Variability in Soil Moisture in a Temperate Deciduous Forest Using Electrical Resistivity and Throughfall Data

In deciduous forests, soil moisture is an important driver of energy and carbon cycling, as well as ecosystem dynamics. The amount and distribution of soil moisture also influences soil microbial activity, nutrient fluxes, and groundwater recharge. Consequently, accurate characterization of interactions and interdependencies between vegetation and soil moisture is critical to forecast water resources and ecosystem health in a changing climate. Such relationships and processes are nevertheless difficult to measure, both in time and space because of our limited ability to monitor the subsurface at necessary scales and frequencies. Several recent studies have shown that electrical resistivity tomography (ERT), using an array of minimally invasive surface electrodes, is a promising method for in-situ soil moisture monitoring. To this point, however, only few studies have used ERT to investigate spatial variability of soil moisture in temperate deciduous forests and to explore any links between soil water and above ground ecosystem variables. In our study in a central Michigan (USA) maple forest during the 2012 growing season, we combined ERT with detailed vegetation surveys and throughfall measurements to obtain better insight into spatial variations in rainwater input and soil water patterns. Resistivity data were collected on a weekly basis along an array of 84 electrodes with a spacing of 1.5 m. The inversion results were temperature corrected, converted to soil moisture, and differenced to obtain 2D images of soil moisture changes. The throughfall data were obtained using a novel method based on dissolution of plaster-of-paris tablets that were positioned below funnels, at 19 locations in the forest. Our results show that: 1) resistivity changes spatially with vegetation distribution, 2) in-season temporal changes in resistivity are related to plant characteristics, in particular to tree count and basal area, and 3) our low-budget throughfall method was capable of measuring precipitation variations over periods of several weeks but not sensitive enough to pick up in-forest variability in detail. The combination of above-ground throughfall measurements, vegetation surveys, and below-ground measurements of soil moisture using ERT did allow for a better understanding of the environmental variables important for studies of forest ecohydrology and atmosphere-plant-soil continuum of temperate forests.