Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
6 t~ ~ - - This chapter describes the basic techniques for quantifying ground water recharge used by soil scientists, engineers, hydrologists, and hydrogeologists. The standard methods available to quantify ground water recharge are discussed, and their applications to ground water recharge of surface mining sites are assessed. TECHNIQUES FOR ESTIMATING GROUND WATER RECHARGE Ground water recharge at a mine site can be estimated, in principle, by a variety of techniques. Tracing water that enters the surface soils as it percolates through the soils and underlying material (vadose zone) to the water table is one example of a direct measurement technique. Other indirect techniques involve such approaches as evaluation of water budgets of root zones and/or aquifers, analysis of surface-water-discharge hydrographs, and interpretation of soil water and ground water chemistry. Attempts at determining recharge in small research plots typically require measurement of many climatic, geologic, soil, and ground water parameters (Table 6.1~. For mine scale -81-
-82- TABLE 6.1 Experimental Measurements Required to Determine Ground Water Recharge Measurement Instrument or Technique Onsite data--atmospheric Precipitation Atmospheric pressure Air temperature Relative humidity Wind speed Net radiation Onsite data--subsurface Hydraulic head Water table Water content of soil Bulk density Soil temperature Water characteristic Hydraulic conductivity Laboratory data Soil texture Particle density Tipping-bucket rain gauge Barometric-pressure transducer Thermistor probe Relative-humidity probe Anemometer Fritschen-type net radiometer Piezometer and duplicate tensiometer nests Water-table well Neutron probe, gravimetric method, and gypsum blocks Gamma probe and core method Thermocouples Tensiometers and neutron probe Determined from changes in water content and tension of a bounded soil volume during drainage and/or evaporation. Hydrometer method Pycnometer method
-83- Table 6.1 continued Bulk density Water characteristic Hydraulic conductivity Clod method Hanging column, pressure plate, pressure membrane Constant-head permeameter, water- characteristic-based methods SOURCE: Adapted from Sophocleous and Perry, 1985. evaluations, spatial and temporal variability in recharge and differences between pre- and post-mining site hydrology are important considerations in the selection of measurement techniques (Allison, 1988~. Interpretation of recharge data collected under ideal conditions requires an appreciation of possibl sources of measurement error and uncertainty. Measurement error can usually be estimated or controlled, but uncertainty is more difficult to quantify. For instance, uncertainty is introduced when short-term hydrologic conditions are assumed to be a valid representation of the long-term site data base, which in fact may be quite different (Court, e 1960; McKay, 19659. Uncertainty is also introduced when recharge is measured for a small area and then extrapolated to a larger area. The sources of these uncertainties are temporal and spatial variability. Comparative studies have shown that different methods can give different estimates of recharge even when the study site and time period are the same (Johansson, 1987; Uma and Egboka, 1988~. Direct measurement and observation of the migration of water from the land surface to the water table require installation of instruments to detect variations in the water content of soil with depth, from the land surface to the water table, over an extended period of time. Generally, these methods
-84- require banks of instrumentation and a large data-measurement program (Sophocleaus and Perry, 1985~. Interpretation of the fluctuation of a water table can also be direct evidence of net recharge. Hydrologic-budget calculations are based on the continuity equation, which states that all water entering and leaving a system must be accounted for (i.e., inflow minus outflow equals change in storage). This concept is applied over a selected area (i.e., a mine site or basin) for a specific time interval. A budget equation written in terms of ground water recharge from the vadose zone is where interflow; P = ET + RO + GWR + ASoil Storage, p ET RO precipitation; evapotranspiration; surface runoff and lateral vadose GWR = ground water recharge; and Roil Storage = change in soils water storage. An equation written in terms of ground water recharge for the saturated zone is where GWin GWR GWout AGW Storage GWin + GWR = GWOut + ~GW Storage ground water inflow rate; ground water recharge; ground water outflow rate; and change in ground water storage. (The units of all above terms are length/time.) Use of the above equations assumes that other system inputs and outputs are negligible. Accurately quantifying ground water recharge by water-budget calculation is more difficult than it may appear because it requires measurement of all the terms in the equation except ground water recharge. Most hydrology and hydrogeology text books--including Kirkby (1978), USDI (1977a, b), and Fetter (1988~--address the method for calculating the water
-85- budget. Knott and Olimpio (1986) compiled water-budget estimates of the average annual recharge rate for Nantucket, Massachusetts, and also made estimates from water table fluctuation and isotope data (Table 6.2~. Their recharge rate calculations have a standard deviation of 39 percent. Typically, the parameter most difficult to estimate in water-budget calculations is evapotranspiration (ET) (Bouwer, 1989~. Lysimeters have been used to document the seasonal changes of ET and the resulting downward flux of water that escapes ET (Bouwer, USDI, 1977a, b; Williams and Hammond, 19881. _ _ ~ 1989; However, given the relatively shallow placement of most lysimeters, the assumption that all the draining water would make it to the water table should be carefully considered. Another method for estimating ground water recharge involves the measurement of water flow through the vadose zone. The flux of water through the vadose zone is a function of the ability of the medium to transmit water, its unsaturated hydraulic conductivity, and the driving force, the hydraulic gradient. In order to use this method the unsaturated hydraulic conductivity and total head distribution must first be quantified in three dimensions and through time. The hydraulic conductivity may be measured using field or laboratory techniques. Hydraulic gradients are estimated from measurements of total head in a series of vertical profiles (Wilson, 1979~. These techniques require sophisticated instrumentation and the expertise of trained personnel, and therefore these techniques are most commonly applied to small research plots and are not typically employed in large basin studies. Some studies have used infiltration rates to infer the occurrence of ground water recharge. These infiltration rates have been found to be helpful in assessing rainfall-runoff relationships (Wells et al., 1982~. However, field measurement techniques may not provide values representative of natural rates (Bouwer, 1989~. Infiltration values alone will not allow calculation of recharge without
-86- TABLE 6.2 Comparison of Recharge Rates (Centimeters Per Year) Derived from Tritium, Water Table Fluctuation, and Water-Budget (Thornthwaite) Methods for Southeastern Massachusetts Study Method Water Water Location Tritimn Table Budget This study [Knot and Site 1 66.3 -- -- Ol~mpio, 1986] Site 2 >42.4 -- -- Site 3 -- 52.1 -- Guswa and LeBlanc (1985) Cape Cod -- -- 45.7 LeBlanc (1984) Falmouth, Cape Cod -- -- 53.3 Olimpio and de Lima (1984) Mattapoisett -- -- 40.4 G. J. Larson (1982) Truro, Cape Cod 27.9-40.6 -- -- Walker (1980) Nantucket ~~ __ 46.0 Delaney (1980) Martha' s Vineyard -- ~- 56.4 Williams and Tasker (1974) Mattapoisett -- -- 45.7 Delaney and Cotton (1972) Truro, Cape Cod -- -- 46.2-49.3 43.9-46.7 Magnusen and Strahler (1972) Tours, Cape Cod Strahler (1972) Cape Cod 30.5 - - -- 44.4 aMichigan State University, written communication, 1982. bRanges based on values of the water-holding capacity of the root zone between 5 and 10 cm, respectively. SOURCE: Knot and Olimpio, 1986.
-87- corresponding measurement of ET, changes in soil water storage, and rates of vertical movement through the vadose zone. Measurement of a change in the position of a water table in response to precipitation events or snowmelt is evidence of ground water recharge, but other possible influences on water table elevations must first be eliminated (Table 6.3) (Freeze and Cherry, 1979; Todd, 1980~. - ~ ~ ~ annual or storm-event To estimate the total recharge from water table hydrographs, both the quantity of water added to storage during the period of rise and that quantity v ~ , of water flowing away from the water table during the event must be determined (Johansson, 1987~. Rasmussen and Anderson (1959) developed a method to estimate seasonal recharge using an estimated recession level from which to calculate the water table rise and the ground water recharge (Figure 6.1~. To compute recharge from such plots, the estimated total change in the water table, Ah, is multiplied by the specific yield and the surface area over which the change is estimated to occur. Analysis of stream baseflow recession curves has also been utilized to estimate basin ground water recharge (Fetter, 1988; Figure 6.2~. This method requires streamflow hydrographs for two or more consecutive years and assumes that all recharge is reflected in the stream hydrographs. The method should not be used for cases where the stream recharges the ground water system (losing streams) Ground water recharge rates have also been . inferred from geochemical studies of water in the unsaturated and saturated zones (Stone, 1985; Bouwer, 1989; Knott and Olimpio, 1986; Colville, 1984; see Table 6.2~. Measurement of thermonuclear tritium, chlorine-36, and chloride mass balance in vadose zone water profiles have been used to infer recharge. Numerical modeling of ground water systems can also be used to estimate areal recharge rates. The inverse method is used to determine a recharge necessary to calibrate the model to measured fluctuations in ground water level (Wang and
-88- TABLE 6.3 S,~mmary of Mechanisms That Lead to Fluctuations in Ground Water Levels Uncon- Hunan- Short- Long- Climatic fined Confined Natural induced lived Diurnal Seasonal term influence Ground water recharge X X X X (infiltration to the water table ) Air entrapment during X X X X ground water rechargo Evapotranspiration and X X X X phreatophytic con~umption Bank-storage effects X X X X near streams Tidal effects near X X X X oc eans Atmospheric pressure X X X X X effects External loading of X X X confined aquifers Earthquakes X X X Ground water pumpage X X X X Deep-well injection X X X Artificial recharge; X X X leakage from ponds, lagoons, and land- ft lls Agricultural irriBati on X X X X and drainaSe Geotechnical drainaBe X X X of open pit mines, slopes, tunnels SOURCE: Freeze and Cherry, 1979.
-89- _1 ~ . it: At: u, c ~Q ¢ u, ;z 4 it > At: 3 20 10 ~ 1] :,'1' 11 11 U l Well NEW 228 ~ ~\1\' 1 Estimated 1',4, '::recession level - ~\ 11 \ - ~ ma_ J ~ J O 1978 J i J~O~ J ~ ~ O ~ ~ J O . 1979 1 1980 1981 J O 1 982 L J A J O 1 983 FIGURE 6.1 Hydrograph of monthly ground water levels and bar graph of monthly precipitation. SOURCE: Knot and Olimpio, 1986 .
-9o- 2(10c' 1 5(~, 10~} 9 70t~ ;00 4()O 300 20() 100 q() 7() A) 4(} JO 'art ~1 .L ., O ~ f .: .K It '.N 10 ~ ~ . 1 . . 1 1 . · I . I I .u I I ~ S O i~ D I F `\ ~ `, I I ~ S O ~ ~ I f ~ ~ ~ I I ~ S O ~ D I f Runoff Year 1 Runoff Year 2 Runoff Year 3 FIGURE 6.2 Semilogarithmic stream hydrographs showing base flow recessions. SOURCE: Fetter , 1988.
-91- Anderson, 1982~. Such ground water modeling requires input of aquifer geometry, boundary conditions, initial conditions, aquifer parameters, and other inflows and outflows at each model node. Water level data spanning the time period of interest are required for calibration. APPLICATION OF TECHNIQUES TO SURFACE MINING SITES Attempting to quantify ground water recharge at mine sites requires an appreciation for the strengths and limits of monitoring techniques, the impact of the selected mining technique on the hydrologic system, and the constraints of the climatic and geologic setting. Successful evaluation of recharge in western settings, such as those found in the Northern Plains Coal Province, requires an assessment of recharge parameters associated with thick coal seams covering thousands of hectares, semiarid climates (observations likely to be highly variable over short distances), seasonally frozen ground, thin soils overlying bedrock, coal aquifers, and clinker outcroppings in potential recharge areas. In contrast, recharge evaluations in the Appalachian Coal Province usually encompass sites of a few hundred hectares, steep wooded terrains, small perennial streams, temperate climates, thin soils overlying fractured bedrock, and small recharge areas. Mining in either of these coal provinces produces a different topography, soil profile, and stratigraphy than existed prior to mining. The mining process also disrupts the local ground water system, which can make identifying post-mining ground water recharge trends difficult (Western Water Consultants, Inc., 1985~. It also has been demonstrated that ground water recharge varies spatially and temporally in both pre-mining and post-mining areas (see Figure 3.5) (Rehm et al., 1982; Van Voast and Reiten, 1988; Kipp et al. ? 1983~. The choice and success of a particular method or combination of techniques
-92- to estimate recharge will depend on site conditions and the desired post-mining hydrology, and on the desired degree of resolution--criteria that have not yet been set. Water-Budget Methods The water-budget techniques for both unsaturated and saturated zones are generally applicable in principle to pre- and post-mining conditions. The equations appear to be simple, with the requisite number of input parameters limited. In practice, however, accurately quantifying these few parameters, such as soil water storage and ground water flow rates (input and output), poses a formidable task. The uncertainty associated with estimating recharge using water-budget techniques can be quite significant and must be assessed on a site-by-site basis. Further, if included, water-budget measurements for the vadose zone are data intensive and have been predominantly applied in research situations only. Vadose Zone Flux Measurements Most mine sites in the United States exist in areas of thin unconsolidated soils that overlie coal, sandstone, shale, and limestone. Such sites do not lend themselves to detailed vadose zone monitoring, principally because unsaturated consolidated formations and secondary fracture porosity and permeability make standard instrumentation unuseable or unreliable. While there may be special situations where the technique would be valuable (e.g., monitoring the effectiveness of spoil-segregation techniques), mine-scale problems generally are not amenable to the use of this type of instrumentation and analysis.
-93- Water Table Fluctuation Measurements Water table recession analysis requires monitoring recessions and quantifying specific yield. Analysis is complicated by several factors. The specific yield and the points of system recharge (natural and post-mining) and discharge are spatially variable and are influenced by mining. Post-mining analysis is further h~mn~red he the time ~ ~ delay in the ground water system recovery period, which may mask recharge events. Implementation requires measurement of the level of the water table, which may be quite difficult to obtain in zones of secondary fractures. Stream Hydrograph Separation Stream hydrograph separation may have the widest applicability in the eastern United States. This method is only appropriately applied to areas large enough to support perennial streams draining mined watersheds. The data-collection techniques use well-established methods. Data analysis is complicated by climatic trends and variability, so meaningful analysis demands either a prolonged data-acquisition phase or comparisons limited to similar climatic conditions. Stream hydrographs are sensitive to both the size and type of mining disturbance. Increases in the post-mining baseflow component of the hydrograph can be attributed to enhanced water storage in reclaimed spoils accompanied by slow release (Minear and Tschantz, 1976). Geochemical Techniques Geochemical techniques have some applicability to pre-mining conditions, particularly in arid regions, because they reflect long-term averages (Bouwer, 1989; Stone, 1985; Knott and Olympia,
-94- 1986). Because of the long-term averaging aspect of this technique, it is generally not applicable to post-mining evaluations. Also, leaching of chemicals from the mine spoils may interfere with the chemistry of ground water recharge. Numerical Modeling Numerical methods can be used for comparison or optimization of proposed techniques. Because of the typically data-intensive nature of ground water models, they are usually employed only in conjunction with other techniques. In the absence of an extensive data-collection program, numerical methods to infer recharge are not particularly useful. CONCLUS ION In conclusion, recharge estimation is fraught with difficulty and uncertainty. Recharge cannot be measured directly. Uncertainty in the measurement of relevant parameters results in an uncertainty in recharge estimates that likely exceeds changes in long-term recharge due to mining if proper reclamation practices are followed. Also, it must be recognized that at some mine sites that are small or are located in dry climates, reliable recharge quantification may be outside the realm of current technology. As an example, consider Van Voast and Reiten's (1988) comment that at some wells associated with mine sites in the Montana coal fields ". . . no local recharge has been observed over 15 years of (water level) record; at others, recharge has been evident only during occasional periods of unusually high snow melt or springtime precipitation."