USGS Ground-Water Availability Studies in the Great Lakes Basin
In January, 2005, the U.S. Geological Survey (USGS) initiated a Water Availability and Use study in the Great Lakes Basin as a pilot for a National Assessment of Water Availability and Use Program. The Great Lakes Basin Pilot: 1) integrates analysis of ground water, surface water, and water use to quantify water availability at a regional scale; 2) demonstrates local-scale assessment; and 3) identifies data needs. The ground-water studies within the project include reviews assessing ground-water in storage and ground-water divides for the basin, estimates of shallow ground-water recharge, and the development of a regional ground-water-flow model for the Lake Michigan Basin. This regional model is large (in excess of two million cells spread over parts of four states) and incorporates the transient response of multiple aquifers to multiple pumping centers. The regional model simulates the exchange between a dense surface-water network and notably heterogeneous glacial deposits underlain by a series of overlapping bedrock units dipping from the Wisconsin Arch into the saline Michigan Basin. The model was used to test new methods to estimate regional recharge and to calibrate large complex models with many parameters. Where a coarse regional approach could not capture important local-scale features, local-scale models were embedded within the regional model for water-availability analysis. The multi-model approach helped quantify changes in the ground-water system in response to variations in recharge and pumping from predevelopment to current conditions. Moreover, it piloted techniques that are transferable to other regional water-availability studies.
Baseflow due to Groundwater Discharge in the Ontario Portion of the Great Lakes Basin
Estimates of baseflow calculated using streamflow monitoring information and mathematical methods of hydrograph separation have been used as input for a series of regional groundwater studies in the Ontario portion of the Great Lakes basin. Some of the studies have subsequently been extended across the full Great Lakes region, and some are being adapted for other areas of Canada that have a similar physiographic setting. Baseflow due to the discharge of groundwater to surface water is an integrated but indirect measure of groundwater recharge, flow, and discharge within gauged watersheds that can be interpreted relative to societal and ecological requirements. Summaries of this discharge, and extrapolation of the summaries from gauged to un-gauged watersheds, inform land and water management activities, including source water protection, as well as the discussion surrounding issues such as the Great Lakes Water Quality Agreement. Studies completed to date include development of methods of watershed characterization and modelling, estimation of the impacts of climate change, and geological interpretation and mapping of baseflow as groundwater discharge. Initial attempts to detect trends and changes in baseflow that indicate human and climatic influences have also been undertaken. A key finding for southern Ontario is that roughly one-half of streamflow is the result of groundwater discharge. Locally, groundwater discharge contributes between 10 and 80 percent of streamflow in areas characterized by fine to coarse textured geological materials. Differences among estimates of baseflow calculated using varying methods of hydrograph separation reveal significant uncertainty, even when summarized as long-term averages. Much of the focus of the studies has been on these long-term averages; however, the dynamics of baseflow due to groundwater discharge, particularly recession during periods of otherwise low flow, have important implications and warrant additional research and interpretation.
Groundwater Management Along Lake Ontario's North Shore
A large stretch of the north shore of Lake Ontario is characterized by a till plain that slopes down from the Oak
Ridges Moraine, a 160 km long ridge of sand, silt and gravel deposits oriented in an approximately east-west
direction north of Lake Ontario.
Since 2000, an ongoing collaborative, multi-faceted program has been underway to better characterize the
groundwater flow system on the Lake's north shore. The program is a collaborative effort between
Conservation Authorities (Ontario's watershed management bodies), and several large municipalities (City of
Toronto, Regional Municipalities of Peel, York and Durham). The program has three main components:
database, geology and groundwater flow modeling; each of which is being actively managed and updated.
In Ontario, as in many jurisdictions in North America, water and environmental data has long been neglected.
Studies that involve the measurement of hydrological parameters and the collection of useful data are
commonly required for approval of land use change by provincial, regional and/or local government agencies.
So although data is frequently collected (at a considerable cost), it has never been rigorously assembled into a
comprehensive database that can be used for future reference. Rather, the data is collected by consultants,
reported through various studies, and then simply lost in archived files. In a similar fashion, individuals at
many government agencies have collected water related data that now reside in locations unknown and, thus,
unavailable to others in the organization.
With this in mind, a comprehensive digital database was assembled to establish the foundation for long term
successful groundwater management. The data model design incorporates information required for
groundwater modeling purposes, thus extending beyond that of traditional groundwater information. The key
data sources include borehole geology, water levels, pumping rates, surface water flows, climate data and
water quality data.
A second component of the overall groundwater management program has been the translation of a
conceptual geological understanding, developed over years of study, into the digital rendering of regional
geological surfaces. This included a focus on understanding geological processes and the depositional
setting of the Quaternary sediments. In addition to using geological data from well records to construct
geological surfaces, an emphasis was placed on incorporating hydrogeological data (e.g. well screen
placements) as well as "expert knowledge" in the form of digitized interpretation lines into the interpolation
process. This approach assisted in rendering the continuity of valley systems and allowed for layer pinch-outs
to be more effectively represented.
The hydrostratigraphy of the Oak Ridges Moraine provided many challenges to the construction of regional and
sub-regional groundwater flow models. Foremost was how to model an extremely large and complex area
without sacrificing the detail needed for stream-aquifer interaction and wellhead protection. Models of the
groundwater flow system were developed that matched observed water levels, flow direction and groundwater
discharge to streams. The models are being utilized by the regional municipal governments to guide land use
decisions. They provide powerful tools for wellhead capture zone analysis, water balance analysis, and
quantifying impacts of increased water takings. The models show that groundwater discharge is focused at
the streams that drain off of the Oak Ridges Moraine. Very little direct groundwater discharge occurs to Lake
River bank filtration as a dynamic biogeochemical and hydrogeological system: evolution with time of reduced zones in an alluvial aquifer
A well field in the alluvial plain of the Lot River (France) was studied for 3 years. The origin of water (river versus aquifer) can be identified using chloride as a natural tracer. The pumping pattern changed significantly at one point during the survey resulting in new groundwater flow lines. In many boreholes, the water chemistry varies seasonally due to bacterial activity and the resulting degradation of organic carbon, which creates a reduced zone (best indicated by dissolution of manganese). This reduced zone seems to evolve spatially with time. Three hypotheses that might explain this phenomenon are tested. (1) Variations in the DOC (dissolved organic carbon) load of the infiltrating river do not produce any changes in groundwater chemistry near the river bank. (2) In some cases, the conditions might still be reducing but aquifer solids are depleted in available Mn. (3) In other cases, changing groundwater chemistry might be caused by changes in water flow patterns (involving therefore different water supply zones within the aquifer).
Impacts of an Upstream Dam and Ground-Water Pumping on Stream Temperature
The combined effects of the presence/absence of an upstream boundary dam and in-reach groundwater pumping on stream temperature under different climate conditions were analyzed over the entire year. Stream temperatures were simulated using the CE-QUAL-W2 water quality model over a 110-kilometer model grid, with the presence/absence of a dam at the top of the reach and groundwater pumping in the lower 60- kilometers of the reach. Measured hourly meteorological data from three representative locations in the western United States were used as model input to replicate humid, semiarid, and arid climate conditions. For each climatic condition four hypothetical flow scenarios were simulated, which included: (1) natural, (2) upstream dam only, (3) upstream dam with in-reach groundwater pumping, and (4) upstream dam removed with in-reach groundwater pumping. Analysis of simulations indicated the impact of dam removal with or without groundwater pumping resulted in significant changes in stream temperature throughout the year for all three climate conditions. From March to August 2001 the presence of a dam caused monthly-mean stream temperatures to decrease 3.0, 2.5, and 2.0 C on average compared with natural conditions for the humid, semiarid, and arid conditions, respectively; however, stream temperatures generally increased relative to natural conditions from September to December. Due to its smaller impact on streamflow, the cessation of groundwater pumping resulted in more subtle temperature changes compared with temperature changes caused by the dam. During the summer and winter stream temperatures were respectively cooler and warmer by generally less than 0.5 C compared with natural conditions. Of potential ecological significance, presence of either a dam or ground-water pumping significantly altered the spatial extent of streamflow by creating transient dry reaches for several semiarid- and arid-climate simulated conditions.
Application of Integral Pumping Tests to estimate the influence of losing streams on groundwater quality
Urban streams receive effluents of wastewater treatment plants and untreated wastewater during combined sewer overflow events. In the case of losing streams substances, which originate from wastewater, can reach the groundwater and deteriorate its quality. The estimation of mass flow rates Mex from losing streams to the groundwater is important to support groundwater management strategies, but is a challenging task. Variable inflow of wastewater with time-dependent concentrations of wastewater constituents causes a variable water composition in urban streams. Heterogeneities in the structure of the streambed and the connected aquifer lead, in combination with this variable water composition, to heterogeneous concentration patterns of wastewater constituents in the vicinity of urban streams. Groundwater investigation methods based on conventional point sampling may yield unreliable results under these conditions. Integral Pumping Tests (IPT) can overcome the problem of heterogeneous concentrations in an aquifer by increasing the sampled volume. Long-time pumping (several days) and simultaneous sampling yields reliable average concentrations Cav and mass flow rates Mcp for virtual control planes perpendicular to the natural flow direction. We applied the IPT method in order to estimate Mex of a stream section in Leipzig (Germany). The investigated stream is strongly influenced by combined sewer overflow events. Four pumping wells were installed up- and downstream of the stream section and operated for a period of five days. The study was focused on four inorganic (potassium, chloride, nitrate and sulfate) and two organic (caffeine and technical-nonylphenol) wastewater constituents with different transport properties. The obtained concentration-time series were used in combination with a numerical flow model to estimate Mcp of the respective wells. The difference of the Mcp's between up- and downstream wells yields Mex of wastewater constituents that increase downstream of the stream. In order to confirm the obtained Mcp's concentrations of additional measurements in the investigated stream were compared with the concentrations in the groundwater up- and downstream of the stream section. The results revealed increased Mcp's downstream of the stream section for chloride, potassium and nitrate, whereas Mcp of sulfate was decreased. Micropollutants caffeine and technical-nonylphenol showed decreased Mcp's downstream of the stream section in 75 % of the cases. Values of Mex could only be given for chloride, potassium, nitrate and caffeine. The comparison of concentrations in the stream with those in the groundwater points to the streambed as a zone where mass accumulation and degradation processes occur. The obtained results imply that the applied method can provide reliable data about the influence of losing streams on groundwater quality.
Screening for Groundwater Contaminants Discharging to Urban Streams
Groundwater contaminated with urban pollutants can adversely affect freshwater aquatic ecosystems where it discharges to streams, lakes or wetlands. Generally such occurrences have been revealed following the discovery of contaminated groundwater plumes at a particular site or from wells in the area. Thus, this contaminant pathway tends to be dealt with on a site-specific and aquifer-focused basis. In contrast, surface water contaminant monitoring typically relies on bulk water concentrations from one or a small set of locations, thus ignoring the spatial variation in contaminant loading, potential losses to sediment or the atmosphere, and the full range of benthic components of the aquatic ecosystem. There are few studies outlining the extent of this contamination from the perspective of the surface water body's deeper benthic community, which might be expected to experience the greatest contaminant concentrations, on more than a local-scale. In this study, we report on an approach to stream-reach-screening for urban contaminants in discharging groundwater, with the focus on detection rather than accurate quantification. The methodology consists of a drive-point technique for sampling groundwater from below the stream bed (e.g. typically 50 cm) along a chosen reach at intervals of about 10 m. Groundwater samples were then analyzed for a wide range of common urban contaminants and general chemistry. This screening method was performed in three urban settings in Canada with known groundwater contamination, covering sections of about 140 to >500 m. The known contaminant plumes at each site were detected and roughly delineated. In addition, potential areas of previously-unknown groundwater contamination were also identified at each site. Contaminants included BTEX and other petroleum hydrocarbons, various chlorinated solvent compounds, nitrate, 1,4-dioxane, MTBE and elevated chloride (likely indicating road salt). These preliminary findings suggest that this approach may be useful for quickly assessing the cumulative threat to aquatic ecosystems of potentially multiple groundwater contaminant sources discharging to surface water bodies in urban settings.