Abstract: 【Objective】High priority has been paid on development of soil geo data base based on soil legacy data by international scientific community in recent years. Temporal and spatial soil information with high resolution, as a new and indispensable tool for researchers and managers, can be applied to soil and environmental quality assessment, arable land conversation, flood and draught mitigation, erosion control, non-point source pollution control, soil and agricultural product pollution control and many other areas. In last decades, several nation-wide soil surveys have been conducted in China. Large number of soil data has been obtained. However, the valuable soil survey data, still in hand drafting map or paper printing format, has been unavailable to researchers and public shares, as the data have been stored at different locations and have not been processed, digitalized and integrated. The purpose of the study is to develop China Digital Soil Maps (CDSM) through extracting soil information from legacy data collected from the whole country. 【Method】Data model of CDSM was constructed by analyzing the characteristics of soil legacy data from different time periods and with consideration of future trends to topics of soil survey. The data model confined the kinds and scopes of information to be extracted and compiled for CDSM. A data treatment working flow was developed, that composed of about 180 separated processes for extracting, analyzing and integrating heterogeneous and massive soil information of different regions and periods. CDSM has been completed by using the methodology for processing massive soil geo data and a computer software program IMAT (Intelligent Mapping Tools) developed in the study. 【Result】CDSM was mainly composed of two kinds of soil geo information. One kind was vector maps of the whole country at 1﹕50 000 scale with polygon objects to describe soil types, soil organic matter, total nitrogen, available phosphorus, available potassium and pH-value. The second kind was point data of about 100 000 soil profiles and 80 000 plough layer samplings with description of sampling point coordinates, sampling year, and soil physical and chemical properties obtained from soil survey works since 1980s. The information contains both the slow-changing soil features such as soil types, parent material, soil mechanical composition, texture and other morphological features and the fast-changing properties such as concentrations of soil carbon and nutrient concentrations. 【Conclusion】The study shows that the CDSM data model is fit both to accept different kind of legacy data from soil surveys across the country over 30 years and to further extend and expand in future. With the S90PT point layer, information of soil sampling points from the new round soil survey works can be added into CDSM. According to time series, space sequence or thematic order, required soil information of the whole country with a map resolution of 100 m×100 m can be inquired and provided by CDSM. The information has been applied by a number of agricultural, environmental and land resource organizations, divisions and research institutes for knowing current soil quality status and understanding changes in soil and environment quality. Along with the extension in time, space and themes in the future, CDSM spatial and temporal soil information will have broader application in various fields.

Key words: China digital soil maps , spatial and temporal information , data model , soil quality , soil profile data