AI Science Tracker
EN

Forest models link land-use patterns to recorded envenomation incidence

Federal University of Rio Grande; Federal University of Alagoas · 3 Nov 2026 · climate

Discovery

Forest models link land-use patterns to recorded envenomation incidence

Source check: Source checked. Not independent scientific confirmation.

Finding: Random forests analyzed monthly reported scorpion, snake and spider envenomation across 645 São Paulo municipalities from 2010 to 2024. Urban land proportion had the highest permutation importance for observed incidence in all three models, ahead of weather variables. This is an ecological predictive association, not evidence that urbanization causes individual incidents. Random train/test splitting ignored spatial and temporal dependence, reporting may be incomplete, and hypothetical warming outputs are unmeasured sensitivity scenarios.

Technology: Random forest; permutation feature importance

Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.

AI contribution
Forest ensembles learn nonlinear relationships between municipal land-use/weather measurements and reported incidence; permutation increases in out-of-bag error identify the predictors most informative within each fitted model.
Test and result
Random-forest models of existing municipal notification records, with 80:20 random train/test splitting and out-of-bag permutation importance. The represented result concerns predictor importance for observed 2010–2024 incidence.
Limitation / still unknown
Causal individual risk, generalization under spatially or temporally separated validation and actual effects of future warming remain unestablished. Socioeconomic and surveillance differences were omitted; underreporting can affect recorded incidence.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC13731446/

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 21 Nov 2026. Review scope: Primary indexed PMC full text: data/model Methods, measured-incidence feature-importance results and limitations, exact canonical citation identity/date, current primary PubMed notice relationships. Warming scenarios reviewed as predictions; supplementary municipality rankings not inspected..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse climate discoveries.