Relationship between Boro Rice Production and MODIS Derived NDVI for Rice Production Forecasting: A Case Study on Bangladesh

Abstract

The present investigation illustrates an inclusive approach to extract remotely sensed Normalized Difference Vegetation Index (NDVI) from Moderate Resolution Imaging Spectroradiometer (MODIS) (AQUA/TERRA) imageries to find out a relationship with Boro rice production for forecasting crop production in the context of Bangladesh. This study utilizes AQUA/TERRA MODIS reflectance data (250 m resolution) for the month of March (Peak-greenness period) to calculate the average NDVI values by following MODIS based algorithm at district level during 2011-2016. The linear regression analysis of calculated average NDVI and BBS estimated Boro rice production statistics reveals a significant positive relationship due to maximize photosynthetic activities. Among the regression equations from (2011-2016), the highest regression coefficients R2=0.87 and R2=0.85 for AQUA and TERRA MODIS data have been found respectively in 2015. Therefore this regression equation can be used for future estimation of Boro rice production at country scale. However, further testing and simulation of this regression model is required to generate Boro rice production forecasting dataset on timely basis. Hence this study summarizes that, NDVI based regression equation may be an effective process to forecast the Boro rice production which can play an important role in decision-making process relevant to the food security issues of Bangladesh.

The Dhaka University Journal of Earth and Environmental Sciences, Vol. 8(1), 2019, P 33-40