Ecosystem mapping in mountainous areas by fusing multi-source data and the related knowledge

Mapping and modeling the complex ecosystems and their changes over time are key issues in spatial ecology, biogeography, ecosystem ecology and biodiversity researches. This paper attempts to propose a simple, practical and automatic method to produce the ecosystem map in mountainous areas by fusing multi-source data and the related knowledge. The multi-source data included the 30m-resolution land cover map and the vegetation map of China (1:1 000 000). Three fusion strategies were contained in the proposed approach: hard matching, buffer matching and merged categories matching. Meanwhile, the related spatial distribution knowledge and the law of spatial distance decay were used to determine the optimal vegetation type, when more than two vegetation types are matched simultaneously. Taking the Southwestern China as study area, a new 30m-resolution ecosystem map with 144 ecosystem types was generated by the proposed method, which was used to establish the red list of ecosystems and evaluate the condition of biodiversity in the White Paper: China's Biodiversity.