Demographic rates and population viability of ... - Wiley Online Library

Wildlife Monographs 194:1–37; 2016; DOI: 10.1002/wmon.1018

Demographic Rates and Population Viability of Black Bears in Louisiana JARED S. LAUFENBERG,1,2 Department of Forestry, Wildlife and Fisheries, University of Tennessee, 274 Ellington Plant Sciences Building, Knoxville, TN 37996, USA JOSEPH D. CLARK, U.S. Geological Survey, Northern Rocky Mountain Science Center, Southern Appalachian Research Branch, University of Tennessee, 274 Ellington Plant Sciences Building, Knoxville, TN 37996, USA MICHAEL J. HOOKER,3 Department of Forestry, Wildlife and Fisheries, University of Tennessee, 274 Ellington Plant Sciences Building, Knoxville, TN 37996, USA CARRIE L. LOWE,4 Department of Forestry, Wildlife and Fisheries, University of Tennessee, 274 Ellington Plant Sciences Building, Knoxville, TN 37996, USA KAITLIN C. O’CONNELL-GOODE,5 Department of Forestry, Wildlife and Fisheries, University of Tennessee, 274 Ellington Plant Sciences Building, Knoxville, TN 37996, USA JESSE C. TROXLER,6 Department of Forestry, Wildlife and Fisheries, University of Tennessee, 274 Ellington Plant Sciences Building, Knoxville, TN 37996, USA MARIA M. DAVIDSON, Louisiana Department of Wildlife and Fisheries, 646 Cajundome Boulevard, Suite 126, Lafayette, LA 70506, USA MICHAEL J. CHAMBERLAIN, Warnell School of Forestry and Natural Resources, University of Georgia, 180 E Green Street, Athens, GA 30602, USA RICHARD B. CHANDLER, Warnell School of Forestry and Natural Resources, University of Georgia, 180 E Green Street, Athens, GA 30602, USA

ABSTRACT The Louisiana black bear (Ursus americanus luteolus) was reduced to a few small, fragmented, and isolated subpopulations in the Lower Mississippi Alluvial Valley by the mid-twentieth century resulting from loss and fragmentation of habitat. In 1992, the United States Fish and Wildlife Service (USFWS) granted the Louisiana black bear threatened status under the United States Endangered Species Act of 1973. Since that time, a recovery plan was developed, a reintroduced population was established, and habitat recovery has occurred. The Recovery Plan states that a minimum of 2 populations must be viable (i.e., persistence probabilities over 100 years >0.95), 1 in the Tensas River Basin and 1 in the Atchafalaya River Basin. Consequently, our objectives were to 1) estimate demographic rates of Louisiana black bear subpopulations, 2) develop data-driven stochastic population projection models, and 3) determine how different projection model assumptions affect population trajectories and predictions about long-term persistence. Our overall goal was to assess long-term persistence of the bear subpopulations in Louisiana, individually and as a whole. We collected data using varying combinations of non-invasive DNA sampling, live capture, winter den visits, and radio monitoring from 2002 to 2012 in the 4 areas currently supporting breeding subpopulations in Louisiana: Tensas River Basin (TRB), Upper Atchafalaya River Basin (UARB), Lower Atchafalaya River Basin (LARB), and a recently reintroduced population at the Three Rivers Complex (TRC). From 2002 to 2012, we radio monitored fates of 86 adult females within the TRB and 43 in the TRC. Mean estimates of annual adult survival for the TRB and TRC were 0.997 and 0.990, respectively, when unknown fates were assumed alive and 0.970 and 0.926 when unknown fates were assumed dead. From 2003 to 2013, we observed 130 cub litters from 74 females in the TRB, and 74 cub litters from 45 females in the TRC. During the same period, we observed 43 yearling litters for 33 females in the TRB and 21 yearling litters for 19 females in the TRC. The estimated number of cubs and number of yearlings produced per breeding adult female was 0.47 and 0.20, respectively, in the TRB and 0.32 and 0.18 in the TRC. On the basis of matrix projection models, asymptotic growth rates ranged from 1.053 to 1.078 for the TRB and from 1.005 to 1.062 for the TRC, depending on how we treated unresolved fates of adult females. Persistence probabilities estimated from stochastic population models based on telemetry data ranged from 0.997 to 0.998 for the TRC subpopulation depending on model assumptions and were >0.999 for the TRB regardless of model assumptions. We extracted DNA from hair collected at baited, barbed-wire enclosures in the TRB, UARB, and LARB to determine individual identities for capture-markrecapture (CMR) analysis. We used those detection histories to estimate apparent survival (w), per-capita recruitment (f), abundance (N), realized growth rate (l), and long-term viability, based on Bayesian

Received: 25 February 2015; Accepted: 4 February 2016 1

Present address: Department of Fish, Wildlife, and Conservation Biology, Colorado State University, Fort Collins, CO 80523, USA. E-mail: [email protected] 3 Present address: Warnell School of Forestry and Natural Resources, University of Georgia, 180 E Green Street, Athens, GA 30602, USA. 4 Present address: Washington Department of Fish and Wildlife, 2315 North Discovery Place, Spokane Valley, WA 99216, USA. 5 Present address: Florida Fish and Wildlife Conservation Commission, 350 Carrol Street, Eastpoint, FL 32328, USA. 6 Present address: Tennessee Valley Authority, 400 West Summit Hill Drive, WT 11C-K, Knoxville, TN, 37902, USA. 2

Laufenberg et al.




Population Viability of Louisiana Black Bears

1

hierarchical modeling methods that allowed estimation of temporal process variance and parameter uncertainty. Based on 23,312 hair samples, annual N for females in the TRB ranged from 133 to 164 during 2006–2012, depending on year and how detection heterogeneity was modeled. Geometric mean of l ranged from 0.996 to 1.002. In the UARB, we collected 11,643 hair samples from 2007 to 2012, from which estimates of N for females ranged from 23 to 43 during the study period, depending on detection heterogeneity model. The geometric mean of l ranged from 1.038 to 1.059. Estimated N for females in LARB ranged from 69 to 96, and annual l ranged from 0.80 to 1.11 based on 3,698 hair samples collected during 2010–2012, also depending on year and heterogeneity model. Probabilities of persistence over 100 years for the TRC and TRB based on stochastic matrix projection models that used vital rate estimates from telemetry data were >0.95 for all scenarios. Probability of persistence at the TRB and the UARB based on projection models that used vital rate estimates from CMR analyses ranged from 0.928 to 0.954 and from 0.906 to 0.959, respectively, depending on model assumptions. Data from the LARB were insufficient for a viability assessment. Thus, individual persistence probabilities for TRB and UARB did not meet the strict definition of viability (i.e., >0.95) under some model assumptions. However, the joint probability of bears persisting either in the TRB or UARB was >0.993 assuming individual population dynamics were independent and was >0.958 assuming dynamics were perfectly correlated. Furthermore, including the TRC increased the joint probability of bears persisting somewhere in the TRB, UARB, or TRC to >0.999 based on the most pessimistic individual persistence estimates from those subpopulations. Therefore, if the intent of specifying that 2 subpopulations should be viable was to ensure the persistence of Louisiana black bears somewhere within its historical range, then the viability threshold was met. Ó 2016 The Wildlife Society. KEY WORDS Bayesian, capture-mark-recapture, demographics, hierarchical model, Louisiana black bear, population viability analysis, threatened species, Ursus americanus luteolus.

Tasas Demogr
aficas y Poblaciones Viables de Osos Negros en Louisiana RESUMEN A mediados del siglo XX, la p
erdida y fragmentaci
on del h
abitat redujo las poblaciones de oso negro

de Louisiana (Ursus americanus luteolus) a unas pocas subpoblaciones peque~ nas, fragmentadas y aisladas en el Valle aluvial del bajo Mississippi. En 1992, el Servicio de Pesca y Vida Silvestre de Estados Unidos (USFWS) otorg
o al oso negro de Louisiana el estatus de Amenazado bajo la Ley de Especies en Peligro de 1973. Desde entonces, un plan de recuperaci
on fue desarrollado, una poblaci
on reintroducida fue establecida, y el h
abitat ha sido restaurado. El plan de recuperaci
on establec
ıa que por lo menos dos poblaciones, una en la cuenca del r
ıo de Tensas y la otra en la cuenca del r
ıo Atchafalaya deb
ıan ser viables (es decir, que la probabilidad de persistencia de una poblaci
on por m
as de 100 a~ nos sea >0.95). En consecuencia, nuestros objetivos fueron: 1) estimar las tasas demogr
aficas de las subpoblaciones de oso negro de Louisiana, 2) desarrollar modelos estoc
asticos basados en datos de proyecci
on de poblaciones, y 3) determinar c
omo los diferentes supuestos del modelo de proyecci
on afectan las trayectorias de poblaci
on y predicciones sobre la persistencia a largo plazo. Nuestro objetivo general fue evaluar la persistencia a largo plazo de subpoblaciones de osos en Louisiana, en forma individual y en conjunto. Los datos fueron colectados desde 2002 hasta el 2012 utilizando muestreos de ADN no invasivos, capturas de individuos, visitas a cuevas de invierno, y telemetr
ıa, en las 4
areas que actualmente contienen subpoblaciones reproductoras en Louisiana: cuenca del rio Tensas (TRB), cuenca alta del r
ıo Atchafalaya (UARB), cuenca baja del r
ıo Atchafalaya (LARB), y una poblaci
on recientemente reintroducida en el complejo de Tres R
ıos (TRC). Desde 2002 a 2012, hemos monitoreado el destino de 86 hembras adultas en el TRB y 43 en el TRC. Estimaciones medias anuales de supervivencia de adultos para el TRB y TRC fueron de 0.997 y 0.990 respectivamente, bajo el supuesto que osos con destino desconocido estaban vivos, y 0.970 y 0.926 cuando se asumieron que osos con destino desconocido hab
ıan muerto. Desde 2003 a 2013, se observaron 130 camadas de oseznos de 74 hembras en el TRB, y 74 camadas de oseznos de 45 hembras en el TRC. Durante el mismo per
ıodo, se observ
o 43 camadas de oseznos de un a~ no de 33 hembras en el TRB y 21 camadas de oseznos de un a~ no de 19 hembras en el TRC. El n
umero estimado de oseznos y n
umero de oseznos de un a~ no producidos por una hembra adulta en el TRB fue de 0.47 y 0.20, respectivamente, y de s 0.32 y 0.18 en el TRC. A partir de modelos de proyecci
on de matriz, las tasas de crecimiento asint
oticas variaron de 1.053 hasta 1.078 para el TRB y de 1.005 a 1.062 para la TRC, dependiendo de c
omo se trat
o el destino de hembras adultas. Probabilidades de persistencia estimadas a partir de los modelos estoc
asticos de poblaci
on basados en datos de telemetr
ıa oscilaron entre 0.997-0.998 para la subpoblaci
on TRC dependiendo en los supuestos del modelo, y de > 0.999 para el TRB independientemente de los supuestos del modelo. Se extrajo ADN de pelo colectado de trampas de alambre de puas de recintos cebados en el TRB, UARB y LARB para determinar las identidades individuales para el an
alisis de captura-marcado y recaptura (CMR). 2

Wildlife Monographs




194

Utilizamos historias de detecci
on para estimar la supervivencia aparente (w), el reclutamiento per c
apita (f), la abundancia (N), la tasa de crecimiento real (l), y la viabilidad a largo plazo basado en m
etodos de modelado jer
arquicos bayesianos que permitieron estimar la varianza temporal y par
ametros de incertidumbre. Sobre la base de 23,312 muestras de pelo, el N anual para hembras en el TRB oscil
o entre 133-164 durante 2006-2012, dependiendo del a~ no y c
omo la heterogeneidad de detecci
on fue modelada. La media geom
etrica de l oscil
o 0.9961.002. En el UARB, se colectaron 11,643 muestras de pelo durante el periodo de 2007 a 2012, a partir del cual las estimaciones de N para las hembras oscilaron desde 23 hasta 43 durante el per
ıodo del estudio, y dependiendo del modelo de heterogeneidad de detecci
on. La media geom
etrica de l vari
o de 1.038 a 1.059. El N estimado para hembras en LARB vari
o de 69 a 96, y l anuales de 0.80 a 1.11 basado en 3,698 muestras de pelo recogidos durante desde 2010 hasta 2012, tambi
en dependiendo de a~ no y del modelo de heterogeneidad. Las probabilidades de persistencia de m
as de 100 a~ nos de la TRC y TRB basados en modelos de proyecci
on matriz estoc
asticos que utilizaron estimaciones de tasas vitales a partir de datos de telemetr
ıa fueron de > 0.95 para todos los escenarios. Probabilidad de persistencia en el TRB y el UARB basado en modelos de proyecci
on que utilizaron estimaciones de tasas vitales del an
alisis de CMR variaron de 0.928 hasta 0.954 y 0.906 hasta 0.959, respectivamente, en funci
on de las hip
otesis del modelo. Los datos de la LARB fueron insuficientes para una evaluaci
on de viabilidad. Por lo tanto, las probabilidades de persistencia individuales para TRB y UARB no cumplieron con la definici
on estricta de viabilidad (es decir > 0.95) bajo algunos supuestos del modelo. Sin embargo, la probabilidad conjunta de la persistencia de, ya sea en el TRB o UARB fue de > 0.993 asumiendo que las din
amicas poblaciones individuales sean independientes, y fue de > 0.958 asumiendo que las din
amicas estaban perfectamente correlacionados. Adem
as, la inclusi
on del TRC aumento la probabilidad conjunta de persistencia de osos en alg
un lugar del TRB, UARB o TRC a >0.999 basados en las estimaciones de persistencia individuales m
as pesimistas de esas subpoblaciones. Por lo tanto, si la intenci
on de tener 2 subpoblaciones viables fue para asegurar la persistencia de los osos negros de Louisiana en alg
un lugar de su
area de distribuci
on hist
orica, entonces si se cumpli
o con el umbral de viabilidad.

Taux D
emographiques et Viabilit
e de la Population des Ours Noirs en Louisiane

RESUM EN L’ours noir de Louisiane (Ursus americanus luteolus) a
et
e r
eduit a quelques petites souspopulations decoup
ees et isol
ees dans la vall
ee alluviale du Mississippi des la moiti
e du 20ieme siecle, r
esultat de pertes et d
ecoupages de leur habitat. En 1992, le U.S. Fish and Wildlife Service (USFWS) a accord
e a l’ours noir de Louisiane le statut d’espece menac
ee dans le d
ecret des especes en danger de 1973. Depuis ce temps, un plan de r
eg
en
eration a
et
e developp
e, une population r
eintroduite a
et
e
etablie, et une r
eg
en
eration de l’habitat a eu lieu. Le plan de r
eg
en
eration indique qu’au moins 2 populations doivent ^etre viables (i.e., probablilit
es de persistance sur 100 ans >0,95), une dans le bassin de la riviere Tensas et une dans le bassin de la riviere Atchafalaya. Par cons
equent, nos objectifs
etaient de 1) estimer les taux d
emographiques des sous-populations d’ours noirs de Louisiane, 2) d
evelopper des modeles stochastiques de projection de population a partir de donn
ees, et 3) d
eterminer comment diff
erentes hypotheses des modeles de projection influencent les trajectoires de populations et les pr
edictions sur la persistance a long-terme. Notre but principal
etait d’estimer la persistance a long-terme des sous-populations d’ours en Louisiane, individuellement et dans leur ensemble.Nous avons collect
e des donn
ees en utilisant des combinaisons de pr
elevements non-invasifs d’ADN, captures, visites de tanieres pendant l’hiver, et surveillance radio de 2002 a 2012 dans les 4 zones qui supportent la reproduction des sous-populations en Louisiane: le bassin de la riviere Tensas (Tensas River Basin, TRB), le bassin sup
erieur de la riviere Atchafalaya (Upper Atchafalaya River Basin, UARB), le bassin inf
erieur de la riviere Atchafalaya (Lower Atchafalaya River Basin, LARB), et une population r
ecemment r
eintroduite au complexe des trois rivieres (Three Rivers Complex, TRC). De 2002 a 2012, nous avons radio surveill
e le sort de 86 femelles adultes dans le TRB et 43 dans le TRC. Les estimations moyennes de survie annuelle des adultes dans le TRB et le TRC
etaient 0,997 et 0,990, respectivement, quand les sorts ind
ecis
etaient consid
er
es comme vivants, et 0,970 et 0,926 quand les sorts ind
ecis
etaient consid
er
es comme morts. De 2003 a 2013, nons avons observ
e 130 port
ees d’oursons par 74 femelles dans le TRB, et 74 port
ees d’oursons par 45 femelles dans le TRC. Pendant la m^eme p
eriode, nous avons observ
e 43 port
ees d’un an par 33 femelles dans le TRB et 21 port
ees d’un an par 19 femelles dans le TRC. Les nombres estim
es d’oursons et d’oursons d’un an produits par des femelles adultes reproductrices dans le TRB
etaient de 0,47 et 0,20, respectivement. Dans le TRC, ces nombres
etaient 0,32 et 0,18. Bas
e sur les modeles de projection matricielle, les taux de croissance asymptotiques variaient de 1,053 a 1,078 pour le TRB et de 1,005 a 1,062 pour le TRC, d
ependant de la faSc on dont nous avons trait
e les sorts ind
ecis des femelles adultes. Les probabilit
es de persistance estim
ees par des modeles de population

Laufenberg et al.




Population Viability of Louisiana Black Bears

3

stochastiques bas
es sur les donn
ees de t
el
em
etries variaient de 0,997 a 0,998 pour la sous-population de TRC en fonction des hypotheses du modele et
etaient sup
erieures a 0,999 pour le TRB ind
ependamment des hypotheses du modele. Nous avons extrait de l’ADN de poils collect
es avec des app^ats dans des enclos de fil-barbel
es dans le TRB, UARB, et le LARB afin de d
eterminer les identit
es individuelles pour l’analyse de capture-marquagerecapture (capture-mark-recapture, CMR). Nous avons utilis
e les historiques de d
etections pour estimer la survie apparente (w), le recrutement par individu (f), l’abondance (N), le taux de croissance r
ealis
e (l), et la viabilit
e a long-terme a partir de m
ethodes bay
esiennes de mod
elisation hi
erarchique qui ont permis l’estimation  partir de 23 312
echantillons de poils, de la variance temporelle du processus et de l’incertitude des parametres. A le N annuel pour les femelles du TRB variait de 133 a 164 pendant la p
eriode 2006-2012, en fonction de l’ann
ee et de la faSc on dont l’h
et
erog
en
eit
e de d
etection
etait modelis
ee. La moyenne g
eom
etrique de l variait de 0,996 a 1,002. Dans le UARB, nous avons collect
e 11 643
echantillons de poils de 2007 a 2012 a partir desquels les estimations de N pour les femelles variaient de 23 a 43 pendant la p
eriode de l’
etude, en fonction de l’h
et
erog
en
eit
e de d
etection du modele. La moyenne g
eom
etrique de l variait de 1,038 a 1,059. Les valeurs estim
ees de N pour les femelles du LARB variaient de 69 a 96, et l annuel, bas
e sur la collecte de 3 698
echantillons de poils, variait de 0,80 a 1,11 pendant la p
eriode 2010-2012, en fonction de l’ann
ee et de l’h
et
erog
en
eit
e du modele. Les probabilit
es de persistance sur 100 ans pour le TRC et le TRB estim
ees par les modeles de projection matricielle stochastiques calcul
es a partir d’estimations de taux vitaux par des donn
ees de t
el
em
etrie
etaient sup
erieures a 0,95 pour tous les sc
enarios. Les probabilit
es de persistance dans le TRB et le UARB estim
ees par de modeles de projection calcul
es a partir d’estimations de taux vitaux par les analyses CMR variaient de 0,928 a 0,954 et de 0,906 a 0,959, respectivement, en fonction des hypotheses des modeles. Les donn
ees du LARB
etaient insuffisantes pour une estimation de viabilit
e. Ainsi, les probabilit
es de persistance individuelles pour le TRB et le UARB n’ont pas satisfait la d
efinition stricte de viabilit
e (i.e., >0,95) pour certaines hypotheses des modeles. Cependant, la probabilit
e jointe d’ours persistant dans le TRB ou le UARB
etait sup
erieure a 0,993 en assumant que les dynamiques de population
etaient ind
ependantes, et
etaient sup
erieures a 0,958 en assumant que les dynamiques
etaient parfaitement corr
el
ees. De plus, inclure le TRC a augment
e la probabilit
e jointe d’ours pouvant persister quelque part dans le TRB, le UARB, ou le TRC a plus de 0,999 a partir des estimations de persistances individuelles les plus pessimistes de ces sous-populations. Ainsi, si le but de sp
ecifier que 2 sous-populations devraient ^etre viables pour assurer la persistance des ours noirs de Louisiane quelque part a l’int
erieur son habitat historique, alors le seuil de viabilit
e a
et
e atteint.

Contents INTRODUCTION ................................................................................. 4 STUDY AREA......................................................................................... 6 METHODS............................................................................................. 7 Data Sources......................................................................................... 7 Capture and monitoring .................................................................... 7 Non-invasive DNA sampling............................................................. 7 DNA extraction and microsatellite genotyping....................................... 8 Marker selection for individual identification........................................ 8 Demographic Rate Analysis .................................................................. 9 General approach .............................................................................. 9 Survival rates of radio-collared adult female bears ................................. 9 Reproductive rates of radio-collared adult female bears ......................... 10 Demographic rates from capture-mark-recapture .................................. 11 Population Viability Analysis .............................................................. 13

RESULTS.............................................................................................. 15 Demographic Rate Analysis ................................................................ 15 Survival rates of radio-collared adult female bears ............................... 15 Reproductive rates of radio-collared adult female bears ......................... 15 Demographic rates from capture-mark-recapture data........................... 17 Population Viability Analysis .............................................................. 20 DISCUSSION ....................................................................................... 21 Demographic Rate Analysis ................................................................ 22 Population Viability Analysis .............................................................. 24 MANAGEMENT IMPLICATIONS .................................................... 26 SUMMARY ........................................................................................... 26 ACKNOWLEDGMENTS..................................................................... 27 LITERATURE CITED......................................................................... 27 APPENDIX ........................................................................................... 31

INTRODUCTION

becoming increasingly scarce because of human-caused habitat loss and fragmentation. Furthermore, additional threats, such as competition with introduced species, human-caused mortality, or disrupted community dynamics may directly affect population dynamics, resulting in overall population declines. Therefore, recovery of such species can be complex depending on the level of understanding of species biology, causes of decline, and management options for mitigating threats. Once those aspects

Large mammal species are important for the roles they play as keystone species in community dynamics, as contributors to ecosystem functioning, and as umbrella species in the conservation of other species (Maehr et al. 2001). However, restoration of large-mammal populations can be challenging because largebodied species often require large expanses of habitat that are 4

Wildlife Monographs




194

of recovery are well understood and management actions have been implemented, rigorous population assessments are needed to evaluate overall recovery status. The American black bear (Ursus americanus) is the most common of the North American ursids and once occurred throughout the continent from northern Canada into Mexico (Pelton 2003). Since European settlement, the historical range of the black bear has been reduced by 25–35% with most of that reduction occurring in the contiguous United States (Scheick and McCown 2014). Large contiguous populations continue to persist in mountainous regions such as the Rocky and Appalachian mountains, largely because these rugged topographies are less prone to human development and exploitation. In contrast, human development in the Southeastern Coastal Plain has reduced bear populations, which now exist in small vestigial patches of forests consisting of mixtures of bottomland hardwood swamps, pocosins, and pine (Pinus spp.; Wooding et al. 1994). Such small populations are subject to increased probabilities of extinction compared with larger ones simply because of stochastic demographic processes (MacArthur and Wilson 1967, Shaffer 1987, Lande 1993). Hellgren and Vaughan (1994) identified alleviation of negative demographic and genetic consequences caused by habitat loss and fragmentation as conservation and management priorities for southeastern bear populations. The Louisiana black bear (Ursus americanus luteolus) once ranged throughout Louisiana, southern Mississippi, and eastern Texas, and occurred in greatest numbers in the bottomland hardwoods of the Lower Mississippi Alluvial Valley (LMAV; St. Amant 1959). By the 1950s, much of the bottomland hardwoods had been converted to agriculture and, in Louisiana, the bear population was estimated to be 80–120 bears equally distributed between the Tensas River Basin and the coastal portion of the Atchafalaya River Basin (St. Amant 1959). In response to low population numbers, the Louisiana Wildlife and Fisheries Commission (now Louisiana Department of Wildlife and Fisheries [LDWF]) initiated a reintroduction program from 1964 to 1967 during which 161 bears were captured in Cook County, Minnesota and released in Louisiana, 31 in the Tensas River Basin (TRB) and 130 in the Upper Atchafalaya River Basin (UARB; Taylor 1971). Genetic analyses conducted by Triant et al. (2004) and Csiki et al. (2003) indicated that the UARB bears were similar to bears from Minnesota, suggesting that the UARB reintroduction was successful. In 1992, the United States Fish and Wildlife Service (USFWS) granted the Louisiana black bear threatened status under the United States Endangered Species Act of 1973, listing loss and fragmentation of habitat as the primary threats (USFWS 1992). In 1995, a Recovery Plan was drafted and approved that outlined recovery goals designed to meet the objective of reducing threats to the Louisiana black bear and supporting habitat (USFWS 1995). The 1995 Recovery Plan listed the following criteria for delisting: 1) at least 2 viable subpopulations, 1 each in the Tensas and Atchafalaya River Basins; 2) establishment of immigration and emigration corridors between the 2 viable subpopulations; and 3) long-term protection of the habitat and interconnecting corridors that support each of the 2 viable subpopulations used as justification for delisting. A viable subpopulation was defined as Laufenberg et al.




Population Viability of Louisiana Black Bears

having a 95% chance of persistence over 100 years, despite random effects of demography, environment, genetics, and natural catastrophes. A number of studies have been conducted since the Recovery Plan was published and before initiation of our work that have contributed to understanding Louisiana black bear ecology and life-history requirements. Research has focused on movement patterns (Marchinton 1995, Anderson 1997, Wagner et al. 2001, Hightower 2003, Benson and Chamberlain 2007a), habitat needs (Weaver 1990, Nyland 1995, Stinson 1996, Bowman 1999), taxonomy (Warrillow et al. 2001, Kennedy et al. 2002, Csiki et al. 2003, Triant et al. 2004), denning ecology (Weaver and Pelton 1994, Hightower et al. 2002, Crook and Chamberlain 2010), public attitudes (Bowman et al. 2001, Van Why and Chamberlain 2003a), mortality (Pace et al. 2000, Van Why and Chamberlain 2003b), and population abundance (Beausoleil 1999, Boersen et al. 2003, Triant et al. 2004). Most recently, Hooker (2010), Lowe (2011), and Troxler (2013) estimated bear abundances at the Tensas River Basin (TRB), Upper Atchafalaya River Basin (UARB), and Lower Atchafalaya River Basin (LARB), respectively. Additionally, O’Connell-Goode et al. (2014) updated population estimates and evaluated the effects of flooding the Morganza Spillway on bear demographics at UARB and Clark et al. (2015) assessed movement characteristics within and connectivity among Louisiana black bear subpopulations in Louisiana. Management activities have improved recovery prospects for the Louisiana black bear. In 2009, the USFWS designated approximately 484,000 ha of federal, state, and privately owned lands as Critical Habitat for the Louisiana black bear under the Endangered Species Act (USFWS 2009; Fig. 1). Since listing in 1992, 22,263 ha of potential bear habitat were created under the Federal Wetland Reserve Program and 3,654 ha were protected through the establishment of Bayou Teche National Wildlife Refuge, adding to the existing 115,500 ha of federal and state lands within the boundaries of the Critical Habitat designation (USFWS 2009). Additionally, a reintroduction program was conducted from 2001 to 2009 to reestablish a subpopulation in the Three Rivers Complex (TRC) located in east-central Louisiana between the TRB and the UARB (Fig. 1). The primary objective of this program was to translocate breeding-age females with cubs from the TRB to suitable but vacant habitat, thereby establishing another breeding subpopulation to strengthen the network of bear subpopulations in the region. Since inception of the reintroduction program, 48 adult females with 104 cubs have been translocated to the TRC. To determine the probability of persistence of the Louisiana black bear, a unified evaluation of Louisiana black bear recovery throughout the entire LMAV of Louisiana was needed. Thus, our goal was not only to address recovery Criterion 1 in the 1995 Recovery Plan but to use the best available science to assess longterm viability of the assemblage of bear subpopulations within the historical range of Louisiana black bear. With that goal in mind, our objectives were to 1) estimate demographic rates of Louisiana black bear subpopulations, 2) develop data-driven stochastic population projection models, and 3) determine how different assumptions about projection model structure and parameter values affect population trajectories and long-term persistence 5

Figure 1. Map of the study area showing each of the 4 subpopulations of Louisiana black bear (black polygons) and United States Fish and Wildlife Service designated Critical Habitat (cross-hatched areas) located in the Tensas River Basin (TRB), Three Rivers Complex (TRC), Upper Atchafalaya River Basin (UARB), and Lower Atchafalaya River Basin (LARB) within the Lower Mississippi Alluvial Valley in Louisiana, USA. Natural land cover is in gray.

probabilities. For the purposes of this analysis, we included bear subpopulations within the historical range regardless of source (i.e., Minnesota reintroduction).

STUDY AREA Our analysis area included the entire LMAV of Louisiana and western Mississippi, but field data collection was restricted to TRB, TRC, LARB, and UARB (Fig. 1). Most of Louisiana is Outer Coastal Plain Mixed Forest (i.e., uplands) and Lower Mississippi Riverine Forest (i.e., alluvial; U.S. Forest Service 2004). The uplands consisted of prairie and woodlands, whereas the alluvial region included swamps, coastal marshes, beaches, and barrier islands. Elevations ranged from sea level at the coast 6

to 163 m at Driskill Mountain in the uplands. An extensive riverine system consisted of >6,400 km of navigable waterways. The study area had a humid subtropical climate, with long, hot, humid summers and short, mild winters. Average annual temperatures ranged from 168 C to 218 C (http://www. usclimatedata.com/climate/baton-rouge/louisiana/united-states /usla0033, accessed 15 Nov 2014). Rainfall was abundant and well distributed throughout the year; annual precipitation ranged from 102 cm to 153 cm. Historically, much of Louisiana was covered by bottomland deciduous forest with an abundance of ash (Fraxinus spp.), elm (Ulmus spp.), cottonwood (Populus spp.), sugarberry (Celtis laevigata), sweetgum (Liquidambar styraciflua), water tupelo (Nyssa aquatica), oak (Quercus spp.), and bald cypress Wildlife Monographs




194

(Taxodium distichum). Upland areas consisted of loblolly (Pinus taeda) and shortleaf pine (Pinus echinata). Much of the alluvial area had been converted to agriculture, primarily consisting of corn, soybeans, and wheat (Neal 1990).

METHODS Data Sources We used data collected from 4 primary research activities: 1) live capture, 2) winter den visits, 3) radio monitoring of individuals fitted with very high frequency (VHF) transmitters, and 4) non-invasive DNA sampling. Additional data were opportunistically collected from sightings, road mortalities, and human-bear conflict management activities throughout the LMAV of Louisiana. Data collection was conducted by the University of Tennessee, LDWF, U.S. Geological Survey, and Louisiana State University from 2002 to 2012 in the 4 areas supporting breeding subpopulations. Capture and monitoring.—We captured black bears each year from 2002 to 2011 as part of several projects with various research and management objectives including investigations of habitat use, denning ecology, reproduction, survival, movement patterns, and translocation. We trapped bears using modified Aldrich spring-activated foot snares (Aldrich Animal Trap Company, Clallam Bay, WA, USA) or culvert traps. We checked traps once daily except during extremely hot weather (i.e., >358 C) when we checked traps twice daily or disabled them during diurnal hours. We immobilized bears using 4.4 mg of ketamine hydrochloride and 2.2 mg of xylazine hydrochloride per kg or using 45 mg of 1 Telazol (Fort Dodge Animal Health, Fort Dodge, IA, USA) per kg of estimated body mass. After latency, we placed bears in lateral or sternal recumbency, applied sterile ophthalmic lubricant to prevent corneal desiccation, and secured blindfolds to reduce visual stimulation and prevent retinal damage. We monitored rectal temperature, respiration, and pulse throughout each immobilization. We intravenously administered yohimbine hydrochloride at a dosage of 0.2 mg per kg of estimated body mass as an antagonist for xylazine. We equipped females 36 kg (2002–2005) and 45 kg in mass (2006–2011) with mortality-sensitive VHF radio collars (Advanced Telemetry Systems, Isanti, MN, USA; Telonics, Mesa, AZ, USA). All collars incorporated a leather spacer soaked in oil to serve as a release mechanism. Unmarked individuals received unique lip tattoos, plastic ear tags, and passive integrated transponder (PIT) tags. We recorded existing marks, morphometric measurements, estimated age class, general condition, and

reproductive status for all bears. We extracted first upper premolars for age determination by cementum annuli analysis (Willey 1974). Animals were handled according to University of Tennessee Institutional Animal Care and Use Committee (IACUC) protocol number 1716 or Louisiana State University IACUC protocol number A-03-04. We located radio-collared females in the TRB and TRC by VHF signal during January–March 2003–2013 to determine reproductive state (solitary, with cubs, or with yearlings) and to record observed litter size. When feasible, we immobilized females to replace VHF collars approaching the end of battery life using the same immobilization drugs and procedures as those used for live capture. Cubs were weighed, sexed, and implanted with PIT tags. We collected hair samples for DNA analysis. Additionally, we selected and immobilized females for translocation to the TRC from 2001 to 2009 as part of the reintroduction program. We conducted radio monitoring in the TRB and TRC at various intensities and durations according to a variety of research objectives during the time span of this study. From 2003 to 2005, adult females were located by ground telemetry 3 times per week during the active months (i.e., non-denning) from April to November to determine space use of resident bears in the TRB and of bears recently released into the TRC (Benson and Chamberlain 2007a, b). Radio monitoring resumed in the TRC in 2006 and the TRB in 2007 and continued through 2012 with bi-monthly or monthly telemetry flights during active months to monitor survival of adult females. From 2002 to 2012, we located collared females by ground telemetry throughout the nondenning period to conduct post-den emergence observations of family groups in the TRB and TRC to verify reproductive status. Once located, we approached females on foot to determine reproductive state and to record observed litter size. Non-invasive DNA sampling.—Non-invasive DNA sampling was based on the use of molecular markers to obtain unique, multilocus genotypes of individual animals. We used DNA extracted from hair collected at baited, barbed-wire enclosures (hereafter referred to as hair snares) at TRB, UARB, and LARB to determine individual identities and record capture histories for capture-mark-recapture (CMR) analysis (Woods et al. 1999). To ensure that all bears would have opportunities to be sampled, hair snares were spaced with the goal that 4 sites would be available per adult female home range (Otis et al. 1978). Site density, number of sites, and sampling area varied among study areas depending on home range size, area of forested habitat, and accessibility (Table 1).

Table 1. Distribution of black bear hair-collection sites for study areas in the Tensas River Basin (TRB), Upper Atchafalaya River Basin (UARB), and Lower Atchafalaya River Basin (LARB) of Louisiana, USA, 2006–2012. Study area

Home range size (km2)a

Site densityb

Sites per home rangec

n sites

Sampling area size (km2)d

Total hair samples

Sites producing samples per week

Samples per week

TRB UARB LARB

10.0 15.7 11.8

1/3.8 1/5.0 1/5.2

2.63 3.14 2.27

209 115 118

795 575 613

23,312 11,643 3,698

138–206 65–115 90–104

98–1,382 15–607 53–281

a

Adult female home range size estimates for TRB, UARB, and LARB obtained from Smith and Pelton (1990), Wagner (1995), and Murrow et al. (2013), respectively. b Site density ¼ No. of sites/km2. c Sites per home range ¼ Home range  site density. d Sampling area was estimated by circumscribing each site by a circle with a radius equal to that of an adult female home range, merging those circles into a single polygon, and calculating the area contained within that polygon. Note: non-forested habitat was not excluded from these estimates. Laufenberg et al.




Population Viability of Louisiana Black Bears

7

We established hair snares in 2006–2009 consisting of a single strand of 4-point, 15.5-gauge barbed wire stretched around 3–5 trees at 40–50 cm above ground and enclosing an area approximately 5  5 m. Beginning in 2010, we constructed sites using 2 strands of barbed wire, 1 located at 3540 cm and 1 at 6570 cm above ground, to increase the likelihood of collecting hair from bears that avoided detection by crawling under or stepping over the single wire. We baited each site with small amounts of bakery products (e.g., sweet rolls, donuts) and a scent attractant (artificial raspberry or honey flavoring; Mother Murphy’s Laboratories, Greensboro, NC, USA). We checked all sites for hair samples and rebaited them every 7 days for 8 weeks each year. We placed hair samples collected from individual barbs in individually labeled coin envelopes and stored them in a dry location at room temperature until we performed DNA extraction. To ensure sufficient DNA for sequence analysis, we collected only samples with 5 hairs. To prevent contamination with future hair samples, we used a cigarette lighter or propane torch to burn any remaining hair from the barbs and forceps after sample collection. We collected hair using this protocol for 6 years in the TRB (2006–2011), 6 years in the UARB (2007– 2012), and 3 years in the LARB (2010–2012). In 2012 in the TRB, we sampled sites for only 3 weeks because research objectives changed to less intensive long-term monitoring of subpopulation trends. Greater densities of hair snares to ensure all bears have a nonzero probability of being captured are sometimes required for bear populations with relatively small home ranges and high population densities (Settlage et al. 2008). Such site densities can produce a large number of hair samples and genotyping all of them can be cost-prohibitive. Therefore, we selected only a portion of the total number of samples collected (i.e., subsample) for DNA analysis. Our objective was to genotype a subset of collected samples that represented a random sample and ensured spatial coverage and adequate capture probabilities for CMR analyses. Because population densities and size of surveyed areas differed, the number of subsamples and the method of selection varied by subpopulation. Our subsampling objective was to submit a number of viable samples, defined as those samples containing adequate material (i.e., 5 guard hairs or combination of guard hairs and underfur hairs), for DNA analysis on each study area that would achieve an overall weekly capture probability (p) 0.2 (Laufenberg et al. 2013). That number was initially based on our assumptions of what would be required to reach that objective and was increased at later dates as necessary for each year if preliminary analyses indicated estimates of p were inadequate. In the TRB, we ultimately selected 1 viable sample from 75 randomly selected sites each week from those sites that produced 1 collected sample. Within each selected site, we examined those samples in random order to select the first viable sample. If no viable samples were available for a given site, we passed over that site. If the number of unique sites that produced 1 viable sample in a given week was 2 consecutive months but were subsequently re-encountered or recollared (i.e., temporary loss of signal) were right-censored to the month of the last active signal (i.e., removed from the data set for all subsequent months) and re-entered the data set as new individuals during the month they were re-encountered or recollared. Including the missing months could have biased survival estimates high because bears that die usually do not have the opportunity to be re-encountered, unlike bears that live (White and Garrott 1990). Because non-parturient females can be active during the winter den season in Louisiana (Crook and Chamberlain 2010), we did not assume survival was 1.0 during that period and applied the same censoring rule throughout the year.

9

Radio-collared adult female bears with cubs were translocated from the TRB to the TRC during the winter den season from 2001 to 2009 in an effort to reestablish a breeding subpopulation. From the perspective of the TRB subpopulation, those females essentially were losses. However, treating those animals as losses can negatively bias estimates of survival because only radiocollared individuals were exposed to translocation (Clark and Eastridge 2006). Therefore, we right-censored translocated TRB females to the month of translocation and re-entered them into the data set as new TRC females the same month. Loss of radio signal caused by battery depletion, malfunction, or inaccessibility occasionally prevented collar recovery and fate determination. When fates of individuals are unknown, a maximum survival estimate treating missing animals as alive and a minimum survival estimate treating missing animals as dead at time of signal loss can be obtained that provide an upper and lower bound for survival (Heisey and Fuller 1985, Pollock et al. 1989). To bound survival estimates, we constructed 2 data sets: 1 assuming radio-collared bears with unknown fates were alive at the time of signal loss and right censoring those to the month of last active signal (i.e., assumed alive, AA) and 1 assuming they died with mortality assigned to the month of last active signal (i.e., assumed dead, AD). The latter scenario is relevant because poachers sometimes destroy radio collars after killing the animal and, assuming that signal loss is not related to mortality, can produce positively biased estimates of survival. We used both estimates in the population projections to provide pessimistic and optimistic estimates of growth. We used a parametric exponential model of survival time with a constant discrete hazard rate function and a hierarchical modeling approach to estimate subpopulation-specific annual survival rates for female black bears in the TRB and TRC. Annual survival (S) was defined as S i;j ðtÞ ¼ eH i;j ðtÞ ;

P where H i;j ðtÞ ¼ 12 t¼1 hi;j ðtÞ was the cumulative discrete hazard and hi;j ðtÞ ¼ eðdi;j Þ was the unit (monthly) hazard rate. Subscripts i and j indexed years and subpopulations, respectively, and d was defined as the baseline log hazard rate held constant across months within years. To estimate mean annual survival rates and temporal process variances, we treated annual survival rates for each subpopulation as random effects by imposing a hierarchical model structure whereby annual log hazard rates for each study area were modeled as dt ¼ md þ et et  Normal ð0; s 2d Þ

;

random realizations from independent normal hyperdistributions. For each subpopulation, we used Uniform (15, 0) and Uniform (0, 3) priors for md and sd, respectively. Reproductive rates of radio-collared adult female bears.—To estimate reproductive rates for the TRB and TRC, we used reproductive-state data from radio-collared adult females collected during winter den visits. First, we used a multi-state transition modeling approach to estimate the probability that a female was in reproductive state solitary (S), with cubs (C), or with

10

yearlings (Y) during winter given her reproductive state during the previous winter (Schwartz and White 2008). This approach assumes transitions between states are first-order Markovian processes and differs from the classical multi-state CMR modeling approach (Arnason 1972) in that apparent survival and detection probabilities are assumed to be 1. We made that assumption because we analyzed only data from females that survived the transition period and because detection rate was based on radio telemetry. We separately estimated transition probabilities for the TRB and TRC to compare rates between the 2 subpopulations. Because transitions from S to Y and from Y to Y were not biologically possible, we fixed those transition probabilities at 0. To ensure transition probabilities were restricted to the interval [0, 1] and met the unit sum constraint requirement for transitions from 1 state to all other states, we indirectly imposed a Dirichlet prior for transition probabilities by specifying hyperpriors ai,j  gamma(1,1) and the relationship ci;j ¼ ai;j =

Xn j¼1

ai;j ;

where ci,j is the probability of transitioning from state i to state j (Royle and Dorazio 2008, K
ery and Schaub 2012). We assumed that transition probabilities were constant across time and age classes because our data were not adequate to model temporal or individual effects. We next estimated the posterior distributions of stable-state probabilities (i.e., proportion of females in each reproductive state) using matrix decomposition (Caswell 2001). We compared the distributions of stable-state probabilities for females in different reproductive states from the TRB to those of the TRC to identify potential differences in overall reproduction and litter survival processes, averaged across years and age classes. From the estimated stable reproductive-state probabilities, the proportion of females with cubs and proportion with yearlings can be multiplied by the mean litter size of the respective age class to obtain estimates of fecundity and yearling recruitment per breeding-age female, which can be used to project the number of new cubs or recruited yearlings in population projection models. We independently modeled litter-size probabilities for cub and yearling litters as a multinomial process whereby each possible litter size was treated as a categorical response variable on the nominal scale. We used observed litter-size data for females with cubs collected during winter den visits in the TRB and TRC and multinomial logistic regression to estimate the probability of a female producing a 1-, 2-, 3-, or 4-cub litter conditional on the female being in the C reproductive state. Similarly, we used observed litter-size data for females with yearlings to estimate litter-size probabilities for that age class of dependent offspring. We separately estimated litter size probabilities for the TRB and TRC. We used a Dirichlet prior via gamma hyperpriors to ensure probabilities were restricted to the interval [0, 1] and met the unit sum constraint requirement. We derived an estimate for mean litter size by first calculating the posterior distribution for mean litter size as i ¼ L

4 X Pr ðLi;j Þ  Li;j ; j¼1

Wildlife Monographs




194

where Pr(Li,j) is the probability of litter size j and Li,j is litter size j for the ith sample, and then calculating the mode of that distribution. To derive an estimate of the number of female cubs produced per breeding age female (c) for each study area, we first multiplied the Hadamard (i.e., element-wise) product of the mean cub litter size and vectors of the C stable state by 0.5 based on an assumed 1:1 sex ratio for cubs (Reynolds and Beecham 1980, Elowe and Dodge 1989, Hellgren and Vaughan 1989). We then derived a point estimate (i.e., mode) and credible interval of c by from the resulting vector. Similarly, we derived an estimate of the number of yearlings produced per breeding age female (y) by the same process but using posterior distributions for mean yearling litter size and Y stable-state probability. Demographic rates from capture-mark-recapture.—The complete CMR data set consisted of DNA-based detection records of individual bears obtained from hair-collection surveys conducted across arrays of hair snares in the TRB, UARB, and LARB subpopulations. We conducted surveys in a robust-design format consisting of primary sampling occasions (i.e., years) between which each subpopulation was open to gains and losses and secondary occasions (i.e., 8 weeks) within primary occasions during which each subpopulation was geographically and demographically closed (Pollock 1982). Selection of >1 sample for DNA analysis from the same site in a given week occasionally resulted in the same bear being detected more than once during the same week. Additionally, individuals were often detected at >1 site within a given week. We consolidated those multiple within-week detections into single binary detection records. The final CMR data set used for analysis consisted of binary records (yi,k,t) indicating whether individual i (i ¼ 1,. . .,n) was detected during week k (k ¼ 1,. . .,K) of year t (t ¼ 1,. . .,T) where n is the total number of individuals ever detected, K is the number of detection occasions within each year, and T is the number of years. Our general approach to data analysis was to use a hierarchical CMR modeling framework based on a state-space parameterization of the Jolly-Seber model (Royle and Dorazio 2008, Link and Barker 2010) to estimate abundance (N), annual apparent survival (w), annual per-capita recruitment (f), annual realized population rate-of-change (l), and weekly detection probabilities (p) for females in the TRB, UARB, and LARB. Per-capita recruitment was the ratio of the number of new recruits (i.e., in situ births or immigrants large enough in size to encounter the barbed wire) to the total number of current residents (i.e., breeding or non-breeding age) in the subpopulation. As such, it differed from the number of cubs (c) or yearlings (y) that we estimated from the telemetry data. We restricted our analysis to females because vital rates of females are more important determinants of population growth than those of males (Beston 2011) and because per-capita female recruitment rates are more intuitive and comparable to c and y from the telemetry data. We considered annual values of w and f for the TRB and UARB as random quantities arising from separate probability distributions to directly estimate temporal process variation s 2f and s 2f , respectively, while accounting for imperfect detection and sampling variation. We did not attempt to estimate temporal variance in vital rates for the LARB because the number of years of data collection was insufficient for reliable estimation. Laufenberg et al.




Population Viability of Louisiana Black Bears

Additionally, we modeled the relationship between f and N to test for density dependence in that vital rate. We used a parameter-expanded data augmentation methodology to avoid technical problems caused by changes in the parameter space with each draw of the MCMC estimation procedure (Royle et al. 2007, Royle and Dorazio 2012). This approach artificially inflates the number of individuals in the observed data set with a fixed, known number of all-zero detection histories and includes an estimable zero-inflation parameter that represented the probability of inclusion in the population at the beginning of the study (Royle et al. 2007). The basic structure of the state-space model formulation included 2 model components that described 3 ecological state processes of interest (i.e., initial abundance, survival, and recruitment) and 1 model component that described the observation state process (i.e., detection) as follows: 1) a model for initial abundance during the first study year in each subpopulation; 2) a model for the change in abundance over time as a function of survival and recruitment; and 3) a model for the observation (i.e., CMR) data. We first defined a latent state variable matrix Z of dimension M  T where element zi,t indicates whether individual i is alive and has not permanently emigrated from the study area (zi,t ¼ 1) or is dead or has permanently emigrated (zi,t ¼ 0) at time t, M is the sum of the total number of detected individuals across all study years (n) and the number of all-zero detection histories used to augment the data set, and T is the number of study years. We selected a number of individuals with which to augment the observation data for each subpopulation that would result in M >> n and avoid upper truncation of the posterior distribution for N, yet small enough to avoid excess computation time. We selected M values of 400 for the TRB, 150 for the UARB, and 200 for the LARB. We modeled the initial state of each individual in the augmented data set as zi;1  Bernoulli ðcÞ

where zi,1 indicates if individual i is alive and a member of the sampled population at the beginning of the study and c ¼ EðN 1 Þ=M is the inclusion probability (Royle and Dorazio 2008). As a result, initial abundance (N1) was defined as N1 ¼

XM i¼1

zi;1 :

The second component of the ecological state process modeled abundance in years t ¼ 2,. . .,T as zi;t  Bernoulli ðzi;t1 wt þ ð1  ai;t1 Þbt Þ;

where ai;t1 ¼ max ðzi;t ; . . . ; zi;t1 Þ indicates if individual i has already been recruited to the population. The state process equation defines the probability that an individual is alive and a member of the sampled population at time t as w given it was alive and on the study area at time t  1 and as b if the individual had not previously been a member of the sampled population. Note that w ¼ SF, where S is the true annual probability of survival and 11

F is annual probability of fidelity to the study area, because deaths and permanent emigration cannot be distinguished without ancillary information. The parameter b is considered the probability of being recruited into the population either via in situ reproduction or permanent immigration. However, that probability is influenced by M and has no direct biological interpretation. Per-capita recruitment (f) is related to b and is a more intuitive vital rate, which we defined as ft ¼

bt V t1 ; t ¼ 2; . . . ; T ; N t1

P where V t1 ¼ M  i ai;t1 is the number of available recruits. We defined f as the number of bears new to the population (i.e., recruits) divided by the number of resident bears. This definition does not distinguish between in situ recruits that are born in the study area and immigrant recruits that disperse to the study area from adjacent but unsampled areas. However, we limited our analysis to females, which typically display strong natal site fidelity (Elowe and Dodge 1989, Schwartz and Franzmann 1992, Costello 2010). We described the model component for the detection data as yi;j;t  Bernoulli ðzi;t  pi;t Þ;

where pi,t is the weekly detection probability for individual i in year t. To separate sampling variance from process variance for w and f, we treated annual values for each of those vital rates as random variables coming from a common hyperdistribution using an appropriate link function. We described the hyperdistribution for w as logitðwt Þ ¼ mw þ et et  Normalð0; s 2f Þ;

Where mw is the overall mean annual apparent survival on the logit scale, et is the annual deviation from the mean in year t, and s 2f is the temporal process variance governing annual deviations. Similarly, we modeled temporal process variation of f as logðf t Þ ¼ mf þ et et  Normal ð0; s 2f Þ

;

where mf is the overall mean annual recruitment on the log scale, et is the annual deviation from the mean in year t, and s 2f is the temporal process variance. Individual heterogeneity in p is a well-known and prevalent issue when estimating vital rates for black bears from DNA-based CMR data (Tredick et al. 2007, Clark et al. 2010, Laufenberg et al. 2013). However, the most appropriate family of distributions (e.g., beta, log-normal, or finite mixture) used to model individual heterogeneity is not identifiable using databased selection criteria because different families can produce nearly identical data distributions parameterized by different values of N (Link 2003). An alternative approach to selecting a single distribution family is to consider multiple families and base inference on the entire set of models. Therefore, we considered 2 common families of distributions, the logistic-normal (Coull and 12

Agresti 1999, Dorazio and Royle 2003) and the finite-mixture distribution (Pledger 2000). For the logistic-normal distribution (model 1), we defined p for individual i in year t as logit ðpi;t Þ ¼ mt þ ei ei  Normal ð0; s 2 Þ;

where mt is the mean weekly detection probability in year t on the logit scale, ei is the individual deviation from the mean, and s2 is the variance governing random variation among individuals (i.e., random effects). For the finite-mixture distribution (model 2), we defined p for individual i in year t as pi;t ¼ pt;g g  Categorical ðA; pÞ;

where pt,g is the weekly detection probability for mixture g in year t, A is the number of mixtures, and p is a vector defining the probability of an individual belonging to mixture g. This model assumes individuals belong to one of several groups of detectability (i.e., mixtures). For our analysis, we restricted A to 2 mixtures because of data limitations. Detection probabilities likely differed across years in response to annual variation in abundance, distribution of food resources, weather, or other unknown factors. Therefore, we modeled mt and pt,g for the logit-normal and finite-mixture distributions with fixed-effect differences among years. In 2010, the hair-site configuration was modified from a 1-wire system to a 2-wire system that likely influenced the distribution of individual differences in p. To account for this change in sampling methodology, we modeled pre- and post-modification differences in s2 and p as fixedeffects model terms. We assumed no temporal variation or behavioral effects in detection probabilities across weeks within years. To model density dependence in per-capita recruitment, we defined a log-linear model for the relationship between f and N (Lebreton and Gimenez 2013) as logðf t Þ ¼ b0 þ b1 N t þ et ei  Normal ð0; s 2f Þ;

where b0 and b1 are the intercept and slope parameters, respectively, et is the annual deviation from the mean in year t, and s 2f is the temporal process variance governing annual deviations. We imputed information about z for all years between the first and last year of observation to improve efficiency (K
ery and Schaub 2012). We accomplished this by creating a data matrix of known latent states where we recorded a 1 for all years we knew an individual to be alive (e.g., 1 0 0 1 now becomes 1 1 1 1) and used NAs for years for which we had no information (e.g., 1 1 0 0 becomes 1 1 NA NA). A rapidly developing approach for analyzing different types of population data in a single unified framework is integrated population modeling (Besbeas et al. 2002, Brooks et al. 2004, Schaub et al. 2007). This approach combines information collected from different sampling methods into a single population model facilitating simultaneous estimation of Wildlife Monographs




194

multiple vital rates and population processes that could not have been achieved if data sets were separately analyzed. Furthermore, use of integrated population models increases accuracy and precision when different types of data collected on the same vital rate (e.g., CMR and known-fate data) are concurrently analyzed. Although integration of our CMR and known-fate analyses was possible for the TRB study area, we instead chose to directly incorporate known-alive status information into the known latent state matrix for the TRB by matching genotypes of females occurring in the CMR and radio-monitoring data sets. Similarly, we incorporated known-alive status information from bears handled during live-capture efforts for radiocollaring, den-season captures for reproduction assessment, and conflict management activities in the TRB, UARB, and LARB. In addition to including ancillary information about knownalive status, the known latent state matrix can also incorporate information from known mortalities by entering zeros at known times of death. Therefore, we used genotype data collected from known mortalities documented during radio monitoring and road mortality recoveries to incorporate known times of death into the analysis. The TRB females removed and translocated as part of the reintroduction efforts in the TRC were also represented by zeros in the known latent state matrix from the time of their removal to the end of the study. Because our goal was to use posterior distributions estimated from our CMR analysis in a Bayesian population viability analysis, we used knowledge about bear population dynamics to specify informative priors for mw, b1, s 2f , and s 2f to ensure biologically realistic estimates. We used a Uniform(0,5) prior distribution for mw to restrict mean annual apparent survival (w) to plausible values between 0.50 and 0.99 on the probability scale. To avoid values of b1 corresponding to a positive linear relationship between N and f characteristic of an Allee effect, we used a Uniform(5,0) prior distribution. Bayesian estimation of variance parameters for random effects models can be difficult when data are sparse (e.g., short time series of CMR data; White et al. 2009) and often results in overly positive-skewed posterior distributions. Therefore, we imposed an upper bound on temporal variance parameters by using a Uniform(0,1) prior for sw and a Uniform(0,2) prior for sf corresponding to approximate coefficients of variation of 0.5 and 1.0 for logit(w) and log(f), representing biologically realistic magnitudes of variation for those demographic rates. For the remaining demographic parameters, we used the following vague priors in all CMR models: b0Normal(0,0.0001) and cBeta(0.000001,1). We used the median of the posterior distributions as point estimators for s 2f and s 2f because performance of the median typically is better for sparse CMR data sets (White et al. 2009). We also estimated the geometric mean of realized annual population G Þ by calculating the mode of the posterior growth rate ðl G , which we derived from the joint posterior distribution for l distribution of annual estimates of l. We excluded initial estimates of l from the time series used in the derivation procedure because of potential bias associated with those estimates (Hines and Nichols G . 2002) that would cause bias in corresponding estimates of l Determining the demographic segment of the population that is being sampled is important in CMR-based studies of population dynamics because vital-rate estimates and inferences of population dynamics drawn from those estimates pertain only Laufenberg et al.




Population Viability of Louisiana Black Bears

to the sampled population and may not reflect population segments that are not sampled. For DNA-based hair snare studies, young bears may never be detected because they are too small to encounter the barbed wire. To determine whether young bears were part of the sampled population, we performed a search of our DNA-based CMR data set for genotypes that matched genotypes of cubs that were handled on the TRB study area during years of hair sampling. We then searched our DNA-based CMR data set for genotype matches between live-captured bears 1 year old and non-invasively detected bears and determined the age at which each bear was first detected at a hair snare. We tallied the numbers of cubs detected at hair snares and the ages of first detection in hair snares for live-captured bears. Population Viability Analysis We adopted a Bayesian approach to population viability analysis based on stochastic population projection methods to assess probability of persistence of Louisiana black bears in the LMAV of Louisiana. A population projection simply is 1 of many possible population trajectories, some of which are more likely to occur than others, based on a stochastic model with a number of simplifying assumptions that govern population dynamics. Probability of persistence can be inferred from those trajectory outcomes most likely to occur (i.e., extinction vs. persistence) while accounting for uncertainty caused by stochastic population processes. Our goal was to develop a set of models based on a range of biologically reasonable model parameters and assumptions to project subpopulation trajectories and characterize persistence probabilities. We constructed 2 stochastic population projection models for the TRB and UARB subpopulations using CMR-based demographic rate estimates—1 for each method used to model detection heterogeneity (i.e., model 1 and model 2). We used Monte Carlo methods to include 2 types of random variation into each model: temporal (environmental) stochasticity in annual vital rates and demographic stochasticity in individual life-history events. We incorporated temporal process variation by drawing random values each year for transformed values of f and w (i.e., log and logit transforms, respectively) from normal probability distributions defined by estimated means and variances. We incorporated demographic variation by drawing random values for the number of recruits and survivors using appropriate probability distributions (i.e., Poisson and binomial, respectively) defined by the annual values of f, w, and N. We specified a density dependence of f based on the log-linear relationship estimated from the CMR analyses. We included parameter estimation uncertainty into our projections by simulating a population trajectory for each MCMC sample (i.e., n ¼ 30,000) of the joint posterior distribution from our CMR analyses. For all projections, we placed an upper bound on simulated values of f equal to the largest observed annual estimate from each subpopulation for each CMR model. We included this upper limit to avoid overly optimistic effects of extremely large values that could be generated by the density-dependent relationship if sudden declines in abundance occurred during the simulations or by the stochastic process by which temporal variation in f was incorporated. For each projection model, we simulated trajectories over a 100-year period. We derived an estimate of the 13

probability of persistence for each subpopulation under each simulation model by dividing the number of trajectories that remained extant (i.e., N  1) by 30,000. We assumed population projections and subsequent evaluations of long-term persistence applied only to females in age classes 1. We did not project growth for the LARB subpopulation because we had estimates of only 2 interannual periods for each demographic rate (i.e., f, w, and l). We constructed a stochastic age-structured population model (Caswell 2001) for the TRC subpopulation, which was not sampled using the CMR methods, using demographic rates and temporal process variances estimated from the telemetry-based survival and reproduction data. We restricted the population model to age classes 1 because data for individual cub survival were insufficient and we wanted to project population dynamics for the same age classes for the TRC as we did for the 2 other subpopulations (see Results). We simulated population trajectories over a 100-year period using Monte Carlo methods. We assumed a sex ratio for yearling litters of 1:1, age of primiparity for rearing yearling litters as 4, and maximum age as 24. We used estimates of adult female survival rates and temporal process variance from our known-fate survival analysis for all females 2 years of age. For yearling survival, we obtained estimates from the published literature on southeastern bears (Hellgren 1988 [0.78], Lombardo 1993 [0.53], Maddrey 1995 [0.78], Beausoleil 1999 [0.57]). We then calculated the mean and sample standard deviation of those estimates to be used in the simulations. For recruitment of females into the yearling age class, we used a 3step approach that simulated the number of female yearlings produced by each breeding-age female. First, we simulated whether a female had a yearling litter using our estimate of the Y stable reproductive state probability (hereafter referred to as R). We then simulated the number of yearlings produced by females in the Y state using our estimates of yearling litter-size probabilities. Finally, we simulated the number of female yearlings occurring in each litter by treating the sex of each litter member as a Bernoulli trial with probability ¼ 0.5 (i.e., 1:1 sex ratio). Information about the standing-age distribution must be available to specify the starting conditions for simulations to forecast population trajectories using age-structured population models. Because live-capture data were not available for the TRC subpopulation, we constructed a stochastic population model that was individual-based to simulate life-history events (i.e., survival and recruitment of female yearlings) of translocated females that were censored from the radio-monitoring data set or had missing reproductive data before 2013 (i.e., the starting date of the simulations). We combined those simulated events with known life-history events of monitored females and separately repeated that process for each population trajectory to directly incorporate uncertainty about the initial standing-age distribution into our viability assessment. We incorporated multiple sources of stochasticity in vital rates to model population trajectories in the TRC for 100 years similar to the CMR-based simulations. More specifically, the approach we used accounted for uncertainty caused by temporal variation and demographic stochasticity, uncertainty in adult survival rates caused by the 2 ways we handled unknown fates (i.e., AA and 14

AD), and uncertainty in the form and strength of density dependence in reproductive rates. We also incorporated parameter estimation uncertainty by simulating a population trajectory for each MCMC sample of the posterior distributions (i.e., 30,000) from our telemetry-based survival and reproduction data analyses. We derived 4 approximate values of temporal process variation (sR) for R from coefficients of variation calculated for the estimated means and process variances of f based on each CMR model from the TRB and the UARB because we did not have data-based estimates of temporal process variance for reproductive rates in the TRC. These derivations allowed us to incorporate temporal stochasticity for recruitment in our model that reflected observed variation in recruitment within the LMAV. Moreover, because we did not have empirical estimates of density-dependent relationships between N and reproduction in the TRC, we incorporated density dependence by assuming a relationship between N and R based on the Michaelis-Menten function for enzyme kinetics as used by Taylor et al. (2006) and defined as Rt ¼ ðRMAX Þ

CC N t1  1 CC N t1  1 þ

u

;

where RMAX is the value for R estimated from our reproductive state transition analysis, CC is the carrying capacity for the TRC, and u is a shape parameter governing the strength of non-linearity of the density-dependent relationship. We chose this function for its simplicity and flexibility and because it is governed by only 2 parameters, 1 that describes strength of the density-dependent relationship and 1 that describes the density required to zero the affected rate (Taylor 1994). This function also allowed us to compare the effects that the strength of the density-dependent relationship might have on our projections. We derived possible values for CC using an estimate of current bear habitat in the TRC and density estimates from the TRB and UARB. We first quantified the amount of current bear habitat on state or federally owned land or on private land within areas designated as Critical Habitat (848.4 km2; USFWS 2009) based on habitat classification categories reported by Murrow et al. (2013). We then used abundance estimates from the TRB and UARB and effective sampling areas calculated by placing a 1,600m buffer (approximate radius of annual female home range) around respective trapping arrays to obtain density estimates. Those density estimates were then multiplied by the current habitat available at TRC to estimate CC. By deriving 4 estimates of CC (i.e., 2 for TRB-derived estimates and 2 for UARBderived estimates) and simulating separate projections for each, we could assess uncertainty in our projections regarding different possible population densities that the TRC could support. To account for uncertainty in the strength of non-linearity of the density-dependent relationship, we considered simulations with u¼0.1 and u¼0.5. To account for differences in survival-rate estimates caused by assuming unknown fates as right-censored (AA) or mortalities (AD), we ran simulations using both estimates. We restricted combinations of the values of CC and values of sR to come from the same subpopulation and same CMR model from which each of those values were derived, Wildlife Monographs




194

which resulted in 4 combinations. In total, we developed 16 projection models each representing a different combination of model assumptions (e.g., estimates of adult female survival based on unknown fates assumed alive, u ¼ 0.1, and CC and sR based on CMR model 1 from TRB). Similar to the TRB and UARB projections, we placed an upper bound on simulated values of R equal to the largest estimated value observed in the posterior distribution for the Y stable reproductive state probability to avoid overly optimistic effects of extremely large values that could be generated by the stochastic process. We summarized the outcomes of the 30,000 trajectories for each of those models the same as we did for the CMR-based simulations. We repeated this process for the TRB subpopulation to compare persistence probabilities with those from the CMR analyses. Procedures were identical except that we used a standing- age distribution as starting conditions that was derived by multiplying a stable age distribution from a deterministic population matrix model (described below) by an estimate of total abundance for year 2012 from our CMR analysis. We considered values for CC of the density dependence function and values of temporal process variation for R based on estimates from the CMR analysis for TRB. Similar to the TRC, we developed 8 projection models that represented different combinations of model assumptions (e.g., estimates of adult female survival based on unknown fates assumed alive, u ¼ 0.1, and CC, sR, and starting abundance based on CMR model 1 from TRB. We used deterministic population matrix models (Caswell 2001) assuming independence between recruitment and survival to estimate the asymptotic rate of population growth (lAsym) for the TRC and TRB. We parameterized a matrix for each sample from the posterior samples of yearling recruitment (y) and adult female survival (S) to include parameter uncertainty and used the mean of estimates of yearling survival obtained from the published literature on southeastern bears for all matrices. We derived 2 estimates of lAsym each for the TRC and TRB based on mean value for all matrix models that used estimates of S that were either based on assuming unknown fates were alive or dead. We performed all calculations using Program R and we report the mean. Assuming dynamics of individual subpopulations in Louisiana as independent, we calculated the joint probability that Louisiana black bears would persist in 1 subpopulation (i.e., the TRB, TRC, and UARB) as: 1

n Y 1  Pr ðP i Þ; i¼1

where Pr(Pi) is the probability of persistence for subpopulation I and n is the number of subpopulations. We also calculated the joint probability of persistence that bears would persist in either the TRB or UARB by constructing a synchronized projection model that assumed perfect temporal correlation between the 2 subpopulations. We did not include the TRC subpopulation because different parameters were estimated at TRC based on telemetry. We incorporated perfect temporal correlation by synchronizing random annual variations in demographic rates for each subpopulation. Specifically, we defined probability Laufenberg et al.




Population Viability of Louisiana Black Bears

distributions describing random annual fluctuations in survival and recruitment based on temporal variance estimates from a single MCMC sample for 1 subpopulation (i.e., reference) and then used those distributions to generate random values for each year of that subpopulation’s trajectory. For the complement subpopulation trajectory, we rescaled the random values by multiplying each one by the ratio of the estimated mean annual apparent survival for the reference subpopulation to that of the complement subpopulation. We did this to account for differences between the dispersion of annual vital rates of each subpopulation. We repeated the simulation procedure for each MCMC sample resulting in 30,000 trajectories and calculated the probability of persistence as the proportion of synchronized trajectories that resulted in a combined population abundance 1. We calculated the probability that either the TRB or UARB would persist based on 4 different synchronized projection models: 1) TRB as the reference subpopulation and variance estimates based on CMR model 1, 2) TRB as the reference subpopulation and variance estimates based on CMR model 2, 3) UARB as the reference subpopulation and variance estimates based on CMR model 1, and 4) UARB as the reference subpopulation and variance estimates based on CMR model 2.

RESULTS Demographic Rate Analysis Survival rates of radio-collared adult female bears.—From 2002 to 2012, we monitored 86 radio-collared females 2 years of age within the TRB for 305 bear-years and 43 females 2 years of age within the TRC for 208 bear-years. We recorded 4 known mortalities in the TRB and 9 in the TRC. The causes of 3 known mortalities in the TRB were human-related (i.e., vehicle collision or research-related) and the cause of 1 mortality was unknown. In the TRC, 8 mortalities were human-related (i.e., vehicle collision, illegal kill, or capture-related) and 1 mortality was due to natural causes. Assuming unknown fates were mortalities (AD), 10 mortalities would have occurred in the TRB and 16 in the TRC. In general, annual survival rate estimates were higher and less variable for the TRB compared with the TRC regardless of censoring method (Figs. 2 and 3). When unknown fates were right censored (AA), the mean annual survival rate estimate was 0.997 (95% CI ¼ 0.967–0.999) for TRB and 0.990 (95% CI ¼ 0.911–0.997) for the TRC (Fig. 2). Estimates of the mean and temporal process variance for the baseline log hazard rate (d) in the TRB were 7.02 (95% CI ¼ 9.84 to 5.89) and 0.95 (95% CI ¼ 0.07–2.87), respectively, and 5.89 (95% CI ¼ 8.38 to –4.86) and 1.16 (95% CI ¼ 0.20–2.87) in the TRC. Assuming unknown fates were mortalities (AD), mean annual survival rates were 0.970 (95% CI ¼ 0.931–0.992) in the TRB and 0.926 (95% CI ¼ 0.845–0.983) in the TRC (Fig. 3). Estimates of the mean and temporal process variance for d in the TRB was 5.93 (95% CI ¼ 7.33–5.13) and 0.18 (95% CI ¼ 0.03–2.30), respectively, and 4.87 (95% CI ¼ 6.52 to 4.27) and 0.68 (95% CI ¼ 0.07–2.51) in the TRC. Reproductive rates of radio-collared adult female bears.—From 2003 to 2013, we observed 142 transitions among reproductive states for 58 females in the TRB and 74 transitions for 29 females 15

Figure 2. Annual estimates (diamonds) and 95% credible intervals (bars) of adult female survival for Louisiana black bears within the Tensas River Basin (top) and Three Rivers Complex (bottom) in Louisiana, USA, 2002–2012. Estimates assume bears with unresolved fates were alive at time of last contact. Thick dashed lines are mean annual survival estimates and thin dashed lines are 95% credible intervals.

in the TRC. Females in the TRB were more likely to transition to having a litter of cubs from any previous state (i.e., ci,2 for i ¼ 1, 2, 3) compared with females in the TRC (Table 2). Conversely, females in the TRC were more likely to successfully raise a litter of cubs to a litter of yearlings (i.e., c2,3) than females in the TRB. Furthermore, females in the TRC were nearly twice as likely to remain without offspring and half as likely to produce cubs in the year immediately following losing an entire litter of cubs as were females in the TRB. The estimated stable-state probability of females being without offspring was greater in the TRC (Pr(S) ¼ 0.46, 95% CI ¼ 0.31–0.64) than in the TRB (Pr(S) ¼ 0.27, 95% CI ¼ 0.19–0.36; Fig. 4), whereas the estimated probability of females having a litter of cubs was greater in the TRB (Pr(C) ¼ 0.51, 95% CI ¼ 0.45–0.57) compared with the TRC (Pr(C) ¼ 0.34, 95% CI ¼ 0.23–0.43). The proportion of females with yearling litters was nearly identical for the TRB (Pr(Y) ¼ 0.22, 95% CI ¼ 0.16–0.28) and the TRC (Pr(Y) ¼ 0.20, 95% CI ¼ 0.12–0.27). From 2003 to 2013, we observed 130 litters consisting of cubs for 74 females in the TRB and 74 litters for 45 females in the TRC. During the same period, we observed 43 litters consisting of yearlings for 33 females in the TRB and 21 yearling litters for 19 females in the TRC. Although estimated probabilities of females having litters of 1 or 2 cubs were greater in the TRB than in the TRC and probability estimates for 3- or 4-cub litters were greater in the TRC, strong evidence of a true difference existed only for the 3-cub litter category (i.e., minimal overlap of 95% CIs; Fig. 5). Similarly, females in the TRB were more likely to have single-yearling litters and females in TRC were more likely to have 2- and 4-yearling litters, although there was substantial overlap among 95% credible intervals. Mean cub and yearling litter sizes were 1.85 (95% CI ¼ 1.72–1.99) and 1.42 (95% CI ¼ 1.26–1.64) in the TRB, whereas estimates for the TRC were 2.14 (95% CI ¼ 1.94–2.37) and 1.84 (95% CI ¼ 1.55–2.28). Estimates of the number of cubs (c) and number of yearlings (y) produced per breeding-age female for the TRB were 0.47 (95% CI ¼ 0.41–0.54) and 0.20 (95% CI ¼ 0.14–0.28), respectively, whereas estimates for the TRC were 0.32 (95% CI ¼ 0.21–0.40) and 0.18 (95% CI ¼ 0.11–0.27). Estimated asymptotic growth rates based on a deterministic matrix model for the TRB when unresolved fates were AA and AD were 1.046 (95% CI ¼ 0.999– 1.087) and 1.026 (95% CI ¼ 0.974–1.073) and for the TRC were

Table 2. Estimated transition rates between reproductive states for adult female black bears in the Tensas River Basin (TRB) and Three Rivers Complex (TRC), Louisiana, USA, 2003–2013. Transition to reproductive state Current reproductive state

Figure 3. Annual estimates (diamonds) and 95% credible intervals (error bars) of adult female survival for Louisiana black bears within the Tensas River Basin (top) and Three Rivers Complex (bottom) in Louisiana, USA, 2002–2012. Estimates assume bears with unresolved fates were dead at time of last contact. Thick dashed lines are mean annual survival estimates and thin dashed lines are 95% credible intervals. 16

TRB Solitary With cubs With yearlings TRC Solitary With cubs With yearlings a

Solitary

With cubs

With yearlings

0 0.35 (0.23–0.46)a 0.65 (0.54–0.77) 0.25 (0.16–0.38) 0.30 (0.20–0.43) 0.43 (0.31–0.56) 0.20 (0.09–0.39) 0.80 (0.61–0.91) 0 0.66 (0.47–0.82) 0.25 (0.14–0.41) 0.35 (0.15–0.59)

0.34 (0.18–0.53) 0 0.14 (0.06–0.29) 0.60 (0.42–0.73) 0.65 (0.41–0.85) 0

95% credible intervals in parentheses. Wildlife Monographs




194

Figure 4. Posterior distributions for proportions of adult female Louisiana black bears with no litters (top), cubs (center), and yearlings (bottom) within the Tensas River Basin (dark gray) and Three Rivers Complex (light gray) in Louisiana, USA, 2002–2012. Dashed lines are posterior distribution modes.

1.042 (95% CI ¼ 0.969–1.104) and 0.999 (95% CI ¼ 0.912– 1.028), respectively. Demographic rates from capture-mark-recapture data.—Hair collection resulted in a large number of samples for each of the 3 areas (Table 1). Expected heterozygosity for individual microsatellite loci ranged from 0.16 to 0.78 for the TRB, 0.30 to 0.77 for the UARB, and 0.31 to 0.73 for the LARB across all 23 available loci (Table 3). Based on those values, we selected marker sets consisting of 9 loci in the TRB, 7 loci in the UARB, and 8 loci in the LARB for identification of individual bears (Table 3). The overall PIsibs for the TRB was 1.5  103, corresponding to a 1 in 673 chance that a bear shared its multilocus genotype with another bear. Using the Dunn-Sidak method to control the experimentwise error rate, 3 of 9 microsatellite loci violated Hardy-Weinberg expectations (a ¼ 0.006) and 15 of the associations among 36 pairs of loci exhibited linkage disequilibrium (a ¼ 0.001). For the UARB, the overall PIsibs was 3.6  103, corresponding to a 1 in 274 chance that 2 bears shared the same multilocus genotype. None of the 7 microsatelLaufenberg et al.




Population Viability of Louisiana Black Bears

lite loci violated Hardy-Weinberg expectations (a ¼ 0.007) and 2 of 21 loci pairs exhibited linkage disequilibrium (a ¼ 0.002). The overall PIsibs for the LARB was 3.0  103, or a 1 in 337 chance that 2 bears shared the same multilocus genotype. One of 8 loci violated Hardy-Weinberg expectations (a ¼ 0.006) and 2 of 28 locus pairs exhibited linkage disequilibrium (a ¼ 0.002). Extrapolation of mismatch distribution plots indicated that the expected numbers of 0MM-pairs were 1 for the TRB, UARB, and LARB. From 2006 to 2012, we attempted DNA extraction and microsatellite genotyping for 3,821, 1,755, and 1,599 hair samples from hair snares surveyed in the TRB, UARB, and LARB, respectively. The average annual genotyping success rate ranged from 79% to 87% by study area and the average annual percentage of samples that contained hairs from >1 bear ranged from 0.999 (Table 6). Abundance at year 100 for AA-based models averaged 17.8 (SD ¼ 0.14) more female bears (21.4% difference, 101.0 vs. 83.3; Table 6) than AD-based models, which demonstrates the effects of lower survival estimates used in the AD-based models. Average abundance at year 100 for the TRB was less than all abundance estimates from our CMR analysis, which implies our simulations projected a gradual decline of the TRB subpopula20

tion over time. However, our simulation models incorporated components (e.g., upper bound on reproduction) explicitly designed to prevent unrestricted exponential population growth from occurring in our projections that would have resulted in overly optimistic persistence probabilities. Therefore, lower abundance after 100 years is to be expected given incorporation of such preventative measures into our simulations. The estimate of probability of persistence for the TRB subpopulation was 0.954 based on vital rate estimates from model 1 of the CMR analysis and 0.928 based on vital rate estimates from model 2. For the UARB, probabilities of persistence were 0.959 and 0.906 based on vital rate estimates from model 1 and model 2, respectively. Because no cubs-of-theyear handled during winter den captures were ever detected at hair snares and were not part of the sampled population for CMR-based demographic rate analyses, our projections for the TRB and UARB pertain to bears 1 year of age. Assuming dynamics of the TRB and UARB subpopulations were independent and using the most pessimistic persistence probabilities for each, the overall probability that bears would persist in 1 subpopulation after 100 years was >0.993. The probability that bears would persist in either the TRB or UARB assuming perfectly correlated temporal stochasticity was 0.993 from model 1 and 0.988 from model 2 based on TRB temporal variance estimates, and 0.983 from model 1 and 0.958 from Wildlife Monographs




194

Figure 9. Population parameter estimates and 95% credible intervals (error bars) from model 1 (individual capture heterogeneity modeled with logistic-normal distribution, black diamonds) and from model 2 (individual capture heterogeneity modeled with 2-point finite mixture distribution, gray circles) for female Louisiana black bears within the Lower Atchafalaya River Basin in Louisiana, USA, 2010–2012.

model 2 based on UARB temporal variance estimates. Including the TRC, assuming dynamics of all 3 subpopulations were independent, and using the most pessimistic persistence probabilities for each, the overall probability that bears would persist in 1 subpopulation after 100 years was >0.999.

DISCUSSION Our projections for the TRC and TRB based on telemetry data indicated that those Louisiana black bear subpopulations were viable as defined in the Recovery Plan under all model assumptions, whereas the probability of persistence for the TRB and UARB subpopulations based on CMR data met the viability definition for model 1 but not model 2. Persistence probabilities for the TRB subpopulation based on telemetry data were higher (>0.999) than probabilities based on the CMR data (0.928–0.954). The relative differences between telemetry- and CMR-based persistence estimates were likely caused in part by the inclusion of age structure in the telemetry-based projection models through yearling-specific survival rates, which accounted for some heterogeneity in demographic rates caused by agespecific differences among individuals. Incorporating such heterogeneity into population projections would increase overall persistence (White 2000, Morris and Doak 2002). Furthermore, telemetry-based projections did not include emigration processes that could have negatively affected population projections, whereas the CMR-based projections implicitly included those Laufenberg et al.




Population Viability of Louisiana Black Bears

processes by using apparent survival, which represents the probability of surviving and remaining a member of the population. However, the effects of emigration would likely be minimal because our projections were exclusively based on females, which are unlikely to emigrate (Elowe and Dodge 1989, Schwartz and Franzmann 1992, Costello 2010). Regardless, the probability that Louisiana black bears would persist somewhere within the TRB or UARB, even when dynamics of those subpopulations were considered perfectly correlated, was 0.958. This viability analysis does not include persistence probabilities for Louisiana black bears in the TRC, LARB or elsewhere in Louisiana or Mississippi. Inclusion of the TRC increased the joint probability of persistence to >0.999 and including additional subpopulations would only further increase the likelihood of long-term viability of bears in the LMAV. Population viability analysis has come under scrutiny because generic software packages to conduct such analyses are often misused, sufficient time series of data to account for environmental variation are often lacking, and estimation of parameter uncertainty is often ignored or inappropriate (Taylor 1995, Beissinger and Westphal 1998). However, Brooks et al. (2004) conducted separate population viability analyses for 21 species for which sufficient data were available, using half the data for parameter estimation and population forecasting and half for model validation. They found that estimates of extinction risk were reliable regardless of software package used. Our projection 21

Table 5. Summary of 30,000 simulated population trajectories over a 100-year period for female Louisiana black bears in the Three Rivers Complex (TRC), Louisiana, USA. Simulations were based on adult survival rates estimated from radio-telemetry data and reproductive rates estimated from den visit data, incorporated parameter estimation uncertainty, and included different strengths of density dependence (i.e., u ¼ 0.1 or 0.5) using the Michaelis-Menten function for enzyme kinetics. We derived possible values for carrying capacity using an estimate of current bear habitat in the TRC and density estimates from the Tensas River Basin (TRB) and Upper Atchafalaya River Basin (UARB). Simulations were conducted separately for estimates of adult survival rates that treated unresolved radio losses as censored (i.e., assumed alive) and estimates that treated those losses as mortalities (i.e., assumed dead), Louisiana, USA, 2002– 2012. TRC simulation model Assumed alive TRB model 1e, u ¼ 0.1 TRB model 1, u ¼ 0.5 TRB model 2f, u ¼ 0.1 TRB model 2, u ¼ 0.5 UARB model 1g, u ¼ 0.1 UARB model 1, u ¼ 0.5 UARB model 2h, u ¼ 0.1 UARB model 2, u ¼ 0.5 Assumed dead TRB model 1, u ¼ 0.1 TRB model 1, u ¼ 0.5 TRB model 2, u ¼ 0.1 TRB model 2, u ¼ 0.5 UARB model 1, u ¼ 0.1 UARB model 1, u ¼ 0.5 UARB model 2, u ¼ 0.1 UARB model 2, u ¼ 0.5

Mean

LCL

UCL

Pr(N100 > 0)

72.1 71.8 71.8 71.8 71.8 71.9 71.6 72.1

20.0 20.0 20.0 20.0 19.0 20.0 20.0 20.0

128.0 129.0 128.0 128.0 128.0 128.0 128.0 128.0

0.997 0.997 0.997 0.998 0.997 0.997 0.997 0.997

52.7 52.7 52.6 52.5 52.7 52.8 52.7 52.8

19.0 19.0 18.0 18.0 19.0 19.0 18.0 19.0

94.0 95.0 95.0 95.0 95.0 95.0 94.0 95.0

0.997 0.997 0.997 0.997 0.997 0.998 0.997 0.997

a

b

c

d

Table 6. Summary of 30,000 simulated population trajectories over a 100-year period for female Louisiana black bears in the Tensas River Basin (TRB), Louisiana, USA. Simulations were based on adult survival rates estimated from radio-telemetry data and reproductive rates estimated from den visit data, incorporated parameter estimation uncertainty, and included different strengths of density dependence (i.e., u¼0.1 or 0.5) using the Michaelis-Menten function for enzyme kinetics. Simulations were conducted separately for estimates of adult survival rates that treated unresolved radio losses as censored (i.e., assumed alive) and estimates that treated those losses as mortalities (i.e., assumed dead), Louisiana, USA, 2002–2012. TRB simulation model Assumed alive TRB model 1e, u¼0.1 TRB model 1, u¼0.5 TRB model 2f, u¼0.1 TRB model 2, u¼0.5 Assumed dead TRB model 1, u¼0.1 TRB model 1, u¼0.5 TRB model 2, u¼0.1 TRB model 2, u¼0.5

Meana

LCLb

UCLc

Pr(N100 > 0)d

101.0 101.1 101.1 101.0

37.0 38 38 39

151 151 151 151

>0.999 >0.999 >0.999 >0.999

35 34 35 35

129 129 128 128

>0.999 >0.999 >0.999 >0.999

83.4 83.2 83.3 83.2

a

Mean female abundance after 100 years. 2.5% percentile of distribution of abundances after 100 years. c 97.5% percentile of distribution of abundances after 100 years. d Probability of persistence after 100 years. e Carrying capacity (CC), temporal process variance for reproduction (sR), and initial abundance derived from capture-mark-recapture analyses for the TRB, modeling capture heterogeneity with a logit-normal distribution (i.e., model 1). f Carrying capacity (CC), temporal process variance for reproduction (sR), and initial abundance derived from capture-mark-recapture analyses for the TRB, modeling capture heterogeneity with a finite mixture distribution (i.e., model 2). b

a

Mean female abundance after 100 years. 2.5% percentile of distribution of abundances after 100 years. c 97.5% percentile of distribution of abundances after 100 years. d Probability of persistence after 100 years. e Carrying capacity (CC) and temporal process variance for reproduction (sR) derived from capture-mark-recapture analyses for the TRB, modeling capture heterogeneity with a logit-normal distribution (i.e., model 1). f Carrying capacity (CC) and temporal process variance for reproduction (sR) derived from capture-mark-recapture analyses for the TRB, modeling capture heterogeneity with a finite mixture distribution (i.e., model 2). g Carrying capacity (CC) and temporal process variance for reproduction (sR) derived from capture-mark-recapture analyses for the UARB, modeling capture heterogeneity with a logit-normal distribution (i.e., model 1). h Carrying capacity (CC) and temporal process variance for reproduction (sR) derived from capture-mark-recapture analyses for the UARB modeling capture heterogeneity with a finite mixture distribution (i.e., model 2). b

models were tailored to black bear life-history processes, our analysis took place over an extended period, and the analysis explicitly included parameter uncertainty. Our projections were based on the assumption that the environmental and demographic mechanisms regulating population dynamics during our study remain the same for the next 100 years and on assumptions built into the population projection models themselves. The inclusion of covariances among vital rates and subpopulations, the exclusion of density effects, and any number of other modeling choices could change that. However, we attempted to take a conservative (pessimistic) approach and we believe our projections were reasonable and defendable. Demographic Rate Analysis Survival of adult females is a key determinant in population persistence for black bears (Freeman et al. 2003, Hebblewhite et al. 2003, Mitchell et al. 2009), brown bears (Ursus arctos; Mace and Waller 1998, Wielgus et al. 2001, Kovach et al. 2006), 22

and polar bears (Ursus maritimus; Taylor et al. 1987, Eberhardt 1990) given its proportionally greater contribution to population growth rates relative to other vital rates (e.g., subadult survival, cub survival, litter size). Survival rate estimates for Louisiana black bears before federal protection do not exist; thus, we could not compare pre-listing and current mortality risk. However, our telemetry-based annual survival rate estimates for adult females in the TRB were similar to or higher than those from other stable to increasing non-hunted black bear populations in the southeastern United States (Clark and Smith 1994, Bales et al. 2005, Dobey et al. 2005, Wear et al. 2005, Clark and Eastridge 2006), which suggests current mortality rates for females in the TRB are at a sustainable level. Moreover, temporal variation of adult female survival in the TRB was low, which is consistent with predictions suggesting demographic rates most important to population stability should be robust to varying environmental conditions and population densities in long-lived mammals (Fowler 1981, Pfister 1998, Gaillard and Yoccoz 2003). High survival rates with low temporal variation undoubtedly contributed to our finding that persistence probabilities were high in the TRB. Telemetry-based estimates of mean annual adult female survival for the TRC, assuming unknown fates were censored, also were similar to those of stable non-hunted bear populations in the southeastern United States although slightly lower than the TRB (i.e., 0.990 [TRC] vs. 0.997 [TRB]). When unknown fates were assumed dead, the difference between the TRB and TRC doubled in magnitude and mortality risks appeared greater in the TRC. Differences between survival rates in the TRB and TRC likely reflect the added effects of mortality caused by Wildlife Monographs




194

illegal kills; nearly half (4 of 9) the documented mortalities at the TRC were attributed to poaching compared with no poaching-related mortalities documented in the TRB. This finding is in contrast to Benson and Chamberlain (2007a), who recorded no illegal kills of 21 reintroduced adult females during the first 5 years of the reintroduction effort in the TRC. All 4 illegal kills occurred after the Benson and Chamberlain (2007a) study was concluded. A potential explanation is that competition for space and resources increased on protected state and federal lands as subpopulation numbers increased, leading some bears to spend more time on less protected private properties where poaching threats may have been higher. However, such range expansion would only account for half the illegal kills of radio-collared females because 2 of 4 occurred on state-owned Wildlife Management Areas. Alternatively, increasing numbers of illegally killed bears in the TRC may simply be related to a greater number of bears occurring in that area in later years and an increased likelihood of human-bear encounters. Regardless of the underlying cause, the rate of mortality caused by illegal kill in recent years may be an important factor limiting future population growth in the TRC. We chose the mode as our point estimator for most vital rates because it typically is less likely to produce overly pessimistic estimates when posterior distributions are skewed and because we wanted to be as consistent as possible regarding which summary statistic we reported as point estimates. However, the mode may be overly optimistic for some vital rates when most of the probability mass of a posterior distribution is located near the upper bound of distributional support. This situation was observed for adult female survival rates in the TRB and TRC based on known-fate telemetry data, which explains the apparent discrepancy between our point estimates and the number of mortalities relative to the number of monitored bear-years. In practice, selection of the most appropriate summary statistic for Bayesian point estimation should be a decision based on a loss function that takes into account the risk associated with using each potential estimator (Berger 1993). However, our demographic rate analyses were designed to provide the necessary information to conduct a Bayesian population viability analysis, which was based on the entire posterior distribution and not only point estimates. Therefore, we viewed the risk associated with our choice of point estimator to be minimal relative to the main objective of our study. We provide additional summary statistics (i.e., mean, standard deviation, and the 2.5th, 50th, and 97.5th percentiles) for each model parameter from each demographic rate analysis (Appendices A–I). The higher likelihood of females in the TRC successfully raising litters of cubs to yearlings (i.e., transition from C to Y) suggests that cub survival was lower in the TRB. Lower cub survival may be caused by greater competition for resources and greater potential for intraspecific killing because the TRB subpopulation may be closer to carrying capacity. Evidence of density-dependent regulation of dependent offspring survival has been previously reported in ursids (Bunnell and Tait 1981, Clark and Smith 1994, Czetwertynski et al. 2007). In contrast, females in the TRC were less likely to transition from any of the 3 reproductive states to producing a litter of cubs and more likely to Laufenberg et al.




Population Viability of Louisiana Black Bears

remain solitary, indicating breeding success was lower in the TRC. Lower breeding success in the TRC may be related to fewer breeding opportunities caused by few resident breeding males in the area at the onset of the recent reintroduction project (i.e., Allee effect). Mate finding is one of several potential Allee effects that can decrease reproduction at low population levels (Boukal and Berec 2002, Berec et al. 2007). This interpretation was also evidenced by the litter-size data, with females in the TRC producing more cubs and yearlings per litter but with a lower overall litter production rate than females in the TRB. Although our data suggest an Allee effect at TRC resulting in low reproductive rates, this effect was likely due to the preponderance of female founders used in the reintroduction effort, and would not result from natural processes. Despite these differences, stable-state probabilities for yearling litter status for the TRB and TRC were similar, suggesting that the positive effect of higher cub survival was largely offset by potential Allee effects resulting in a higher proportion of non-bred females in the TRC (Courchamp et al. 2008). Estimates of lAsym from the telemetry data for the TRB were 0.999 regardless of the model used for adult female survival rate (i.e., AA vs. AD) compared with stable to slightly decreasing G ). The discrepancy between growth based on CMR estimates (l G and lAsym is to be expected because l G inherently includes l temporal stochasticity in vital rates that cause lower overall future growth rates, whereas lAsym assumes stationary, ergodic conditions and a stable age distribution resulting in higher growth rates (Morris and Doak 2002, Mills 2007). Adult female survival rates and the number of yearlings (y) produced per breeding-age female were slightly lower at the TRC than at the TRB, which contributed to lower lAsym. Such apparent Allee effects could dissipate as cubs born at TRC grow older and reach maturity and as more males immigrate to the TRC. Other researchers have documented low initial growth rates of reintroduced bear populations that dramatically increased in subsequent years (Eastridge and Clark 2001, Murphy et al. 2015). Detection heterogeneity is a common challenge to overcome when estimating population abundance for bears from CMR data because estimates can be greatly influenced by how heterogeneity is modeled. We found that all vital-rate estimates were affected by how we modeled detection heterogeneity and that the specific pattern and extent of those effects differed by population and rate. Estimates differed least for UARB where detection probabilities were highest. Given the sensitivity of vital rates to how detection heterogeneity was modeled, we conclude that fitting multiple heterogeneity models and basing inference on estimates from the complete set was a reliable and conservative approach. Another potential difficulty for DNA-based CMR studies is determining whether all age cohorts within a population are being sampled because age data generally cannot be obtained from DNA. We found that cubs known to be alive and present on the study area were not detected at hair sites. This result is in contrast to Kendall et al. (2009) who concluded their abundance estimates for grizzly bears included all age cohorts. Body size differences between grizzly and black bear cubs and different wire configurations likely lead to this species contrast. Because field 23

collection methods were standardized across all study areas, abundance estimates and demographic rate estimates from CMR analyses for the TRB, UARB, and LARB should be interpreted as pertaining to age cohorts 1 year old. Although the specific patterns of variation in abundance and G indicated stable recruitment differed between models, l population growth for the TRB and positive growth for the UARB; lAsym indicated positive growth at TRB and TRC. Greater variability of growth rates in the UARB may reflect greater environmental variation in recruitment or higher demographic variability caused by the smaller population size (Shaffer 1987, White 2000, Mills 2007). Apparent survival based on the CMR analysis (0.87–0.92) was lower than the estimate from the telemetry data (0.97–0.99) in the TRB. That difference is expected because w from CMR includes emigration, whereas survival based on known-fate analysis does not. This factor contributed to our finding that persistence probabilities based on CMR data were consistently lower for the TRB and UARB compared with the persistence probabilities based on telemetry data for the TRB and TRC. Additionally, the sampled population represented in the CMR data set included younger age classes that typically have lower survival rates (i.e., 1- and 2year-old females) which, when compared with the known-fate data set (i.e., primarily females 3 years old), also may have contributed to the observed differences between CMR-based and telemetry-based survival estimates. Our analysis of CMR data for the TRB and UARB showed evidence for a negative relationship between f and N, which suggests that density-dependent regulatory factors influence dynamics of those subpopulations. Eberhardt (1977) described an ordered sequence of mechanisms by which large mammal populations are regulated as density approaches carrying capacity. Initially, increased intra-specific competition for resources and direct conspecific-caused mortality would be expected at higher densities and cause reductions in survival of dependent offspring and independent juveniles. For bears, several studies have reported evidence supporting such population regulation through an inverse relationship between cub survival and population density (Lindzey et al. 1983, Miller et al. 2003, Schwartz et al. 2006, Czetwertynski et al. 2007, Garrison et al. 2007); such regulation is likely operating in the TRB based on the lower cub survival in that subpopulation compared with the TRC. However, recruitment estimates for our study should be interpreted as in situ recruitment of yearling bears. That interpretation prohibits a clear understanding of which vital rates were being influenced by population density because our data do not separate yearling recruitment into its demographic components (i.e., female reproductive rate and cub survival). Nevertheless, multiple mechanisms likely operate simultaneously to regulate populations that are near carrying capacity (Eberhardt 1977). Such synergistic effects may explain why we detected a density-dependent relationship in our study because per-capita yearling recruitment represents the cumulative effects of multiple demographic processes that may be regulated by density. Also, the longer time period of our study likely contributed to our ability to detect density dependence. Estimating process variation of demographic rates over time is critical for incorporating temporal stochasticity into population

24

projection models used for population viability assessments (White 2000). Reliability of variance estimates in terms of bias and precision for CMR analyses is linked to the number of animals sampled within each year and to the number of years of sampling (Burnham and White 2002, White et al. 2009). Using simulated data sets and Bayesian estimation methods, White et al. (2009) found that estimates of temporal variance for apparent survival based on the mean of the posterior distribution generally were positively biased when the number of animals released was small, the time series of CMR data was relatively short, and the true variance was 0. Conversely, those authors found that the median outperformed the posterior mean and mode except when the true variance was 0, a case that is unlikely to occur in real data sets. Therefore, we chose to base our estimates of temporal process variances on posterior medians, which typically fall between the mean and mode of skewed distributions, thereby minimizing potential bias. We estimated f and l for only 2 intervals at the LARB, and both substantially differed by interval. Whether the large difference across intervals is because population dynamics at LARB truly are more variable or because our study occurred over an unusual sequence of extreme dynamics cannot be determined without a longer time series of data. Moreover, w was considerably lower than at the TRB or UARB, likely because of greater exposure to anthropogenic causes of mortality (e.g., road kills) compared with other Louisiana black bear subpopulations (Pace et al. 2000). Population Viability Analysis The estimate of the probability of persistence for the TRB based on CMR estimates from model 1 met the viability threshold (>95%), whereas the persistence estimate based on demographic rate estimates from model 2 did not. Regardless, several other lines of evidence suggest long-term stability and sustainability of G based on either CMR model the TRB subpopulation. First, l indicated that the subpopulation was stationary during our study period despite the removal and translocation of 25 reproducing adult females and 49 offspring from the TRB to the TRC in 2006–2009 as part of the reintroduction effort. Second, we found evidence of a density-dependent relationship in per-capita recruitment suggesting the TRB subpopulation was at or near a stable, self-regulating population level. Lastly, lAsym and probabilities of persistence from the telemetry data were 0.999, regardless of how unresolved fates of radio-collared adult females were treated. Similar to the TRB, the UARB projections based on CMR estimates from model 1 met the viability threshold, unlike the persistence probability for model 2. We also found evidence of density dependence operating in the UARB and a geometric mean population growth rate of 1.04 during our study. Bears have persisted in the UARB for almost 50 years since the reintroduction in the 1960s, including the years after the opening of the Morganza Spillway in 1973 (Russell 1977), which flooded a large portion of bear habitat. Moreover, negligible effects on apparent survival and site fidelity were found when a more severe flooding event took place in the UARB in 2011 after the Morganza Spillway was again opened (O’Connell-Goode et al. 2014). Although that study did not evaluate the impact of the

Wildlife Monographs




194

2011 flood on recruitment, long-term growth rates of bear populations typically are not greatly affected by occasional recruitment failures characteristic of periodic food shortages because of subsequent synchronized breeding of reproductive females (McLaughlin et al. 1994, Clark et al. 2005, McLellan 2015). Taken collectively, those findings suggest the UARB subpopulation is robust and sustainable. Our projections for the TRC indicated that this subpopulation was viable under all projection modeling assumptions that we tested, but comparisons of persistence probabilities using CMR data at the TRB suggest that the telemetry-based estimates were more optimistic. Nevertheless, survival was high and greater than other increasing reintroduced (0.91; Wear et al. 2005) or naturally recolonizing (0.90; Bales et al. 2005) black bear populations in the southeastern United States. Natural reproduction has occurred at TRC and male immigrants from UARB have been documented (Laufenberg and Clark 2014). We suspect that this population will increase more rapidly once more males become established in the area. Variation of demographic rates among individuals within a population can be caused by phenotypic variation, such as when rates differ across age classes or nutritional condition gradients, or can be caused by genetic variation, wherein fitness varies among individuals in response to natural selection pressure. Whatever the underlying cause, individual heterogeneity often results in some animals having exceptionally high fitness that can shield small populations from extinction during periods of unfavorable environmental conditions and can increase the overall probability of long-term persistence (White 2000). Our DNA-based CMR data did not contain information on age or other individual-level factors that would have allowed us to detect and estimate individual demographic heterogeneity. However, age- and individual-specific differences in vital rates have been documented in other black bear populations (Bunnell and Tait 1985, Elowe and Dodge 1989, Clark and Smith 1994) and were likely operating in the TRB and UARB. Therefore, we conclude our estimates of persistence probabilities based on projections models that did not include individual heterogeneity likely were conservative. We did not explicitly include temporal correlations among subpopulation-specific vital rates in our projections for the TRB and UARB because the length of our time series of CMR data was insufficient to reliably estimate among-parameter covariances. Such correlations can decrease persistence probabilities for the same general reasons as those for among subpopulation correlations (Morris and Doak 2002). However, high means and low variances of adult female survival rates and relatively higher variation in per-capita recruitment indicate population dynamics are primarily driven by recruitment processes rather than survival processes. This relationship would dampen potential covariance effects. Nonetheless, CMR-based monitoring efforts in the TRB and UARB are expected to continue, which should allow estimation of among-parameter covariances and their effects on population dynamics in the future. Incorporating negative density dependence into projection models causes compensatory mechanisms to return populations to equilibrium levels and reduces the overall risk of Laufenberg et al.




Population Viability of Louisiana Black Bears

extinction (Ginzburg et al. 1990). Furthermore, inference about long-term population persistence is sensitive to the specific form of the density-dependent relationship included in a projection model used for population viability analysis (Mills et al. 1996). Therefore, density-dependent relationships should be based on empirical data collected from the population of interest rather than be assumed from population theory or extrapolated from other populations or species. To accomplish that, we estimated the functional relationship between per-capita recruitment and abundance directly from our CMR data set for the TRB and UARB. This analysis allowed us to incorporate data-based regulatory mechanisms into the population projections. However, parameter estimation uncertainty prevented conclusive determination of the form of density dependence, which could result in misleading conclusions about population persistence if that uncertainty was ignored. To account for that uncertainty, our projections for the TRB and UARB explicitly incorporated parameter uncertainty, including the density dependence parameter, into our simulations. We conclude that incorporating density dependence into our projection models was justified and that inferences about the long-term persistence of those subpopulations were reliable. Regulatory mechanisms in large mammals are expected to operate only when populations are near carrying capacity (Eberhardt 1977, Fowler 1981) and may not be realistic for a recently re-established subpopulation such as the TRC. However, population projection models that do not include regulatory mechanisms would result in exponential growth given sufficient vital rates, and also would be unrealistic. Conversely, not incorporating density dependence in demographic rates could eliminate any compensatory response for a small declining population that could mistakenly increase probabilities of extinction. Therefore, we chose to include density dependence to avoid overly optimistic or overly pessimistic probabilities of persistence. Because simulation results showed that long-term persistence was not sensitive to the form of population regulation, we conclude that the forms of density dependence we used did not result in misleading inferences about the viability of the TRC. The possible Allee effects detected in the TRC subpopulation could have implications for estimating extinction probabilities. Small populations are inherently more vulnerable to extinction than larger ones, Allee effects being a potential cause. More explicitly, if Allee effects are operating, small populations can reach a minimum threshold whereby negative density effects begin to operate leading to an extinction vortex from which the population cannot recover (Mills 2007). We did not incorporate Allee effects into our projections because the effect at TRC was likely caused by a skewed sex ratio of the reintroduced bears, which would not likely be duplicated for any of the subpopulations we studied. However, choosing a larger extinction threshold (i.e., pseudo-extinction threshold) of N < 10, for example, might help alleviate the risks of not including Allee effects in our models. We did not do this because there was no biological justification for which value to choose as a pseudoextinction threshold and because a larger threshold was not specified in the Recovery Plan. 25

Finally, our analysis was strictly a demographic assessment; we did not investigate the potential effects of inbreeding or other deleterious genetic effects resulting from small population size and isolation. Likewise, we did not address issues of taxonomy, which are largely unresolved (Warrillow et al. 2001, Csiki et al. 2003, Triant et al 2004, Van Den Busshe et al. 2009, Puckett et al. 2015). We assumed that any bear population within the historical range of U. a. luteolus was potentially part of the overall metapopulation from a demographic standpoint, regardless of origin or taxonomic status. That assumption is supported by Puckett et al. (2015), who suggested that U. a. americanus may be the most accurate subspecies designation for bears across the eastern range based on nuclear and mitochondrial genetic analyses.

MANAGEMENT IMPLICATIONS Large mammals are important components of wildlife communities and recovery of imperiled species is critical to conserving functional ecosystems. Understanding species biology, identifying causes of population declines, and implementing management actions to abate declines are important steps on the road to recovery of at-risk species. Primary causes of Louisiana black bear population decline were identified at the time the subspecies was granted federal protection. Subsequently, recovery objectives were defined and life-history requirements have been extensively studied. Therefore, our main objective was to provide a comprehensive assessment of population status that would provide information essential to evaluating Louisiana black bear recovery. To achieve that goal, we not only addressed recovery Criterion 1 (e.g., at least 2 viable subpopulations, 1 each in the Tensas and Atchafalaya River Basins) in the 1995 Recovery Plan but also used the best available science to assess long-term viability of the assemblage of bear subpopulations within the historical range of Louisiana black bear. Our results indicate that the individual probabilities of persistence of the TRB and UARB subpopulations were slightly >0.95 based on demographic rate estimates from CMR model 1 and from telemetry data, but slightly 0.999), and the presence of the LARB and subpopulations elsewhere in Louisiana and Mississippi further enhance prospects for persistence of the Louisiana black bear. Use of modern Bayesian methods to account for process variation (i.e., temporal and demographic variance) and estimation uncertainty, as we did in our study, has yet to become common practice in evaluating population viability. Explicitly incorporating parameter uncertainty through the use of Bayesian posterior distributions is advantageous because it results in a wider distribution of extinction times that is more likely to

26

contain the true distribution (Wade 2002). Moreover, Bayesian population viability analyses have the added benefit of expressing extinction risk in terms of a frequency-based framework that is more readily incorporated into delisting decisions and adaptive management components of recovery plans (Goodman 2002). Our Bayesian approach to population viability analysis explicitly incorporated multiple sources of variation including parameter uncertainty and process variation, which provided straightforward estimates of persistence probabilities that fully accounted for uncertainty. Decisions about removing the Louisiana black bear from the list of threatened species will ultimately depend on the assumptions of the projections deemed most reasonable, and the level of uncertainty that authorities and stakeholders determine are acceptable. Regardless, monitoring of these bear subpopulations should continue for indications of changes in demographics and to collect more data for estimating temporal process variance. We suggest an active adaptive management approach for Louisiana black bears whereby alternative ways to meet management objectives are explored, the outcomes of alternatives based on the current state of knowledge are predicted, one or more of these alternatives are selected, impacts of management actions are monitored, and knowledge is updated and management actions adjusted (Williams et al. 2007).

SUMMARY














Annual survival rate estimates for adult females in the Tensas River Basin (TRB) and the Three Rivers Complex (TRC) were similar to or higher than those from other non-hunted black bear populations in the southeastern United States. Temporal variation was also low, which is consistent with predictions about life-history strategies in long-lived mammals. Reproductive transitions of females with cubs to females with yearlings (a surrogate of cub survival) were lower in the TRB than TRC, possibly caused by greater competition for resources and manifestation of density dependence at TRB, which may be closer to carrying capacity. In contrast, females in the TRC were more likely to remain solitary, possibly because of Allee effects caused by few resident breeding males in the area at the onset of the reintroduction project. Population growth was stationary at the TRB and was positive in the Upper Atchafalaya River Basin (UARB). Greater variability of growth rates in the UARB may reflect greater environmental variation in recruitment or higher demographic variability caused by the smaller population size. For the TRC, population growth was stable to increasing regardless of how unknown fates were treated in our adult female survival analysis. Our analysis of CMR data for the TRB and UARB showed evidence of a negative relationship between f and N, which suggested that density-dependent regulatory factors influenced dynamics of those subpopulations. Estimates of the probability of persistence for the TRB and UARB subpopulations met the definition of viable as stated in the Recovery Plan based on population projections using demographic rate estimates from CMR model 1, which modeled latent heterogeneity in detection as finite mixtures.

Wildlife Monographs




194

Persistence probabilities based on estimates from CMR model 2, which modeled detection heterogeneity as individual random effects, did not meet the viability criterion for either subpopulation.
Projections for the TRC and TRB based on telemetry data indicated that those subpopulations were viable under all projection model assumptions. A viability analysis was not performed for the LARB subpopulation because our time series was too short to reliably estimate temporal process variance.
Despite uncertainties about the viability of the TRB and UARB subpopulations, the probability that Louisiana black bears would persist somewhere within either of those subpopulations assuming perfectly correlated population dynamics and the most pessimistic parameter estimates was 0.958.
Inclusion of the TRC subpopulation increased persistence probabilities to >0.999 and presence of bears in the LARB and elsewhere in Louisiana and Mississippi increase these persistence probabilities for the overall bear population.

ACKNOWLEDGMENTS Funding was provided by the Louisiana Department of Wildlife and Fisheries, U.S. Fish and Wildlife Service, U.S. Army Corps of Engineers, Roy O. Martin Timber Company, Black Bear Conservation Coalition, the CoyPu Foundation, Louisiana State University, University of Tennessee, and U.S. Geological Survey. Special thanks go to D. Fuller (U.S. Fish and Wildlife Service) for logistical and contractual assistance and to T. White (University of Tennessee) who provided administrative support throughout this project. We gratefully acknowledge K. Van Why, J. F. Benson, S. Ginger, J. Yarkovich, A. C. Crook, and D. Gammons who collected much of the data for the study. We thank the many technicians and land owners that contributed to this project for their hard work and hospitality, respectively. Thanks also go to D. Paetkau for guidance with population genetics analyses and to J. Goad for conducting aerial telemetry. N. Andr
e and X. VelezLiendo provided translations of the abstract. Finally, we extend our appreciation to the editor and 2 anonymous referees for their constructive and insightful comments. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

LITERATURE CITED Anderson, D. R. 1997. Corridor use, feeding ecology, and habitat relationships of black bears in a fragmented landscape in Louisiana. Thesis, University of Tennessee, Knoxville, USA. Arnason, A. N. 1972. Parameter estimates from mark-recapture experiments on two populations subject to migration and death. Researches on Population Ecology (Tokyo) 13:97–113. Bales, S. L., E. C. Hellgren, D. M. Leslie, and J. Hemphill, Jr. 2005. Dynamics of a recolonizing population of black bears in the Ouachita Mountains of Oklahoma. Wildlife Society Bulletin 33:1342–1351. Beausoleil, R. A. 1999. Population and spatial ecology of the Louisiana black bear in a fragmented bottomland hardwood forest. Thesis, University of Tennessee, Knoxville, USA. Beissinger, S. R., and M. I. Westphal. 1998. On the use of demographic models of population viability in endangered species management. Journal of Wildlife Management 62:821–841. Benson, J. F., and M. J. Chamberlain. 2007a. Space use, survival, movements, and reproduction of reintroduced Louisiana black bears. Journal of Wildlife Management 71:2393–2403.

Laufenberg et al.




Population Viability of Louisiana Black Bears

Benson, J. F., and M. J. Chamberlain. 2007b. Space use and habitat selection by female Louisiana black bears in the Tensas River Basin of Louisiana. Journal of Wildlife Management 71:117–126. Berec, L., E. Angulo, and F. Courchamp. 2007. Multiple Allee effects and population management. Trends in Ecology and Evolution 22:185–191. Berger, J. O. 1993. Statistical decision theory and Bayesian analysis. Second edition. Springer-Verlag, New York, New York, USA. Besbeas, P., S. N. Freeman, B. J. T. Morgan, and E. A. Catchpole. 2002. Integrating mark-recapture-recovery and census data to estimate animal abundance and demographic parameters. Biometrics 58:540–547. Beston, J. A. 2011. Variation in life history and demography of the American black bear. Journal of Wildlife Management 75:1588–1596. Boersen, M. R., J. D. Clark, and T. L. King. 2003. Estimating black bear population density and genetic diversity at Tensas River, Louisiana using microsatellite DNA markers. Wildlife Society Bulletin 31:197–207. Boukal, D. S., and L. Berec. 2002. Single-species models of the Allee effect: extinction boundaries, sex ratios and mate encounters. Journal of Theoretical Biology 218:375–394. Bowman, J. L. 1999. An assessment of habitat suitability and human attitudes for black bear restoration in Mississippi. Dissertation, Mississippi State University, Starkville, USA. Bowman, J. L., B. D. Leopold, F. J. Vilella, and D. A. Gill. 2001. Attitudes of landowners toward American black bears compared between areas of high and low bear populations. Ursus 12:153–160. Brooks, S. P., R. King, and B. J. T. Morgan. 2004. A Bayesian approach to combining animal abundance and demographic data. Animal Biodiversity and Conservation 27:515–529. Bunnell, F. L., and D. E. Tait. 1981. Population dynamics of bears—implications. Pages 75–98 in C. W. Fowler, and T. D. Smith, editors. Dynamics of large mammal populations. John Wiley and Sons, New York, New York, USA. Bunnell, F. L., and D. E. Tait. 1985. Mortality rates of North American bears. Arctic 38:316–323. Burnham, K. P., and G. C. White. 2002. Evaluation of some random effects methodology applicable to bird ringing data. Journal of Applied Statistics 29:245–264. Caswell, H. 2001. Matrix population models: construction, analysis, and interpretation. Second edition. Sinauer Association, Sunderland, Masachusettes, USA. Clark, J. D., and R. Eastridge. 2006. Growth and sustainability of black bears at White River National Wildlife Refuge, Arkansas. Journal of Wildlife Management 70:1094–1101. Clark, J. D., R. Eastridge, and M. J. Hooker. 2010. Effects of exploitation on black bear populations at White River National Wildlife Refuge. Journal of Wildlife Management 74:1448–1456. Clark, J. D., J. S. Laufenberg, M. Davidson, and J. L. Murrow. 2015. Connectivity among subpopulations of Louisiana black bears as estimated by a step selection function. Journal of Wildlife Management 79:1347–1360. Clark, J. D., and K. G. Smith. 1994. A demographic comparison of 2 black bear populations in the Interior Highlands of Arkansas. Wildlife Society Bulletin 22:593–603. Clark, J. D., F. T. van Manen, and M. R. Pelton. 2005. Bait stations, hard mast, and black bear population growth in Great Smoky Mountains National Park. Journal of Wildlife Management 69:1633–1640. Costello, C. M. 2010. Estimates of dispersal and home-range fidelity in American black bears. Journal of Mammalogy 91:116–121. Coull, B. A., and A. Agresti. 1999. The use of mixed logit models to reflect heterogeneity in capture-recapture studies. Biometrics 55:294–301. Courchamp, F., J. Berec, and J. Gascoigne. 2008. Allee effects in ecology and conservation. Oxford University Press, Oxford, United Kingdom. Crook, A. C., and M. J. Chamberlain. 2010. A multiscale assessment of den selection by black bears in Louisiana. Journal of Wildlife Management 74:1639–1647. Csiki, I., C. Lam, A. Key, E. Coulter, J. D. Clark, R. M. Pace, K. G. Smith, and D. D. Rhoads. 2003. Genetic variation in black bears in Arkansas and Louisiana using microsatellite DNA markers. Journal of Mammalogy 84:691–701. Czetwertynski, S. M., M. S. Boyce, and F. K. Schmiegelow. 2007. Effects of hunting on demographic parameters of American black bears. Ursus 18:1–18. Dobey, S., D. V. Masters, B. K. Scheick, J. D. Clark, M. R. Pelton, and M. E. Sunquist. 2005. Ecology of Florida black bears in the Okefenokee-Osceola Ecosystem. Wildlife Monographs 158.

27

Dorazio, R. M., and J. A. Royle. 2003. Mixture models for estimating the size of a closed population when capture rates vary among individuals. Biometrics 59:351–364. Eastridge, R., and J. D. Clark. 2001. Experimental reintroduction of black bears to the Big South Fork area of Kentucky and Tennessee. Wildlife Society Bulletin 29:1163–1174. Eberhardt, L. L. 1977. Optimal policies for conservation of large mammals, with special reference to marine ecosystems. Environmental Conservation 4:205– 212. Eberhardt, L. L. 1990. Survival rates required to sustain polar bears. Journal of Wildlife Management 54:587–590. Elowe, K. D., and W. E. Dodge. 1989. Factors affecting black bear reproductive success and cub survival. Journal of Wildlife Management 53:962–968. Ennis, S., and T. F. Gallagher. 1994. A PCR-based sex-determination assay in cattle based on the bovine amelogenin locus. Animal Genetics 25:425–427. Fowler, C. W. 1981. Density dependence as related to life history strategy. Ecology 62:602–610. Freeman, A. H., K. M. Portier, and M. E. Sunquist. 2003. Life history analysis for black bears (Ursus americanus) in a changing demographic landscape. Ecological Modelling 167:47–64. Gaillard, J. M., and N. G. Yoccoz. 2003. Temporal variation in survival of mammals: a case of environmental canalization? Ecology 84:3294–3306. Garrison, E. P., J. W. McCown, and M. K. Oli. 2007. Reproductive ecology and cub survival of Florida black bears. Journal of Wildlife Management 71:720–727. Ginzburg, L. R., S. Ferson, and H. R. Akcakaya. 1990. Reconstructibility of density dependence and the conservative assessment of extinction risk. Conservation Biology 4:63–70. Goodman, D. 2002. Predictive Bayesian population viability analysis: a logic for listing criteria, delisting criteria, and recovery plans. Pages 447–469 in S. R. Beissinger, and D. R. McCullough, editors. Population viability analysis. University of Chicago Press, Chicago, Illinois, USA. Hebblewhite, M., M. Percy, and R. Serrouya. 2003. Black bear (Ursus americanus) survival and demography in the Bow Valley of Banff National Park, Alberta. Biological Conservation 112:415–425. Heisey, D. M., and T. K. Fuller. 1985. Evaluation of survival and cause-specific mortality rates using telemetry data. Journal of Wildlife Management 49:668–674. Hellgren, E. C. 1988. Ecology and physiology of a black bear (Ursus americanus) population in Great Dismal Swamp and reproductive physiology in the captive female black bear. Dissertation, Virginia Polytech Institute and State University, Blacksburg, USA. Hellgren, E. C., and M. R. Vaughan. 1989. Demographic analysis of a black bear population in the Great Dismal Swamp. Journal of Wildlife Management 53:969–977. Hellgren, E. C., and M. R. Vaughan. 1994. Conservation and management of isolated black bear populations in the southeastern coastal plain of the United States. Proceedings of the Annual Conference Southeastern Association of Fish and Wildlife Agencies 48:276–285. Hightower, D. A. 2003. Fine-scaled movements and habitat use of black bears in south central Louisiana. Thesis, Louisiana State University, Baton Rouge, USA. Hightower, D. A., R. O. Wagner, and R. M. Pace. 2002. Denning ecology of female American black bears in south central Louisiana. Ursus 13:11–17. Hines, J. E., and J. D. Nichols. 2002. Investigations of potential bias in the estimation of l using Pradel’s (1996) model for capture-recapture data. Journal of Applied Statistics 29:573–587. Hooker, M. J. 2010. Estimating population parameters of the Louisiana black bear in the Tensas River Basin, Louisiana, using robust design capture-markrecapture. Thesis, University of Tennessee, Knoxville, USA. Kendall, K. C., J. B. Stetz, J. Boulanger, A. C. Macleod, D. Paetkau, and G. C. White. 2009. Demography and genetic structure of a recovering grizzly bear population. Journal of Wildlife Management 73:3–17. Kennedy, M. L., P. K. Kennedy, M. A. Bogan, and J. L. Waits. 2002. Geographic variation in the black bear (Ursus americanus) in the eastern United States and Canada. Southwestern Naturalist 47:257–266. K
ery, M., and M. Schaub. 2012. Bayesian population analysis using WinBUGS: a hierarchical perspective. Academic Press, Amsterdam, Netherlands. Kovach, S. D., G. H. Collins, M. T. Hinkes, and J. W. Denton. 2006. Reproduction and survival of brown bears in southwest Alaska, USA. Ursus 17:16–29.

28

Lande, R. 1993. Risks of population extinction from demographic and environmental stochasticity and random catastrophes. American Naturalist 142:911–927. Laufenberg, J. S., and J. D. Clark. 2014. Population viability and connectivity of the Louisiana black bear (Ursus americanus luteolus). U.S. Geological Survey Open-File Report 2014–1228, Reston, Virginia, USA. Laufenberg, J. S., F. T. van Manen, and J. D. Clark. 2013. Effects of sampling conditions on DNA-based estimates of American black bear abundance. Journal of Wildlife Management 77:1010–1020. Lebreton, J.-D., and O. Gimenez. 2013. Detecting and estimating density dependence in wildlife populations. Journal of Wildlife Management 77:12–23. Lindzey, J. S., G. L. Alt, C. R. McLaughlin, and W. S. Kordek. 1983. Population response of Pennsylvania black bears to hunting. Bears: Their Biology and Management 5:34–39. Link, W. A. 2003. Nonidentifiability of population size from capture-recapture data with heterogeneous detection probabilities. Biometrics 59:1123–1130. Link, W. A., and R. J. Barker. 2010. Bayesian inference with ecological applications. Academic Press, Burlington, Massachusettes, USA. Lombardo, C. A. 1993. The population ecology of black bears on Camp LeJeune, North Carolina. Thesis, University of Tennessee, Knoxville, USA. Lowe, C. L. 2011. Estimating population parameters of the Louisiana black bear in the Upper Atchafalaya River Basin. Thesis, University of Tennessee, Knoxville, USA. MacArthur, R. H., and E. O. Wilson. 1967. The theory of island biogeography. Princeton University Press, Princeton, Jew Jersey, USA. Mace, R. D., and J. S. Waller. 1998. Demography and population trend in grizzly bears in the Swan Mountains, Montana. Conservation Biology 12:1005–1016. Maddrey, R. C. 1995. Morphology, reproduction, food habits, crop depredation, and mortality of black bears on the Neuse-Pamlico Peninsula, North Carolina. Thesis, University of Tennesse, Knoxville, USA. Maehr, D. S., R. F. Noss, and J. L. Larkin, editors. 2001. Large mammal restoration: ecological and sociological challenges in the 21st century. Island Press, Washington, D.C., USA. Marchinton, F. B. 1995. Movement ecology of black bears in a fragmented bottomland hardwood habitat in Louisiana. Thesis, University of Tennessee, Knoxville, USA. McLaughlin, C. R., G. J. Matula, Jr. and R. J. O’Connor. 1994. Synchronous reproduction by Maine black bears. International Conference on Bear Research and Management 9:471–479. McLellan, B. N. 2015. Some mechanisms underlying variation in vital rates of grizzly bears on a multiple use landscape. Journal of Wildlife Management 79:749–765. Miller, S. D., R. A. Sellers, and J. A. Keay. 2003. Effects of hunting on brown bear cub survival and litter size in Alaska. Ursus 14:130–152. Mills, L. S. 2007. Conservation of wildlife populations. Blackwell Publishing, Malden, Massachusettes, USA. Mills, L. S., S. G. Hayes, C. Baldwin, M. J. Wisdom, J. Citta, D. J. Mattson, and K. Murphy. 1996. Factors leading to different viability predictions for a grizzly bear data set. Conservation Biology 10:863–873. Mitchell, M. S., L. B. Pacifici, J. B. Grand, and R. A. Powell. 2009. Contributions of vital rates to growth of a protected population of American black bears. Ursus 20:77–84. Morris, W. F., and D. F. Doak. 2002. Quantitative conservation biology: theory and practice of population viability analysis. Sinauer Associates, Inc., Sunderland, Massachusetts, USA. Murphy, S. M., J. J. Cox, J. D. Clark, B. C. Augustine, J. T. Hast, M. Strunk, D. Gibbs, and S. Dobey. 2015. Rapid growth and genetic diversity retention in an isolated reintroduced black bear population in the Central Appalachians. Journal of Wildlife Management 79:807–818. Murrow, J. L., J. D. Clark, S. Flanagan, and C. Thatcher. 2013. A tool for estimating effects of land-use changes on Louisiana black bear habitat. Final report to the Fish and Wildlife Service, Ecological Services Field Office, Lafayette, Louisiana, USA. Neal, W. A. 1990. Proposed threatened status for the Louisiana black Bear. Federal Register 55:25341–25345. Nyland, P. D. 1995. Black bear habitat relationships in coastal Louisiana. Thesis, Louisiana State University, Baton Rouge, USA. O’Connell-Goode, K. C., C. L. Lowe, and J. D. Clark. 2014. Effects of opening the Morganza Flood Control Structure on a threatened black bear population in Louisiana. Animal Conservation 17:476–485.

Wildlife Monographs




194

Otis, D. L., K. P. Burnham, G. C. White, and D. R. Anderson. 1978. Statistical inference from capture data on closed animal populations. Wildlife Monographs 62:7–135. Pace, R. M. III, D. R. Anderson, and S. Shively. 2000. Sources and patterns of black bear mortality in Louisiana. Proceedings of the Annual Conference Southeastern Association of Fish and Wildlife Agencies 54:365–373. Paetkau, D. 2003. An empirical exploration of data quality in DNA-based population inventories. Molecular Ecology 12:1375–1387. Paetkau, D. 2004. The optimal number of markers in genetic capture-markrecapture studies. Journal of Wildlife Management 68:449–452. Pelton, M. R. 2003. Black bear. Pages 547–555 in G. A. Feldhamer, B. C. Thompson, and J. A. Chapman, editors. Wild mammals of North America. Johns Hopkins University Press, Baltimore, Maryland, USA. Pfister, C. A. 1998. Patterns of variance in stage-structured populations: evolutionary predictions and ecological implications. Proceedings of the National Academy of Sciences 95:213–218. Pledger, S. 2000. Unified maximum likelihood estimates for closed capturerecapture models using mixtures. Biometrics 56:434–442. Plummer, M. N. 2011. rjags: Bayesian graphical models using mcmc. R package version 3–8. http://CRAN.R-project.org/package¼rjags. Accessed 29 Nov 2012. Plummer, M. N. B., K. Cowles, and K. Vines. 2006. CODA: convergence diagnosis and output analysis for MCMC. R News 6:7–11. Pollock, K. H. 1982. A capture-recapture design robust to unequal probability of capture. Journal of Wildlife Management 46:752–757. 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 53:7–15. Puckett, E. E., P. D. Etter, E. A. Johnson, and L. S. Eggert. 2015. Phylogeographic analyses of American black bears (Ursus americanus) suggest four glacial refugia and complex patterns of postglacial admixture. Molecular Biology and Evolution 23:2414–2427. Raymond, M., and F. Rousset. 1995. GENEPOP (Version 1.2)  Population genetics software for exact tests and ecumenicism. Journal of Heredity 86:248–249. Reynolds, D. G., and J. J. Beecham. 1980. Home range activities and reproduction of black bears in west-central Idaho. Bears: Their Biology and Management 4:181–190. Roon, D. A., L. P. Waits, and K. C. Kendall. 2005. A simulation test of the effectiveness of several methods for error-checking non-invasive genetic data. Animal Conservation 8:203–215. Royle, J. A., and R. M. Dorazio. 2008. Hierarchical modeling and inference in ecology: the analysis of data from populations, metapopulations and communities. Academic Press, New York, New York, USA. Royle, J. A., and R. M. Dorazio. 2012. Parameter-expanded data augmentation for Bayesian analysis of capture-recapture models. Journal of Ornithology 152: S521– S537. Royle, J. A., R. M. Dorazio, and W. A. Link. 2007. Analysis of multinomial models with unknown index using data augmentation. Journal of Computational and Graphical Statistics 16:67–85. Russell, M. 1977. Apparent effects of flooding on distriubtion and landings of industrial bottomfish in the northern Gulf of Mexico. Northeast Gulf Science 1:77–78. Schaub, M., O. Gimenez, A. Sierro, and R. Arlettaz. 2007. Use of integrated modeling to enhance estimates of population dynamics obtained from limited data. Conservation Biology 21:945–955. Scheick, B. K., and W. McCown. 2014. Geographic distribution of American black bears in North America. Ursus 25:24–33. Schwartz, C. C., and A. W. Franzmann. 1992. Dispersal and survival of subadult black bears from the Kenia Peninsula, Alaska. Journal of Wildlife Management 56:426–431. Schwartz, C. C., M. A. Haroldson, and G. C. White. 2006. Survival of cub and yearling grizzly bears in the Greater Yellostone Ecosystem, 1983–2001. Pages 18–23 in C. C. Schwartz, M. A. Haroldson, G. C. White, R. B. Harris, S. Cherry, K. A. Keating, D. S. Moody, and C. Servheen, editors. Temporal, spatial, and environmental influences on the demographics of grizzly bears in the Greater Yellowstone Ecosystem. Wildlife Monographs 161. Schwartz, C. C., and G. C. White. 2008. Estimating reproductive rates for female bears: proportions versus transition probabilities. Ursus 19:1–12. Settlage, K. E., F. T. van Manen, J. D. Clark, and T. L. King. 2008. Challenges of DNA-based mark-recapture studies of American black bears. Journal of Wildlife Management 72:1035–1042.

Laufenberg et al.




Population Viability of Louisiana Black Bears

Shaffer, M. 1987. Minimum viable populations: coping with uncertainty. Pages 69–86 in M. E. Soule, editor. Viable populations for conservation. Cambridge University Press, Cambridge, United Kingdom. Smith, T. R., and M. R. Pelton. 1990. Home ranges and movements of black bears in a bottomland hardwood forest in Arkansas. Ursus 8:213–218. Sokal, R. R., and F. J. Rohlf. 1995. Biometry. Third edition. W. H. Freeman and Company, San Francisco, California, USA. St. Amant, L. S. 1959. Louisiana wildlife inventory and management plan. Louisiana Wildlife and Fisheries Commision, Baton Rouge, USA. Stinson, P. H. 1996. Black bears in the Tunica Hills: present status and habitat suitability. Thesis, Louisiana State University, Baton Rouge, USA. Taberlet, P., and G. Luikart. 1999. Non-invasive genetic sampling and individual identification. Biological Journal of the Linnean Society 68:41–55. Taylor, B. L. 1995. The reliability of using population viability analysis for risk classification of species. Conservation Biology 9:551–558. Taylor, E. F. 1971. A radio-telemetry study of the black bear (Euarctos americanus) with notes on its history and present status in Louisiana. Thesis, Louisiana State University, Baton Rouge, USA. Taylor, M. K., editor. 1994. Density-dependent population regulation in black, brown, and polar bears. International Conference on Bear Research and Management Monograph Series No. 3. Taylor, M. K., D. P. DeMaster, F. L. Bunnell, and R. E. Schweinsburg. 1987. Modeling the sustainable harvest of female polar bears. Journal of Wildlife Management 51:811–820. Taylor, M., K. Miroslaw, M. Obbard, H. D. Cluff, and B. Pond. 2006. RISKMAN: stochastic and deterministic population modeling RISK MANagement decision tool for harvested and unharvested populations. Government of Nunavut, Iqaluit, Nunavit Territory, Canada. Tredick, C. A., M. R. Vaughan, D. F. Stauffer, S. L. Simek, and T. Eason. 2007. Sub-sampling genetic data to estimate black bear population size: a case study. Ursus 18:179–188. Triant, D. A., R. M. Pace III, and M. Stine. 2004. Abundance, genetic diversity and conservation of Louisiana black bears (Ursus americanus luteolus) as detected through noninvasive sampling. Conservation Genetics 5:647–659. Troxler, J. C. 2013. Population demographics and genetic structure of black bears in coastal Louisiana. Thesis, University of Tennessee, Knoxville, USA. U.S. Fish and Wildlife Service. 1992. Endangered and threatened wildlife and plants; determination for threatened status for U. a. luteolus (Louisiana black bear). Federal Register 57:588–595. U.S. Fish and Wildlife Service. 1995. Louisiana black bear recovery plan. U.S. Fish and Wildlife Service, Jackson, Mississippi, USA. U.S. Fish and Wildlife Service. 2009. Endangered and threatened wildlife and plants; designation of critical habitat for the Louisiana black bear (Ursus americanus luteolus). Federal Register 74:10350–10409. U.S. Forest Service. 2004. Bailey’s ecoregions and subregions of the United States, Puerto Rico, and the U.S. Virgin Islands: national atlas of the United States. Reston, Virginia, USA. Van Den Busshe, R. A., J. B. Lack, D. P. Onorato, L. C. Gardner-Santana, B. R. McKinney, J. D. Villalobos, M. J. Chamberlain, D. White, and E. C. Hellgren. 2009. Mitochondrial DNA phylogeography of black bear (Ursus americanus) in central and southern North America: conservation implications. Journal of Mammalogy 90:1075–1082. Van Why, K. R., and M. J. Chamberlain. 2003a. Sportsmen knowledge and opinions on restoring the Louisiana black bear. Proceedings of the Southeastern Association of Fish and Wildlife Agencies 57:299–307. Van Why, K. R., and M. J. Chamberlain. 2003b. Mortality of Louisiana black bears associated with elevated train trestles. Canadian Field-Naturalist 117:113115. Wade, P. R. 2002. Bayesian population viability analysis. Pages 213–238 in S. R. Beissinger, and D. R. McCullough, editors. Population viability analysis. Chicago University Press, Chicago, Illinois, USA. Wagner, R. O. 1995. Movement patterns of black bears in south central Louisiana. Thesis, Lousiana State University, Baton Rouge, USA. Wagner, R. O., D. A. Hightower, and R. M. Pace III. 2001. Measuring levels and patterns of activity in black bears. Ursus 12:181–188. Waits, L. P., G. Luikart, and P. Taberlet. 2001. Estimating the probability of identity among genotypes in natural populations: cautions and guidelines. Molecular Ecology 10:249–256. Warrillow, J., M. Culver, E. Hallerman, and M. Vaughan. 2001. Subspecific affinity of black bears in the White River National Wildlife Refuge. Journal of Heredity 92:226–233.

29

Wear, B. J., R. Eastridge, and J. D. Clark. 2005. Factors affecting settling, survival, and viability of black bears reintroduced to Felsenthal National Wildlife Refuge, Arkansas. Wildlife Society Bulletin 33:1363–1374. Weaver, K. M. 1990. The ecology and management of black bears in the Tensas River Basin of Louisiana. Dissertation, University of Tennessee, Knoxville, USA. Weaver, K. M., and M. R. Pelton. 1994. Denning ecology of black bears in the Tensas River Basin of Louisiana. Bears: Their Biology and Management 9:427–433. White, G. C. 2000. Population viability analysis: data requirements and essential analyses. Pages 288–331 in L. Boitani, and T. K. Fuller, editors. Research techniques in Animal Ecology. Columbia University Press, New York, New York, USA. White, G. C., K. P. Burnham, and R. J. Barker. 2009. Evaluation of a Bayesian MCMC random effects inference methodology for capture-mark-recapture data. Pages 1119–1127 in D. L. Thomson, E. G. Cooch, and M. J. Conroy, editors. Modeling demographic processes in marked populations. Springer, New York, New York, USA.

30

White, G. C., and R. A. Garrott. 1990. Analysis of wildlife radio-tracking data. Academic Press, New York, New York, USA. Wielgus, R. B., F. Sarrazin, R. Ferriere, and J. Clobert. 2001. Estimating effects of adult male mortality on grizzly bear population growth and persistence using matrix models. Biological Conservation 98:293–303. Willey, C. H. 1974. Aging black bears from first premolar tooth sections. Journal of Wildlife Management 38:97–100. Williams, B. K., R. C. Szaro, and C. D. Shapiro. 2007. Adaptive management: the U.S. Department of the Interior technical guide. Adaptive Management Working Group, U.S. Department of the Interior, Washington, D.C., USA. Wooding, J. B., J. A. Cox, and M. R. Pelton. 1994. Distribution of black bears in the southeastern coastal plain. Proceedings of the Annual Conference Southeastern Association of Fish and Wildlife Agencies 48:270–275. Woods, J. G., D. Paetkau, D. Lewis, B. N. McLellan, M. Proctor, and C. Strobeck. 1999. Genetic tagging of free-ranging black and brown bears. Wildlife Society Bulletin 27:616–627.

Wildlife Monographs




194

Appendices Appendix A. Summary statistics of posterior samples from parametric exponential models of Louisiana black bear survival based on known-fate telemetry data collected in the Tensas River Basin (TRB) and Three Rivers Complex (TRC) in Louisiana, USA, 2002–2012. Mean

SD

Mode

2.5%

25%

50%

75%

97.5%

Assumed alivea STRB, 2002b STRB, 2003 STRB, 2004 STRB, 2005 STRB, 2006 STRB, 2007 STRB, 2008 STRB, 2009 STRB, 2010 STRB, 2011 STRB, 2012 STRB c md,TRBd sd,TRBe STRC, 2002 STRC, 2003 STRC, 2004 STRC, 2005 STRC, 2006 STRC, 2007 STRC, 2008 STRC, 2009 STRC, 2010 STRC, 2011 STRC, 2012 STRC md,TRC sd,TRC

0.976 0.990 0.989 0.968 0.990 0.990 0.959 0.988 0.986 0.985 0.986 0.989 7.377 1.320 0.941 0.965 0.972 0.976 0.954 0.978 0.937 0.834 0.975 0.926 0.964 0.969 6.261 1.497

0.024 0.011 0.013 0.034 0.012 0.011 0.035 0.016 0.018 0.021 0.021 0.009 1.006 0.810 0.100 0.044 0.033 0.027 0.038 0.024 0.040 0.079 0.028 0.068 0.046 0.023 0.914 0.721

0.990 0.999 0.999 0.990 0.999 0.999 0.985 0.999 0.999 0.998 0.999 0.997 7.018 0.954 0.994 0.997 0.997 0.997 0.986 0.998 0.959 0.864 0.997 0.975 0.996 0.990 5.894 1.158

0.906 0.961 0.956 0.868 0.959 0.961 0.862 0.949 0.941 0.931 0.933 0.967 9.837 0.073 0.648 0.848 0.888 0.903 0.852 0.915 0.833 0.654 0.901 0.746 0.839 0.911 8.380 0.204

0.968 0.986 0.985 0.961 0.985 0.985 0.946 0.983 0.983 0.982 0.982 0.985 7.933 0.645 0.936 0.952 0.960 0.965 0.938 0.968 0.918 0.786 0.964 0.902 0.952 0.956 6.785 0.952

0.983 0.993 0.993 0.980 0.993 0.993 0.970 0.992 0.991 0.991 0.991 0.991 7.193 1.248 0.972 0.980 0.983 0.985 0.963 0.987 0.946 0.844 0.984 0.945 0.979 0.974 6.118 1.446

0.991 0.998 0.998 0.990 0.998 0.998 0.984 0.998 0.997 0.997 0.997 0.996 6.665 1.950 0.992 0.994 0.995 0.995 0.981 0.996 0.966 0.895 0.995 0.971 0.994 0.987 5.591 2.031

0.998 1.000 1.000 0.998 1.000 1.000 0.995 1.000 1.000 1.000 1.000 0.999 5.890 2.872 1.000 1.000 1.000 1.000 0.997 1.000 0.990 0.954 1.000 0.995 1.000 0.997 4.860 2.868

Assumed deadf STRB, 2002 STRB, 2003 STRB, 2004 STRB, 2005 STRB, 2006 STRB, 2007 STRB, 2008 STRB, 2009 STRB, 2010 STRB, 2011 STRB, 2012 STRB md,TRB sd,TRB STRC, 2002 STRC, 2003 STRC, 2004 STRC, 2005 STRC, 2006 STRC, 2007 STRC, 2008 STRC, 2009 STRC, 2010 STRC, 2011 STRC, 2012 STRC md,TRC sd,TRC

0.960 0.974 0.972 0.935 0.964 0.965 0.935 0.956 0.968 0.966 0.967 0.967 6.014 0.795 0.897 0.927 0.937 0.897 0.926 0.932 0.877 0.836 0.943 0.817 0.926 0.920 5.101 0.950

0.026 0.018 0.020 0.047 0.022 0.021 0.040 0.031 0.026 0.027 0.028 0.016 0.561 0.598 0.105 0.057 0.047 0.055 0.044 0.041 0.053 0.071 0.042 0.104 0.058 0.036 0.572 0.632

0.970 0.996 0.979 0.960 0.976 0.973 0.959 0.973 0.977 0.976 0.978 0.970 5.928 0.182 0.927 0.988 0.989 0.921 0.934 0.961 0.903 0.890 0.990 0.889 0.990 0.926 4.872 0.680

0.893 0.932 0.926 0.804 0.910 0.913 0.831 0.873 0.906 0.901 0.896 0.931 7.327 0.030 0.589 0.789 0.829 0.764 0.826 0.841 0.748 0.668 0.846 0.557 0.782 0.845 6.524 0.070

0.949 0.963 0.961 0.920 0.954 0.954 0.918 0.945 0.957 0.956 0.956 0.958 6.301 0.325 0.874 0.898 0.908 0.870 0.901 0.908 0.848 0.796 0.915 0.762 0.897 0.898 5.378 0.470

0.965 0.977 0.975 0.949 0.968 0.968 0.946 0.963 0.972 0.971 0.972 0.969 5.939 0.675 0.920 0.936 0.945 0.906 0.931 0.938 0.886 0.851 0.950 0.846 0.935 0.923 5.006 0.840

0.978 0.989 0.987 0.966 0.980 0.980 0.963 0.976 0.986 0.985 0.985 0.978 5.634 1.131 0.960 0.971 0.975 0.935 0.959 0.963 0.914 0.890 0.978 0.894 0.970 0.946 4.716 1.303

0.994 0.999 0.999 0.985 0.994 0.995 0.982 0.993 0.999 0.999 0.999 0.992 5.131 2.300 0.997 0.998 0.998 0.976 0.990 0.991 0.956 0.936 0.998 0.947 0.998 0.983 4.268 2.511

a

Unresolved fates were assumed alive at time of last contact. Annual survival rate. Mean annual survival rate. d Mean unit (monthly) hazard rate. e Temporal process variance of unit hazard rate expressed as standard deviation. f Unresolved fates were assumed dead at time of last contact. b c

Laufenberg et al.




Population Viability of Louisiana Black Bears

31

Appendix B. Summary statistics of posterior samples from categorical regression model of litter sizes for cub and yearling litters based on den visit data collected for Louisiana black bears in the Tensas River Basin (TRB) and Three Rivers Complex (TRC) in Louisiana, USA, 2003–2013. Mean

SD

Mode

2.5%

25%

50%

75%

97.5%

Cubs uTRB,1a uTRB,2 uTRB,3 uTRB,4  TRB b L uTRC,1 uTRC,1aa uTRC,1 uTRC,1  TRC L

0.38 0.42 0.16 0.04 1.85 0.32 0.29 0.30 0.09 2.15

0.04 0.04 0.03 0.02 0.07 0.05 0.05 0.05 0.03 0.11

0.38 0.42 0.15 0.03 1.85 0.32 0.29 0.30 0.08 2.14

0.30 0.34 0.10 0.01 1.72 0.22 0.20 0.20 0.04 1.94

0.35 0.40 0.14 0.03 1.80 0.28 0.26 0.26 0.07 2.08

0.38 0.42 0.16 0.04 1.85 0.32 0.29 0.29 0.09 2.15

0.41 0.45 0.18 0.05 1.90 0.36 0.33 0.33 0.11 2.23

0.46 0.51 0.22 0.07 1.99 0.43 0.40 0.40 0.16 2.37

Yearlings uTRB,1 uTRB,2 uTRB,3 uTRB,4  TRB L uTRC,1 uTRC,1 uTRC,1 uTRC,1  TRC L

0.66 0.28 0.04 0.02 0.40 0.44 0.04 0.12 1.43 1.88

0.07 0.06 0.03 0.02 0.10 0.10 0.04 0.06 0.10 0.19

0.66 0.27 0.02 0.00 0.39 0.42 0.01 0.10 1.42 1.84

0.52 0.16 0.01 0.00 0.22 0.26 0.00 0.03 1.26 1.55

0.61 0.23 0.02 0.01 0.33 0.37 0.01 0.07 1.36 1.75

0.66 0.27 0.04 0.02 0.40 0.44 0.03 0.11 1.42 1.87

0.71 0.32 0.06 0.03 0.47 0.51 0.06 0.16 1.49 2.00

0.79 0.41 0.11 0.08 0.59 0.63 0.15 0.27 1.64 2.28

a b

Probability of litter size i, where i ¼ 1, 2, 3, or 4. Mean litter size.

Appendix C. Summary statistics of posterior samples from multi-state transition model of reproductive states based on den visit data collected for Louisiana black bears in the Tensas River Basin (TRB) and Three Rivers Complex (TRC) in Louisiana, USA, 2003–2013.

TRB cS,Sa cC,S cY,S cS,C cC,C cY,S cC,Y SSPSb SSPC SSPY cc yd TRC cS,S cC,S cY,S cS,C cC,C cY,C cC,Y SSPS SSPC SSPY c y

Mean

SD

Mode

2.5%

25%

50%

75%

97.5%

0.34 0.26 0.22 0.66 0.31 0.78 0.43 0.27 0.51 0.22 0.47 0.21

0.06 0.06 0.08 0.06 0.06 0.08 0.06 0.04 0.03 0.03 0.03 0.04

0.35 0.25 0.20 0.65 0.30 0.80 0.43 0.27 0.51 0.22 0.47 0.20

0.23 0.16 0.09 0.54 0.20 0.61 0.31 0.19 0.45 0.16 0.41 0.14

0.30 0.22 0.17 0.62 0.27 0.73 0.39 0.24 0.49 0.20 0.45 0.18

0.34 0.25 0.22 0.66 0.31 0.78 0.43 0.27 0.51 0.22 0.47 0.20

0.38 0.29 0.27 0.70 0.35 0.83 0.47 0.30 0.53 0.24 0.49 0.23

0.46 0.38 0.39 0.77 0.43 0.91 0.56 0.36 0.57 0.28 0.54 0.28

0.65 0.26 0.35 0.35 0.16 0.65 0.58 0.47 0.34 0.20 0.31 0.18

0.09 0.07 0.11 0.09 0.06 0.11 0.08 0.08 0.05 0.04 0.05 0.04

0.66 0.25 0.35 0.34 0.14 0.65 0.60 0.46 0.34 0.20 0.32 0.18

0.47 0.14 0.15 0.18 0.06 0.41 0.42 0.31 0.23 0.12 0.21 0.11

0.59 0.21 0.27 0.28 0.12 0.57 0.53 0.41 0.30 0.17 0.28 0.15

0.66 0.26 0.35 0.34 0.15 0.65 0.58 0.46 0.34 0.19 0.31 0.18

0.72 0.31 0.43 0.41 0.20 0.73 0.63 0.52 0.37 0.22 0.35 0.21

0.82 0.41 0.59 0.53 0.29 0.85 0.73 0.64 0.43 0.27 0.40 0.27

Transition probability from reproductive state i to state j, where i and j ¼ S (solitary), C (cubs), or Y (yearling) states. Stable reproductive state probability. c Number of female cubs produced per breeding female. d Number of female yearlings produced per breeding female. a

b

32

Wildlife Monographs




194

Appendix D. Summary statistics of posterior samples for demographic rates from capture-mark-recapture model 1 based on noninvasive DNA data collected from Louisiana black bears in the Tensas River Basin (TRB) in Louisiana, USA, 2006–2012. Detection heterogeneity among individuals modeled using a logit-normal distribution. N2006a N2007 N2008 N2009 N2010 N2011 N2012 NTotalb EB2007c EB2008 EB2009 EB2010 EB2011 EB2012 f2007d f2008 f2009 f2010 f2011 f2012 b0e b1f sfg w2007h w2008 w2009 w2010 w2011 w2012 mwi swj l2007k l2008 l2009 l2010 l2011 l2012 G l l p2006 m p2007 p2008 p2009 p2010 p2011 p2012 s p1wire n s p2wire

Mean

SD

Mode

2.5%

25%

50%

75%

97.5%

143.78 154.22 164.92 161.79 159.29 162.94 157.91 248.43 33.28 24.44 11.47 8.81 16.58 9.81 0.23 0.16 0.07 0.06 0.10 0.06 8.03 0.07 0.55 0.86 0.90 0.91 0.93 0.91 0.91 2.31 0.45 1.10 1.06 0.98 0.98 1.01 0.97 1.00 0.13 0.12 0.14 0.11 0.16 0.16 0.05 1.43 1.06

9.86 8.87 8.51 6.18 6.51 6.99 9.40 7.70 9.44 7.93 5.49 4.85 5.46 4.63 0.07 0.06 0.04 0.03 0.04 0.03 5.00 0.03 0.43 0.04 0.02 0.02 0.03 0.03 0.03 0.28 0.26 0.08 0.06 0.04 0.04 0.04 0.04 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.14 0.11

143.34 152.21 163.90 161.00 159.13 162.84 156.11 246.70 31.85 22.16 11.02 8.83 14.23 9.62 0.22 0.14 0.07 0.05 0.09 0.06 7.16 0.06 0.12 0.89 0.90 0.91 0.92 0.91 0.91 2.23 0.40 1.08 1.05 0.97 0.98 1.00 0.97 1.00 0.13 0.12 0.14 0.11 0.16 0.16 0.05 1.43 1.04

125.00 138.00 150.00 150.00 148.00 150.00 140.00 235.00 16.48 10.79 1.36 0.87 7.76 1.53 0.11 0.07 0.01 0.01 0.05 0.01 0.15 0.15 0.02 0.78 0.85 0.86 0.88 0.85 0.84 1.81 0.03 0.95 0.96 0.90 0.91 0.93 0.88 0.96 0.09 0.09 0.10 0.08 0.12 0.12 0.04 1.18 0.85

137.00 148.00 159.00 158.00 155.00 158.00 152.00 243.00 26.67 18.88 7.60 4.91 12.73 6.55 0.18 0.12 0.05 0.03 0.08 0.04 4.67 0.08 0.21 0.84 0.89 0.89 0.91 0.89 0.89 2.13 0.24 1.04 1.02 0.95 0.95 0.99 0.94 0.99 0.12 0.11 0.13 0.10 0.15 0.15 0.05 1.34 0.98

144.00 154.00 164.00 162.00 159.00 163.00 158.00 248.00 32.69 23.68 11.28 8.61 15.91 9.60 0.23 0.15 0.07 0.05 0.10 0.06 7.51 0.06 0.44 0.87 0.90 0.91 0.93 0.91 0.91 2.29 0.43 1.09 1.06 0.98 0.98 1.01 0.97 1.00 0.13 0.12 0.14 0.11 0.16 0.16 0.05 1.43 1.05

150.00 160.00 170.00 166.00 163.00 167.00 164.00 253.00 39.24 29.34 15.00 12.30 19.84 12.72 0.28 0.19 0.09 0.08 0.13 0.08 10.73 0.05 0.78 0.89 0.92 0.92 0.95 0.93 0.93 2.47 0.64 1.15 1.10 1.01 1.01 1.04 1.00 1.01 0.15 0.14 0.16 0.12 0.17 0.17 0.06 1.52 1.13

164.00 173.00 183.00 175.00 173.00 177.00 176.00 265.00 53.55 42.02 23.10 18.62 29.01 19.63 0.40 0.28 0.15 0.12 0.19 0.12 21.78 0.01 1.64 0.92 0.95 0.95 0.98 0.96 0.97 2.94 0.95 1.27 1.19 1.06 1.07 1.11 1.05 1.04 0.18 0.16 0.19 0.15 0.20 0.20 0.07 1.72 1.29

a

Annual female abundance. Total number of females alive during study. c Expected number of annual female recruits. d Per-capita recruitment rate for females. e Intercept term of log-linear model for density dependence of per-capita recruitment for females. f Slope term of log-linear model for density dependence of per-capita recruitment for females. g Temporal process variance of log-scale per-capita recruitment on for females expressed as standard deviation. h Annual apparent survival rate for females. i Mean annual apparent survival rate for females on logit scale. j Temporal process variance of logit-scaled apparent survival rate for females. k Realized annual population growth rate for females. l Geometric mean of realized annual population growth rates for females. m Mean detection probability on logit scale. n Variance of individual detection probabilities on logit scale expressed as standard deviation. b

Laufenberg et al.




Population Viability of Louisiana Black Bears

33

Appendix E. Summary statistics of posterior samples for demographic rates from capture-mark-recapture model 2 based on noninvasive DNA data collected from Louisiana black bears in the Tensas River Basin (TRB) in Louisiana, USA, 2006–2012. Detection heterogeneity among individuals modeled using a finite-mixture distribution. N2006a N2007 N2008 N2009 N2010 N2011 N2012 NTotalb EB2007c EB2008 EB2009 EB2010 EB2011 EB2012 f2007d f2008 f2009 f2010 f2011 f2012 b0e b1f sfg w2007h w2008 w2009 w2010 w2011 w2012 mwi swj l2007k l2008 l2009 l2010 l2011 l2012 G l l pA,2006m pA,2007 pA,2008 pA,2009 pA,2010 pA,2011 pA,2012 pB,2006 pB,2007 pA,2008 pB,2009 pB,2010 pB,2011 pB,2012 pA,1-wiren pA,2-wire

Mean

SD

Mode

2.5%

25%

50%

75%

97.5%

159.29 136.16 133.72 142.16 142.03 149.69 136.76 240.19 3.67 15.89 23.35 12.69 19.42 5.56 0.02 0.12 0.18 0.09 0.14 0.04 13.11 0.11 0.64 0.85 0.88 0.89 0.90 0.89 0.88 2.04 0.34 0.88 1.00 1.06 0.99 1.03 0.92 1.00 0.06 0.10 0.15 0.12 0.13 0.11 0.07 0.52 0.55 0.61 0.43 0.49 0.51 0.15 0.73 0.74

11.83 7.38 3.74 4.27 4.01 6.02 8.53 6.79 4.99 7.61 6.56 5.36 5.77 3.98 0.04 0.06 0.05 0.04 0.04 0.03 7.51 0.05 0.47 0.04 0.03 0.02 0.03 0.03 0.04 0.23 0.24 0.05 0.07 0.06 0.05 0.05 0.05 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.04 0.05 0.05 0.04 0.04 0.05 0.04 0.05 0.03

160.50 136.97 133.01 141.02 140.96 148.85 137.21 238.57 0.34 15.13 21.43 11.55 16.75 0.96 0.00 0.11 0.16 0.08 0.12 0.01 11.08 0.09 0.27 0.87 0.88 0.89 0.89 0.89 0.89 2.02 0.16 0.88 1.00 1.05 0.98 1.02 0.91 1.00 0.05 0.10 0.15 0.12 0.13 0.11 0.07 0.51 0.54 0.60 0.43 0.49 0.51 0.14 0.73 0.75

135.00 122.00 127.00 135.00 136.00 139.00 120.00 229.00 0.01 2.02 11.89 2.61 9.57 0.17 0.00 0.01 0.09 0.02 0.07 0.00 0.66 0.24 0.03 0.77 0.82 0.84 0.85 0.83 0.79 1.61 0.02 0.78 0.88 0.96 0.90 0.94 0.81 0.95 0.03 0.07 0.12 0.09 0.09 0.08 0.04 0.45 0.46 0.52 0.35 0.41 0.42 0.09 0.63 0.69

152.00 131.00 131.00 139.00 139.00 145.00 131.00 235.00 0.41 10.47 18.73 9.02 15.23 2.37 0.00 0.08 0.14 0.06 0.11 0.02 8.04 0.14 0.27 0.83 0.86 0.87 0.88 0.87 0.86 1.90 0.15 0.85 0.95 1.02 0.95 0.99 0.89 0.98 0.05 0.09 0.14 0.11 0.12 0.10 0.06 0.49 0.51 0.58 0.40 0.46 0.47 0.12 0.70 0.73

160.00 136.00 133.00 142.00 142.00 149.00 137.00 240.00 1.68 15.66 22.82 12.47 18.92 4.92 0.01 0.12 0.17 0.09 0.13 0.03 12.29 0.11 0.54 0.86 0.88 0.89 0.89 0.89 0.88 2.03 0.29 0.88 0.99 1.06 0.98 1.03 0.92 1.00 0.05 0.10 0.15 0.12 0.13 0.11 0.07 0.52 0.54 0.61 0.43 0.49 0.50 0.15 0.73 0.75

167.00 141.00 136.00 145.00 144.00 154.00 142.00 244.00 4.89 20.94 27.46 16.09 23.10 8.06 0.03 0.16 0.21 0.12 0.16 0.05 16.97 0.07 0.92 0.88 0.89 0.90 0.91 0.91 0.90 2.17 0.49 0.90 1.04 1.10 1.02 1.06 0.95 1.01 0.06 0.12 0.16 0.14 0.14 0.12 0.08 0.54 0.58 0.64 0.46 0.52 0.54 0.17 0.76 0.76

182.00 151.00 142.00 152.00 151.00 163.00 154.00 255.00 17.85 31.37 37.53 23.93 32.01 14.63 0.13 0.25 0.29 0.17 0.23 0.10 30.55 0.02 1.77 0.91 0.92 0.93 0.95 0.96 0.95 2.56 0.89 1.00 1.13 1.18 1.09 1.14 1.01 1.04 0.09 0.14 0.19 0.16 0.17 0.15 0.10 0.59 0.65 0.70 0.52 0.59 0.60 0.22 0.82 0.80

a

Annual female abundance. Total number of females alive during study. c Expected number of annual female recruits. d Per-capita recruitment rate for females. e Intercept term of log-linear model for density dependence of per-capita recruitment for females. f Slope term of log-linear model for density dependence of per-capita recruitment for females. g Temporal process variance of log-scale per-capita recruitment on for females expressed as standard deviation. h Annual apparent survival rate for females. i Mean annual apparent survival rate for females on logit scale. j Temporal process variance of logit-scaled apparent survival rate for females. k Realized annual population growth rate for females. l Geometric mean of realized annual population growth rates for females. m Detection probability for mixture A or B. n Probability of individual coming from mixture A with detection probability ¼ pA. b

34

Wildlife Monographs




194

Appendix F. Summary statistics of posterior samples for demographic rates from capture-mark-recapture model 1 based on noninvasive DNA data collected from Louisiana black bears in the Upper Atchafalaya River Basin (UARB) in Louisiana, USA, 2007–2012. Detection heterogeneity among individuals modeled using a logit-normal distribution. N2007a N2008 N2009 N2010 N2011 N2012 NTotalb EB2008c EB2009 EB2010 EB2011 EB2012 f2008d f2009 f2010 f2011 f2012 b0e b1f sfg w2008h w2009 w2010 w2011 w2012 mwi swj l2008k l2009 l2010 l2011 l2012 G l l p2007 m p2008 p2009 p2010 p2011 p2012 s p1wire n s p2wire

Mean

SD

Mode

2.5%

25%

50%

75%

97.5%

26.84 35.48 42.21 39.36 37.39 44.65 65.79 11.62 10.29 3.44 3.37 9.99 0.45 0.29 0.08 0.09 0.27 3.62 0.15 0.86 0.89 0.88 0.87 0.89 0.89 2.09 0.35 1.34 1.17 0.96 0.97 1.16 1.06 0.29 0.37 0.37 0.21 0.17 0.31 0.88 1.47

3.19 1.97 1.69 2.97 3.54 3.51 2.42 4.02 3.40 2.37 2.25 3.72 0.18 0.10 0.06 0.06 0.11 3.56 0.10 0.51 0.04 0.04 0.05 0.04 0.04 0.36 0.26 0.18 0.11 0.08 0.07 0.11 0.05 0.06 0.05 0.05 0.05 0.04 0.06 0.19 0.27

25.00 34.01 41.00 39.01 35.98 43.01 64.98 10.91 8.97 1.90 1.55 7.79 0.42 0.26 0.04 0.04 0.22 1.65 0.09 0.59 0.90 0.89 0.89 0.89 0.90 1.99 0.08 1.31 1.15 0.94 0.95 1.11 1.06 0.29 0.37 0.37 0.20 0.17 0.31 0.82 1.40

22.00 33.00 40.00 34.00 32.00 39.00 62.00 4.48 4.61 0.10 0.27 4.25 0.14 0.12 0.00 0.01 0.10 1.21 0.42 0.05 0.80 0.79 0.76 0.80 0.80 1.44 0.01 1.02 0.98 0.81 0.85 0.98 0.97 0.18 0.27 0.28 0.12 0.09 0.19 0.55 1.01

24.00 34.00 41.00 37.00 35.00 42.00 64.00 8.78 7.87 1.56 1.58 7.21 0.32 0.22 0.04 0.04 0.19 1.16 0.20 0.44 0.87 0.86 0.85 0.86 0.86 1.86 0.14 1.21 1.10 0.91 0.92 1.08 1.03 0.25 0.33 0.34 0.18 0.14 0.27 0.74 1.27

26.00 35.00 42.00 39.00 37.00 44.00 65.00 11.37 9.92 3.10 2.98 9.50 0.43 0.28 0.07 0.08 0.26 2.75 0.13 0.80 0.89 0.88 0.88 0.89 0.89 2.07 0.30 1.32 1.16 0.95 0.97 1.15 1.06 0.29 0.37 0.37 0.21 0.17 0.31 0.86 1.44

29.00 37.00 43.00 41.00 40.00 47.00 67.00 14.13 12.37 4.90 4.79 12.24 0.56 0.36 0.12 0.12 0.34 5.36 0.08 1.23 0.92 0.91 0.91 0.91 0.92 2.31 0.53 1.45 1.24 1.01 1.02 1.23 1.09 0.33 0.40 0.40 0.25 0.20 0.35 1.00 1.63

35.00 40.00 46.00 46.00 45.00 53.00 71.00 20.27 17.82 8.83 8.46 18.52 0.83 0.52 0.21 0.22 0.52 13.06 0.02 1.89 0.96 0.95 0.95 0.95 0.96 2.87 0.92 1.73 1.41 1.12 1.12 1.42 1.15 0.40 0.46 0.46 0.32 0.27 0.44 1.31 2.08

a

Annual female abundance. Total number of females alive during study. c Expected number of annual female recruits. d Per-capita recruitment rate for females. e Intercept term of log-linear model for density dependence of per-capita recruitment for females. f Slope term of log-linear model for density dependence of per-capita recruitment for females. g Temporal process variance of log-scale per-capita recruitment on for females expressed as standard deviation. h Annual apparent survival rate for females. i Mean annual apparent survival rate for females on logit scale. j Temporal process variance of logit-scaled apparent survival rate for females. k Realized annual population growth rate for females. l Geometric mean of realized annual population growth rates for females. m Mean detection probability on logit scale. n Variance of individual detection probabilities on logit scale expressed as standard deviation. b

Laufenberg et al.




Population Viability of Louisiana Black Bears

35

Appendix G. Summary statistics of posterior samples for demographic rates from capture-mark-recapture model 2 based on noninvasive DNA data collected from Louisiana black bears in the Upper Atchafalaya River Basin (UARB) in Louisiana, USA, 2007–2012. Detection heterogeneity among individuals modeled using a finite-mixture distribution. N2007a N2008 N2009 N2010 N2011 N2012 NTotalb EB2008c EB2009 EB2010 EB2011 EB2012 f2008d f2009 f2010 f2011 f2012 b0e b1f sfg w2008h w2009 w2010 w2011 w2012 mwi swj l2008k l2009 l2010 l2011 l2012 G l l pA,2007m pA,2008 pA,2009 pA,2010 pA,2011 pA,2012 pB,2007 pB,2008 pA,2009 pB,2010 pB,2011 pB,2012 pA,1-wiren pA,2-wire

Mean

SD

Mode

2.5%

25%

50%

75%

97.5%

26.41 34.54 41.22 35.65 33.70 41.19 64.15 10.97 10.48 3.06 3.25 9.81 0.45 0.31 0.07 0.09 0.29 4.22 0.18 0.89 0.88 0.86 0.83 0.88 0.88 1.96 0.43 1.33 1.17 0.91 0.97 1.18 1.05 0.21 0.24 0.27 0.14 0.07 0.23 0.65 0.68 0.64 0.62 0.70 0.64 0.66 0.66

4.57 1.60 1.30 1.86 2.33 2.21 1.76 5.06 3.51 1.98 2.27 3.42 0.23 0.11 0.05 0.06 0.11 5.25 0.16 0.51 0.05 0.04 0.05 0.05 0.05 0.36 0.27 0.24 0.12 0.07 0.08 0.12 0.05 0.08 0.05 0.04 0.04 0.03 0.05 0.11 0.06 0.08 0.06 0.07 0.06 0.06 0.09

22.97 32.99 40.00 34.99 32.98 40.00 62.99 11.22 9.10 2.56 1.59 8.80 0.49 0.27 0.06 0.04 0.28 2.02 0.10 0.63 0.88 0.87 0.86 0.87 0.88 1.94 0.10 1.37 1.15 0.91 0.94 1.17 1.04 0.25 0.24 0.27 0.13 0.06 0.22 0.62 0.68 0.63 0.62 0.71 0.63 0.66 0.68

22.00 33.00 40.00 33.00 30.00 38.00 62.00 0.69 4.69 0.03 0.23 4.22 0.02 0.13 0.00 0.01 0.12 1.27 0.57 0.07 0.78 0.77 0.71 0.78 0.79 1.32 0.02 0.89 0.97 0.76 0.83 0.97 0.96 0.05 0.14 0.18 0.07 0.02 0.13 0.46 0.56 0.50 0.50 0.55 0.51 0.53 0.48

23.00 33.00 40.00 34.00 32.00 40.00 63.00 7.78 7.96 1.59 1.46 7.30 0.29 0.23 0.04 0.04 0.21 1.30 0.22 0.49 0.85 0.84 0.80 0.85 0.85 1.72 0.19 1.17 1.09 0.86 0.91 1.09 1.02 0.14 0.20 0.24 0.11 0.05 0.19 0.57 0.64 0.59 0.58 0.66 0.59 0.62 0.60

25.00 34.00 41.00 35.00 33.00 41.00 64.00 11.15 10.13 2.82 2.81 9.44 0.45 0.30 0.07 0.08 0.28 2.94 0.14 0.84 0.88 0.87 0.84 0.88 0.89 1.94 0.40 1.33 1.16 0.91 0.96 1.17 1.05 0.21 0.24 0.27 0.13 0.07 0.23 0.64 0.68 0.64 0.62 0.71 0.63 0.66 0.66

28.00 35.00 42.00 37.00 35.00 42.00 65.00 14.31 12.64 4.24 4.65 11.90 0.60 0.37 0.10 0.13 0.36 5.48 0.09 1.27 0.91 0.89 0.87 0.91 0.92 2.18 0.64 1.49 1.24 0.96 1.02 1.25 1.08 0.27 0.27 0.30 0.16 0.09 0.26 0.71 0.72 0.70 0.66 0.75 0.68 0.70 0.72

38.00 39.00 44.00 40.00 39.00 46.00 68.00 20.80 18.25 7.55 8.42 17.40 0.90 0.54 0.19 0.24 0.54 17.51 0.02 1.90 0.96 0.94 0.92 0.96 0.97 2.78 0.95 1.80 1.42 1.05 1.14 1.44 1.14 0.35 0.34 0.36 0.22 0.12 0.32 0.87 0.81 0.81 0.75 0.83 0.76 0.77 0.81

a

Annual female abundance. Total number of females alive during study. c Expected number of annual female recruits. d Per-capita recruitment rate for females. e Intercept term of log-linear model for density dependence of per-capita recruitment for females. f Slope term of log-linear model for density dependence of per-capita recruitment for females. g Temporal process variance of log-scale per-capita recruitment on for females expressed as standard deviation. h Annual apparent survival rate for females. i Mean annual apparent survival rate for females on logit scale. j Temporal process variance of logit-scaled apparent survival rate for females. k Realized annual population growth rate for females. l Geometric mean of realized annual population growth rates for females. m Detection probability for mixture A or B. n Probability of individual coming from mixture A with detection probability ¼ pA. b

36

Wildlife Monographs




194

Appendix H. Summary statistics of posterior samples for demographic rates from capture-mark-recapture model 1 based on noninvasive DNA data collected from Louisiana black bears in the Lower Atchafalaya River Basin (LARB) in Louisiana, USA, 2010–2012. Detection heterogeneity among individuals modeled using a logit-normal distribution. N2010a N2011 N2012 NTotalb EB2011c EB2012 f2011d f2012 w2011e w2012 l2011f l2012 p2010 g p2011 p2012 sph

Mean

SD

Mode

2.5%

25%

50%

75%

97.5%

101.14 81.77 89.11 121.60 0.24 20.19 0.00 0.25 0.80 0.83 0.80 1.08 0.09 0.19 0.18 1.21

10.99 8.58 8.61 11.24 1.21 6.47 0.01 0.09 0.06 0.06 0.06 0.11 0.02 0.04 0.04 0.21

96.03 77.49 85.56 116.26 0.00 18.74 0.00 0.23 0.82 0.84 0.82 1.07 0.09 0.19 0.18 1.14

85.00 69.00 76.00 105.00 0.00 8.90 0.00 0.10 0.68 0.71 0.69 0.89 0.05 0.10 0.10 0.86

93.00 76.00 83.00 114.00 0.00 15.62 0.00 0.19 0.76 0.79 0.77 1.01 0.08 0.16 0.15 1.07

99.00 80.00 88.00 120.00 0.00 19.74 0.00 0.24 0.80 0.84 0.81 1.08 0.09 0.19 0.18 1.19

107.00 86.00 94.00 128.00 0.00 24.12 0.00 0.30 0.84 0.88 0.84 1.15 0.11 0.21 0.21 1.34

128.00 103.00 109.00 149.00 3.09 34.37 0.03 0.44 0.90 0.94 0.91 1.30 0.14 0.27 0.25 1.67

a

Annual female abundance. Total number of females alive during study. c Expected number of annual female recruits. d Per-capita recruitment rate for females. e Annual apparent survival rate for females. f Realized annual population growth rate for females. g Mean detection probability on logit scale. h Variance of individual detection probabilities on logit scale expressed as standard deviation. b

Appendix I. Summary statistics of posterior samples for demographic rates from capture-mark-recapture model 2 based on noninvasive DNA data collected from Louisiana black bears in the Lower Atchafalaya River Basin (LARB) in Louisiana, USA, 2010–2012. Detection heterogeneity among individuals modeled using a finite-mixture distribution. a

N2010 N2011 N2012 NTotalb EB2011c EB2012 f2011d f2012 w2011e w2012 l2011f l2012 pA,2010g pA,2011 pA,2012 pB,2010 pB,2011 pB,2012 pAh

Mean

SD

Mode

2.5%

25%

50%

75%

97.5%

89.12 70.62 79.45 109.36 0.14 19.90 0.00 0.29 0.79 0.83 0.79 1.12 0.06 0.14 0.16 0.36 0.62 0.47 0.69

6.39 4.24 6.06 5.64 0.82 6.10 0.01 0.09 0.06 0.07 0.06 0.13 0.02 0.03 0.04 0.05 0.05 0.05 0.05

86.25 69.02 76.69 107.81 0.00 18.91 0.00 0.27 0.80 0.84 0.80 1.11 0.06 0.13 0.16 0.35 0.63 0.46 0.70

79.00 64.00 70.00 100.00 0.00 8.76 0.00 0.11 0.66 0.68 0.66 0.88 0.03 0.09 0.09 0.27 0.52 0.38 0.59

85.00 68.00 75.00 105.00 0.00 15.83 0.00 0.22 0.75 0.78 0.75 1.03 0.05 0.12 0.13 0.32 0.59 0.44 0.66

88.00 70.00 79.00 109.00 0.00 19.61 0.00 0.28 0.79 0.83 0.79 1.11 0.06 0.14 0.16 0.35 0.62 0.47 0.69

93.00 73.00 83.00 113.00 0.00 23.69 0.00 0.34 0.83 0.89 0.83 1.20 0.07 0.15 0.19 0.39 0.66 0.50 0.73

104.00 80.00 93.00 122.00 1.54 32.69 0.02 0.48 0.90 0.96 0.90 1.38 0.10 0.19 0.24 0.45 0.72 0.58 0.78

a

Annual female abundance. Total number of females alive during study. c Expected number of annual female recruits. d Per-capita recruitment rate for females. e Annual apparent survival rate for females. f Realized annual population growth rate for females. g Detection probability for mixture A or B. h Probability of individual coming from mixture A with detection probability ¼ pA. b

Laufenberg et al.




Population Viability of Louisiana Black Bears

37