a period of 82 days for 11 American Kestrels, from the first observation of a fledgling out of the box until the last bird’s final location before migration. The mean number of locations per American Kestrel was 25.4 (SD = 19.2, range = 9–71). Three American Kestrels were alive on 13 September 2013 and subsequently left the study area, presumably upon initiation of migration. Table 1. Demographic characteristics of radio-tagged American Kestrels in Connecticut from 24 June to 13 September 2013. Sex = sex of radio-tagged American Kestrel, radio = date radio was deployed, fledge = fledge date (first observed out of box), mortality = estimated mortality date, days alive = days alive after fledging, death = estimated cause of death, first observation = first observation off natal site, + = death was not observed during the study period, and * = censored from the analyses. Clutch Date Days First Kestrel ID ratio (M|F) Sex Radio Fledge Mortality alive Death observation 216.266 3|1 M 6/19 6/24 - 82+ - 7/15 216.294 2|3 F 6/24 6/26 7/18 23 Unknown Deceased 216.333 2|3 F 6/25 6/26 - 82+ - 7/20 216.521 2|3 M 6/27 7/3 - 73+ - 7/29 216.576a 1|4 F 7/2 7/4 7/11 8 Predation Deceased 216.802 3|2 F 7/5 7/13 7/16 4 Exposure Deceased 216.826 2|3 F 7/5 7/6 7/16 11 Predation Deceased 151.270 3|2 F 7/8 7/12 - 12* - 7/25 216.868 4|1 M 7/8 7/14 7/17 4 Predation Deceased 151.259 2|3 F 7/11 7/16 8/3 19 Predation Deceased 216.576b 2|2 F 7/14 7/20 7/24 5 Exposure Deceased Northeastern Naturalist Vol. 22, No. 1 A.E. Stupik, T. Sayers, M. Huang, T.A.G. Rittenhouse, and C.D. Rittenhouse 2015 25 Causes of mortality We classified deaths as due to predation (n = 4), exposure (n = 2), and unknown (n = 1). We classified a death as predation if we discovered the transmitter near or attached to a bloodied or mangled carcass. Excessive piles of feathers near the carcass also provided evidence of predation, likely by an avian predator. On one rainy day, we found a bird soaked and shivering on the ground; the following day, we found its intact carcass less than 2 m from its previous location. We found another American Kestrel within 100 m of its natal box, lying on top of the grass. Its carcass, though rotted, showed no signs of predation. We classified both of these deaths as exposure. In another case, we found an intact transmitter and harness within 1 m of a stream with no sign of the bird itself. This American Kestrel was actively flying around the natal area the day before. It is extremely unlikely that the bird slipped the transmitter due to the fit and style of the harness. Therefore we classified this event as an unknown death. Survival of post-fledging American Kestrels At the conclusion of the study, daily survival was 0.27 (95% CI = 0.01, 0.53) on 15 September 2013 (Fig. 2). All deaths occurred within a 3-week window, and 4 of the deaths were within 3 days of each other. We also compared survival of early (prior to 7 July) to late-fledging American Kestrels. Daily survival of American Kestrels that fledged early (n = 6) was 0.50 (95% CI = 0.10, 0.90; Fig. 3A), and daily survival of late-fledging individuals (n = 5) was 0 (Fig. 3B). Movement of post-fledging American Kestrels For each individual, we calculated average daily distance, minimum distance, maximum distance, total distance, and net distance from nest box to last known location (Table 2). Movement paths constructed from relocations are provided as supplementary materials (see Supplemental File 1, available online at https://www.eaglehill.us/NENAonline/suppl-files/n22-1-N1288-Rittenhouse-s1, and, for BioOne subscribers, at http://dx.doi.org/10.1656/N1288.s1). Five birds did not move at all for at least 2 consecutive days; thus the minimum daily Figure 2. Survival of post-fledging American Kestrels (8 females and 3 males) from 24 June to 13 September 2013, in Connecticut. The black line is the mean survival rate and the gray lines are 95% confidence intervals. Northeastern Naturalist 26 A.E. Stupik, T. Sayers, M. Huang, T.A.G. Rittenhouse, and C.D. Rittenhouse 2015 Vol. 22, No. 1 movement of all birds was 0 m. The maximum daily movement of all tracked birds was 11.1 km. The smallest total distance (the sum of all daily movements) was 0.47 km, and the largest was 41.9 km. The smallest net distance was 0.04 km, and the largest was 16.1 km. The average daily distance moved did not differ between males and females (F = 3.608, P = 0.090), but the sample size was small; a larger sample may have yielded different results. Likewise, there was no difference for total distances moved by sex (F = 2.371, P = 0.158), or for net distance moved by sex (F = 2.671, P = 0.137). The rate of increase in daily movement distance was 14.08 (SE = 6.93) m per day (Fig. 4). Many American Kestrels exhibited little movement within the first week of fledging from the box. Our efforts to obtain visual observations during this time period usually resulted in homing to the base of a tree or a patch of very thick vegetation. Approximately 1 week after each American Kestrel fledged, we were able to observe the bird perched in a more open space such as on a snag or fencepost. We frequently observed American Kestrels from the same sibling group perched in the same tree, on the same branch, or on nearby fence posts. We observed one bird (ID Figure 3. Survival of post-fledging American Kestrels that (A) fledged prior to 7 July 2013 (n = 6) and (B) fledged after 7 July 2013 (n = 5), in Connecticut. The black lines are the mean survival rates and the gray lines are 95% confidence intervals. Northeastern Naturalist Vol. 22, No. 1 A.E. Stupik, T. Sayers, M. Huang, T.A.G. Rittenhouse, and C.D. Rittenhouse 2015 27 216.266) with another male American Kestrel for weeks after it fledged, including when it started to make larger movements outside of the nest-box area. Another fledgling (ID 216.333) flew directly over the observer at a nest-box location surrounded by at least 3 other American Kestrels. It is unknown whether interaction between the sibling groups occurred at this location. We lost the signals from 2 of the 3 American Kestrels for more than 30 days due to movements within and outside of the study area. Attempts to relocate the birds via systematic searching by vehicle in an ever-widening grid around the last known location failed. However, we relocated the birds by helicopter on 15 August 2013 and then resumed monitoring movements via vehicle until the initiation of migration. For the 3 American Kestrels that made movements off the natal area and were successfully re-located, approximately 40% (27 of 67 observations; ID 216.266), 43% (18 of 42 observations; ID 216.521), and 58% (22 of 38 observations; ID 216.333) of observations occurred off the natal area. Discussion We provide the first known-fate survival estimates for post-fledging American Kestrels in the northeastern US. Juvenile mortality rates are difficult to obtain due to the challenges of documenting movements and distinguishing mortality from dispersal. Survival estimates from banding returns indicated that the first-year American Kestrel survival rate is 31% (Henny 1972), with at least 24% of the deaths caused by humans. The survival rate of 27% that we report is comparatively lower than this first-year survival estimate and considerably lower than adult survival estimates (Henny 1972, Hinnebusch et al. 2010, Roest 1957). We also recorded American Kestrel movements after they fledged from the nest box. Movements of post-fledging American Kestrels ranged from relatively small-scale movements of birds staying within 100 m of their natal box to sudden, large movements exceeding 10 km in one day. Table 2. Movements by post-fledging American Kestrels from 24 June to 13 September 2013 in Connecticut. Total distance was cumulative for the study period, and net distance was from natal box to last known location. Number of Daily Distance (m) Distance (km) Kestrel ID Sex observations Average Minimum Maximum Total Net (km) 216.266 M 67 593 0 3.1 39.7 3.10 216.294 F 23 81 0 0.4 1.9 0.30 216.333 F 38 508 0 4.7 19.3 1.80 216.521 M 42 998 0 11.1 41.9 16.10 216.576a F 8 100 14 0.3 0.8 0.10 216.802 F 4 159 102 0.2 0.6 0.04 216.826 F 14 79 4 0.2 1.1 0.05 151.270 F 11 145 42 0.3 1.6 0.10 216.868 M 4 117 61 0.2 0.5 0.04 151.259 F 19 208 0 0.6 4.0 0.40 216.576b F 4 213 53 0.4 0.9 0.10 Northeastern Naturalist 28 A.E. Stupik, T. Sayers, M. Huang, T.A.G. Rittenhouse, and C.D. Rittenhouse 2015 Vol. 22, No. 1 The studies mentioned above calculated survival rates based on banding returns and used larger samples. Our sample size was relatively small, but we followed all of the birds closely and observed them on a daily basis. Using radio-telemetry, we were able to locate birds when they died, even when concealed in brush or buried underground. Studies done with banding returns also may have only received banding returns from the birds that were relatively mobile—most of the birds in this study died before making substantial movements away from the nest box. Our small sample size precluded a test for differences in survival among female and male American Kestrels during the post-fledging period. One study found only a 1% difference between adult male (75%) and adult female (74%) survival (Hinnebusch et al. 2010). Other studies on American Kestrel survival do not report different survival estimates for males and females (Farmer et al. 2006, Henny 1972). However, we noticed that birds that fledged early in the season (by 7 July) had higher survival than birds that fledged late in the season. Based on prior attempts to track post-fledging American Kestrels at this study site, we hypothesized that American Kestrels would stay close to their natal box before initiating migratory movements (T. Sayers and M. Huang, unpubl. data). With the use of a helicopter, we relocated 2 individuals whose signals had been lost when tracking by motor vehicle. The largest net displacement of an American Kestrel was 16.1 km. However, the other 2 birds that made substantial total movements had a net displacement of less than 3.5 km from their nest boxes. American Kestrels typically stayed deep in vegetation during the first week after fledging. The species is known to be inactive during this time (Varland et al. 1993), and fledglings remain out of view of predators in dense undergrowth or tree branches (T. Sayers and M. Huang, unpubl. data). We also observed Figure 4. Fitted values of daily movements by postfledging American Kestrels (individual as random effect). Mean movements in Connecticut increased with fledging age by approximately 14 m per day. Symbols and gray scale used to denote individual kestrels. Northeastern Naturalist Vol. 22, No. 1 A.E. Stupik, T. Sayers, M. Huang, T.A.G. Rittenhouse, and C.D. Rittenhouse 2015 29 American Kestrels associating with siblings, which has been widely documented in other studies (Smallwood and Bird 2002). Although we recognize that our study reports movements of post-fledging American Kestrels prior to migration, the movement patterns we documented appear similar to a study on natal dispersal (net displacement from natal origin to location the following breeding season) that recorded small-scale movements by most American Kestrels, with a few individuals demonstrating much greater movements (Miller and Smallwood 1997). In our study, 3 American Kestrels that we followed after initial movements away from the natal area demonstrated different methods of post-fledging dispersal. One American Kestrel (Kestrel ID 216.521) stayed within a few hundred meters of its nest box and then moved outside the natal area; it was later found 11.1 km from the natal box. Another bird (Kestrel ID 216.333) shifted the majority of its activity from its natal box area to a neighboring hay field, just across the tree-line. It was located once about 1 km from its natal box and later by helicopter less than 2 km away from its natal box. Finally, one American Kestrel (Kestrel ID 216.266) moved almost daily between its natal box area and nearby fields, all less than 1 km away. Thus, this bird’s total distance traveled was almost as high as that of Kestrel 216.521, but its net displacement was more than 5 times smaller than that of Kestrel 216.521. Our reported total distances traveled by individuals 216.333 and 216.521 are likely extremely conservative because we lacked at least 24 days of observed movements for each bird. Our post-fledging survival estimate of 27% conveys the important point that although adult American Kestrels demonstrate relatively high annual survival (approximately 59%; Henny 1972, Hinnebusch et al. 2010, Roest 1957), most juveniles of the species (73%) may die within 4 weeks of leaving the nest box. This study emphasizes the importance of understanding all avian life stages, especially species of conservation concern. Our results indicate that while an American Kestrel population may appear to be stable due to use of nest boxes and a high reproductive output, the population may still be declining due to low survival in the juvenile life stage. Thus, efforts to conserve American Kestrel populations should also seek to understand factors limiting survival during the post-fledging pe riod. Acknowledgments This study received financial support from a summer undergraduate research fellowship (obtained by A. Stupik) from the Office of Undergraduate Research at the University of Connecticut; the Wildlife and Fisheries Conservation Center at the University of Connecticut; Jean and Bruce Johnson of St. Paul, MN; the Environmental Professionals of Connecticut; Eastern Massachusetts Hawk Watch; Menunkatuck Audubon and the Hartford Audubon Societies. We also thank J. Vokoun for technical assistance. Literature Cited Beyer, H.L. 2012. Geospatial modelling environment (version 0.7.2.0). (software). Available online at http://www.spatialecology.com/gme. Accessed 16 October 2013. Northeastern Naturalist 30 A.E. Stupik, T. Sayers, M. Huang, T.A.G. Rittenhouse, and C.D. Rittenhouse 2015 Vol. 22, No. 1 Buehler, D.A., J.D. Fraser, M.R. Fuller, L.S. McAllister, and J.K.D. Seegar. 1995. Captive and field-tested radio-transmitter attachments for Bald Eagles. Journal of Field Ornithology 66:173–180. Catry, I., A.M.A. Franco, P. Rocha, R. Alcazar, S. Reis, A. Cordeiro, R. Ventim, J. Teodosio, and R. Moreira. 2013. Foraging-habitat quality constrains effectiveness of artificial nest-site provisioning in reversing population declines in a colonial cavity nester. Plos One 8(3):e58320. doi:10.1371/journal.pone.0058320. Cochran, W.W., and R.D. Lord, Jr. 1963. A radio-tracking system for wild animals. Journal of Wildlife Management 27:9–24. Connecticut Department of Environmental Protection (CT DEP). 2010. Connecticut's Endangered, Threatened, and Special Concern Species. Connecticut DEP Wildlife Division, Hartford, CT. Cox, D.R., and D. Oakes. 1984. Analysis of Survival Data. Chapman and Hall, New York, NY. 201 pp. Ecological Software Solutions, LLC. 2013. LOAS version 4.0. Available online at http:// www.ecostats.com/. Accessed 21 October 2013. Farmer, C.J., and J.P. Smith. 2009. Migration monitoring indicates widespread declines of American Kestrels (Falco sparverius) in North America. Journal of Raptor Research 43(4):263–273. Farmer, G.C., K. McCarty, S. Robertson, B. Robertson, and K.L. Bildstein. 2006. Suspected predation by accipiters on radio-tracked American Kestrels (Falco sparverius) in eastern Pennsylvania. Journal of Raptor Research 40(4):294–297. Henny, C.J. 1972. An analysis of the population dynamics of selected avian species. US Fish and Wildlife Service Research Report 1. US Government Printing Office, Washington, DC. 99 pp. Hinnebusch, D.M., J. Therrien, M. Valiquette, B. Robertson, S. Robertson, and K.L. Bildstein. 2010. Survival, site fidelity, and population trends of American Kestrels wintering in southwestern Florida. Wilson Journal of Ornithology 122(3):475–483. Hipkiss, T., J. Gustafsson, U. Eklund, and B. Hornfeldt. 2013. Is the long-term decline of Boreal Owls in Sweden caused by avoidance of old boxes? Journal of Raptor Research 47(1):15–20. Kaplan, E.L., and P. Meier. 1958. Nonparametric estimation from incomplete observations. Journal of the American Statistical Association 53(282):457–481. Katzner, T., S. Robertson, B. Robertson, J. Klucsartis, K. McCarty, and K.L. Bildstein. 2005. Results from a longer-term nest-box program for American Kestrels: Implications for improved population monitoring and conservation. Journal of Field Ornithology 76(3):217–226. Lambrechts, M.M., K. Wiebe, P. Sunde, T. Solonen, F. Sergio, A. Roulin, A.P. Møler, B.C. Lopéz, J.A. Fargallo, K. Exo, G. Dell’Omo, D. Costantini, M. Charter, M.W. Butler, G.R. Bortolotti, R. Arlettaz, and E. Korpimäki. 2011. Nest-box design for the study of diurnal raptors and owls is still an overlooked point in ecological, evolutionary, and conservation studies: A review. Journal of Ornithology 153:23–34. Mech, L.D. 1983. Handbook of Animal Radio-tracking. University of Minnesota Press, Minneapolis, MN. 108 pp. Miller, K.E., and J.A. Smallwood. 1997. Natal dispersal and philopatry of southeastern American Kestrels in Florida. The Wilson Bulletin 109(2):226–232. Newton, I. 1994. Experiments on the limitation of bird-breeding densities: A review. Ibis 136(4):397–411. Northeastern Naturalist Vol. 22, No. 1 A.E. Stupik, T. Sayers, M. Huang, T.A.G. Rittenhouse, and C.D. Rittenhouse 2015 31 Pollock, K.H., S.R. Winterstein, C.M. Bunck, and P.D. Curtis. 1989. Survival analysis in telemetry studies: The staggered-entry design. Journal of Wildlife Management 53(1):7–15. R Core Development Team. 2012. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Robillard, A., D. Garant, and M. Bélisle. 2013. The swallow and the sparrow: How agricultural intensification affects abundance, nest-site selection, and competitive interactions. Landscape Ecology 28:201–215. Roest, A.I. 1957. Notes on the American Sparrow Hawk. Auk 74:1–19. Smallwood, J.A., and D.M. Bird. 2002. American Kestrel (Falco sparverius). Number 602, 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/602. Accessed 26 October 2013. Smallwood, J.A., and M.W. Collopy. 2009. Southeastern American Kestrels respond to an increase in the availability of nest cavities in north-central Florida. Journal of Raptor Research 43(4):291–300. Smallwood, J.A., M.F. Causey, D.H. Mossop, J.R. Klucsarits, B. Robertson, S. Robertson, J. Mason, M.J. Maurer, R.J. Melvin, R.D. Dawson, G.R. Bortolotti, J.W. Parrish, Jr., T.F. Breen, and K. Boyd. 2009. Why are American Kestrel (Falco sparverius) populations declining in North America? Evidence from nest-box programs. Journal of Raptor Research 43(4):274–282. Steenhof, K., and B. Peterson. 2009. Site fidelity, mate fidelity, and breeding dispersal in American Kestrels. Journal of Ornithology 121(1):12–21. Turchin, P. 1998. Quantitative Analysis of Movement: Measuring and Modeling Population Redistribution in Animals and Plants. Sinauer Associates, Inc., Sunderland, MA. 396 pp. Varland, D.E., E.E. Klaas, and T.M. Loughlin. 1993. Use of habitat and perches, causes of mortality, and time until dispersal in post-fledging American Kestrels. Journal of Field Ornithology 64:169–178. Verts, B.J. 1963. Equipment and techniques for radio-tracking Striped Skunks. Journal of Wildlife Management 27:325–339.