Improving the validation of ecological niche models with remote sensing analysis
2018
José-Silva, Leandro | dos Santos, Reginaldo Carvalho | de Lima, Bruna Martins | Lima, Mendelson | de Oliveira-Júnior, José Francisco | Teodoro, Paulo Eduardo | Eisenlohr, Pedro V. | da Silva Junior, Carlos Antonio
Ecological Niche models (ENMs) are tools that allow us to approximate the area of suitability for a species, thereby allowing elaboration of conservation strategies. The validation of these models in situ is not always possible due to costly access remote areas where conserved species are often found. The goal of our study was to provide a new validation concept for ENMs by applying remote sensing (SR) techniques, such as Geographic Object-Based Image Analysis (GEOBIA), which enables mapping of large areas and provides detailed information on land use. To assess the GEOBIA validation technique, we selected the species Bertholletia excelsa (Brazil nut), a tree that has great importance as a non-timber forest product and is considered vulnerable by the International Union for Conservation of Nature (IUCN). Models were built on the ‘biomod2’ package, and evaluation was conducted using the area under the receiver operating characteristic curve (AUC) and True Skill Statistics (TSS) metrics. Images were obtained from the orbital Operational Land Imager (OLI) on board the Landsat-8 satellite and the thematic maps were evaluated using Kappa and Overall Accuracy Statistics. We calculated vegetation indices (EVI, SAVI, LAI, and NDVI) and applied them to the GEOBIA technique. A total of 693 possible sites of B. excelsa were detected. Of these, 25 accessible sites were used for validation, and 45 new records of B. excelsa were added in the study area. GEOBIA was demonstrated to have high potential for validating ENMs, as well as in the extraction of arboreal species from medium-resolution spatial images.
Afficher plus [+] Moins [-]Mots clés AGROVOC
Informations bibliographiques
Cette notice bibliographique a été fournie par National Agricultural Library
Découvrez la collection de ce fournisseur de données dans AGRIS