Climate-Adaptive Intelligent Irrigation Model for Water-Efficient Cultivation of Horticultural Crops

Abstract

Climate variability and water shortage have emerged as the significant problems of sustainable growing of horticultural crops and irrigation. In many cases, the traditional irrigation techniques tend to lead to wasteful use of water, poor timing and minimal yield in the varying climatic conditions. This research aims to overcome these shortcomings by defining a smart irrigation model that is climate-adaptive to water-saving horticultural crop growing based on the Internet of Things (IoT) sensors and random forest regression methods. The suggested framework combines real-time monitoring of the environment, and machine learning-based irrigation forecasting to optimize water use and enhance irrigation decision-making. The environmental conditions such as soil moisture, temperature, humidity, rainfall, solar radiation and wind speed are measured with the help of IoT-based monitor devices and are used to train the model and predict. Random Forest regression model was used to determine the irrigation needs of crops in different climatic conditions because it is robust, has nonlinear learning capability, and is capable of determining the needs of crops with high accuracy. The regression performance in terms of minimized Root Mean Square Error (RMSE) and minimized Mean Absolute Error (MAE) and high coefficient of determination (R 2 Score) presented in experimental evaluation represented accurate prediction of the irrigation through the use of the model. Over-irrigation was also reduced and the irrigation efficiency was enhanced as well in the smart irrigation system, which led to a high level of water-saving performance over traditional irrigation methods. The suggested system promotes real-time climate-adaptive irrigation control, minimizes human control, boosts sustainable horticultural production, and leads to precision farming and smart water resources conservation in contemporary farm settings.