Study On Integration Of Remote Sensing Information And Crop Model To Estimate Rice Production In Jiangsu Province

Accurate dynamic monitoring and production forecast of crop to food security and the development of sustainable agriculture is very important.Crop model is a powerful tool in monitoring crop and forecasting production.But it is difficult to satisfy in the region for its single-point mode.The combination of remote sensing information and crop model using the data of remote sensing information and the simulation of crop model can solve this problem,so that crop model can play to its advantages in a Region scale.In order to estimate the rice planting area and production of Jiangsu province using the remote sensing information combined with the crop model, which is directly useful to have a dynamic monitoring on rice growth to the region of Jiangsu, this study combined the remote sensing information with the crop model ORYZA2000, established numerical simulation model of remote sensing, estimated the rice planting area and yield in Jiangsu province from the single point to the area. This study is divided into the following sections:(1)for proving the applicability of the model in the local,using the experimental data of Zhenjiang to revise and validate the ORYZA2000model,then through the data of eight agricultural test site in Jiangsu province to make an appropriate adjustment and simulation the rice growth in this area;(2)using the geometric correction,radiometric calibration,atmospheric correction and image fusion cutting processing to calculate Normailized Difference Vegetation Index(NDVI)and Enhanced Vegetation Index(EVI) and establish the regression analysis of the vegetation index and Leaf Area Index(LAI) through ENVI software.The results show that the quadratic fit pattern of EVI and LAI by HJ environmental satellite data is optimal;(3)LAI is the contact link remote sensing information and crop model combination:assume that some pixel of LAI conforms the changes of typical growth of rice,so we think this pixel as rice,also conforms a certain type of rice growth.A variety of local rice sowing time from the point of view of crop simulation assumptions,in order to include the variety of rice,floating up and down the simulation of LAI by10%to comply with the type of simulation,in line with established remote sensing image retrieval LAI time series to find out conditions of a pixel,that is consistent with some sort of rice growth simulation,the growth conditions of such rice,so we realize the simulation of rice growth in the region;(4) Simulate the rice of the experimental plot in Zhenjiang,according to the above research ideas.The results show that the rice area and production of the simulation and the actual value of the difference is small,the error is less than7%,and then estimate and monitor the rice in Zhenjing,select the HJ environmental satellite remote sensing data throuth the simulation accuracy of comparison of HJ and MODIS data;(5) Apply the same approach on other sites in Jiangsu province to the rice growth monitoring and yield forecasting.Receive the rice distribution and the total output of Jiangsu province from the result of the eight regional integration to achieve the region simulation of the combination of rice production model and remote sensing information.Comparison with the actual value,higher precision,better effect.