Studying on Crop Response Model for Grapes under Climate Change Scenario: Statistical Study Approach

Abstract

Grapes vine originally a temperate fruit crop and it’s also grown successfully under tropical conditions. Grape is one of the economically important fruit crops grown in India. As Theni district is the leading producer of grape in Tamil Nadu, followed by Coimbatore and Dindigul, this study is centred on the Theni-Kambum block area. In this region, Muskat Humbug is a well-liked cultivar that yields more than other varieties. In this study, this cultivar was employed. In this study, an artificial neural network (ANN), multiple linear regression (MLR), and elastic net (ELNET) regression methods were used to construct a yield prediction model (ANN) using twelve years secondary data. Additionally, we evaluate our model over a two-year period using field-level data from GRS and neighbouring farms. Finding the best-fit model for predicting grape yield and PDI using meteorological parameters in the Theni district is the goal of this communication. The model is chosen according to many performance indicators including RMSE, MAPE, MAE, and R2. Among the three techniques developed in the study, the Artificial Neural Network is found to be best for prediction of grape yield based on weather and disease incidence for the available data in the studied region.