Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Artículo científico |
| Language: | en |
| Published: |
Conservation biology : the journal of the Society for Conservation Biology
2026
|
| Subjects: | |
| Online Access: | https://pubmed.ncbi.nlm.nih.gov/41028974/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- Integrating expert range maps and opportunistic occurrence records of marine fish species in range estimates. Zhang, Zhixin Bede-Fazekas, Ákos Molinos, Jorge García Mammola, Stefano Kass, Jamie M Qu, Junmei Oeser, Julian Yuan, Songxi Zhang, Chongliang Gu, Jiqi Ding, Liuyong Lin, Qiang Animals Conservation of Natural Resources Fishes Animal Distribution Biodiversity Models, Biological Species distribution models (SDMs) are commonly used to estimate species' geographic distributions to inform biodiversity assessments and conservation planning. However, despite their growing popularity, range predictions of SDMs are affected by biases in opportunistic occurrence records and the lack of information on range limits. Integration of expert range maps in SDMs could help, but this strategy is still rarely used, especially for marine species. We built SDMs for 196 marine fish species with global distributions of Epinephelidae and Syngnathidae, 4 modeling algorithms, and opportunistic occurrence data. We then developed 2 types of SDM ensembles (i.e., combined predictions of multiple individual SDMs): with and without integration of expert range maps. We quantified the level of dissimilarity in range estimates between the 2 ensembles and explored the effects of taxonomic identity, geographic attributes, and conservation status on dissimilarity in model predictions. Although both types of ensembles had good predictive performance, ensembles informed by expert range maps avoided overpredictions of ranges past geographical barriers. Moreover, the dissimilarity between predictions of the 2 ensembles depended on multiple factors, including the number and extent of opportunistic occurrences, distance of occurrences to the expert range polygons, and fish family. Based on our findings, we recommend that researchers combine complementary information provided by expert range maps and opportunistic occurrences when predicting marine species distributions with SDMs.