Methods of Groundwater Quantifications and Their Limitations

Water Freedom System

Survive Global Water Shortages

Get Instant Access

Shiklomanov (1997) first estimated groundwater stocks for each continent by multiplying the total area of the continent by the expected groundwater depth, a water loss factor, and effective porosity. The maximum depth of groundwater storage was taken to 2000 m with three zones of groundwater storage, which were characterized by different hydrodynamics. They were attributed different values of effective porosity ranging from 5% for the lowest to 15% for the uppermost layer. According to Shiklomanov and Rodda (2003), the uppermost zone of continental crust stores 3.6 million km3, the intermediate depth zone 6.2 million km3, and the lowest zone hosts 13.6 million km3 of groundwa-ter. Shiklomanov (1997) differentiated between saline and fresh groundwater stocks and estimated the latter with 10.5 million km3.

Other methods used to quantify global groundwater resources, found in the available literature, can be divided into modeling, remote sensing, GIS, geo-physical/geochemical (i.e., isotope) techniques, and surface water discharge determinations (Zektser 2002). Most estimates are made by assuming that climate parameters are stationary (i.e., the climatic patterns observed in the past will be observed in the future with insignificant variations; Gleick 1993), an assumption that cannot always be made with natural water dynamics and rapid climate change. Ideally, groundwater evaluations should be based on reliable, continuous, and dense data of geological, hydrological, and meteorological parameters (Dragoni and Sukhija 2008; Shiklomanov and Rodda 2003). On the other hand, groundwater volumes are often performed by multiplying surface areas with estimated average depths of groundwater saturated strata (Jacobson 2000; Shiklomanov and Rodda 2003). For this, a critical question is the depth of groundwater, which depends on the hydrogeological structure, recharge hydrodynamics, rock types, and geothermal gradients. For example, porosity can only be estimated over large volumes due to the small-scale complexity of the geological formations in the subsurface (Arnell 2002). Therefore, any estimate of groundwater stocks is highly uncertain.

A prominent example for large-scale modeling is the WaterGAP Global Hydrology Model (Doll et al. 2003), which aims to assess global availability of water stocks and usage along with the estimates of long-term global change impacts on water resources. Two submodels, Global Hydrology Model and Global Water Use Model, simulate the components of global continental water and water use for households, industry, and agriculture, respectively. All simulations cover the continents (except for Antarctica) and have a 0.5° x 0.5° spatial resolution with daily time steps (Doll and Fiedler 2008; Fiedler et al.

2008). WaterGAP calculates daily "vertical" water balances for both land area and open water bodies. For land surface, it consists of a canopy water balance and a soil water balance, derived as functions of land cover, soil water capacity, and monthly temperature, solar radiation, and precipitation. For open water bodies, the vertical water balance is the difference between precipitation and evaporation. The sum of the runoff generated within a cell and the discharge flowing from a cell is then transported through storage compartments including groundwater, lakes, artificial water reservoirs, wetlands, and rivers (Doll et al. 2003). Finally, the total cell discharge is taken to the next downstream cell following a global drainage direction map to compute river discharge values (Doll and Lehner 2002). Calibration of discharge is performed for 724 drainage basins and at 1235 gauging stations worldwide that cover 50% of the global land area and 70% of the actively discharging area. Results show the long-term modeled average annual discharge within 1% of the measured discharge (Doll and Fiedler 2008). Theoretically, this volume can be annually withdrawn from global aquifers without threat of depletion. The standard WaterGAP groundwater recharge algorithm was modified for semiarid and arid regions, based on independent estimates of direct groundwater recharge. This yields long-term global groundwater recharge (i.e., renewable groundwater) with 12,666 km3 yr-1 for the calibration period of 1961-1990. Comparison of continents in WaterGAP shows that South America has the largest renewable groundwater resources in the world. As this continent relies predominantly on surface water for public supply with associated sanitary problems, the largest potential of providing better quality water exists for this region of the world (Reboucas 1999).

Other hydrological models were introduced by Yates (1997), Klepper and van Drecht (1998), and Arnell (1999), all of whom focused primarily on atmospheric and land surface waters and largely ignored groundwater. For better quantitative estimates of the distribution and fluxes of energy and water on Earth and to compare models, the Global Energy and Water Cycle Experiment (GEWEX) was initiated and run by the World Climate Research Programme in 1988 (Potter and Colman 2003). The goal of this project was to predict the global hydrologic cycle and water and energy fluxes with the help of remote sensing methods, with a focus on land surface and upper oceans. The Global Soil Wetness Project (GSWP), on the other hand, produced data sets of hydro-logical and hydrogeological information related to the fluxes of water on the continents (Oki and Kanae 2006). It also tests and compares outcomes of other models that look at global magnitude and distribution of terrestrial water fluxes with 23 participating models.

The efficiency of models depends on input parameters and hydrological data. Their insufficiency may compromise outcome. Many parameters are obtained by derivation from existent information because the required spatial and temporal information about groundwater-containing bodies is often not available at continental scales. For instance, WaterGAP is based on river basins

(Doll and Fiedler 2008; Fiedler et al. 2008), which may not match aquifer boundaries. In addition, calibration requires historical data that may lack reliability due to the short time periods of monitoring, changing measurements techniques, limited geographical coverage, and growing anthropogenic impact. For example, data for WaterGAP cover only half of the land surface, thus requiring excessive extrapolation over the uncovered areas and leading to potential over- or underestimations. The model also uses groundwater by vertical seepage via soils and ignores groundwater recharge via rivers and lakes due to the absence of reliable data. However, in arid and semi-arid regions such surface waters can significantly contribute to groundwater recharge (Arnell 2002). Omission of such mechanisms may therefore lead to underestimation of total renewable groundwater, especially in arid regions where accurate figures on groundwater dynamics are crucial.

Remote sensing techniques have recently been used to evaluate the magnitude of spatial and temporal variations in water storage and fluxes on Earth. For example, the Gravity Recovery and Climate Experiment (GRACE) twin satellite project was initiated in 2002 (Rodell and Famiglietti 2002) and is expected to allow observation of water storage changes in aquifers on a monthly basis (Guntner et al. 2007a, b). To optimize results, measurements are taken by two satellites within a distance of ~220 km in a polar orbit at an altitude of ~500 km. They continuously record variations in gravity and supply multiple measurements at the same areas over various seasons to yield high-resolution data with 1 cm height changes on spatial resolutions of 200-300 km (Dragoni and Sukhija 2008; Swenson et al. 2003). Initial publications on the GRACE mission focused on surface water (Swenson et al. 2003; Syed et al. 2008) and groundwater storage change measurements, with a link to the WaterGAP model (Fiedler et al. 2008; Guntner et al. 2007b). Measurements by GRACE can produce groundwater recharge and discharge rates and assess their spatial distribution. Such measurements are, however, expensive and still under development (Guntner et al. 2007a; Jacobson 2000). For instance, GRACE is so far not capable of measuring short-term variations in water and energy storage and fluxes, and this may result in systematic errors due to tides, wave motion, or soil moisture fluctuations.

Various studies of GIS application for groundwater-related issues have a primary focus on regional scales (Chen et al. 2005; Johnson and Njuguna 2002; Kharad et al. 1999). Globally, the World-wide Hydrogeological Mapping and Assessment Programme (WHYMAP) and the International Groundwater Resources Assessment Center (IGRAC) projects are mainly used for better visualization, but they lack quantifi cation or modeling of groundwater resources (IGRAC 2008; Struckmeier and Richts 2006, 2008). GIS-based water evaluations depend on reliable and high frequency input data on surface and aquifer systems, and input often requires signifi cant data mining and programming skills.

Seismic refraction and reflection techniques can be applied to establish estimates of maximum depths where groundwater is present. Despite their advantages of good reliability, quick implementation, and easy interpretation, a considerable drawback lies in the need to calibrate with present boreholes (Sorensen and Asten 2007). Ground-penetrating radar (GPR) has been proven to be a useful tool in calculating groundwater table depths (Doolittle et al. 2006); however, it is limited, for instance, by aquifer heterogeneities and associated up-scaling (Huggenberger and Aigner 1999). Estimates of groundwater recharge, residence times, surface water interactions, and origin of baseflow can be arranged via isotope tracers (Clark and Fritz 1997). Isotope methods are predominantly applied at a local or regional scale (Kendall and McDonell 2000), while globally the Global Network on Isotopes in Precipitation may be also be useful for groundwater recharge studies (IAEA/WMO 2008).

Was this article helpful?

0 0

Post a comment