Abstract: Desertification is one of most serious environmental and social-economic problem in the world, which has been suffering China for a long bine. Desertification brought about environmental deterioration and land degradation, and caused heavy losses of economy. Therefore, it is necessary to cap out binely remote sensing monitoring and assessing desenification to study its mechanism and process of development and to understand its situation and trend, and to provide scientific basis for the state govermment in developing the national programs for combating desertification. In this paper, we took sandy desertification in northern China as an object of study.According to research and practices for nearly 20 years, we consider that the sandy desertification is land degradation mainly caused by excessive human activities through wind erosion. Repeated monitoring is reqinred to establish the space and time variability of the sandy desertification parameters, and to idetify the regions at risk of severe desertification. Remote sensing has proven its potential for the environmental observation and has become a powerful tool in many fields as well as in sandy desertification monitoring. The main remotely-sensed dare for the research project were,a) black and white aerial photographs on 1: 50 000 scale during 1975～1977 and in 1985, and b) Landsat-MSS satellite images of 1975～1977, Landsat-TM satellite images of 1984～1988 and 1993～1994 generated by a combination of band 2, 3, 4or 3, 4, 5 on 1: 200 000 or 1: 250 000 scales. The studied region's topographical maps of if 50 000 and if 100 000 scales and background maps, such as the map of vegetation, soil, climatic, hydrology, geology, geomorphology etc. and the map of administrative area. land use, and the statistics such as population, livestock, products of agricultute and animal husbandry varied in years, and relevant reports were used as basic information source. Based on above materials, we have identified the dominant and associate processes of sandy desertification and their seventy from the available field information and visual interpretation of remotely sensed data. The most important for monitoring is to determine a system of the classification of sandy desertification. Sandy desertification indicated by wind erosion activities and the processes of physical and chemical deterioration of the desertified land will continue with the intensity of the activities. Those processes can be divided into four classes, such as sligt(L), moderate(M), severe(S) and very severe(VS), according to the different status based on quantitative and qualitative assessments of the desertified land, which are helpful not ohly for understanding the processes and their situations but also for employing control measures. The potential criteria should be representative and practical for monitoring the sandy desertification, which indicate the existence of desertification through statistics and phenomena related closely to the environmental conditions of the desertified processes. They also can be classed into direct and indirect criteria to be identified respectively as essential factors of diagnosis and supplementary indexes. On the basis of extensive field investigation and remote sensing data of different periods, we established a preliminary degree class and index system of sandy desertification and developed a suitable method for remote sensing monitoring and assessing sandy desertification. The results indicate that during last decades the sandy desertification in northern China has been caused mainly by irrational human economic activities, and its general trend is development with a few areas changing in reverse.For example, since the middle 1970's to the late 1980's the sandy desertified land has expanded at least 2 100 km2 annually, which happened principally in the interlocking area between cultivated land and grazing land, and rainfeed land and pastoral area. During the same bine, about 10% of the sandy desertified land had been controlled. The spread rate of sandy desertified land in