Study of soil moisture remote sensing retrieval system based on Interactive Data Language

It has long been recognized that soil moisture is a significant parameter of climatological, hydrological and ecological systems. In energy exchange and matter exchange between land surface and atmosphere, soil moisture is also a key state variable. It is very important to monitor soil moisture in large range for hydrological and meteorological process and laws research, as well as for agriculture production. The monitoring methods of soil moisture include site observation on the ground and monitoring by remote sensing. The application of site observation is suitable for small range monitoring because it is labor-intensive and equipment-intensive. While, remote sensing has the positive aspects in monitoring surface soil moisture(within several centimeters) status in large range. However, the popular soil moisture data products from remote sensing are so rough in spatial resolution that it is more suitable for global scale rather than regional scale. Being affected by complicated factors, such as meteorology, topography, soil properties and vegetation, soil moisture is a highly variable parameter both temporally and spatially. From the view of regional scale research, it would lead much non-determinacy to relevant research, no matter using soil moisture data from ground-based observation constrained in small range or using those already existing data products from satellite observation rough in spatial resolution. For improving research reliability of regional scale, it demands observation data not only with high spatial resolution but also in large scale. Aiming at this issue, this paper adopting MODIS(moderate-resolution imaging spectrometer) data with middle resolution(1km) as data source, employing thermal inertia model as retrieval model, using Interactive Data Language(IDL) as platform, develops a system of soil moisture retrieval for regional scale research. This system provides functions including fast data preprocessing for massive remote sensing data, processing for measured data, soil moisture data retrieval as well as the rapid output of retrieval result. It would support the soil moisture fast retrieval in regional scale, make up the defects of rough resolution of existing soil moisture data production, realize the fast automatic processing for massive remote sensing data.