Premium
Using remote sensing to model tree species distribution in Peruvian lowland Amazonia
Author(s) -
Pérez Chaves Pablo,
Ruokolainen Kalle,
Tuomisto Hanna
Publication year - 2018
Publication title -
biotropica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.813
H-Index - 96
eISSN - 1744-7429
pISSN - 0006-3606
DOI - 10.1111/btp.12597
Subject(s) - amazon rainforest , thematic mapper , amazonian , remote sensing , elevation (ballistics) , thematic map , geography , vegetation (pathology) , digital elevation model , species distribution , environmental science , physical geography , satellite imagery , ecology , forestry , cartography , habitat , mathematics , biology , medicine , geometry , pathology
Species distribution models for Amazonian trees have mostly been produced at scales and resolutions that are too broad and coarse for practical use in either conservation or forestry. On the other hand, several studies have shown that elevation and the medium‐resolution remote sensing data available via Landsat imagery can be successfully used to detect differences in plant species composition in Amazonia. Therefore, it seems likely that the same data can also be used to predict geographical distributions of individual taxa. Here we use remotely sensed data and a maximum entropy algorithm (MaxEnt) to generate landscape‐scale distribution models at 30‐m‐resolution for five economically important timber tree genera ( Apuleia , Amburana , Crepidospermum , Dipteryx, and Manilkara ). Individual Landsat Thematic Mapper bands and normalized difference vegetation index yielded acceptable model performance, and the use of averaging filters (3 × 3 and 5 × 5 pixel low‐pass filters) improved model performance further. Including elevation as a predictor also improved model performance for all the genera. Our results suggest that it is possible to use Landsat bands and elevation as predictors for modeling the potential distribution of tree species in lowland Amazonia at a fine enough resolution to facilitate the practical management of forest resources.