Spatio-temporal evolution and potential analysis of cropping index in Sichuan Province during 2000-2016

Cropping index is one of the basic indexes used to measure use intensity, space-time distribution and changes of farmlands at regional scale, which is essential for basis agricultural decision-making. To investigate cropping indexes, the related changes and potential cropping indexes of farmlands in Sichuan Province, a long-term (2000-2016) series of MODIS-NDVI remote sensing data products were used to reconstruct NDVI time-series curve using Asymmetric Gauss Fitting Function (AGFF). The peak frequencies of the NDVI time-series curves were used for determining the spatial distributions and temporal changes of annual cropping indexes by using the second order difference method. Then the spatial pattern of potential and promotable potential cropping indexes of farmland in Sichuan Province were simulated using the temperature-rainfall model, meteorological data and land-use maps to refine farmland dynamics in the region. The aim of the study was to provide scientifically-drive recommendations to local governments on the regulation of intensive development of farmlands in the study area. The results showed that single cropping pattern, which made up 78.05% of cultivated land, was the main cropping system in Sichuan Province for the period 2000-2016. This mainly distributed in the Northwest Sichuan Plateau region, Northeast Sichuan, Southwest Sichuan Hilly area and Panxi Valley region. The double cropping system mainly distributed in Chengdu region and the surroundings. Overall, the spatial distribution of cropping index (MCI) in Sichuan Province decreased in trend from Chengdu and the surroundings to other regions. According to the promotable and potential cropping indexes, Sichuan Province had 89.7% of cultivated land with higher promotable potential in multiple cropping. Chengdu and the surroundings had lower promotable potential cropping index (PMCI_p < 50%) and higher inter-annual fluctuation (standard deviation of cropping index > 60%) than other regions in Sichuan Province. Cultivated land in the northeastern and southern areas of Sichuan (which accounted for 65.6% of the total area of Sichuan), was higher in promotable potential cropping index (PMCI_p ≥ 50%) and less inter-annual fluctuation compared with Chengdu and the surroundings. There was the likelihood for this to become the best suitable land with the highest cropping index in Sichuan Province. Accounting for 10.3% of cultivated land area of Sichuan in the northwest and northeast mountainous areas (with negative promotable cropping index) formed the main region of agricultural over-development in the study area. There was therefore the need to classify these farmlands into ecological restoration areas. The cropping indexes extracted by integrating the phenology of crops and long-term series of remote sensing data were significantly correlated with the statistics values in Sichuan, indicating the method was applicable in evaluation of cropping index in provincial scale. The result was beneficial reference to policy decision of regional planting structure.