IoT-BASED EVAPOTRANSPIRATION ESTIMATION OF PEANUT PLANT USING DEEP NEURAL NETWORK

Abstract

The water availability in soil strongly influences crop growth by sustaining photosynthesis, respiration, and the maintenance of plant temperature. The water availability will decrease due to crop evapotranspiration (ETc) which is influenced by reference evapotranspiration (ETo) and crop coefficient (Kc). During water shortage, Kc is strongly influenced by soil evaporation coefficient (Ke) and basal crop coefficient (Kcb) which can be calculated using the Blue Red Vegetation Index (BRVI). The purpose of this study was to apply and evaluate a new method of estimating ETo, Ke, and Kcb at a research site using a Deep Neural Network (DNN) with minimum requirements. The results of the ETo estimation using DNN shows a good output with a determinant coefficient (R2) being 0.774. Meanwhile, the estimates of Ke and Kcb show excellent results with the determinant coefficient (R2) being 0.9496 and 0.999 respectively.