Interpreting the spatial distribution of soil properties with a physically-based distributed hydrological model

Authors: Libohova, Zamir; MANCINI, MARCELO - University Of Arkansas; Winzeler, Hans - Edwin; Read, Quentin; SUN, NING - Department Of Energy; BEAUDETTE, DYLAN - Natural Resources Conservation Service (NRCS, USDA); WILLIAMS, CANDISS - Natural Resources Conservation Service (NRCS, USDA); Blackstock, Joshua; SILVA, SERGIO - Federal University Of Lavras; CURI, NILTON - Federal University Of Lavras; Adhikari, Kabindra; Ashworth, Amanda; MINAI, JOSHUA - Argonne National Laboratory; Owens, Phillip

Submitted to: Geoderma Regional
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/6/2024
Publication Date: 9/11/2024
Citation: Libohova, Z., Mancini, M., Winzeler, H.E., Read, Q.D., Sun, N., Beaudette, D., Williams, C., Blackstock, J.M., Silva, S., Curi, N., Adhikari, K., Ashworth, A.J., Minai, J., Owens, P.R. 2024. Interpreting the spatial distribution of soil properties with a physically-based distributed hydrological model. Geoderma Regional. https://doi.org/10.1016/j.geodrs.2024.e00863.
DOI: https://doi.org/10.1016/j.geodrs.2024.e00863

Interpretive Summary: The current soil property maps are static and coarse scale for any precision management applications. Detailed maps of soil properties are needed for precision agriculture applications, especially for small farms with limited resources. Yet, such maps may be expensive due to high cost associated with field sampling. We implemented a hydrological model combined with field data and instruments for monitoring soil water dynamic and produced high resolution daily soil moisture and water table depth maps for two catchments under forest and pasture. The daily soil moisture and water table maps summarized for the four seasons and annually related with measured trends of soil moisture instruments and observed patterns of soil properties such as soil thickness, depth to water limiting layers and drainage. This allowed for high resolution soil property maps that can be expanded to other larger farms and watersheds through hydrological models with limited field observations. In addition, the daily soil moisture for surface and subsurface layers as well as water table depth generated by hydrological models provide farmers with detailed spatial and temporal soil water movement to support year around precision agriculture activities.

Technical Abstract: Digital soil maps are commonly data-driven. The development of physically based models for soil mapping is difficult due to the complexity of soils. However, physically-based hydrological models have been successful in simulating water dynamics. Since water movement is a major driver of pedogenesis, the physical rules that govern water movement might help explain the occurrence of soil properties in space. We propose the use of distributed hydrological models to map soil properties. The Distributed Hydrology Soil Vegetation Model (DHSVM) was utilized to simulate soil moisture content (SM) and water table depth (WTD) in two hillslope catchments under pasture and forest. SM sensors and wells were installed in both catchments to validate simulations via Nash-Sutcliffe Efficiency (E). In-situ observations were made in 87 sites within both catchments to study the connection between simulations and observed soil properties. The simulated time series of SM and WTD were clustered per season using Dynamic Time Warping (DTW). Simulations of surficial SM (0–20 cm) were the most accurate (E=0.45). Subsurface SM (45–60 cm) and WTD did not present satisfactory simulations. Clusters of simulated SM could separate in-situ soil observations in statistically different populations, showing that hydrologic models could identify areas that followed different pedogenic trajectories driven by water dynamics. The thickness of Btx horizons were spatially grouped into different populations by SM clusters from every season except spring. For other properties, only SM dynamics of specific seasons could group significantly different population of observed soil properties, suggesting that the explanatory power of simulated water movement varies seasonally. Spatiotemporal physically-based modeling of water dynamics can help soil mapping by linking water movement to its effects in pedogenesis.