Estimating the population size and density of Italian wall lizards (Podarcis siculus) through photo identification capture-recapture

The population density of three lacertid lizards (Podarcis sicula, Podarcis muralis, Lacerta viridis) was studied along several transects crossing agro-forest habitats in Mediterranean central Italy. Overall, seven transects, in three different wooded patches, were walked for lizards. Distance sampling (with uniform model design) was applied to the dataset in order to calculate population size, dispersion, and coefficient of variation at each site. In order to detect which factors may influence lizard density, a Generalized Linear Model (GLZ; multinomial distribution and cumulative log link function) was built, with environmental variables and density of predators’ variables being included in the model as covariates (scale predictor). Density of the three lizard species differed significantly among study sites, evidencing species-specific responses to local patch conditions. None of the environmental variables taken separately in the GLZ model influenced significantly the lizard densities, whereas lizards densities showed species-specific response to the considered environmental variables. The largest species (L. viridis) showed the highest density in the fragment with the most irregular shape and largest wood size, whereas the two smaller Podarcis species presented their highest population density in the site with the smallest wood patch and with a very low snake density. Density of P. sicula was negatively correlated with both the woodland area and Colubridae density, and was positively correlated with woodland shape (i.e. with circularity).

Bobcats (Lynx rufus) are valuable indicators of connectivity in the highly fragmented landscape of coastal southern California, yet their population sizes and densities are largely unknown. Using noninvasive scat sampling in a capture–recapture framework, we estimated population sizes for 2 similar areas of natural habitat with differing levels of isolation by human development in Santa Monica Mountains National Recreation Area, California. We used scat transects with geographic information system land-use layers and home-range sizes of bobcats to estimate effective sampling area and population densities. Estimates of population size in the study area connected to a much larger habitat area (26–31 individuals) were similar to estimates for the area that was completely surrounded by development (25–28 individuals). Bobcat densities for the 2 study areas also were similar (ranging from 0.25 to 0.42 bobcat/km 2 ) and likely represent recent population declines because of notoedric mange likely interacting with toxicants. These methods proved effective despite particularly low densities of bobcats and may be especially useful when study areas are geographically isolated, reducing the uncertainty in size of the sampling area.

The estimation of population density animal population parameters, such as capture probability, population size, or population density, is an important issue in many ecological applications. Capture–recapture data may be considered as repeated observations that are often correlated over time. If these correlations are not taken into account then parameter estimates may be biased, possibly producing misleading results. We propose a generalized estimating equations (GEE) approach to account for correlation over time instead of assuming independence as in the traditional closed population capture–recapture studies. We also account for heterogeneity among observed individuals and over-dispersion, modelling capture probabilities as a function of covariates. The GEE versions of all closed population capture–recapture models and their corresponding estimating equations are proposed. We evaluate the effect of accounting for correlation structures on capture–recapture model selection based on the quasi-likelihood information criterion (QIC). An example is used for an illustrative application and for comparison to currently used methodology. A Horvitz–Thompson-like estimator is used to obtain estimates of population size based on conditional arguments. A simulation study is conducted to evaluate the performance of the GEE approach in capture-recapture studies. The GEE approach performs well for estimating population parameters, particularly when capture probabilities are high. The simulation results also reveal that estimated population size varies on the nature of the existing correlation among capture occasions.

Science, statistics and surveys: a herpetological perspective

Mark-recapture estimators are commonly used for population size estimation, and typically yield unbiased estimates for most solitary species with low to moderate home range sizes. However, these methods assume independence of captures among individuals, an assumption that is clearly violated in social species that show fission-fusion dynamics, such as the Asian elephant. In the specific case of Asian elephants, doubts have been raised about the accuracy of population size estimates. More importantly, the potential problem for the use of mark-recapture methods posed by social organization in general has not been systematically addressed. We developed an individual-based simulation framework to systematically examine the potential effects of type of social organization, as well as other factors such as trap density and arrangement, spatial scale of sampling, and population density, on bias in population sizes estimated by POPAN, Robust Design, and Robust Design with detection heterogeneity. In the present study, we ran simulations with biological, demographic and ecological parameters relevant to Asian elephant populations, but the simulation framework is easily extended to address questions relevant to other social species. We collected capture history data from the simulations, and used those data to test for bias in population size estimation. Social organization significantly affected bias in most analyses, but the effect sizes were variable, depending on other factors. Social organization tended to introduce large bias when trap arrangement was uniform and sampling effort was low. POPAN clearly outperformed the two Robust Design models we tested, yielding close to zero bias if traps were arranged at random in the study area, and when population density and trap density were not too low. Social organization did not have a major effect on bias for these parameter combinations at which POPAN gave more or less unbiased population size estimates. Therefore, the effect of social organization on bias in population estimation could be removed by using POPAN with specific parameter combinations, to obtain population size estimates in a social species.

Human impact is considered the major threat to the global decline of biodiversity, especially for threatened endemic species in karst ecosystems. Studies assessing a species' demography based on temporal and spatial indicators of population size, density and structure are expected to evaluate the level of impact of threats and are therefore becoming increasingly important for species conservation efforts. Goniurosaurus huuliensis, an endemic species in Vietnam, is one of the most threatened reptiles in the world. This karst–adapted species is classified by the IUCN Red List as Critically Endangered and listed under CITES Appendix II due to habitat loss and over–exploitation for the international pet trade. Here we provide the first evaluation of the population status of G. huuliensis. We applied a 'capture mark–recapture' method to estimate the population size and identify the population density and structure. The total population size was estimated to comprise a maximum of 1,447 individuals in integrated suitable habitats, possibly reaching up to 2,855 individuals exclusively in karst habitats within the total extension of occurrence. This is exceedingly lower than the threshold for a minimum viable population. Furthermore, G. huuliensis is documented to occur in extremely small mean population densities of only 6.4 indiv./km and 2.5 indiv./km/day along the surveyed transects. Based on the demographic information, the ongoing severe human impact (e.g. wildlife exploitation and limestone quarrying) is driving G. huuliensis to the brink of extinction. In situ conservation measures are therefore urgently required. We recommend that in-situ actions should be increased, and a plan should be developed to establish a species and habitat conservation area for G. huuliensis.

Capture-recapture methods were used to estimate population size, survival rate, and recruitment to a breeding population of female wood ducks (Aix sponsa) in 1979-86. A total of 181 females was captured in nest boxes and banded during the 8-year period. Population size averaged 44 individuals. Mean annual survival rate was 0.55, and annual recruitment averaged 22 females. Data were used to illustrate the philopatric behavior of female wood ducks during the breeding season. J. WILDL. MANAGE. 51(2):401-404 Estimation of waterfowl survival rates has been improved by developing and using a series of models for analyzing band-recovery data (Brownie et al. 1978). Studies using these analytical methods have examined the relationship between survival and harvest rate (reviewed in Nichols et al. [1985]), population density (Anderson 1975, Conroy and Eberhardt 1983), and habitat conditions on breeding (Nichols et al. 1982) and wintering (Nichols and Hines, in press) areas. Capture-recapture methods allow estimation of population size and recruitment as well as survival, but are used infrequently to study avian population dynamics (Nichols et al. 1981, Pollock 1981). Open population models that allow gains and/or losses to occur between sampling periods seem particularly well-suited for long-term studies of waterfowl populations (Anderson and Sterling 1974, Sulzbach and Cooke 1979). In this paper we use 8 years of capture-recapture data from a breeding population of female wood ducks to estimate population size, recruitment, and survival and to determine whether wood duck hens are philopatric. We thank J. D. Nichols and J. E. Hines for helping with data analysis. J. D. Nichols provided an especially helpful review of an early draft of the paper. We also are grateful to the many people who helped in various ways durIPresent address: Indiana Department of Natural Resources, R.R. 4, Box 47, Peru, IN 46970. This content downloaded from 157.55.39.238 on Sat, 02 Jul 2016 05:33:15 UTC All use subject to http://about.jstor.org/terms 402 WOOD DUCK HEN SURVIVAL * Hepp et al. J. Wildl. Manage. 51(2):1987 ing the study, particularly L. D. Vangilder. This study was supported by a U.S. Dep. Energy Contract, DE-AC09-76SR00-819, with the Univ. Georgia (SREL).

Many large marine species are vulnerable to anthropogenic pressures, and substantial declines have been documented across a range of taxa. Many of these species are also long-lived, have low individual resighting rates and high levels of individual heterogeneity in capture probability, which complicates assessments of their conservation status with capture-�mark-�recapture (CMR) models. Few studies have been able to apply CMR models to whale sharks Rhincodon typus, the world’s largest fish. One of their aggregation sites off Mafia Island in Tanzania is characterised by unusually high residency of this Endangered species, making it an ideal target for CMR methods. Three different CMR models were fitted to an 8 yr photo-identification data set to estimate abundance, population trend and demographic parameters. As anticipated, resighting rates were unusually high compared to other aggregations. Different CMR models produced broadly similar parameter estimates, showing a stable population trend with high survivorship and limited recruitment. Tagging and biopsy sampling for concurrent research did not negatively affect those sharks’ apparent survival or capture probabilities. Scenario-based power analyses showed that only pronounced abundance trends (±30%) would be detectable over our study period, at a 90% level of probability, even with the relatively high precision in yearly abundance estimates achieved here. Other, more transient whale shark aggregations, with reduced precision in abundance estimates, may only be able to confidently detect a similar trend with CMR models after 15-20 yr of observations. Precautionary management and long-term monitoring will be required to assist and document the recovery of this iconic species.

The Strait of Gibraltar is an important habitat for cetaceans due to its high marine productivity. However, it is also the second most navigated channel in the world, subjecting cetaceans to a high level of vessel traffic, including an established whale‐watching fleet. Both maritime traffic and whale‐watching activities have been shown to impact the behaviour of cetaceans, but little is known about their impact on the demography and dynamics of cetacean populations. The aim of this study was to evaluate the impacts of both ferry traffic and whale‐watching vessels on the apparent survival probability of a bottlenose dolphin (Tursiops truncatus) population that occurs in the Strait. A Bayesian hierarchical mark–recapture modelling approach was applied to 8 years of photo‐identification data (2002–2009). Apparent annual survival probability was negatively correlated with ferry traffic, which explained >70% of temporal variation in survival, in contrast to whale‐watching, the effect of which was almost negligible. Despite these results, other natural and human‐related factors are likely to drive apparent survival in the study area. Abundance increased between 2002 and 2005, and then decreased between 2006 and 2009, while local per‐capita recruitment decreased from 2004 until the end of the study period. These shifts correlated temporally with the construction of a large harbour on the Moroccan coast, which increased maritime traffic significantly, including a 40% increase in ferry traffic that regularly transited the Strait. These results highlight the need to better understand the impact of maritime traffic on the demography of the dolphin population in the Strait, in order to implement evidence‐based conservation regulations in a region of high cetacean occurrence.

Evaluation of the Schnabel, geometric and nonparametric estimators of population size was performed on a model population of 100 individuals which possessed a bivariate normal home range model with ox = 1 and a, = 2, a random spatial pattern, an average initial probability of capture of p = 0.10 or p = 0.40 and either a trap happy or trap shy learning process. Data were generated from a model which simulated 10-sample mark-recapture experiments consisting of 25 traps. Evaluation of these estimators reveals the magnitude of bias and variability present in such circumstances. The results indicate that the nonparametric estimator was the most robust under these conditions. J. WILDL. MANAGE. 43(2):474-483 Estimating the size of an animal population presents complex problems to population ecologists. General reviews of some of the methods were presented by Overton (1969) and Seber (1973). Estimators which have been developed are based on various assumptions, some of which are not fully satisfied in practice. When assumptions are not satisfied, wide variation may result in the estimates. Since many different estimators are available, it is desirable to evaluate their performance under varying degrees of violation of the assumptions. This would provide the investigator a measure of the degree of robustness of the estimators and shed light into which one would be most suitable for a specific problem. Field evaluation of the estimators is possible if mark-recapture experiments are performed in areas where the true sizes of the populations are known. However, this information is rarely available. An alternate approach for determining the robustness of the estimators is to develop simulation models to mimic populations exposed to mark-recapture experiments. In such models the parameters that control the animals' response to trapping can be varied, resulting in differential magnitudes of assumption violation. Recently, simulation has been used in the study of estimators of population parameters. Gates (1969) used computer simulation to evaluate the bias of a variety of line transect density estimators. Burnham and Overton (1969) and Manly (1970) employed computer simulation in evaluating the behavior of various markrecapture estimators. In another paper, Manly (1971) used simulation to examine Jolly's (1965) variance formulas for estimators of population parameters. In the present study, computer simulation techniques were also used. The objective of this work was to evaluate the sample expectation, variance, bias, and mean square error of mark-recapture estimators of population size when the assumption of equal probability of capture of animals is violated in response to a learning process related to the previous capture history. Appreciation is extended to the Division of Forestry and Wildlife Resources, Virginia Polytechnic Institute and State University, Blacksburg, Virginia, and to the Department of Fisheries and Wildlife, Michigan State University, East Lan' Present address: Department of Fisheries and Wildlife, Michigan State University, East Lansing, MI 48824. 474 J. Wildl. Manage. 43(2):1979 This content downloaded from 157.55.39.80 on Fri, 22 Apr 2016 05:57:59 UTC All use subject to http://about.jstor.org/terms SIMULATION OF LEARNED TRAP RESPONSE *Zarnoch 475 sing, Michigan, for computer time used in the development of the model and evaluation of the estimators of population size. ESTIMATORS OF POPULATION SIZE Letting N = number of individuals in the population and S = number of samples taken, the following estimators were evaluated. Schnabel.-The Schnabel estimator, modified by Chapman (1952), is defined

SummaryEvaluating the status of endangered wildlife depends upon well‐designed field studies. Finances and logistics often constrain field studies to finite (limited‐sized) areas where inductive inferences are needed to extrapolate results to populations. Although available quantitative techniques for analysing data are robust to many aspects of field investigations, few investigators assess the influence of their study area size on estimators of population parameters and subsequent inferences derived from those estimators.We used mark–recapture to monitor an entire population of spotted owlsStrix occidentalisin the San Bernardino Mountains of southern California (2140 km2) for which we knew the approximate true rate of survival. We defined hypothetical study areas of varying size by subsampling the population in increments of five territories; we then estimated apparent survival and emigration for non‐juvenile and juvenile owls within each of these sample study areas to assess the influence of study area size on estimators of survival and population trends.Estimated survival rates of juvenile spotted owls increased approximately fourfold from the smallest sample area to the largest (min = 0·08,SÊ = 0·03;max = 0·33,SÊ = 0·03). In contrast, estimates of apparent survival for non‐juvenile owls did not vary with study area size (rangenon‐juvenile = 0·80–0·82,SÊ = 0·01–0·03).Juvenile emigration was extremely high in the smallest study area (juvenile = 0·77,SÊ = 0·09) and remained above 10% until > 62% of our study area (approximately 900 km2) was encompassed by a sample study area. Non‐juvenile owls had low annual emigration probabilities from all sample study area sizes (rangenon‐juvenile = 0·00–0·02).Although estimates of λ (finite population growth rate) increased gradually from 0·828 to 0·903 as the subsample increased from 20 to 143 territories, these estimates were similar to the ‘true’ value.Synthesis and applications.We provide direct estimates of the bias that sampling limited study areas has on emigration and mark–recapture estimators of survival. Our results demonstrate that permanent emigration from limited study areas can lead to underestimates of survival and population growth rates. In addition, our approach illustrates a technique for using multistate models to assess study design and estimator assumptions.

Podarcis siculus (Rafinesque-Schmaltz, 1810) or Italian wall lizard is one of the most invasive reptile-species. Recently, this lacertid lizard has been introduced to Mediterranean areas of southern Europe, South-West Asia (Turkey) and North America (USA). An abundant population of P. siculus was discovered on one of the sites of the Natural Ornithological Park in the Imeretinskaya Lowland, on an area of over 0.22 km2 (Sochi, Russia). The data were collected in the May of 2020 in a strip survey method in the Imeretinskaya Lowland. To identify the colonization area of the invader, we examined all 8 sections of the Natural Ornithological Park in the Imeretinskaya Lowland and adjacent urbanized areas. More than 150 animals were observed. These Italian wall lizards, undoubtedly, belong to the northern-central Italian morphotype (presumably P. s. campestris). This is the first record of this species in the former USSR area and, also, this is the species’ north-easternmost locality. The population inhabits secondary natural biotopes and urban area. Among them are the banks of artificial water bodies, areas with cultivated trees and shrubs, as well as parks, and house lawns in the urban area. Population density was estimated from eight to 40 specimens per 100 m of the transect. A moderate proportion of young specimens (more than a 40%) would indicate a healthy and continued growth of the emerging population. To determine the possible period of the species introduction, space images of the Imeretinskaya Lowland were analyzed beginning from the transformation of its landscape for the Winter Olympic Games of Sochi 2014 until the May of 2020. The introduction of the species presumably occurred with the delivery of large-sized ornamental trees and shrubs from Italy in 2012–2013. Podarcis siculus should be included in the list of herpetofauna of Russia and particularly of the Caucasus. This is an alien species with a proven ability to become an invasive species, what will lead to a greater undesirable and unavoidable contact with native small lizards of the genus Darevskia Arribas, 1997. On the other hand, as it is often observed with new invaders, a sudden rise in population abundance could be followed by a sharp decline. A continuous monitoring of the area in question and of the number of local Italian wall lizards is necessary to confirm or refute the assumed scenarios of further invasion of P. siculus on the Black Sea Coast of the Caucasus. Further action plans for this population should be developed depending on supposed future trends.

In this paper, we describe a method for estimating animal population density per unit area from mark, release, recapture (MRR) data. Standard procedures are available for estimating population size from MRR data, but not density per unit area, which is often of more practical value. If data on dispersal movements are available in addition to capture/recapture records, they can be used to estimate the area over which the study population ranges. These area estimates can then be combined with MRR abundances to yield estimates of population density. As a concrete example of how these methods can be applied, we calculated population density estimates for seven MRR data sets for the screwworm, Cochliomyia hominivorax (Diptera: Calliphoridae), from Costa Rica. As a standard of comparison, we also used a different method to calculate estimates of screwworm density for five sets of USDA APHIS Screwworm Eradication Program data from Mexico. Mean densities for both sets of analyses were similar, ranging from 10 to 120 adult flies/km2. General agreement between the Mexican and Costa Rican estimates lends support to the dispersal approach we propose here. Although we have applied these methods to the screwworm as a specific example, they are entirely general and can be applied to virtually any organism for which both MRR records and dispersal data are available.

Determining the factors that limit and regulate population density is a key problem in population ecology, but often difficult to test in practice. The northern red-backed vole ( Clethrionomys rutilis Pallas, 1779 sec. Kryštufek et al. 2020) occurs throughout the boreal forests of the northern hemisphere and the factors influencing their populations are still being elucidated. A large-scale experiment was conducted in the Yukon from 1986 to 1996 in which food availability, nutrient availability, predation, and interspecific competition were experimentally manipulated. Paradoxically, there was no evidence that these experimental treatments altered vole population density. Here, we apply a capture–mark–recapture modeling framework to test for experimental treatment effects on recruitment and apparent survival. Experimental manipulations resulted in opposing effects on recruitment and apparent survival. These contrasting demographic effects meant that the population growth rate and density remained unaffected by experimental treatments, upholding the original results. We suggest that as some experimental areas became more saturated with voles, this forced higher levels of vole dispersal or attracted mustelid or avian predators, either of which could result in the contrasting effect on apparent survival. Our results highlight that manipulation of environmental factors can alter demographic parameters without substantially affecting overall abundance.

Habitat loss, introduced disease, and government-sponsored eradication programs have caused population declines in all 5 species of prairie dogs. Black-tailed prairie dogs (Cynomys ludovicianus) currently occupy only about 2% of an extensive geographic range (160 million hectares) and were recently considered for listing under the United States Endangered Species Act. Accurate estimates of density for populations of prairie dogs would be valuable for estimating range-wide abundance and for determining threats to species persistence, yet estimates of density using robust approaches (e.g., complete enumeration or mark–recapture) are rarely undertaken. We introduce a novel approach to estimating density of prairie dogs using mark–resight methods. Using mark–resight, mark–recapture, and 3 other indices, we estimated the abundance of prairie dogs on 3 reintroduced colonies over a 3-year period (2003–2005). We show that mark–resight is a superior approach to estimating abundance of prairie dogs, that average density estimates from the southern extremity of the species' range are considerably lower (11.3 prairie dogs/ha) than estimates from more northerly climes (X̄ = 18.3–90.3 prairie dogs/ha), and that population densities can fluctuate widely in accordance with local environmental conditions. We propose that resource agencies use mark–resight methods to obtain density estimates of prairie dog populations within diverse ecoregions, and couple these estimates with an assessment of the area occupied by prairie dog colonies to determine range-wide abundance.