Forecasting Models for Bengal Gram Using Weather Parameters for Bidar District of Karnataka State, India

Abstract

The weather parameters play a very important role in agriculture and data on these is most useful in evaluation of potential aspects of crop production. The present study is focused to develop pre-harvest forecast models for Bengal gram yield based on agro-meteorological data in Bidar district of Karnataka State. The correlation coefficient (r) was calculated to study the week wise association between Bengal gram yield and weather parameters and the value of ‘r’ ranges from -0.62 to +0.65. It is observed that trend of Bengal gram yield with respect to year is almost linear. The forecast models were developed using the weather parameters namely Maximum Temperature, Minimum Temperature, Morning Relative Humidity, Evening Relative Humidity, Wind Speed and Rainfall. The weather indices were developed using the data from 2006 to 2017 and the validation of the developed models were confirmed using data of the year 2018 to 2020. The study found that high value of adjusted R2 in the developed forecast model and for the year 2019 the predicted values were between 0.5 to 6 % lower than actual values. The study shows that pre-harvest forecasting can be done using the best fitted model in the 10th SMW (i.e., 5th March to 11th March) before harvest of the Bengal gram crop.