Random Forest models compare environmental and spatial predictors of Thai Paphiopedilum distributions to prioritise surveys and conservation. Sampling bias, imperfect detection and changing habitat can make static suitability maps appear more certain than they are.
Key findings
- Random Forest models compare environmental and spatial predictors of Thai Paphiopedilum distributions to prioritise surveys and conservation. Sampling bias, imperfect detection and changing habitat can make static suitability maps appear more certain than they are.
Why this matters globally
This work adds internationally comparable evidence in Environment and defines questions for replication in other populations or systems. Its global value lies in the evidence and transferable reasoning, not in a single impact score.
Thai researcher contribution
Thailand-linked authors and Burapha University, Mahidol University contribute to the research network behind this work. Thai participation is identified from bibliographic affiliations and should be checked against the author list and source article.
Limitations to consider
Performance can fall under dataset shift; external validation, leakage checks, calibration and post-deployment monitoring are needed.