Developing Daily Cloud-Free Snow Composite Products From MODIS and IMS for the Tienshan Mountains

Abstract

Daily cloud-free snow cover products are important premise and foundation for hydrological simulation, climate system research, and snow disaster monitoring in the Tienshan Mountains, Central Asia. In this study, partial clouds appearing in the Moderate Resolution Imaging Spectroradiometer (MODIS) were removed by temporal and spatial filtering, and the remaining cloud pixels were replaced by Interactive Multisensor Snow and Ice Mapping System (IMS). The results show that the retrieved annual mean snow cover obtained from the improved MODIS product data is 31.5% higher than the original MODIS product. Through validation performed via in situ observations, the overall accuracy, land accuracy, and snow accuracy also increased to 88.2%, 91.9%, and 81.9%, respectively. However, the overall accuracy shows a lower accuracy in transitional months compared to other months, which derives mainly from the difference in spatial scales between IMS and in situ observations.