Smart Weather Data Management System for Sustainable Agriculture in Maharashtra Using Machine Learning

Abstract

Agriculture will benefit greatly from efficient weather data sustainable management. Data can be utilized to schedule the prediction yield, crop growth, and irrigation. This paper collects real-time meteorological station data via APIs, primarily focusing on the weather data sustainable management system and data related to Maharashtra Wardha, India, which is involved in a live series period. Agriculture can make potential through the growth of Smart Weather Data Management (SWDM) platforms. Real-time data from India, combined with intelligent insights, can be used to propose innovative and sustainable solutions. Four layers are involved: i. Acquisition ii. Storage iii. Application iv. Processing. ML checks for errors and missing values based on Land-Wardha reanalysis once the information is obtained. To assess the accuracy of the Temperature, Humidity, and Rainfall approximations with the use of some evaluation metrics with coefficients of determination (R2-Scores), (RMSE), and (MSE). The system's services include weather forecast time series, meteorological data visualization and analysis, and ML-based evaluation. Real-time services for temperature, humidity, wind speed, and cloud cover given the prediction for the Previous /Next five days in Maharashtra, Wardha, India. The platform is built with the aim of providing services and solutions that can assist both farmers and representatives.