Generate integrated land cover product for regional climate model by fusing different land cover products

Abstract

Land cover and its change play an important role as a critical variables in global change studies and environment sciences. With many satellite-derived land cover products have become available at global and regional scales, it is possible to improve the quality of the product by integrating existing ones. In this paper, we propose a methodology to generate integrated land cover product for regional climate model over China. The integration rules were established based on the accuracies information from the ground truth. The accuracy of the integrated product was greatly improved, with overall accuracy of 68.7%, and class-specific accuracy ranging from 25.7% to 91.2%. Additionally, the spatial patterns of the land cover over China were well captured, and good agreement with the Landsat-based classification was achieved. Further, our results implicates the quality of the land cover products for integration are significant critical in our approach, the accuracy of the integrated land cover product is dependant on accuracy of the original products used for integration, especially of the local products. This integrated product could potentially improve the performance of regional climate models by providing better estimates of key land surface parameters over the region.