Determining the spatial distribution of groundwater and surface water exchange using heat as a tracer
MetadataShow full item record
The magnitude, location, and timing of groundwater and surface water (GWSW) interaction (both as groundwater discharge and hyporheic cycling) in streams have implications stream ecosystems, nutrient and contaminant reactions, and stream restoration work. In many areas of the world, high phosphorus and nitrate agricultural runoff is a large threat to water quality. The study location, Elton Creek in Cattaraugus County, NY, is located in glacial outwash sediments and is typical of streams in the Great Lakes watershed. We evaluate four general controls of the indicators (stream morphology, stream gradient, bank slope, and in-stream features) of groundwater/surface water (GWSW) interaction using an analytical GIS model of groundwater discharge. In order to identify locations of GWSW interaction along a 500 m stream reach, a variety of methods were used (including differential streamflow gaging, solute tracers (or channel water balance), and distributed temperature sensor (DTS) monitoring. A GIS analytical model based on the superimposed indicators was compared to the DTS standard deviation in stream temperature derived gaining and losing portions of the stream. The relative correlation of the individual indicators with groundwater discharge areas was identified for the studied section. It was found that the superposition of indicators was able to delineate areas of groundwater discharge with increasing accuracy. The GIS model of the mapped locations of superimposed indicators is expected to be applicable in a wide range of stream systems to locate areas of potential groundwater discharge, groundwater contaminant discharge, and biogeochemical hotspots. In addition to identifying the spatial location of groundwater discharge we applied a coupled heat/mass balance model to DTS stream temperature to determine the location and magnitude of groundwater discharge at high spatial resolution. Previous studies using heat/mass balance modeling of GWSW interaction have either averaged temperature over time and distance, or used multiple parameters which are difficult to quantify. We used a simple heat/mass balance model to determine high spatiotemporal resolution groundwater discharge from DTS stream temperature. A rating curve was developed establishing the relationship between head and stream discharge at cross sections using stilling wells with pressure transducers. The upstream discharge was used as the initial condition (for each time step) to model the groundwater discharge at the study location. Additional downstream discharges were used to determine the effectiveness of the model to predict stream discharge. In this case, it was found that the measurement error in temperature and stream discharge was greater than the variation in predicted downstream streamflow. In addition, the volume of groundwater discharge was not substantial enough to significantly evaluate the model prediction. We suggest that this methodology would be more appropriately applied in shallow streams, with known significant groundwater inputs, and dynamic stream discharge over the studied section.