Estimation of Seasonal Crop Water Requirement Using Support Vector Regression in India's Arid Zone

In addition to irrigation scheduling and planning, quantifying crop water need is now essential for conserving fresh water. The study estimates the seasonal crop water requirement (ETc) for wheat in Barmer, Rajasthan, a region in western India with a hot, dry climate and little annual rainfall. A large amount of water is needed for agriculture, which is the local population's principal line of work. In this region, groundwater, the primary irrigation resource, is decreasing dramatically. One useful method for ensuring precise irrigation is to determine the crop's water demand. For this purpose, climate statistics from 2000 to 2021 of Barmer, Rajasthan are taken into account. The intent of this period is to track how crop evapotranspiration is affected by climate change in the early decades of the twenty-first century. This study presents a support vector regression based model that may be a useful tool for estimating crop water demand. The observed values estimated by FAO-PM56 are compared with the predicted ETc values of the suggested model. The following performance metrics indicate how well the support vector regression is working: the mean absolute error is 0.11 mm/day, the mean squared error is 0.02 mm/day, the root mean square error is 0.1439 mm/day, and the coefficient of determination is 0.9921. The study's findings suggest that stakeholders can utilize this model to boost agricultural yield, control irrigation in agriculture, and conserve water—a resource that is essential for everything, both living and nonliving.