The development of space remote sensing in 1970s was very useful for developing countries such as China, which has a vast territory and unclear resources. The forestry department also attaches great importance to it. It uses MSS, TM and SPOT satellite data to investigate and test the forest resources in the south, northeast and northwest of China. China's "Seventh Five-Year Plan" specifically lists the "Three North" shelter forest remote sensing project, and makes a comprehensive remote sensing and ground comprehensive investigation on various vegetation types and their land potential in the forest grassland and Gobi desert in Northeast China, North China and Northwest China to solve the forest site conditions, grassland livestock carrying capacity and population. The macro, real, fast, multi-band and multi-temporal remote sensing provides favorable conditions for solving this problem. In order to unify the pace and understanding, train cadres, formulate norms and explore practical remote sensing investigation methods before large-scale construction, Pingquan County, Chengde District, Hebei Province was selected as the experimental area in June 1986, and nearly 100 scientific and technological personnel from 37 units were concentrated. In the past two years, the Technical Specification for Remote Sensing Comprehensive Investigation of Shelterbelt in the Three North Areas has been completed, and a series of thematic maps have been compiled and published.
In the Pingquan experiment, a variety of remote sensing images (MSS, TM and SPOT satellite HRV images), China land satellite images and color infrared aerial photographs with various scales (1:130,000,1:75,000,1:30,000) were tested, and their economic comprehensive technology was carried out.
It is the first time in China to use ultra-small scale (1: 1.3 million) color aerial photographs to classify vegetation, and satisfactory results have been achieved. The aerial photograph of 1: 130000 can be enlarged to 1:5000, which can distinguish the ground features of 1-2m, and has a good resolution to Robinia pseudoacacia, Pinus tabulaeformis, poplar, farmland, grassland and orchard.
Remote sensing research of vegetation can not only be used to classify vegetation types, but also further study its seasonal and dynamic changes and crop yield estimation, and make full use of the advantages of multi-temporal to predict the yield of wheat, rice, cotton, soybean and other crops. China has carried out wheat yield estimation in 3 13 counties and cities in Henan, Hebei, Beijing, Tianjin and Huang-Huai-Hai Plain (Xu Xiru, 199 1).
(2) Soil remote sensing research
The application of remote sensing technology in 198 national soil survey not only improved the accuracy, but also saved manpower and financial resources, and achieved gratifying results. Subsequently, a county-level soil survey was carried out, which required more details. Most counties use aerial photographs. Generally speaking, soil interpretation is a multi-factor analysis based on topography, geology, vegetation, hydrology, land use and other data, especially small and medium-sized satellite images for soil interpretation and mapping. The soil interpretation of Datong sheet in Shanxi province makes full use of the advantages of environmental factors and soil geography, and overcomes the weakness of soil display on satellite images. Experiments show that autumn images are the best and have great interpretability. Shanxi agricultural remote sensing has evaluated the1:250,000 soil interpretation map of the whole province, and it is considered that the soil interpretation rate can reach 95%. In many areas of China, remote sensing method is used to carry out soil survey and compile soil type map, and the remote sensing application of soil survey has also developed from static research to dynamic research. Li Tianjie and others used satellite remote sensing technology, digital simulation, routine investigation and geographic information system to monitor and predict the water and salt movement in saline-alkali land in northern Shaanxi Basin.