Estimates Of Apparent Survival Research Articles

Apparent Survival Estimates for Five Species of Tropical Birds in an Endangered Forest Habitat in Western Ecuador1

ABSTRACTEstimates of annual survival are essential for addressing topics in evolutionary and conservation ecology. However, most demographic studies of land birds are based on north temperate species, and few robust estimates of survival based on mark–recapture statistics are available for continental South American birds. We used time‐since‐marking models to estimate apparent survival of adult birds from 7 yr of mist netting data in the Colonche Hills. This site is one of few remaining large tracts of premontane forest in southwestern Ecuador, and an area of high priority for avian conservation. Species with sufficient data for analysis included three hummingbirds (Adelomyia melanogenys, Heliodoxa jacula, Phaethornis baroni), a cotinga (Schiffornis turdinus), and a wren (Henicorhina leucophrys). Our parameter estimates had reduced precision because the number of recaptures was small. Probability of recapture was low in three species , and moderate in two others . Adelomyia and Phaethornis had moderate apparent survival (; probability that a bird neither died nor emigrated from our survey area in a given year). Adults of Adelomyia moved seasonally, and it is possible that permanent emigration from our survey area contributed to low estimates of apparent survival. The other three species had relatively high estimates of adult apparent survival ranging from in Heliodoxa and Schiffornis to a high of in Henicorhina.

Experimental reseeding of juvenile spiny lobsters ( Jasus edwardsii): Comparing survival and movement of wild and naïve lobsters at multiple sites

We used mark–recapture methods to assess the relative survival of naïve and wild-caught juvenile lobsters released to coastal reef in southeast Tasmania, Australia. Naïve lobsters were captured as pueruli (first benthic post-larval stage) and reared in tanks for 12 to 18 months. Naïve and wild-caught lobsters were identified individually by an antennal tag and released by divers at 4 sites. In dive surveys performed over 3 to 5 weeks we resighted similar numbers of wild and naïve lobsters at 2 of the sites, but significantly more naïve lobsters at the remaining sites. Multistate mark–recapture models detected no difference in apparent survival between treatment groups at 3 of the sites. Lower apparent survival of naïve lobsters at the fourth site was attributed to the small size of this reef and the absence of adjacent lobster habitat. Many naïve lobsters did not recognise habitat boundaries and moved away from the release reef onto sand. Acoustic tracking revealed that selecting release sites with contiguous reef extending a minimum of 30 m in all directions should negate losses of this kind. At 3 sites apparent survival estimates were low for wild and naïve lobsters in the 24 h following release. Apparent survival was much higher when lobsters were released into seafloor cages. Acoustic tracking indicated that low apparent survival of lobsters released outside cages was due largely to emigration from our survey areas rather than losses to predation. By following simple criteria in the selection of release sites it is possible to ensure that survival rates among released naïve lobsters are equivalent to those of wild lobsters released in the same manner.

Survivorship in two populations of long‐tailed bats (Chalinolobus tuberculatus) in New Zealand

The survival of New Zealand long‐tailed bats (Chalinolobus tuberculatus) was assessed in two populations, one at Hanging Rock, South Canterbury, South Island, the other at Grand Canyon, central west North Island. Apparent survival of adult females over winter was calculated using mark‐recapture analysis over 5 years. Annual survival varied from 0.75 (95% CI = 0.54–0.88) to 0.89 (0.48–0.99) at Hanging Rock and 0.55 (0.39–0.71) to 0.91 (0.44–0.99) at Grand Canyon. Estimates of apparent survival were consistent with those from a longer‐term study of long‐tailed bats in the Eglinton Valley, Fiordland, South Island. The intrinsic rate of increase (λ) at Hanging Rock was 0.90 (95% CI = 0.4–0.99), indicating that this population may be declining more rapidly than that in the Eglinton Valley. At Grand Canyon, population estimates fluctuated annually, with no clear trend. Predation by introduced mammals and loss of habitat likely caused declines at Hanging Rock. Management of these factors is essential to avoid the local extinction of the species. Long‐term monitoring (greater than 5 years) is required to determine more robust population trends for these long‐tailed bat populations.

Annual survival estimation of migratory songbirds confounded by incomplete breeding site-fidelity: study designs that may help

Many species of bird exhibit varying degrees of site–fidelity to the previous year’s territory or breeding area, a phenomenon we refer to as incomplete breeding site–fidelity. If the territory they occupy is located beyond the bounds of the study area or search area (i.e., they have emigrated from the study area), the bird will go undetected and is therefore indistinguishable from dead individuals in capture–mark–recapture studies. Differential emigration rates confound inferences regarding differences in survival between sexes and among species if apparent survival rates are used as estimates of true survival. Moreover, the bias introduced by using apparent survival rates for true survival rates can have profound effects on the predictions of population persistence through time, source/sink dynamics, and other aspects of life–history theory. We investigated four study design and analysis approaches that result in apparent survival estimates that are closer to true survival estimates. Our motivation for this research stemmed from a multi–year capture–recapture study of Prothonotary Warblers (Protonotaria citrea) on multiple study plots within a larger landscape of suitable breeding habitat where substantial inter–annual movements of marked individuals among neighboring study plots was documented. We wished to quantify the effects of this type of movement on annual survival estimation. The first two study designs we investigated involved marking birds in a core area and resighting them in the core as well as an area surrounding the core. For the first of these two designs, we demonstrated that as the resighting area surrounding the core gets progressively larger, and more “emigrants” are resighted, apparent survival estimates begin to approximate true survival rates (bias < 0.01). However, given observed inter–annual movements of birds, it is likely to be logistically impractical to resight birds on sufficiently large surrounding areas to minimize bias. Therefore, as an alternative protocol, we analyzed the data with subsets of three progressively larger areas surrounding the core. The data subsets provided four estimates of apparent survival that asymptotically approached true survival. This study design and analytical approach is likely to be logistically feasible in field settings and yields estimates of true survival unbiased (bias < 0.03) by incomplete breeding site–fidelity over a range of inter–annual territory movement patterns. The third approach we investigated used a robust design data collection and analysis approach. This approach resulted in estimates of survival that were unbiased (bias < 0.02), but were very imprecise and likely would not yield reliable estimates in field situations. The fourth approach utilized a fixed study area size, but modeled detection probability as a function of bird proximity to the study plot boundary (e.g., those birds closest to the edge are more likely to emigrate). This approach also resulted in estimates of survival that were unbiased (bias < 0.02), but because the individual covariates were normalized, the average capture probability was 0.50, and thus did not provide an accurate estimate of the true capture probability. Our results show that the core–area with surrounding resight–only can provide estimates of survival that are not biased by the effects of incomplete breeding site–fidelity.

ANNUAL SURVIVAL AND RECRUITMENT IN A RUBY-THROATED HUMMINGBIRD POPULATION, EXCLUDING THE EFFECT OF TRANSIENT INDIVIDUALS

We estimated annual apparent survival, recruitment, and rate of population growth of breeding Ruby-throated Hummingbirds (Archilochus colubris), while controlling for transients, by using 18 years of capture-mark-recapture data collected during 1984–2001 at Hilton Pond Center for Piedmont Natural History near York, South Carolina. Resident males had lower apparent survival (0.30 ± 0.05 SE) than females (0.43 ± 0.04). Estimates of apparent survival did not differ by age. Point estimates suggested that newly banded males were less likely than females to be residents, but standard errors of these estimates overlapped (males: 0.60 ± 0.14 SE; females: 0.67 ± 0.09). Estimated female recruitment was 0.60 ± 0.06 SE, meaning that 60% of adult females present in any given year had entered the population during the previous year. Our estimate for rate of change indicated the population of female hummingbirds was stable during the study period (1.04 ± 0.04 SE). We suggest an annual goal of ≥64 adult females and ≥64 immature females released per banding area to enable rigorous future tests for effects of covariates on population dynamics. Development of a broader cooperating network of hummingbird banders in eastern North America could allow tests for regional or metapopulation dynamics in this species. Sobrevivencia Anual y Reclutamiento en una Población del Picaflor Archilochus colubris Excluyendo el Efecto de Individuos Ocasionales Resumen. Estimamos la sobrevivencia anual aparente, reclutamiento y tasa de crecimiento en una población reproductiva del picaflor Archilochus colubris, controlando por la presencia de individuos ocasionales. Utilizamos datos de 18 años de captura-marcaje-recaptura colectados entre 1984 y 2001 en Hilton Pond Center for Piedmont Natural History en cercanías de York, Carolina del Sur. Los machos residentes presentaron una sobrevivencia aparente menor (0.30 ± 0.05 EE) que las hembras (0.43 ± 0.04). Las estimaciones de la sobreviviencia aparente no difirieron entre edades. Estimaciones puntuales sugirieron que los machos marcados por primera vez tuvieron una menor probabilidad de ser residentes que las hembras, pero los errores estándar de estas estimaciones se sobrepusieron (machos: 0.60 ± 0.14 EE; hembras: 0.67 ± 0.09 EE). Para las hembras el reclutamiento estimado fue 0.60 ± 0.06 EE, lo que significa que el 60% de las hembras adultas presentes en un año determinado entraron a la población durante el año anterior. Nuestra estimación de la tasa de cambio indicó que la población de picaflores hembra fue estable durante el período de estudio (1.04 ± 0.02 EE). Para permitir futuras pruebas rigurosas sobre las covariables asociadas a las dinámicas poblacionales, sugerimos una meta anual de ≥64 hembras adultas y ≥64 hembras inmaduras liberadas por área de anillado. La creación de una amplia red cooperativa de anilladores de picaflores en el este de Norteamérica podría permitir estudiar y entender mejor las dinámicas regionales o metapoblacionales de esta especie.

APPARENT SURVIVAL RATES OF WESTERN SANDPIPER (CALIDRIS MAURI) WINTERING IN NORTHWEST BAJA CALIFORNIA, MEXICO

Abstract To estimate annual apparent local survival, we collected capture–resighting data on 256 individually marked male Western Sandpipers (Calidris mauri) wintering at Estero de Punta Banda, Mexico, between 1994–1997. A hierarchical modeling approach was used to address the effect of age class and year on survivorship rates. The best-fit model included a constant apparent survival probability (ϕ = 0.489; 95% CI = 0.410–0.569), but several models fit nearly as well, and averaging among the top five, to account for model uncertainty, suggested that adults had somewhat higher values than juveniles (ϕ = 0.490 ± 0.051 vs. 0.450 ± 0.067). Detection probability was substantially higher for adults than for juveniles (p = 0.741 vs. p = 0.537). Those apparent survival estimates are low compared with those from other studies of Western Sandpipers at breeding and other nonbreeding locations, and substantially lower than the true survivorship rates expected for small sandpipers in general. We interpret these result...

Apparent Survival Rates of Western Sandpiper (Calidris Mauri) Wintering in Northwest Baja California, Mexico

AbstractTo estimate annual apparent local survival, we collected capture–resighting data on 256 individually marked male Western Sandpipers (Calidris mauri) wintering at Estero de Punta Banda, Mexico, between 1994–1997. A hierarchical modeling approach was used to address the effect of age class and year on survivorship rates. The best-fit model included a constant apparent survival probability (ϕ = 0.489; 95% CI = 0.410–0.569), but several models fit nearly as well, and averaging among the top five, to account for model uncertainty, suggested that adults had somewhat higher values than juveniles (ϕ = 0.490 ± 0.051 vs. 0.450 ± 0.067). Detection probability was substantially higher for adults than for juveniles (p = 0.741 vs. p = 0.537). Those apparent survival estimates are low compared with those from other studies of Western Sandpipers at breeding and other nonbreeding locations, and substantially lower than the true survivorship rates expected for small sandpipers in general. We interpret these results as indicating that this site is of below average quality for nonbreeding male Western Sandpipers.