Natural groundwater storage can be improved by constructing a subsurface barrier that is a subsurface dam, in order to capture the subsurface flows and raise the groundwater levels (GWLs) in the sediment layers. Subsurface dams are preferable to surface dams because of lower evaporation, higher functionality, lower cost of construction, lessened risk for contamination and the possibility of utilizing land over the dam. Therefore subsurface dams constitute an affordable and effective method for the sustainable development and management of groundwater resources. The aim of this research project was to develop and test methods for the assessment and location of subsurface dams in water resources management. From previous experiences it has been established that locating suitable sites for construction of subsurface dams plays an important role in the overall success of these dams. Therefore, in order to locate suitable sites, two approaches were followed. The first was the Boolean approach using topographical, geological and landcover data in a geographic information system (GIS) environment for a previously glaciated terrain near Stockholm. The results of the Boolean approach were complemented by a groundwater balance model and a topographic wetness index (TWI). The second approach involved spatial multi-criteria analysis (SMCA) applied to a region with different geological and hydrological conditions. SMCA was applied in Northern Pakistan using factors such as topography, geology, landcover, soil thickness and TWI. Two weighting techniques, the analytic hierarchy process (AHP) and the factor interaction method (FIM), were employed and compared. The Factor removal technique was employed to assess the sensitivity of the model for each factor. Aquifer thickness is an important factor while planning subsurface dams and data regarding the soil thickness is often not available at larger scale. Therefore a simplified regolith model (SRM) was developed for estimating the regolith thickness in previously glaciated terrain with a high frequency of rock outcrops, based on a digital elevation model (DEM) and an optimized outcrop search algorithm. In order to analyse the dynamics of the groundwater flow, a transient 3D groundwater flow model was developed for a subsurface dam. Methods applied to locate suitable sites for the construction of subsurface dams showed some promising results and need to be applied and tested in areas with different hydrological and geological conditions. The Boolean approach is a simple method that could be used during early planning stages for locating suitable sites for the construction of subsurface dams. The SMCA framework enabled the integration of knowledge for decision making, where the weights had a more significant influence on the results than the choice of the weighting method. AHP was considered to be the more robust model for assigning weights in this study. The factor removal technique showed that the modeling results were least sensitive to soil depth and most sensitive to land cover for the construction of subsurface dams. SRM showed reasonable results and could be used in engineering projects prior to detailed field investigations in formally glaciated terrain when borehole data is not available. The groundwater flow modelling results helped to develop some sustainable pumping scenarios to demonstrate the benefits of the subsurface dam. Groundwater flow model results also facilitated the selection of a suitable site for placing a subsurface dam in order to maximize the groundwater storage upstream. It was concluded in this project that the subsurface dams could sustainably be used to mitigate the water supply issues in formerly glaciated humid terrain such as in Sweden and dry climatic areas such as in Pakistan. Moreover, subsurface dams can play an important role in water resources management in coastal areas of formerly glaciated terrain, where saltwater intrusion is a rising environmental issue. Also in dry climatic areas like in Pakistan, methods such as SMCA could make the planning step more robust before the actual construction of dams. Themethods and findings presented in this thesis can be considered to be one tentative step of scientific contribution for better analysis, assessment and the location of subsurface dams. / <p>QC 20160210</p>

The objectives of this research are to measure the aquifer properties (S, T, and K) of selected watersheds delineated to the U.S. Geological Survey gauging stations using streamflow recession and baseflow data and to describe the relations among the properties of shallow aquifers and the physical properties of the basins, such as slope, regolith type and thickness, and land use type. Geographic Information System (GIS) techniques are utilized to investigate critical physiographic controls on transmissivity and storage coefficients on a regional basis. Moreover, the effect of evapotranspiration on recession index is illustrated. Finally, a detailed quantitative comparison of results for the Piedmont and the Blue Ridge Physiographic Provinces in southeast of the U.S. is provided. Recession index, annual groundwater recharge, and annual baseflow data were obtained from 44 USGS-gauging stations with drainage areas larger than 2 (mi2) and less than 400 (mi2). These gauging stations are located in Georgia and North Carolina. Analyses of data focused on GIS techniques to estimate watershed parameters such as total stream length, drainage density, groundwater slope, and aquifer half-width. The hydraulic diffusivity, transmissivity, and storage coefficient of watersheds were computed using hydrograph techniques and the Olmsted and Hely, and Rorabaugh mathematical models. Median recession index values for the Blue Ridge and Piedmont Provinces are 87.8 and 74.5 (d/log cycle), respectively. Median areal diffusivity values for the Blue Ridge and Piedmont are 35,000 and 44,200 (ft2/d), respectively. Median basin-specific estimates of transmissivity for basins in the Blue Ridge and Piedmont are 150 and 410 (ft2/d), respectively. The large values of transmissivity obtained for the Piedmont regolith may be attributed to the thick regolith, low values of basin relief, and voids that develop as a result of fracturing, foliation, weathering, and fractured quartz veins in the saprolite. Median basin-specific estimates of storage coefficient for basins in the Blue Ridge and Piedmont are 0.005 and 0.009, respectively. In general, the results from this study reveal great differences in basin-specific hydraulic parameters of the regolith material within the Piedmont compared to that of the Blue Ridge Physiographic Province.