Regression model to estimate lfood impact on corn yield using MODIS NDVI and USDA cropland data layer

• Published in: 农业科学学报（英文） Vol. 16; no. 2; pp. 398 - 407
• Main Authors: Ranjay Shrestha, Liping Di, Eugene G Yu, Lingjun Kang, SHAO Yuan-zheng, BAI Yu-qi
• Format: Journal Article
• Language: English
• Published: Center for Spatial Information Science and Systems CSISS, George Mason University, VA 22030, USA%State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, P.R.China%Key Laboratory for Earth System Modelling, Ministry of Education/Department of Earth System Science DESS, Tsinghua University, Beijing 100084, P.R.China 2017
• Subjects: agriculture; regression; remote sensing; NDVI; MODIS; corn yield
• ISSN: 2095-3119; 2352-3425
• DOI: 10.1016/S2095-3119(16)61502-2

Flood events and their impact on crops are extremely signiifcant scientiifc research issues; however, lfood monitoring is an exceedingly complicated process. Flood damages on crops are directly related to yield change, which requires accurate assessment to quantify the damages. Various remote sensing products and indices have been used in the past for this purpose. This paper utilizes the moderate resolution imaging spectroradiometer (MODIS) weekly normalized difference vegetation index (NDVI) product to detect and further quantify lfood damages on corn within the major corn producing states in the Midwest region of the US. County-level analyses were performed by taking weighted average of al pure corn pixels (>90%) masked by the United States Department of Agriculture (USDA) Cropland Data Layer (CDL). The NDVI-based time-series difference between lfood years and normal year (median of years 2000–2014) was used to detect lfood occur-rences. To further measure the impact of the lfood on corn yield, regression analysis between change in NDVI and change in corn yield as independent and dependent variables respectively was performed for 30 different lfooding events within growing seasons of the corn. With theR2 value of 0.85, the model indicates statisticaly signiifcant linear relation between the NDVI and corn yield. Testing the predictability of the model with 10 new cases, the average relative error of the model was only 4.47%. Furthermore, smal error (4.8%) of leave-one-out cross validation (LOOCV) along with smaler statistical error indicators (root mean square error (RMSE), mean absolute error (MAE), and mean bias error (MBE)), further validated the accuracy of the model. Utilizing the linear regression approach, change in NDVI during the growing season of corn appeared to be a good indicator to quantify the yield loss due to lfood. Additionaly, with the 250 m MODIS-based NDVI, these yield losses can be estimated up to ifeld level.