USGS Science⌕ Search
Home/Archive/USGS

USGS · 70037583

Predicting species distributions from checklist data using site-occupancy models

M. Kery·B. Gardner·C. Monnerat
View original source ↗Export metadata ↓

Abstract

Aim: (1) To increase awareness of the challenges induced by imperfect detection, which is a fundamental issue in species distribution modelling; (2) to emphasize the value of replicate observations for species distribution modelling; and (3) to show how 'cheap' checklist data in faunal/floral databases may be used for the rigorous modelling of distributions by site-occupancy models. Location: Switzerland. Methods: We used checklist data collected by volunteers during 1999 and 2000 to analyse the distribution of the blue hawker, Aeshna cyanea (Odonata, Aeshnidae), a common dragonfly in Switzerland. We used data from repeated visits to 1-ha pixels to derive 'detection histories' and apply site-occupancy models to estimate the 'true' species distribution, i.e. corrected for imperfect detection. We modelled blue hawker distribution as a function of elevation and year and its detection probability of elevation, year and season. Results: The best model contained cubic polynomial elevation effects for distribution and quadratic effects of elevation and season for detectability. We compared the site-occupancy model with a conventional distribution model based on a generalized linear model, which assumes perfect detectability (p = 1). The conventional distribution map looked very different from the distribution map obtained using site-occupancy models that accounted for the imperfect detection. The conventional model underestimated the species distribution by 60%, and the slope parameters of the occurrence-elevation relationship were also underestimated when assuming p = 1. Elevation was not only an important predictor of blue hawker occurrence, but also of the detection probability, with a bell-shaped relationship. Furthermore, detectability increased over the season. The average detection probability was estimated at only 0.19 per survey. Main conclusions: Conventional species distribution models do not model species distributions per se but rather the apparent distribution, i.e. an unknown proportion of species distributions. That unknown proportion is equivalent to detectability. Imperfect detection in conventional species distribution models yields underestimates of the extent of distributions and covariate effects that are biased towards zero. In addition, patterns in detectability will erroneously be ascribed to species distributions. In contrast, site-occupancy models applied to replicated detection/non-detection data offer a powerful framework for making inferences about species distributions corrected for imperfect detection. The use of 'cheap' checklist data greatly enhances the scope of applications of this useful class of models. ?? 2010 Blackwell Publishing Ltd.

Explore related subjects

Journal of BiogeographyArticle

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

M. Kery, B. Gardner, C. Monnerat. 2010-06-22. Predicting species distributions from checklist data using site-occupancy models. https://doi.org/10.1111/j.1365-2699.2010.02345.x

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2026

Body size, dispersal potential, range size and habitat fragmentation are linked to extinction risk of freshwater mussels

Aim: Trait-based approaches are increasingly recognized as a tool for understanding species’ vulnerabilities to intensifying global change. For species with complex life histories that involve a parasitic life stage, a combination of traits associated with either the host or parasite could influence extinction risk. Location: United States of America Taxon: Freshwater mussels are a highly threatened assemblage of bivalves with an obligate ectoparasitic larval stage requiring a host fish. Adult mussels are sedentary, with long-distance dispersal dictated by the dispersal ability of the host fish during the parasitic larval life phase. Methods: We integrate several open-source spatial and trait datasets for freshwater mussels and host fishes to test hypotheses about the relationship between species-level traits such as body size, area of occupancy (AOO), host fish dispersal, habitat fragmentation, and mussel extinction risk. Results: Imperiled mussels were smaller, had smaller areas of occupancy, used small-bodied, dispersal-limited host fishes, and had higher densities of large dams in their AOO compared to non-imperiled mussels. Conclusions: Body size, along with AOO and dispersal potential, appear to have cascading effects on imperilment where decreases in these traits potentiate population fragmentation via large dams. Thus, in contrast to many other groups, freshwater mussel body size is inversely related to extinction risk and small-bodied mussels should be closely monitored to prevent future extinctions.

Journal of Biogeography↗
USGS2026

Genomics reveals extensive population structure and undescribed phylogenetic relationships in the Cascade torrent salamander (Rhyacotriton cascadae)

Aim Aims of the study are to examine patterns of range-wide genetic differentiation and population structure in a headwater obligate salamander living in a geologically rich region, to identify genetically distinct populations and areas of gene flow between them. Location Oregon and Washington in the Pacific Northwest, United States of America. Time Period Tissue samples were collected in 2022 and 2023. Major Taxa Studied The Cascade torrent salamander Rhyacotriton cascadae. Methods Utilisation of a genome-wide single nucleotide polymorphism (SNP) dataset from across the species range to conduct a principal components analysis (PCA), Bayesian model of population structure, co-ancestry matrix, phylogenetic tree and estimate genetic diversity. Results There are extensive levels of population structure within R. cascadae , including a previously unknown and highly differentiated clade. Structure is characterised by an island-like pattern wherein the species is comprised of six populations that function as independent demographic units, with gene flow largely constrained within populations. Main Conclusions Our findings reveal cryptic population structure within R. cascadae , identifying six distinct populations across the range. The northernmost population in the northwest of the species range in Washington is surprisingly highly divergent from the other five populations, and the divergence was not previously known to science. While major rivers act as phylogeographic boundaries between some populations, these boundaries appear to not always be complete.

Oregon, Washington↗
USGS2024

Invisible hand of sampling for management: Underlying needs to survey a threatened seabird can bias aggregated data

Aim Surveying for a species of concern ahead of proposed activities that alter its habitat is routine practice in conservation and management. Such surveys may accumulate large datasets that could further elucidate trends in abundance and distribution. However, the as-needed surveying of proposed activities may impart a sample site selection bias on the data if used for another purpose. Management of a threatened, forest-nesting seabird offered an example of this. Here we assessed how resource management planning and survey requirements can bias clearance monitoring survey data collected prior to proposed timber harvests, if those data are used for other purposes. Location Oregon and Washington, USA. Taxon Marbled Murrelet ( Brachyramphus marmoratus ). Methods To assess how timber planning and other factors influenced marbled murrelet survey location selection, we used logistic regression models to examine habitat associations of marbled murrelet survey sites ( n = 9178) encompassing proposed timber harvests, and the survey stations ( n = 38,923) therein, across the murrelet's inland range in Washington and Oregon, USA between 1989 and 2021. We then simulated the effect this selective sampling might have on assessments of occupancy trends. Results Most habitat characteristics considered did influence where surveys were located, with distance to roads often being the strongest predictor of survey location. The strength of selection for each covariate changed over time, such that a habitat characteristic strongly influenced location selection in a year but was less influential in another year. The simulation analysis suggested that the non-random selection of survey sites could profoundly bias assessments of occupancy trends. Conclusions When using these clearance monitoring survey data– or any data–beyond their original purpose, careful consideration should be given to the scope of inference provided and analytical methods used, to ensure that observed trends are the product of biological processes and not biased by sampling artefacts.

Oregon, Washington↗