Improved water management in agriculture through an intelligent irrigation system using GPR soil moisture data

Details

Supervisors

Lambot, Sébastien

Faculty

Faculté des bioingénieurs

Degree label

Master [120] : bioingénieur en sciences agronomiques, à finalité spécialisée

Abstract

Currently, irrigation worldwide accounts for 70 % of anthropogenic freshwater consumption. Twenty percent of arable land is irrigated, producing 40 % of the world’s food. With population growth, increased food consumption, and water scarcity, it is crucial to develop new irrigation systems that are more water-efficient while maintaining or even increasing yields. A new drone-borne Ground-Penetration Radar for soil moisture mapping has recently been developed. This technology enables the creation of high-resolution soil moisture maps at the field scale. These maps can be used to improve water management in agriculture. The purpose of this work was to develop two possible applications of this new technology. The first application is the simulation of Variable Rate Irrigation following different resolutions and irrigation strategies. The objective of this model is to estimate the potential benefits of applying a Variable Rate Irrigation based on soil water content maps. An estimation of the water savings achievable through different irrigation methods showed that, depending on the method and resolution of irrigation, water savings could reach up to 30 % for an irrigation resolution of 4 × 4 m compared to conventional or uniform irrigation. Simulations also indicated that the best irrigation strategy depends on the crop’s needs. The choice of irrigation strategy and resolution should depend on the crop’s sensitivity to water stress, water availability, and the cost of Variable Rate Irrigation equipment. The second application involves assimilating GPR data into a crop growth model (AquaCrop) to estimate three soil hydraulic parameters (curve number, field capacity and maximum soil evaporation coefficient), which can then be used to predict yield and devise different irrigation strategies. The advantage of such a model is that it accounts for spatial variability at the field scale and can be used in the context of precision agriculture. The correlation obtained between the yield map modeled by the assimilation of data and the yield map obtained by Irriwatch is 0.302. This is a low positive correlation but encouraging for future developments. The applications of such a model are very promising.