No abstract

No abstract

High-resolution spatial–temporal root zone soil moisture (RZSM) information collected at different scales is useful for a variety of agricultural, hydrologic, and climate applications. RZSM can be estimated using remote sensing, empirical equations, or process-based simulation models. Machine learning (ML) approaches for evaluating RZSM across numerous spatial–temporal scales are less generalizable than process-based models. However, data-driven ML approaches offer a unique opportunity to develop complex models of soil moisture without making assumptions about the processes governing soil water dynamics in a given study region. In this study, comparisons were made between two models, pySEBAL and EFSOIL, which were based on evaporation fraction (EF) and soil properties, and a data-driven model based on the Random Forest (RF) ensemble algorithm. These approaches were evaluated to demonstrate their capabilities for RZSM estimation. The EF obtained from Landsat images was used after validation with eddy covariance measurements as the major input to all three models, along with other meteorological and soil physical properties. The RF model was trained using in situ soil moisture data from Time Domain Reflectometry (TDR) sensors installed in a vineyard from 2018 to 2020. The predictor variables comprised of meteorological, soil properties, EF, and a vegetation index. The results reveal that there was a strong correlation between the in situ measured soil moisture and the RF predicted soil moisture at all sensor locations. Due to the complexity of the physical processes involved in soil water flow, the empirical models pySEBAL and EFSOIL were unable to reliably predict RZSM values at all monitored locations. The high RZSM predicted by pySEBAL demonstrated the presence of possible bias in the model’s algorithm used to estimate soil moisture. We also demonstrated that ML based on the RF algorithm may be used to predict spatially distributed RZSM when a few soil moisture ground measurements are combined with remote sensing to produce soil moisture maps.

No abstract

Accurate quantification of in situ heterogeneity and flow processes through fractured geologic media remains elusive for hydrogeologists due to the complexity in fracture characterization and its multiscale behavior. In this research, we demonstrated the efficacy of tracer-electrical resistivity tomography (ERT) experiments combined with numerical simulations to characterize heterogeneity and delineate preferential flow paths in a fractured granite aquifer. A series of natural gradient saline tracer experiments were conducted from a depth window of 18 to 22 m in an injection well (IW) located inside the Indian Institute of Technology Hyderabad campus. Tracer migration was monitored in a time-lapse mode using two cross-sectional surface ERT profiles placed in the direction of flow gradient. ERT data quality was improved by considering stacking, reciprocal measurements, resolution indicators, and geophysical logs. Dynamic changes in subsurface electrical properties inferred via resistivity anomalies were used to highlight preferential flow paths of the study area. Temporal changes in electrical resistivity and tracer concentration were monitored along the vertical in an observation well located at 48 m to the east of the IW. ERT-derived tracer breakthrough curves were in agreement with geochemical sample measurements. Fracture geometry and hydraulic properties derived from ERT and pumping tests were further used to evaluate two mathematical conceptualizations that are relevant to fractured aquifers. Results of numerical analysis conclude that dual continuum model that combines matrix and fracture systems through a flow exchange term has outperformed equivalent continuum model in reproducing tracer concentrations at the monitoring wells (evident by a decrease in RMSE from 199 to 65 mg/L). A sensitivity analysis on model simulations conclude that spatial variability in hydraulic conductivity, local-scale dispersion, and flow exchange at fracture-matrix interface have a profound effect on model simulations.