Global threat layers do not improve prediction of local threats

Global biodiversity is rapidly declining under the influence of human‐induced environmental change, but our understanding of what is driving change at the local scale is often incomplete. Global datasets such as the Living Planet Database (LPD) provide invaluable site‐based time‐series data on wildlife populations but often lack detailed information on the local threats simpacting them. Here, we test whether global broad‐scale spatial threat data based on the probability of threat impact or threat occurrence data can predict the presence of threats recorded in the LPD for 1668 terrestrial bird and mammal populations. We test six generalised additive model structures, comparing the predictive performance of models using global threat impact probabilities and threat occurrence data to an intercept only and Red List species threats null model. Predictive performance was assessed using 100 repetitions of five‐fold spatial blocked cross‐validation. Across taxa and threat types, the overall ability of models to detect true threat presences was low. Models based on probabilities of threat impact or threat occurrence rarely showed significant gains in predictive performance over the null models. Our findings demonstrate that current broad‐scale threat datasets act as imperfect proxies for on‐the‐ground population impacts, highlighting the essential need for higher‐resolution environmental predictors and increased local reporting to strengthen threat‐biodiversity response analyses.

Authors

Institutions

Publication Details

Journal
Ecography
Published
2026-11-16
DOI
https://doi.org/10.1002/ecog.08400
Primary Topic
Species Distribution and Climate Change
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Global threat layers do not improve prediction of local threats

Michael Hoffmann, Jonas Geldmann, Robin Freeman, Manuela González‐Suárez et al.
Ecography
Species Distribution and Climate Change
article

Global threat layers do not improve prediction of local threats

Michael Hoffmann, Jonas Geldmann, Robin Freeman, Manuela González‐Suárez, Louise McRae, Philippa Oppenheimer
article en

Abstract

Global biodiversity is rapidly declining under the influence of human‐induced environmental change, but our understanding of what is driving change at the local scale is often incomplete. Global datasets such as the Living Planet Database (LPD) provide invaluable site‐based time‐series data on wildlife populations but often lack detailed information on the local threats simpacting them. Here, we test whether global broad‐scale spatial threat data based on the probability of threat impact or threat occurrence data can predict the presence of threats recorded in the LPD for 1668 terrestrial bird and mammal populations. We test six generalised additive model structures, comparing the predictive performance of models using global threat impact probabilities and threat occurrence data to an intercept only and Red List species threats null model. Predictive performance was assessed using 100 repetitions of five‐fold spatial blocked cross‐validation. Across taxa and threat types, the overall ability of models to detect true threat presences was low. Models based on probabilities of threat impact or threat occurrence rarely showed significant gains in predictive performance over the null models. Our findings demonstrate that current broad‐scale threat datasets act as imperfect proxies for on‐the‐ground population impacts, highlighting the essential need for higher‐resolution environmental predictors and increased local reporting to strengthen threat‐biodiversity response analyses.

Ecography
University of Copenhagen (DK), Zoological Society of London (GB), University of Reading (GB)
Species Distribution and Climate Change
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon