Nexus among rice production and environmental factors in the coastal region of Bangladesh: a stochastic modeling approach for future forecasting

Abstract

Sustainable rice production in the coastal region of Bangladesh is threatened due to various climatic impacts. The study investigated the current scenario of rice production and identified the linkage among different input variables related to rice production such as seed, urea, pesticide, etc. in three districts of Patuakhali, Khulna, and Cox’s Bazar, using a log–log model and panel data analysis. The Stochastic Model of multiple regression analysis helped to identify the most important input variables. The input factors found common in all three study areas were seed. The log–log model was used to identify how climatic factors influenced these input variables. Identifying such factors would help to ensure sustainable agriculture production in the future. Seed cost showed a negative elasticity in Patuakhali, while urea cost showed a negative elasticity in Khulna region. Time series forecasting of rice production up to 2030 in the study areas was examined using the best fitted ARIMA model. While rice production showed an increasing trend in two regions, it did not show a satisfactory upward trend in Kutubdia, which is mostly related to their mismanagement of fertilizer use and irrigation water scarcity. The analysis from the research indicated that an area-specific adaptation plan is required to sustain the agricultural production, as the input variables for rice production were not found similar for three different study districts where a specific attention needs to be given. Moreover, to address the nonexistence of quality panel data for rice production, measures have been introduced which includes an online data collection system using a mobile application.