在全数字方式下对宁夏草地覆盖的遥感宏观研究

Macro-scale Study on Grassland Cover of Ningxia Region by Remote Sensing in the Digital Way

以 TM影象为信息源 ,在全数字方式下运用遥感技术与地理信息系统技术结合建立了宁夏草地覆盖的 TM影象数据库、图形数据库和属性数据库。通过分析得出全区现有草地面积 2 4813 4 5 hm2 ,占全区面积的 47.92 % ,其中高覆盖度草地占 3 .93 % ,中覆盖度草地占 43 .45 % ,低覆盖度草地占 5 2 .62 %。

Using digital method of remote sensing and geographical information system techniques, we have interpreted TM images(1996) and established the database of grassland cover in Ningxia Province. The database includes TM image database, graphical database (scale 1∶100 000) and its attributing database. In order to interpret correctly, the interpreting key of all kinds of grassland cover on TM images have been established through analyzing relative reference and images and then these keys have been corrected in the fields. The definition of the grassland is that the vegetation is mainly consisting of grass and its coverage percentage is more than or equal to 5% or grass and shrub land where shrub coverage occupies less than 10%. In the practice, the grassland is classified into three classes according to the vegetation percentage: (1) High coverage grassland: the vegetation coverage percentage is more than or equal to 50%. (2) Moderate coverage grassland: the vegetation coverage percentage is more than or equal to 20% and less than 50%. (3) Low coverage grassland: the vegetation coverage percentage is more than or equal to 5% and less than 20%. Through the study practice and analyzing of grassland databases, the following points can be concluded: (1) In Ningxia Region, there was as much as 2 481 345 hm\+2 grassland and its area percentage is 47.92% in the middle period of 1990's. (2) In the whole grassland, the high coverage grassland's percentage is 3 93%, the moderate coverage grassland's 43 45% and low coverage grassland's 52 62% (3) Comparing with traditional methods, there are many advantages using digital processing method to establish the grassland database such as, mapping rapidly, high accuracy and reducing expenses. (4) With the same interpreting key and graphical database and attributing database that are well matched with image database, we can monitor accurately the grassland in large areas.