A Methodology For Multiple Cropping Index Extraction Of Meijiang Basin Based On MODIS/NDVI

Cultivated land is the important resource to survive and develop for human, andis also the important part of ecological environment system. The multiple cropping isthe way of planting more than a cooked crop in a year on the same land, which canimprove the land-use efficiency and food production. But the cultivated land is usedmore often, strengthening the impact to ecological environment, such as the non-pointpollution to the ecological environment when improving yield. Nowadays, there arevarieties of models that assess the non-point pollution of cultivated land to theecological environment, but these models do not consider the background of intensivecultivation and multiple cropping, such as SWAT model, couldn’t accurately assess thenon-point pollution of those areas to the ecological environment. For this, on the basisof previous studies, the paper set up a model to extract the multiple cropping index offarm belt to provide data support for the improvement to the existing models ofnon-point pollution.This paper chose the typical subtropical multiple cropping area--MeiJiang basin,as the study area, visually interpreted the land use/cover with the2010ALOS remotesensing image of study area, used HANTS(Harmonic Analysis of Time Series) toreconstruct the NDVI time series curve with2006-2011six years of MODIS/NDVItime series data of study area, and then designed model in the GIS softwareworkbench, extracting the time series NDVI curve’s wave information byTwi-differential algorithm, receiving the multiple cropping index spatial distributionbased on single pixel, and at last verified the2011’s multiple cropping index by thesampling data of2011.The main conclusions are as follows:(1) The pixel level’s multiple cropping index can be extracted by modeling. Bymodeling on the GIS software bench, the pixel level’s multiple cropping index can beextracted that is different from the old ones.(2) The multiple cropping index extracted by this model can provide basic datafor quantitative evaluation that intensive cultivation and multiple cropping’s landnon-point pollution to the ecological environment. Indicated by sampling data, thecultivated land multiple cropping index’s accuracy rate is83.3%, which can providemodel support for improvement of excited non-point pollution model.