Toward Low-Cost and Sustainable IoT Systems for Soil Monitoring in Coastal Wetlands

Coastal wetlands and nearshore environments significantly contribute to carbon storage, accounting for nearly half of global blue carbon. However, threats like climate change, invasive species, and sea-level rise make them vulnerable, and hence there is a need for continuous monitoring and treatment. Traditional soil monitoring methods, although thorough, are expensive, labor-intensive, and time-consuming. Remote sensing, as an alternative, also faces challenges, such as weak signals from non-exposed soils and interference from surface water. Additionally, unique challanges in remote wetlands, e.g., the lack of a reliable power grid and rough field conditions, complicate the installation and operations of monitoring equipment. To address these challenges, we present a field -deployable IoT-based soil monitoring system tailored for long-term coastal wetland monitoring. This study mainly introduces two core components of our system: energy management/optimization and the calibration/evaluation of low-cost soil sensors. We develop a novel energy management mechanism powered by multiple predictors, optimizing sensing efficiency and supporting long-term operability, and our contribution incorporates a calibrated set of cost-efficient sensors, ensuring precise soil property esti-mation. We evaluated our energy management with long-term, real-world weather data. The results show our approach allows more frequent sensing without interruptions, outperforming two common energy management mechanisms and ensuring sustained soil sensor operation in wetlands without a power grid. We also explored the use of low-cost sensors to timely and efficiently estimate complex soil health indicators, particularly bulk density, which is traditionally determined via labor-intensive lab tests. While our results affirm the potential of data from these affordable sensors, they also highlight the need for a more comprehensive, long-term validation of our approach.