Remote Sensing Dynamic Estimation of Grass Production in Xilinguole, Inner Mongolia

Abstract

The monitoring of grass production is not only an important measurement index for researching the spatial dynamic patterns of grassland resources, but also an important basis for the optimal use of grassland resources and for monitoring the balance between grassland forage supply and livestock demand. A total of 371 grassland investigation and remote sensing (RS) data of MODIS-NDVI from 2005–2009 were used to establish a relational model between grass production and RS data. The temporal and spatial distributions of grass production in Xilinguole, Inner Mongolia were simulated and analyzed. The main conclusions are as follows: (1) Each model equation had good relativity, among which the power function model was the most appropriate for the calculation of grass production. Through reservation quadrat data, the accuracy of the production estimation model was 78%. It was feasible for the power function model to be applied for RS monitoring. (2) Grass production in the Xilinguole grassland amounted to an average of 34550000 tons over five years, equivalent to 11120000 tons of hay, and the dry weight per unit area was 567.23 kg/hm². The spatial distribution of grass production showed a “high east and low west” pattern. (3) An obvious fluctuation in grass production appeared during 2005–2009. The range of hay was 8000000–14000000 tons, and the coefficient of variation was 20.42%. (4) There were greater differences in the single grass yield and the annual variation of different grass types. There were lower grass yields and greater annual variation in desert grassland, and higher grass yields and relatively lesser annual variation in meadow grassland. The spatial and temporal variation of grass yields also had the close relation with the major climate factors (precipitation, temperature, etc.), especially with the influence of precipitation. This research provides a reference for the protection and utilization of grassland resources.