Agro-Inundation For Maximizing Crop Yield and Water Efficiency

Abstract

Population is increasing at an alarming rate which calls for the improvement of the food resource yield obtained via farming. Irrigation requires tremendous amounts of water, so it is imperative to use water wisely. For this purpose, there is a need for smart irrigation techniques which are designed and proposed in such a manner that the techniques depict the water level and its requirements in a precise and recommended way. As of now, farmers couldn’t find a reliable way to estimate the water needs for their crops. The model proposed uses machine learning techniques to forecast crop water requirements based on the weather attributes for a particular day. In order to make efficient use of this dataset during the prediction and extraction of similar patterns, the features that determine crop water requirements were examined and pre-processed based on our understanding of irrigation and machine learning. Correspondingly, a decision tree technique was applied to predict water requirements to evaluate our data preprocessing technique. We have built a training dataset using the weather attributes of the year 2015. The potential daily water consumption in this training dataset was determined using the Evapotranspiration method. Therefore, this machine learning-based solution is essential for farmers to utilize the available water resources to the fullest.