Development and field installation of smart sensor nodes for quantification of missing water in soil

The article presents an experimental study conducted to mitigate the rising water data uncertainty in the current water monitoring network of NSW, Australia. For this, IoT-enabled multidepth-embedded microcontroller-based smart sensor nodes have been developed comprising of low-cost low-power off-the-shelf sensors and essential electronic components. In the course of development, the developed system has been tested for functional improvements in a controlled environment before open-field installations. The deployed system is used as a real-time monitoring tool for data visualization on an IoT analytics platform. As a foundation, the study implemented a regression model and correlation analysis on the field dataset to understand interdependencies of study variables, such as soil-moisture (SM), soil-temperature (ST), rain-precipitation (P), and environmental-temperature (T). The results obtained demonstrate a strong covariance and reliance of SM and ST on the P volume and the environmental temperature, respectively. It is expected that the real-time acquisition of soil water data through the smart sensor node offers a unique opportunity to further implement the predictive models of potential water recharge via soil profiles as well as to schedule suitable irrigation patterns for agricultural lands.