Groundwater Contributions to Streamflow in Response to Storm Events
Groundwater flow is usually considered to be a relatively slow process. Conversely, many tracer-based hydrograph separation studies interpreted using mass balance equations indicate that pre-event groundwater is often the major contributor to the observed rise in streamflow following a storm event. These studies therefore implicitly (or sometimes explicitly) infer that pre-event groundwater is being rapidly transmitted to the stream channel. The dichotomy generated by separation studies indicating rapid subsurface flow on the one hand versus conceptualizing groundwater movement as a gradual process on the other hand is often referred to as the 'old water paradox'. This work will use a series of numerical experiments to illustrate how the mass balance equations used in many tracer-based hydrograph separation studies can potentially inflate the pre- event groundwater signal measured in the stream due to an inadequate accounting of dispersive processes. Attention will also be given to how other hydrologic factors, such as heterogeneity and the presence of macropores, affect groundwater contributions to streamflow. It is hoped that the results of this work will help resolve the 'old water paradox' and improve our understanding of groundwater's contribution to streamflow during and after storm events.
A Holistic Analysis of the Effects of Discrete Precipitation Events and Temporal Atmospheric Energy Inputs on the Spatio-Temporal Patterns of Temperature in a Streambed
In recent years, there has been an increase in field-based research directed towards characterizing surface water/groundwater interactions using temperature as a tracer. In spite of this effort, relatively little computational work has been performed to provide insight and guidance towards these field-based studies using simulations where the pertinent hydrological, meteorological and surface/variably-saturated subsurface processes are simultaneously taken into account. This paper explores the use of temperature to identify the spatio-temporal patterns of groundwater contributions to streams under transient conditions as driven by discrete precipitation events and as affected by changing atmospheric thermal inputs. To quantify the factors affecting temperature patterns occurring in a stream bed, the HydroGeoSphere numerical model was recently enhanced to include the transport of thermal energy in both the surface and subsurface flow regimes, with full accounting of atmospheric thermal inputs. HydroGeoSphere is a fully-integrated surface/variably-saturated subsurface flow and transport model that is designed to simulate water flow, evapotranspiration/evaporation processes, and advective-dispersive heat and solute transport over the 2D land surface and in the 3D subsurface. A high-resolution 3D numerical simulation of a highly-characterized stream segment in Ontario, Canada was shown to mimic the spatio-temporal thermal patterns observed in the streambed, the surface water and the groundwater. Discrete rainfall events and diurnal fluctuations of atmospheric thermal inputs were found to affect the temperatures throughout the surface and the subsurface, in addition to the thermal energy exchange fluxes between the two regimes. The groundwater exfiltration and infiltration patterns along the stream bed are shown to play a primary role in the regulation of the temperatures in the hyporheic zone and in the surface water which has important implications regarding the health of aquatic habitats. We also explore the significance of geologic heterogeneity and its relation to the spatial patterns of temperature within stream.
Detection and Assessment of Surface Water and Groundwater Interaction Using Streambed Piezometers
Using of streambed piezometers in the quantification and assessment of surface water and groundwater interaction as wells as groundwater recharge estimation is increasingly gaining some attention. To assess surface water and groundwater interactions within the Ganaraska Region Conservation Authority (GRCA) watersheds, an extensive monitoring and data collection program has been initiated. The program is focused mainly on monitoring and analysis of surface water and groundwater vertical hydraulic gradient (VHG) using streambed piezometers installed in a number of locations of high shallow groundwater level that generally serve as discharge areas (gaining stream reaches). VHG is a function of Δ h (the head difference between groundwater elevation inside the piezometer and surface water elevation outside the piezometer) and Δ l (the depth from streambed surface to the top of piezometer screen). The hydraulic gradient is positive in case of groundwater discharge condition (upward gradient) and negative in case of recharge condition (downward gradient). The streambed piezometers were installed in (or in a close proximity to) areas of potential groundwater discharge in sandy and/or gravelly areas. Targeting and identifying of discharge areas was based on inspection, analysis and evaluation of several data sets such as examination and analysis of surficial geology and soils mapping, streams temperature data, potential groundwater discharge mapping, and direct observations of flowing conditions. Based on this initial analysis, several locations were visited and soil samples were taken from streambed using coring samplers and field analyzed to select suitable locations for the installation. Surface water level in the streams and shallow groundwater level fluctuations data collected from the piezometers as well as analysis of climate data from nearby weather stations, baseflow data, fish communities and temperature data have allowed staff to provide a better interpretation of surface water and groundwater interaction in a number of watersheds. Initial data collection and analysis was accomplished by measuring the vertical hydraulic gradients (Δ h divide; Δ l) at these monitoring sites. The VHG data from two groups (eastern and western) piezometer locations was then analyzed and allowed staff to classify these sites as potential groundwater discharge locations. Most of the sites showed variable positive VHG indicating potential groundwater discharge conditions. Two western sites exhibited a weak positive and negative gradient during late summer and early fall of 2006 and 2007 indicating loss of water from the stream to groundwater system. Hydraulic gradient data were also compared to rainfall data for the period from 2005 to 2008 for corresponding months to determine catchments responses to recharge. The precipitation data was collected from two local rain gauges located in eastern and western parts of the GRCA's watersheds. Comparison of the averaged monthly VHG data to summed monthly rainfall data provided compelling evidence that each of the sites analyzed responded at some level to precipitation. This was demonstrated by the fact that most of the sites had matched monthly averaged VHG data and the summed rainfall during the period from 2005 to 2008. For most of the stations, there was a decline in the gradient around August 2006 and all summer of 2007 which corresponded to a lower total rainfall during this time. Some of the sites responded more quickly than others, and again some responded more dramatically reflecting the uniqueness of their catchments where shallow aquifers sustain the discharge areas.
Recharge assessment using daily soil moisture balance and well hydrographs in deltaic unconfined aquifers
Estimation of groundwater recharge to aquifers represents an important step in assessing regional groundwater dynamics and sustainable yield. A recharge estimation method combining daily water balances and well hydrographs was adapted and its applicability to Canadian climatic conditions was tested in the Portneuf deltaic sand unconfined aquifers. Well hydrographs have been recorded at four sites in the aquifers for more than 10 years. Weather stations in the area provide daily temperature and precipitation data to calculate a water balance using conventional methods to estimate hydrologic parameters: potential evapotranspiration, runoff and readily available water supply in soil. Precipitations occurring as snow in winter are accumulated until a threshold temperature above zero is reached, and then accumulated precipitation is made available for infiltration and runoff. The estimated recharge is converted to changes in groundwater level though the use of specific yield. The rate of lowering dh/dt (m/d) of groundwater levels during recession periods, when no recharge occurs, is also taken into account. The approach thus yields a synthetic curve of changes in groundwater level. The matching of the synthetic water level to a well hydrograph is achieved by adjustments in parameters. The daily water balance appears as a simple method providing representative recharge estimates for unconfined aquifers. It has advantages compared to hydrographs and water balance methods used alone. The method also provides restricted ranges of hydrologic parameters for long duration hydrographs.
Watershed-Scale Modelling of Depression-Focussed Groundwater Recharge in the Canadian Prairies
The Canadian prairies is characterized by dry summer and cold winter, undulating topography formed by the Wisconsinian glaciation, and clay-rich soils derived from glacial tills. Previous studies have shown that groundwater recharge is mainly focussed under topographic depressions, which collect snowmelt runoff over frozen soils. The water level in these depressions rises by several hundred millimetres in typical snowmelt years. In contrast, annual precipitation in this region is normally less than 500 mm. The excess water input by snowmelt causes the total infiltration to exceed evaporation demand, resulting in groundwater recharge under depressions. The processes of depression-focussed recharge have been well documented for individual depressions. However, it is not clear how the individual recharge fluxes integrate over a larger scale. Thus, the objective of this study is to: 1) map a large number (>10,000) of individual depressions over the 250-km2 watershed of West Nose Creek, located near Calgary, Alberta, 2) quantify the total depression storage of snowmelt water, and 3) estimate the amount and spatially variability of groundwater recharge over the watershed. Aerial photography and ground-based survey were combined to map the depressions and quantify the storage. A one-dimensional soil water balance model was coupled with a simple transfer algorithm representing the lateral flux of water between depressions and the surrounding uplands. The model was used to estimate average recharge flux over a typical grid cell size (1 km by 1 km) of watershed hydrology models. Preliminary results indicate that recharge flux is sensitive to the amount of depression storage, suggesting that land management practices, such as artificial drainage of depressions, may have major impacts on water resources in the Canadian prairies.