AI-Enabled Analysis of Climate Change on Agriculture and Yield Prediction for Coastal Area

Climate change poses one of the biggest threats to our food security. Hence, analyzing and monitoring the impacts of climate change on agriculture in terms of yield production becomes very important. This project aims to analyze climate change by considering climatic parameters such as rainfall and temperature. Predict the yield using these climatic parameters for one main coastal district. In case the yield is less for a given location of a district, then the project aims to suggest alternate crops which are best suitable for that location. Climatic data is collected from IMD Pune and ZAHRS Brahmavar and stored in CSV format. The data collected is cleaned to get rid of unwanted junk values, null values, etc. The ideal climatic condition to grow crops is also stored in a separate CSV file. Once the user enters the crop, season (Rabi or Kharif), district, and region the machine learning model needs to first predict the climatic conditions for that location entered by the user. The weather forecast is done using the LSTM algorithm using weather data crop can grow or not, and obtain the expected yield if the crop can grow in that region. The model is intended to suggest an alternate crop for the chosen region using the Random Forest algorithm. The data for yield is collected by Crop Production Statistic Information System, Ministry of Agriculture, and Farmer welfare. The accuracy expected for yield production is about 80–85%.