Biology subjects

Dupont, P.

Publications and source records attributed to Dupont, P..

11 recordsLinked to original sources

bioRxiv2025

Predictors of extinction risk in large tropical forest mammals: from global to local

Studies can only guide conservation if their findings are informative at the scales at which practitioners and policymakers operate. Yet, it is rarely tested whether large-scale studies reach similar conclusions to the smaller-scale studies on which conservation traditionally relies. We examine whether predictors of extinction risk are consistent across global, regional, and local scales, for 210 tropical forest mammal species ([≥] 1 kg) that existed during the last 130,000 years, in 64 tropical forests, across three biogeographical realms. We found consistent predictors of extinction risk (body mass, generation length, diet, brain volume and scansoriality) when analyses differed only in their spatial resolution. However, predictors differed when analyses also varied in their temporal extent. Macroecological findings about extinction risk can, thus, inform conservation at smaller scales, but they risk misidentifying threatened species if differences in temporal extent are not recognised.

ecology↗

bioRxiv2022

Wolverine density distribution reflects past persecution and current management in Scandinavia

After centuries of intense persecution, several large carnivore species in Europe and North America have experienced a rebound. Todays spatial configuration of large carnivore populations has likely arisen from the interplay between their ecological traits and current environmental conditions, but also from their history of persecution and protection. Yet, due to the challenge of studying population-level phenomena, we are rarely able to disentangle and quantify the influence of past and present factors driving the spatial distribution and density of these controversial species. Using spatial capture-recapture models and a data set of 742 genetically identified wolverines Gulo gulo collected over [1/2] million km2 across their entire range in Norway and Sweden, we identify landscape-level factors explaining the current population density of wolverines in the Scandinavian Peninsula. Distance from the relic range along the Swedish-Norwegian border, where the wolverine population survived a long history of persecution, remains a key determinant of wolverine density today. However, regional differences in management and environmental conditions also played an important role in shaping spatial patterns in present-day wolverine density. Specifically, we found evidence of slower recolonization in areas that had set lower wolverine population goals in terms of the desired number of annual reproductions. Management of transboundary large carnivore populations at biologically relevant scales may be inhibited by administrative fragmentation. Yet, as our study shows, population-level monitoring is an achievable prerequisite for a comprehensive understanding of the distribution and density of large carnivores across an increasingly anthropogenic landscape.

ecology↗

bioRxiv2022

Spatial variation in red deer density in relation to forest disturbance and ungulate management in a transboundary forest ecosystem

1. ContextForests in Europe are exposed to increasingly frequent and severe disturbances. The resulting changes in the structure and composition of forests can have profound consequences for the wildlife inhabiting them. Moreover, wildlife populations in Europe are often subjected to differential management regimes as they regularly extend across multiple national and administrative borders. The red deer (Cervus elaphus) population in the Bohemian Forest Ecosystem, straddling the Czech-German border, has experienced forest disturbances, primarily caused by windfalls and bark beetle (Ips typographus) outbreaks during the past decades. To adapt local management strategies to the changing environmental conditions and to coordinate them across the international border, reliable estimates of red deer density and abundance are highly sought-after by policymakers, wildlife managers, and stakeholders. 2. Approach and MethodsCovering a 1081-km2 study area, we conducted a transnational non-invasive DNA sampling study in 2018 that yielded 1578 genotyped DNA samples from 1120 individual red deer. Using spatial capture-recapture models, we estimated total and jurisdiction-specific abundance of red deer throughout the ecosystem and quantified the role of forest disturbance and differential management strategies in shaping spatial heterogeneity in red deer density. We hypothesized that (a) forest disturbances provide favourable habitat conditions (e.g., forage and cover), and (b) contrasting red deer management regimes in different jurisdictions create a differential risk landscape, ultimately shaping density distributions. 3. ResultsOverall, we estimated that 2851 red deer (95% Credible Intervals = 2609 - 3119) resided in the study area, with a relatively even overall sex ratio (1406 females, 1229 - 1612 and 1445 males, 1288 - 1626). The average red deer density was higher in Czechia (3.5 km-2, 1.2 - 12.3) compared to Germany (2 km-2, 0.2 - 11). The effect of forest disturbances on red deer density was context dependent. Forest disturbances had a positive effect on red deer density at higher elevations and a negative effect at lower elevations, which could be explained by partial migration and its drivers in this population. Density of red deer was generally higher in management units where hunting is prohibited. In addition, we found that sex ratios differed between administrative units and were more balanced in the non-intervention zones. 4. Synthesis and applicationsOur results show that the effect of forest disturbances on wild ungulates is modulated by additional factors, such as elevation and ungulate management practices. Overall density patterns and sex ratios suggested strong gradients in density between administrative units. With climate change increasing the severity and frequency of forest disturbances, population-level monitoring and management are becoming increasingly important especially for wide-ranging species as both wildlife and global change transcend administrative boundaries.

ecology↗

bioRxiv2022

Modelling spatially autocorrelated detection probabilities in spatial capture-recapture using random effects

Spatial capture-recapture (SCR) models are now widely used for estimating density from repeated individual spatial encounters. SCR accounts for the inherent spatial autocorrelation in individual detections by modelling detection probabilities as a function of distance between the detectors and individual activity centres. However, additional spatial heterogeneity in detection probability may still creep in due to environmental or sampling characteristics. if unaccounted for, such variation can lead to pronounced bias in population size estimates. Using simulations, we describe and test three Bayesian SCR models that use generalized linear mixed models (GLMM) to account for latent heterogeneity in baseline detection probability across detectors using: independent random effects (RE), spatially autocorrelated random effects (SARE), and a twogroup finite mixture model (FM). Overall, SARE provided the least biased population size estimates (median RB: -9 - 6%). When spatial autocorrelation was high, SARE also performed best at predicting the spatial pattern of heterogeneity in detection probability. At intermediate levels of autocorrelation, spatially-explicit estimates of detection probability obtained with FM where more accurate than those generated by SARE and RE. In cases where the number of detections per detector is realistically low (at most 1), all GLMMs considered here may require dimension reduction of the random effects by pooling baseline detection probability parameters across neighboring detectors ("aggregation") to avoid over-parameterization. The added complexity and computational overhead associated with SCR-GLMMs may only be justified in extreme cases of spatial heterogeneity. However, even in less extreme cases, detecting and estimating spatially heterogeneous detection probability may assist in planning or adjusting monitoring schemes.

ecology↗

bioRxiv2022

Estimating spatially variable and density-dependent survival using open-population spatial capture-recapture models

Open-population spatial capture-recapture (OPSCR) models use the spatial information contained in individual detections collected over multiple consecutive occasions to estimate occasion-specific density, but also demographic parameters. OPSCR models can also estimate spatial variation in vital rates, but such models are neither widely used nor thoroughly tested. We developed a Bayesian OSPCR model that not only accounts for spatial variation in survival using spatial covariates, but also estimates local density-dependent effects on survival within a unified framework. Using simulations, we show that OPSCR models provide sound inferences on the effect of spatial covariates on survival, including multiple competing sources of mortality, each with potentially different spatial determinants. Estimation of local density-dependent survival was possible but required more data due to the greater complexity of the model. Not accounting for spatial heterogeneity in survival led to positive bias in abundance estimates (up to 10% relative bias). We provide a set of features in R package nimbleSCR that allow computationally efficient fitting of Bayesian OPSCR models with spatially varying survival. The ability to make population-level inferences of spatial variation in survival is an essential step towards a fully spatially-explicit OPSCR model that can disentangle the role of multiple spatial drivers on population dynamics. Open Research statementcode to reproduce the analysis is available on github; https://github.com/Cyril-Milleret/Public/tree/master/SpatialSurvivalOPSCR

ecology↗

bioRxiv2021

An atlas of white matter anatomy, its variability, and reproducibility based on Constrained Spherical Deconvolution of diffusion MRI

Virtual dissection of white matter (WM) using diffusion MRI tractography is confounded by its poor reproducibility. Despite the increased adoption of advanced reconstruction models, early region-of-interest driven protocols based on diffusion tensor imaging (DTI) remain the dominant reference for virtual dissection protocols. Here we bridge this gap by providing a comprehensive description of typical WM anatomy reconstructed using a reproducible automated subject-specific parcellation-based approach based on probabilistic constrained-spherical deconvolution (CSD) tractography. We complement this with a WM template in MNI space comprising 68 bundles, including all associated anatomical tract selection labels and associated automated workflows. Additionally, we demonstrate bundle inter- and intra-subject variability using 40 (20 test-retest) datasets from the human connectome project (HCP) and 5 sessions with varying b-values and number of b-shells from the single-subject Multiple Acquisitions for Standardization of Structural Imaging Validation and Evaluation (MASSIVE) dataset. The most reliably reconstructed bundles were the whole pyramidal tracts, primary corticospinal tracts, whole superior longitudinal fasciculi, frontal, parietal and occipital segments of the corpus callosum and middle cerebellar peduncles. More variability was found in less dense bundles, e.g., the first segment of the superior longitudinal fasciculus, fornix, dentato-rubro-thalamic tract (DRTT), and premotor pyramidal tract. Using the DRTT as an example, we show that this variability can be reduced by using a higher number of seeding attempts. Overall inter-session similarity was high for HCP test-retest data (median weighted-dice = 0.963, stdev = 0.201 and IQR = 0.099). Compared to the HCP-template bundles there was a high level of agreement for the HCP test-retest data (median weighted-dice = 0.747, stdev = 0.220 and IQR = 0.277) and for the MASSIVE data (median weighted-dice = 0.767, stdev = 0.255 and IQR = 0.338). In summary, this WM atlas provides an overview of the capabilities and limitations of automated subject-specific probabilistic CSD tractography for mapping white matter fasciculi in healthy adults. It will be most useful in applications requiring a highly reproducible parcellation-based dissection protocol, as well as being an educational resource for applied neuroimaging and clinical professionals. O_FIG O_LINKSMALLFIG WIDTH=182 HEIGHT=200 SRC="FIGDIR/small/464139v2_ufig1.gif" ALT="Figure 1"> View larger version (52K): org.highwire.dtl.DTLVardef@18e7c79org.highwire.dtl.DTLVardef@18e2213org.highwire.dtl.DTLVardef@5f124borg.highwire.dtl.DTLVardef@1aa242c_HPS_FORMAT_FIGEXP M_FIG Graphical abstract(Top) shows the FWT pipeline for both CSTs, AF, and motor CC bundles. (Left to right) show the required input structural parcellation maps and a priori atlases for FWT and the resulting virtual dissection include/exclude VOIs. FWT provides two approaches to virtual dissection: (1) is a bundle-specific approach where streamlines are only seeded for the bundle of interest, (2) is a whole brain tractography followed by streamlines segmentation, (top right) shows output tractograms. (Middle) Group-averaged T1 and fODF images are generated from the HCP test-retest data, and FWT is applied to generate the HCP-atlas using the bundle-specific approach (1*). FWTs whole brain tracking and segmentation approach (2*) was applied to the HCP and MASSIVE dataset (right and left) and conducted model-based, and pair-wise similarity analyses and generated voxel-wise cumulative maps per bundle. FWT= Fun With Tracts, FS= FreeSurfer, MSBP= MultiScaleBrainParcellator, PD25= NIST Parkinsons histological, JHU= Johns Hopkins university, Juelich= Juelich university histological atlas, AC/PC= anterior commissure/posterior commissure) UKBB= UK Biobank, SUIT (spatially unbiased cerebellar atlas template), dMRI= diffusion magnetic resonance imaging, CSD= constrained spherical deconvolution, fODF= fiber orientation distribution function, CST= corticospinal tract, AF= arcuate fasciculus, CC= corpus callosum, HCP= human connectome project, MASSIVE= Multiple acquisitions for standardization of structural imaging validation and evaluation. C_FIG

neuroscience↗

bioRxiv2021

Does the punishment fit the crime? Consequences and diagnosis of misspecified detection functions in Bayesian spatial capture-recapture modelling

O_LISpatial capture-recapture (SCR) is now used widely to estimate wildlife densities. At the core of SCR models lies the detection function, linking individual detection probability to the distance from its latent activity center. The most common function (half-normal) assumes a bivariate normal space use and consequently detection pattern. This is likely an oversimplification and misrepresentation of real-life animal space use patterns, but studies have reported that density estimates are relatively robust to misspecified detection functions. However, information about consequences of such misspecification on space use parameters (e.g. home range area), as well as diagnostic tools to reveal it are lacking. C_LIO_LIWe simulated SCR data under six different detection functions, including the half-normal, to represent a wide range of space use patterns. We then fit three different SCR models, with the three simplest detection functions (half-normal, exponential and half-normal plateau) to each simulated data set. We evaluated the consequences of misspecification in terms of bias, precision and coverage probability of density and home range area estimates. We also calculated Bayesian p-values with respect to different discrepancy metrics to assess whether these can help identify misspecifications of the detection function. C_LIO_LIWe corroborate previous findings that density estimates are robust to misspecifications of the detection function. However, estimates of home range area are prone to bias when the detection function is misspecified. When fitted with the half-normal model, average relative bias of 95% kernel home range area estimates ranged between -25% and 26% depending on the misspecification. In contrast, the half-normal plateau model (an extension of the half-normal) returned average relative bias that ranged between -26% and -4%. Additionally, we found useful heuristic patterns in Bayesian p-values to diagnose the misspecification in detection function. C_LIO_LIOur analytical framework and diagnostic tools may help users select a detection function when analyzing empirical data, especially when space use parameters (such as home range area) are of interest. We urge development of additional custom goodness of fit diagnostics for Bayesian SCR models to help practitioners identify a wider range of model misspecifications. C_LI

ecology↗

bioRxiv2020

Consequences of ignoring variable and spatially-autocorrelated detection probability in spatial capture-recapture

ContextSpatial capture-recapture (SCR) models are increasingly popular for analyzing wildlife monitoring data. SCR can account for spatial heterogeneity in detection that arises from individual space use (detection kernel), variation in the sampling process, and the distribution of individuals (density). However, unexplained and unmodeled spatial heterogeneity in detectability may remain due to cryptic factors, intrinsic and extrinsic to the study system. ObjectivesWe identify how the magnitude and configuration of unmodeled, spatially variable detection probability influence SCR parameter estimates. MethodsWe simulated realistic SCR data with spatially variable and autocorrelated detection probability. We then fitted a single-session SCR model ignoring this variation to the simulated data and assessed the impact of model misspecification on inferences. ResultsHighly autocorrelated spatial heterogeneity in detection probability (Morans I = 0.85 - 0.96), modulated by the magnitude of that variation, can lead to pronounced negative bias (up to 75%), reduction in precision (249%), and decreasing coverage probability of the 95% credible intervals associated with abundance estimates to 0. Conversely, at low levels of spatial autocorrelation (median Morans I = 0), even severe unmodeled heterogeneity in detection probability did not lead to pronounced bias and only caused slight reductions in precision and coverage of abundance estimates. ConclusionsUnknown and unmodeled variation in detection probability is liable to be the norm, rather than the exception, in SCR studies. We encourage practitioners to consider the impact that spatial autocorrelation in detectability has on their inferences and urge the development of SCR methods that can take structured unknown or partially unknown spatial variability in detection probability into account.

ecology↗

bioRxiv2020

A hierarchical point process model for spatial capture-recapture data

O_LISpatial capture-recapture (SCR) is a popular method for estimating the abundance and density of wildlife populations. A standard SCR model consists of two sub-models: one for the activity centers of individuals and the other for the detections of each individual conditional on its activity center. So far, the detection sub-model of most SCR models is designed for sampling situations where fixed trap arrays are used to detect individuals. C_LIO_LINon-invasive genetic sampling (NGS) is widely applied in SCR studies. Using NGS methods, one often searches the study area for potential sources of DNA such as hairs and faeces, and records the locations of these samples. To analyse such data with SCR models, investigators usually impose an artificial detector grid and project detections to the nearest detector. However, there is a trade-off between the computational efficiency (fewer detectors) and the spatial accuracy (more detectors) when using this method. C_LIO_LIHere, we propose a point process model for the detection process of SCR studies using NGS. The model better reflects the spatially continuous detection process and allows all spatial information in the data to be used without approximation error. As in many SCR models, we also use a point process model for the activity centers of individuals. The resulting hierarchical point process model enables estimation of total population size without imputing unobserved individuals via data augmentation, which can be computationally cumbersome. We write custom distributions for those spatial point processes and fit the SCR model in a Bayesian framework using Markov chain Monte Carlo in the R package nimble. C_LIO_LISimulations indicate good performance of the proposed model for parameter estimation. We demonstrate the application of the model in a real-life scenario by fitting it to NGS data of female wolverines (Gulo gulo) collected in three counties of Norway during the winter of 2018/19. Our model estimates that the density of female wolverines is 9.53 (95% CI: 8-11) per 10,000km2 in the study area. C_LI

ecology↗

bioRxiv2020

Instrumented individuals are not representative of the population. Survival of wolverines with and without GPS collars

Results from biologging studies are often scaled up to population-level inferences and this begs the question: Are instrumented animals representative of the population given the potential bias in individual selectivity, the influence of capture, handling and wearing bio-loggers? The answer is elusive due to the challenges of obtaining comparable data from individuals without bio-loggers. Using non-invasive genetic data of a large carnivore, the wolverine (Gulo gulo) in Scandinavia, and an open-population spatial capture-recapture model, we were able to estimate and compare survival rates of individuals fitted with GPS collars versus individuals that had never been captured and collared. We found that mortality of GPS collared individuals was 47% lower compared to individuals without GPS collars. While instrumented and non-instrumented wolverines had similar risks of dying from legal culling, GPS collared individuals displayed lower mortality rates due to causes other than legal culling. Reports of positive effects of biologging on the fitness of individuals are rather uncommon and we argue that the most likely explanation is that GPS-collars could shield animals from poaching. Our results highlight the challenges of drawing population level inferences for large carnivores when using data from an instrumented sample of the population.

ecology↗

bioRxiv2019

Estimating abundance with interruptions in data collection using open population spatial capture-recapture models

O_LIThe estimation of population size remains one of the primary goals and challenges in ecology and provides a basis for debate and policy in wildlife management. Despite the development of efficient non-invasive sampling methods and robust statistical tools to estimate abundance, maintenance of field sampling is still subject to economic and logistic constraints. These can result in intentional or unintentional interruptions in sampling and cause gaps in data time series, posing a challenge to abundance estimation, and ultimately conservation and management decisions.\nC_LIO_LIWe applied an open population spatial capture-recapture (OPSCR) model to simulations and a real case study to test the reliability of abundance inferences models to interruption in data collection. Using individual detections occurring over consecutive sampling occasions, OPSCR models allow the estimation of abundance from individual detection data while accounting for lack of demographic and geographic closure between occasions. First, we simulated sampling data with interruptions in field sampling of different lengths and timing. We checked the performance of an OPSCR model in deriving abundance for species with slow and intermediate life history strategies. Finally, we introduced artificial sampling interruptions of various magnitudes and timing to a five-year non-invasive monitoring data set of wolverines (Gulo gulo) in Norway and quantified the consequences for OPSCR model predictions.\nC_LIO_LIInferences from OPSCR models were reliable even with temporal interruptions in monitoring. Interruption did not cause any systematic bias, but increased uncertainty. Interruptions occurring at occasions towards the beginning and the end of the sampling caused higher uncertainty. The loss in precision was more severe for species with a faster life history strategy.\nC_LIO_LIWe provide a reliable framework to estimate abundance even in the presence of sampling interruptions. OPSCR allows monitoring studies to provide contiguous abundance estimates to managers, stakeholders, and policy makers even when data are non-contiguous. OPSCR models do not only help cope with unintentional interruptions during sampling but also offer opportunities for using intentional sampling interruptions during the design of cost-effective population surveys.\nC_LI

ecology↗