Estimate grazing spatial distribution and quantity from MODIS LAI data (Rui Yu)
Author: Rui Yu, CSAP PhD student.
My research is to establish an agent based model exploring a dynamic sustainable rangeland management that balances the socio-economic and ecological function of the rangeland grazing system in Zeku. The first step is to estimate how much grass forage the land could provide. It is almost impossible for me to undertake a large-scale field survey in an 11300km² rangeland, but remote sensing derived datasets provide a continuous, large-scale, cheap and convenient information of rangeland. They are also extensively employed in the field of rangeland productivity estimation. Among them, Leaf Area Index (LAI) and Normalized Difference Vegetation Index (NDVI) are the two commonly used measures to quantify the vegetation status of rangeland.
However, when taking advantage of remote sensing derived data, one important reality has always been most commonly, ignored. This is the remote sensing data can only observe the time slice status of vegetation rather than the whole process of vegetation development. Ignorance of this fact would lead to the totally underestimate or wrong result of rangeland productivity, especially in the field of carrying capability calculation in grazing intensive regions.
Estimated grazed LAI in ecological economic cooperation grazing group area
The LAI value of land patches would always be underestimated due to the LAI consumption by herbivores. This effect would be overwhelming obvious in the grazing intensified area. Remote sensing derived datasets such as Moderate-resolution Imaging Spectroradiometer (MODIS) are widely used and extensively validated around the world, and the ignorance of the effect of herbivore remover of vegetation is acceptable on a global scale of vegetation carbon assimilation or fixation, especially in the forest area. However, in the context of rangeland, especially in grazing intensified areas, the LAI consumption by livestock could have a significant effect on the quantity and quality of grass productivity. In fact, some of the researches even argued that grazing coupled with climate change are the main factors contributed to the regional grassland degradation and even desertification. It may directly lead to the green land to the bare land, thus a rapid decrease of LAI would be observed.
Considering the problems that have been mentioned above, it is of great importance to identify the spatial distribution and quantity of consumed LAI by livestock on rangeland. This is also the fundamental basis of my PhD research, that is, to identify where the land patch has been grazed and how much has been grazed in summer and winter separately. Here, a new empirical growth-defoliation-grazing function is developed to analysis MODIS15A2H006 product. The following figure shows the spatial distribution of the grazed amount of LAI in Zeku, China (2011) by this method. Unfortunately, I don’t have the winter and summer pasture map to validate this result, hope I can get it soon and tell the accuracy of my method. Another problem is the current method cannot deal with the fixed-site grazing, but I think it can be solved with more information we can explore from MODIS dataset.
Author: Rui Yu