Remote Sensing Monitoring Of Crop Growth In Songnen Plain Based On Long-time Series MODIS Images

Remote sensing monitoring of crop growth is one of the key tasks in the field of agricultural remote sensing research.Obtaining crop growth data as early as possible is important for early warning of agricultural production,decision-making of agricultural policies and estimation of crop yield.It is indispensable to monitor the long-term growth status and spatial dynamic evolution of crops in Songnen Plain in Northeast China,which is the main grain producing areas in China.Because all kinds of natural disasters that may occur at any time will have various adverse effects on farmland planting and the structure and function of agro-ecosystem.Traditional crop growth monitoring is mainly carried out by manual investigation or field observation.Although the accuracy is relatively high,their spatial coverage is often insufficient,and at the same time,it wastes time and energy,resulting in low efficiency.In the past three decades,the application of remote sensing technology in crop growth monitoring has been widely studied.Accurate and timely monitoring of regional crop growth and early estimation of crop yield are of great help to agricultural planning and policy making.Based on this,this paper takes Songnen Plain as the research object,adopts a new comprehensive remote sensing monitoring index of crop growth,and develops a largescale method for monitoring crop growth based on the MODIS time series image data from 2000 to 2020,which can accurately analyze the crop growth and forecast the yield,and can quickly obtain the growth information of the latest remote sensing image.The normalized difference vegetation index（Normalized Difference Vegetation Index,NDVI）and enhanced vegetation index（Enhanced Vegetation Index,EVI）are used to construct a comprehensive growth monitoring index GI（Growth Index）.By using the difference model,the growth status of crop growing season shown by the latest remote sensing images is compared with the average growth situation of crop growing season in the past two decades.To evaluate and analyze the crop growth in different months of the latest growing season in Songnen Plain.Through the monitoring and analysis of crop growth in Songnen Plain from 2018 to 2020,the main results are as follows:（1）After the long-term monitoring and analysis of crop growth season in Songnen Plain from 2000 to 2020,the comprehensive GI is generated by using Arc GIS and MODIS Reprojection Tool and combined with NDVI and EVI.Regression analysis is made on the monitoring index of crop growth season in fifty-three counties and cities of Songnen Plain and the yield data of the corresponding areas.According to the determination coefficient,it is determined that the best coefficient value is when NDVI is 0.6 and EVI is 0.4,R~2 is the highest,with a value of 0.578.It has been verified that the comprehensive monitoring index of crop growth at the county scale in Songnen Plain has the highest correlation between GI and per unit yield（R~2 = 0.589）,which is better than the single NDVI（R~2 = 0.341）and EVI（R~2 = 0.458）,and higher than the previous domestic and foreign scholars’ threshold selection method based on a single vegetation index to evaluate crop growth.（2）Through the quantitative analysis and verification of the interannual and annual yield changes in Songnen Plain,it can be concluded that the harvest of autumn grain crops in 2018 is stable but poor.In the time distribution from May to October in 2019,the crop growth in Songnen Plain showed a trend of "worse in the early stage,better in the middle stage and slightly worse in the later stage".The crop growth in Songnen Plain in July and August was better than that in the early period（2000-2018）,and worse in June and October.The growth of crops is generally better than that of the same period in 2000-2018,and crop yields in the Songnen Plain in 2019 remain stable compared with the past two decades,ensuring that China is self-sufficient in grain.The overall growth of crops in 2020 is slightly lower than that of crops in the same period in previous years.（3）Through the spatial quantitative analysis and verification of the interannual and annual yield changes in Songnen Plain,it can be concluded that the crop growth in the central,western,northeast and southwest of Songnen Plain in recent years is better than that in the past 20 years.In some cities in the southeast,the crop growth is basically the same as in the past.Moreover,the crops in Songnen Plain are growing well along the Songhua River basin,specifically in the central part of Harbin,the north of Songyuan City and the southeast of Daqing City,mainly because of the abundant water resources,and the good weather factors in the main growth period of crops are also the main reasons for the better growth of crops in Songnen Plain.The comprehensive GI monitoring results constructed in this study are more accurate and can effectively evaluate the large-scale crop growth and forecast the crop yield in time than a single vegetation index.