Evaluation of descriptive and non-intrusive geophysical methods for the identification of saturation area dynamics and their controls

In the Catskill Mountains of New York State runoff is most commonly generated from groundwater seeps, causing parts of the landscape to saturate. Groundwater seeps and frequently saturated near stream areas are important source waters for streams, but can also be contaminant-contributing areas in agricultural fields. Knowledge of the landscape position of these saturation areas, the spatial/temporal evolution of runoff generation and the connectivity to surface water bodies are important parameters to consider. Many water quality models and risk assessment techniques should rely on these principles when assessing non-point source pollution in agricultural watersheds. However, there is little research that has comprehensively studied the spatial/temporal dynamics of these saturated areas and provides methods that meet the present needs in hydrology and hydrological modelling applications. To characterize the spatial extent of saturated areas a study was conducted on a 2.44 ha hillslope in the Town Brook watershed in the Catskill Mountains. Various non-intrusive methods were used consisting of vegetation surveys, GPS mapping and electromagnetic induction (EMI) and compared to existing information obtained from ground water table measurements and remotely sensed data. To better understand the location of the saturated area we characterized the soil profile with Ground Penetrating Radar (GPR) and Geoseismic surveys. The non-intrusive methods could not agree with the observed patterns of the saturated areas nor with a simple distributed model that used only topography and low-resolution soil information. For a better characterization of these areas additional information about soil characteristics and preferential flow paths was needed. The latter could be obtained with the Geoseismic and Ground Penetrating Radar. These instruments were capable of measuring the local depth to the shallow hard pan and the presence of natural pipes that carried large amounts of water via a subsurface. In summary we found that quick flow through large subsurface pathways needs to be considered for improved prediction of saturated areas.