Low-cost in-situ sensor networks for soil moisture and water table measurements: experiences and recommendations.

Soil moisture is an essential parameter for irrigation management, transport of pollutants and estimation of energy, heat, and water balances. Soil moisture is one of the most important soil spatial-temporal variables due to the highly heterogeneous nature of soils which in turn drives water fluxes, evapotranspiration, air temperature, precipitation, and soil erosion. Recent developments have seen an increasing number of electromagnetic sensors available commercially for soil volumetric water content (θ) and their use is expanding providing decision support and high-resolution data for models and machine learning algorithms.

In this context, two demonstrations of in-situ LoRaWAN sensor networks are presented. The 1^st one is from a grassland site, Johnstown Castle, Wexford Ireland where a network of 10 low-cost soil moisture (SM) sensors has been operating for 12 months. The 2^nd network has been operating for 6 months at a peatland site (Cavemount Bog, Offaly, Ireland) which is currently undergoing a rehabilitation process through re-wetting. At this site, in addition to SM sensors, ultrasonic sensors are used for continuous measurement of the water table at 7 locations. For both sites, the analytical performance of the SM sensors has been determined in the laboratory, through calibrations in liquids of known dielectric permittivity and through field validation via sample collection or time domain-reflectometry instrumentation (TDR). Experiences and recommendations in deploying, maintaining, and servicing the sensor networks, and data management (cleaning, validation, analysis) will be presented and discussed. Emphasis will be placed on the key learnings to date and the performance of the low-cost sensor networks in terms of collected data.

Small-scale sensor networks like these are expected to bridge the gap between the low spatial resolution provided by the satellite-derived products and the single point/field measurements. Within the project, the sensor network will provide spatial observations to complement existing fixed point measurements. It will allow researchers to investigate SM dynamics at field scale in response to different soil types, soil density, elevation, and land cover.