Capturing landscape changes and ecological processes in Nikko National Park (Japan) by integrated use of remote sensing images

摘要

Landscape change is of crucial ecological importance because of its impacts on biodiversity and the sustainability of habitats and landscapes. The ability to infer ecological processes from landscape changes provides a powerful new tool for large-scale and regional ecological research. Accordingly, we investigated landscape changes in Nikko National Park (Japan) from 1981 to 2006 using an integration of remote sensing images to track ecological processes in the park. We showed that natural succession and anthropogenic disturbance have caused an enormous expansion of coniferous forest (mostly Larix kaempferi and Tsuga diverifolia), which has resulted in an alteration of sika deer (Cervus nippon) habitat structure. The deer population has vastly increased in size, with concomitantly negative effects on the biodiversity of forest-floor vegetation. All changes that we detected have degraded landscape and ecosystem sustainability. Our integrated use of remote sensing images in 2006 accurately identified and classified landscape pattern in the study area. We achieved an overall accuracy of 96.48%, demonstrating that the procedure has an enormous potential for synthetic processing of multitemporal and multiresource remote sensing images.