IHIIHIHHI 9:393 ‘IIIHIHHIWIWHIIHHWIHIWHHIHiIIH HTHS rTI’HESIS “um" 1111111111111111 A Michigan State 3129300 University This is to certify that the thesis entitled Rate of recharge to a heteroqeneous aquifer: An investigation using bomb tritium. presented by Manrico Delcore has been accepted towards fulfillment of the requirements for filigree in Maj rofessor _ Graham J. Larson Date W 0-7639 MS U is an Affirmative Action/Equal Oppormm'ly Iufimfion MSU LIBRARIES RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. rte; "3-4 #131, 2 2 19:35 ~ I N. . ? '. a | -~ 4" ' l _ _.._a ' . :9 i .“c'e‘r' ‘4 if; _ q 1 ‘0‘; 3]; ‘ - fififlh*ffiit°‘ SEP 2 0 2000 RATE OF RECHARGE TO A HETEROGENEOUS AQUIFER: AN INVESTIGATION USING BOMB TRITIUM By Manrico Delcore A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Geological Sciences 1985 ABSTRACT RATE OF RECHARGE TO A HETEROGENEOUS AQUIFER: AN INVESTIGATION USING BOMB TRITIUM By Manrico Delcore Bomb tritium was used to calculate the rate of recharge to a heterogeneous drift aquifer. The aquifer is located in Kalamazoo County, Michigan, and is composed predominantely of outwash sands and gravels. Lenses of fine sand, silt and clay are also present. Forty six groundwater samples were obtained from 2 groundwater divides, one near the City of Portage and the other near the City of Richland. Bomb-tritium was found in 41 samples ranging from 2.4 to 29.6 m below the water table. Pre-bomb tritium was identified in 5 samples at depths greater than 21.3 m. Assuming piston type flow at the groundwater divides, the rate of recharge to the aquifer was found to range from 19.5 to 32.0 cm per year. These rates are in close agreement with those previously determined by conventional hydrological techniques. dedicated to my parents and Mary Beth ii ACKNOWLEDGMENTS I would like to express my sincere appreciation to Dr. Grahame J. Larson, my thesis adviser for encouragement and friendship he has so freely this project. I would also like to thank members Dr. David T. Long and Dr. Roger B. their advice and encouragement. Thanks Monaghan for all his help. Special thanks go to Mary Beth and my their constant help and moral support. 111 the advice, given during the committee Wallace for also to Bill parents for TABLE LIST OF FIGURES . . . . . LIST OF TABLES . . . . . INTRODUCTION . . . . . . METHODOLOGY . . . . . . . STUDY AREA . . . . . . . GEOLOGY . . . . . . . HYDROLOGY . . . . . . SOURCES OF TRITIUM DATA . SAMPLING . . . . . . OF CONTENTS TRITIUM IN THE SATURATED ZONE . . . . RECHARGE RATES . . . . . DISCUSSION . . . . . . . CONCLUSION . . . . . . . APPENDIX I . . . . . . . ANALYSIS OF SAMPLES . PRE-DISTILLATION . . ELECTROLYSIS . . . . POST-DISTILLATION . . APPENDIX II . . . . . . . iv LIQUID SCINTILLATION COUNTING . . COUNTING EFFICIENCY AND QUENCHING APPENDIX III . . . . . . . . . . . . TRITIUM CALCULATION . . . . . . . TWO SIGMA ERROR DETERMINATION . . BIBLIOGRAPHY . . . . . . . . . . . . 29 3O 33 33 35 36 List of Figures Figure 1. Schematic representation of groundwater flow at groundwater divides (Not to scale) (After Toth, 1962) . . . . . . . . . . . . . . . . Figure.2. Geologic map of Kalamazoo County (After Monaghan and Larson, 1983). . . . . . . . . . . . Figure 3. Tritium Input Function for the Kalamazoo area . . . . . . . . . . . . Figure 4. Water table map of Kalamazoo county (After D.N.R., 1983). . . . . . . . . . . Figure 5. Tritium concentration as a function of sample depth. . . . . . . . . . . . . Figure 6. Pre-distillation apparatus. . . . . . Figure 7. Post-distillation apparatus. . . . . . Figure 8. Quench curve. . . . . . . . . . 10 12 15 20 28 32 Table 1. List of Tables Raw tritium data. . . . vii 15 INTRODUCTION The use of naturally occurring tritium in hydrological investigations was first introduced by Libby in 1953 (Libby, 1953). Shortly thereafter, it was realized that tritium derived from atmospheric nuclear testing from 1952 to 1962 could also be used to identify water of pre-bomb and post-bomb origin (Eriksson, 1958). For example, meteorically derived water of pre-bomb or pre 1953 origin is characterized today by having less than 2 TU (1 TU = 1 Tritium atom / 10 E 18 Hydrogen atoms), whereas water of post-bomb origin is characterized by tritium concentrations greater than 2 TU (Butlar and Libby, 1955; Begemann and Libby, 1957). By far the majority of groundwater investigations involving bomb tritium have focused on water in the unsaturated zone. For example, it has been used to locate zones of recharge (Verhagen et al., 1970; Allison and Hughes, 1975; Verhagen et al., 1978;; SukhiJa and- Rao, 1983), as well as to obtain information regarding sources of recharge to various types of aquifers (Bredenkamp and Vogel, 1970; Persson, 1974). Munnich et al., (1967) introduced a technique by which the actual rate of recharge could be determined using tritium in the unsaturated zone; this technique has been used by other investigators with excellent results (Andersen and Sevel, 1974; Allison and Hughes, 1974; Bredenkamp et al., 197A; Dincer et al., 1974; Vogel et al., 1974; Sukhija and Shah, 1975; Allison and Hughes, 1978). There has, however, been very little work done with tritium in the saturated zone. The possible sources of groundwater as well as rates of flow have been determined using tritium in the saturated zone by Butlar and Wendt (1958), Maloszewski et a1. (1983), Turner et a1. (1983), Stewart and Downes (1980), Brown and Taylor (1974), Ritter (1980), and Monaghan and Ritter (1982). Also, Atakan et a1. (1974) and Offer (1982) have succeeded in using tritium in groundwater to calculate the rate of recharge to shallow, unconfined, homogeneous sand aquifers. It remained to be determined, however, if tritium could be successfully used to quantify recharge to more complex hydrogeologic settings. The purpose of this investigation is to determine if bomb tritium can be used to Calculate the rate of recharge of a heterogeneous aquifer. ETHODOLOGY The actual dating of groundwater by its tritium content involves first the calculation of a tritium input function which computes the concentration of tritium entering the groundwater system, corrected for decay (Thatcher, 1967; Rabinowitz et al., 1977a; Atakan et al., 1974; Andersen and Sevel, 1974; Maloszewski et al., 1983); and second, the laboratory measurement of the actual quantity of tritium in groundwater (Wyerman, 1976). In an unconfined aquifer, groundwater samples for tritium analyses are generally collected at groundwater divides (Offer, 1982). It is only here (Figure 1) that the flow of meteoric water moving through the aquifer is subvertical and downward (Hubbert, 1940; Eriksson, 1958; Toth, 1962; 1963) so that each precipitation event is preserved as a discrete layer without mixing with adjacent layers (Munnich et al., 1967). In short, a replica of the recharge history is preserved at groundwater divides in a piston type of flow regime (Allison and Holmes, 1973; Allison and Hughes, 1974; Nir, 1964). O ‘U - > - ‘U 3 U -—-' Flow Hm ---- Equipment!“ lines Figure 1. Schematic representation of ground water flow at groundwater divides (Not to scale) (After Toth, 1962). STUDY AREA The major aquifer for the city of Kalamazoo, Michigan was selected for this investigation because it meets the following criteria: 1. The aquifer is heterogeneous. 2. The existing well coverage is sufficient to thoroughly sample the aquifer. 3. The tritium input function for the area can be easily calculated. 4. The rate of recharge has been previously determined by conventional methods, providing a clear test of the results obtained by this investigation. GEOLOGY The Kalamazoo area is underlain almost entirely by glacial drift (Figure 2) that ranges from less than 30 to more than 181 meters thick (Forstat, 1982). Texturally, the drift can be differentiated into outwash and till (Deutch et al., 1960; Monaghan and Larson, 1983; Monaghan, 1984). Outwash forms the most extensive surficial material, covering 78% of the area and serves as the principle aquifer for the region (Allen et al., 1972). It is generally well sorted, and consists of stratified sands and gravels with graded beds, large scale cross bedding, our: 3- Kalamazoo- (EUfl P'ain ‘vflzfia D . IICMIM Galesburg-Vicksburg . ’/ I I Plain 0N Climax-Seats Plain Geleeburg Nicksburg Plain Gill, d I iifee 9 Lu ,1 L0 Kilometers Drumllnized. ground mer eine L A v c 5 1p LOCIflOfl MID Mereines Figure 2. Geologic map of Kalamazoo County (After Monaghan and Larson, 1983). and channel scours. Where eXposed, the outwash also includes occasional lenses or stringers of fine sand, silt and clay (Monaghan and Larson, 1983). The till is generally associated with end moraines and drumlinized ground moraine. In the moraines it is variable in thickness and consists chiefly of cobbles and pebbles in a sandy clay matrix. Sometimes the till shows faint bedding structures. Within the drumlinized ground moraine located in the southeast corner of the study area, the till is structureless and generally contains more clay than till in the end moraines. It is seldom more than a few meters thick and often overlies deposits of coarse sand and gravel (Monaghan and Larson, 1983). HYDROLOGY The initial source of all fresh groundwater in the Kalamazoo area is precipitation (Deutch, et al., 1960). Precipitation averages 83.8 cm (33 inches) per year and is distributed throughout the year with June being the wettest month and January the driest (Stronmen, 1971). Evaporation measurements are only available for the growing season, May to October (Nurnberger, 1982), and average 85.3 cm (33.6 inches) as determined from evaporation pan measurements. Groundwater-surface water interactions in the area are generally very complex because of abundant surface water bodies. The area encompasses 2 major rivers, 9 principal streams, and 356 lakes ranging from less than an acre to 2,050 acres. Groundwater discharges into some of these water bodies while others act as recharge points (Allen et al., 1972). In addition, the cities of Portage and Kalamazoo, as well as several industrial plants, remove large quantities of groundwater from the aquifer, further complicating the groundwater-surface water flow regimes. Allen et al. (1972) have suggested that outwash material forming the major aquifer in the study area can be divided into an upper and a lower aquifer with an intervening aquiclude consisting of fine materials. After reviewing over 300 well log records in the region, the present author concludes that there is no evidence for an extensive confining layer. However, abrupt changes in the texture of the outwash clearly suggest a heterogeneous aquifer. SOURCES OF TRITIUM DATA The average yearly concentrations of tritium added to groundwater in the study area are shown in Figure 3. The input function was generated by identifying the yearly groundwater recharge period for the study area (Roether, 1967), and extrapolating the monthly concentration of tritium (corrected for radioactive decay) recorded for Chicago area precipitation (U.S. Geological Survey, 1982; International Atomic Energy Agency, 1969-1975) from 1953 to 1979. The application of the Chicago data seems reasonable (Andersen and Sevel, 1974; Atakan et al., 1974; Maloszewski et al., 1982) since the study area lies only 100 km northeast of Chicago. In Kalamazoo evaporation during the crop season, May to October, averages 85.3 cm. (33.6 inches). This exceeds precipitation during the same time period by 73% (Stronmen, 1971). Transpiration, which is at a maximum during this time period, also increases the water deficit. As a result, during high evapotranspiration periods most of the water found in the unsaturated zone together with associated tritium would be lost to the atmosphere and not be incorporated into groundwater. Also, enrichment with respect to tritium in the unsaturated zone due to fractionation effects (Stewart, 1972) should be negligible (Thatcher, 1967; Smith et al., 1970). At the same time any increase in the tritium content of soil moisture resulting from the microbial oxidation of HT to HTO (Ehhalt, 1973) would not be introduced to the groundwater since during the recharge period biological activity is low. 17- 16-1 15-4 14- 21155 7- .d Trit in m Concentration (WI 100) Figure 3. oarlglnel amount (Corrected tor decoy I ----- .s‘ '_ II— ‘ I. I! ::- ' --°1:- . 4.. . & {$38 xiii comm 5953? fig was '. -. . t e . « 'y '7 Yea r (1953-1979) Tritium Input Function for the Kalamazoo area. 10 SAMPLING. A water-table map of the study area (Michigan Department of Natural Resources Water Quality Division, 1983) based on 1,200 well records was used to locate two major groundwater divides within the outwash aquifer (Figure 4). On the basis of this map two areas were chosen for sampling, one near the City of Portage and the other near the City of Richland (Figure 4). In both areas waterwells are of varying depth and completely cased except near their base where they draw water through a screened interval which is generally less than one meter long. The existing well coverage in both areas was sufficient to sample the saturated thickness of the aquifer from 2.4 to 33.2 m (8 to 110 feet) at approximately 1 m (3.2 feet) intervals. The tritium content of the samples was determined following the U.S. Geological Survey and International Atomic Energy Agency established analytical procedures outlined in the appendices (Wyerman, 1976; International Atomic Energy Agency, 1967). ll RHW stow 9 law Coo teer Interval ten 0 Centre! o Other Figure 4. Water table map of Kalamazoo County (After D.N.R. 1983). 12 TRITIUM IN THE SATURATED ZONE The results of the tritium analyses of 46 water samples from the Portage and Richland areas are shown on Table 1. The location of each well sampled is shown in Figure 4. In addition, the concentration of tritium measured in each sample is plotted in Figure 5 as a function of sample depth below water table and shows a range from 0 to 173 TU. In the Portage area bomb tritium (> 2 TU) was recognized in 21 samples derived from 2.7 m to 22.9 m below the water table. Pre-bomb tritium (< 2 TU) was identified in 3 samples from depths greater than 21.3 m below the water table (Figure 5a). The data also show 3 samples containing no bomb tritium together with 2 samples containing bomb tritium 22.3 m below the water table. In the Richland area bomb tritium was recognized in 22 samples derived from 2.4 to 29.6 m below the water table (Figure 5b). Pre-bomb tritium was identified in 2 samples from depths greater than 32.0 m below the water table. The data from Richland show no bomb tritium below 29.6 m below the water table. The tritium data in Figure 5 suggests that the depth of penetration of bomb tritium is different in Portage and 13 Table 1. Raw tritium data. 800910 lleotrolyei e H Counting Groe e s emote + T U No erHclency number ettloienoy c pa TU 5T0 0.3815 128 0.075 10.48 108.46 9.21 1 0.6815 142 0.075 3.2 19.10 3.18 2 0.5851 129 0.075 6.74 41.23 5.43 5 0.6775 LS9 J.075 6.72 40.01 3.9: 4 0.6851 125 ).075 6.72 40.61 3.47 J ; 2:1: 119 ..075 6.53 43. L 3.7; : J.::51 137 1.0 0 7.28 “9.73 3.10 5 0.6665 11* =.'7S J :9 3.46 1..$ 574 U.SéL? 1-0 } U75 4.:8 101.1r 5._3 11 0.6461 -20 9.073 3.38 34.05 S ’9 .2 0.5479 115 0.075 «.32 Q.)0 H71 13 0.6337 -19 0.075 .73 ~4.‘v 3.7“ 14 'J :“03 1-5 ').073 2.09 25.-) .I _4 15 0.6430 -13 .075 4.79 J 1' e, is 9.5448 119 ).075 :.41 ii 27 . 3 15 0.5322 114 0.075 3.41 Li-t“ 7 L 20 ).d47§ 126 n.073 5.96 23 21 - 1 570 3.6283 125 ‘.075 10 7: 1 : +1 7 -1 33 J.6762 117 0.073 5.56 -2 61 1 . 24 “07:03 1:5 000;5 3044 29.80 3 1 25 0.7030 117 3.0 3 3.32 50..‘ 3.37 :3 )07192 $17 30075 5.73 57.32 5. 3 2? 007083 122 0.075 :0“O 23.44 107" 28 0.6321 125 3.075 5.94 25.;3 2.3 29 0.6934 124 0.075 6.66 20.1: 5..1 30 0.7175 123 0.075 5.23 18.76 1.6 31 0.6776 123 0.075 6.26 42.25 3.69 STD 0.7454 127 0.075 9.46 108.46 9.2; 34 0.7641 127 0.075 6.42 65.29 E 55 35 0.7416 133 0.070 9.33 13.98 9.60 36 007295 120 0007- 0024 24.41 2007 38 0.7131 137 0.070 8.18 57.32 4.67 39 006638 110 0.075 4056 0000 N/H 43 0.7174 119 0.075 5.48 13.00 1.1. STD 0.6551 115 0.075 10.50 107.84 9.16 45 0.6242 111 0.075 3.34 50.25 4.76 47 0.6933 123 0.075 6.79 43.71 3.7- 48 0.7501 118 0.075 5.18 10.63 0.95 49 0.7470 124 0.075 6.70 45.66 3.88 50 007387 119 00075 5053 17036 1048 52 0.6785 129 0.075 6.59 37.28 3.17 53 0.6631 124 0.075 6.97 40.05 3.41 54 0.7413 111 0.075 5.67 ' 20.59 1.75 STD 0.6154 110 0.075 10.10 108.43 9.22 57 0.6065 119 0.075 6.69 37.30 3.17 58 0.6334 127 0.075 7.84 63.60 5.35 59 0.6323 110 0.075 4.41 0.00 N/A 60 0.5188 116 0.075 5.10 13.21 2.73 61 0.6039 123 0.075 6.88 32.23 2.73 62 0.6415 129 0.075 12.18 173.91 14.70 63 0.6523 120 0.075 6.46 36.34 3.09 64 0.6246 115 0.075 4.40 0.00 N/A l4 18 O 17-1 xx 12. a 11-1 101 F a- ’5 7- " o x a o 0 j o o . 5‘ I: o o c 4. o .2 ° 3 E " 0 ° 2 E 24 o O O O 3 o c '4 o o 0 n A E a .2 CI! 7.1 ': . .- C-J . .7 0 0 a o ' 0 0 . a ’5 . = z- o ' 0 . g 0 0 9 1.1 9 I I r “t I I l s so u so as so as Depth Below Water Table l m ) o nucnuno sums n rennet uni-Lu Figure 5. Tritium concentrations of groundwater as a function of sampled depth. 15 Richland areas. A comparison of the two areas shows that the Portage area is heavily urbanized with a population of 38,157, while the Richland area is predominantly agricultural with a population of 3,536. In an urban area with its proliferation of low permeability materials, artificial drainage, and other alterations of the natural ground surface, the amount of water infiltrating would be less than that in an agricultural area (Jens and McPherson, 196“). Therefore, in an urban area the 1953 marker would not have travelled as far as in a rural one. Likewise, the two anomalous samples from the Portage section are also the result of the same land use variation since they are located outside the urban development area of the city of Portage. RECHARGE RATES Assuming vertical piston-type flow at water-table divides, the volume of water in an aquifer above a specific marker would represent the amount of recharge that the aquifer has received since that marker entered the aquifer. If the age of the groundwater at a particular depth, the saturated thickness of the aquifer above that depth, and the porosity of the aquifer materials are all known, then 16 the rate of recharge to the aquifer can be determined by the following formula (Vogel 1967): R = [P x D] / T where R is the recharge rate, P is the porosity, D the depth below the water table to a specific marker, and T the time in years since that marker entered the groundwater. In the Kalamazoo area the deepest occurrence of bomb tritium below groundwater divides would be associated with . the earliest precipitation event which contains bomb tritium. This would be in 1954 (Figure 3) or about the time of the first atmospheric thermonuclear tests (Begemann and Libby, 1957; Butlar and Wendt, 1958; Eriksson, 1964; 1967). Likewise, groundwater with no bomb tritium would be associated with precipitation events prior to 195“. In the recharge calculations, a porosity of 0.30 ($0.03), has been assumed based on the type of aquifer material (Todd,1959; Walton, 1970; Davis and DeWiest, 1966). Depths of 19.5 to 21.3 m (63 to 70 ft) and 29.6 to 32.0 m (97 to 105 ft) to the 1953 marker layer have been assigned to the Portage and Richland areas, repectively. Also, a time of 31 years has been assigned since introduction of bomb tritium to the aquifer. The recharge rates obtained from this investigation, 18.9 to 20.6 cm (7.4 to 8.1 in) per year for the Portage 17 area and 28.6 to 31.0 cm (11.3 to 12.2 in) per year for the Richland area are in close agreement with an average rate of 22.9 cm (9 in) per year that has been previously determined for the Kalamazoo area by water budget analysis (Allen et al., 1972). DISCUSSION The input function shown in Figure 3 predicts a tritium concentration in Kalamazoo groundwater of several hundred TU. However, the highest concentration in groundwater never exceeded 174 TU. To resolve this discrepancy it is necessary to look at the sources of precipitation for both the study area and the Chicago area. ’Winter-time precipitation occurring in the study area has two origins: large scale atmospheric disturbances and moisture from lake Michigan (Dr. Nurnberger, personal communication). The moisture found in the first case is essentially the same as that of the Chicago area and therefore would have similar tritium concentrations. However, the second component is not generally found in the Chicago area precipitation and would be generally deficient in bomb tritium since it is derived chiefly from Lake Michigan waters (Dincer et al., 197“). Consequently, the tritium concentrations in 18 precipitation in the study area have probably been lower for the past 31 years than those measured for precipitation in Chicago. The input functions also predict that the groundwater in Kalamazoo should contain two spikes of higher than average tritium concentration corresponding to the recharge periods of 1959, and 1963. These spikes were generally not found in the groundwater. Because the sampling interval was approximately 1 m, it is possible that the markers were missed and not sampled. Another possibility is that hydrodynamic dispersion has reduced the intensity of these spikes. Knowing certain parameters we can calculate the expected reduction or smoothing of the spikes resulting from hydrodynamic dispersion. Employing conservative estimates for the diffusion coefficient of tritium, D’ a S E -10 (Allison and Hughes, 197”; Smith et al., 1970) and the following equation (Freeze and Cherry, 1979) C = C0 erfc (x / D* t) where erfc is the complementary error function, 0 and Co the relative concentrations, x the distance diffused in time t.] it can be shown that molecular diffusion can account for tritium moving 0.0002 m per year or only 6.2 mm per 31 years. Based on this small value, molecular diffusion can 19 therefore be safely ignored (Nir, 1964; Bredenkamp et al., 197A). It has been noted that as tritium travels through an aquifer it disperses (Eriksson, 1958; Atakan et al., 197A; Dincer et al., 197a; Andersen and Sevel, 1974; Zimmerman et al., 1966; 1967). This hydraulic dispersion can be calculated using a groundwater flow velocity of 3.0 E -6 cm/sec which is based on the calculated recharge rate, a range of dispersivity coefficients from 8 E -3 to 2.5 E -2 cm (Ogata, 1970),, and the following equation (Ogata and Banks, 1961; Ogata, 1970; Freeze and Cherry, 1979): D =Dm xV where D is the hydraulic dispersion, Dm is the coefficient of dispersion (dispersivity), and V is the groundwater flow velocity. The result of the calculation shows that hydraulic dispersion can acccount for 2.4 E —8 to 7.5 E -8 cm2 per second, or 23 cmz to 78 cm in 31 years. Based on this small value hydraulic diSpersion can also be safely ignored. 2O CONCLUSION The Kalamazoo area in Michigan is underlain by outwash sands and gravels randomly interstratified with stringers of finer material. This variation in texture results in a heterogeneous aquifer. The tritium concentration in 46 samples of groundwater collected beneath two water table divides was determined in order to calculate the rate of recharge to the aquifer. It was found that tritium concentrations in the saturated zone ranged from 0 to 17M TU. Pre-bomb tritium was found to occur at depths of 21.3 to 33.2 m below the water table. Based on this depth, recharge to the aquifer was calculated to be in the ‘range of 19.5 to 32.0 cm per year. This is in close agreement with recharge rates determined using conventional water budget analysis. This investigation has shown that the bomb tritium method can be applied to heterogeneous aquifers in order to determine recharge rates. Furthermore, since the amount and distribution of bomb tritium predicted by the input function was not found in the groundwater, the input function need not be calculated for the entire period from 1952 to the present. Only the date when bomb tritium entered the aquifer has to be known. Moreover. unless the area of interest was experiencing drought, or located in 21 the southern hemisphere, the onset of bomb tritium can be safely assumed to be 1954. The only data necessary to calculate recharge would be the location of the water divide(s), and tritium concentrations in the groundwater. This makes the bomb tritium method ideally suited for the determination of recharge rates in developing areas where there are often few records of hydrological data which are needed when using conventional hydrological methods. Bomb tritium provides a relatively easy and inexpensive way to determine the rate of recharge to heterogeneous aquifers. 22 APPENDICES APPENDIX I ANALYSIS OF SAMPLES The laboratory procedure employed in the determination of the tritium content of groundwater samples is composed of four parts: pre-distillation, electrolysis, post- distillation, and liquid scintillation counting. All samples were analyzed at the tritium laboratory at Michigan State University. The laboratory is located in room 117 of the Natural Sciences Building. PRE-DISTILLATION. The initial step in the analysis consists of the complete distillation of the sample to remove all dissolved solids and suspended particles. The system used consists of a boiling flask, KJeidahl bulb, Allihn condenser and receiving flask (Figure 6). In addition, cooling water, gas, vacuum and dry air are used. The system is first 23 Water out Allihn condenser Boiling flask Heater Water in Vacuum Balloon Receiving f I ask Figure 6. Pre-distillation apparatus. 24 dried under vacuum. Once drying is complete, dry air is admitted and the sample placed in the boiling flask which is immediately returned to the line. Heat is applied and distillation begins. If during distillation the pressure compensator indicates excessive pressure, vacuum is applied momentarily. Just before the sample has been totally distilled the heat is removed to allow any water in the KJeidahl bulb to drain. A heat source is lightly applied to the KJeidahl bulb and connecting glass to totally dry them. The heat is returned to the boiling flask and continued for several minutes beyond apparent dryness to insure the dehydrolyzation of salts. Finally the receiving flask is removed, with the distilled sample, and sealed from the atmosphere. ELECTROLYSIS. Enrichment of the samples must be carried out because of the low levels of tritium found in natural waters. Electrolytic enrichment is used because it allows a large number of samples to be enriched at the same time, it is simple, and has been proven to yield reproducible results (Florkowski, 1981). The electrolysis system consists of: gas exhaust lines, a low temperature bath, Ostlund electrolysis cells, power supply and electrodes. Normally 180 to 200 ml of water are placed in the Ostlund cell followed by an aliquot 25 of an electrolyte. Currently NaOH pellets are used as the electrolyte. Once the NaOH has thoroughly dissolved the cells are placed in the cooling bath and allowed to equilibrate with the temperature of the bath. Since it has been found that the electrolytic process is most efficient at a low temperature the water bath and the cells are kept at 2 degrees Centigrade (Hoffman and Stewart, 1966). The iron (-) and nickel (+) electrodes are next introduced into the cells, the vacuum lines attached to the top of the cells and the exhaust fan turned on. The electrodes are connected in series electrically and connected to the power supply unit. The current used is based on a total electrode area of 35 cm2 and a maximum current density of 0.17 amps per cm2 (Wyerman, 1976). Since electrolytic efficiency decreases as the electrolysis temperature increases, and since heat dissipation is a function of the water level inside the cell, it is necessary to limit the amount of heat produced by the electrolysis as the volume decreases. To this end the current supplied to the electrodes is decreased as a function of the volume remaining in the cells. Electrolysis is carried out until the level of the sample drops below the end of the electrodes and the circuit is broken. At this point the electrolysis heads are replaced by the distillation heads and the samples sealed, 26 POST-DISTILLATION. When electrolysis is complete, 8 to 10 ml of enriched sample normally remain in the cell. This water must be quantitatively removed from the cell and the NaOH removed from the sample. Both of these steps are accomplished by freeze—up distillation. The post-distillation line consists of Ostlund cells, distillation heads, weighing bulbs, heat sources, liquid nitrogen, vacuum, and dry air (Figure 7). The Ostlund cells are removed from the tank and connected to the weighing bulbs via the distillation heads and vacuum is applied. If the system appears vacuum-tight the vacuum is shut off at the stOpcock. The weighing bulb is partially submerged in liquid nitrogen. A flexible heating coil is wrapped around. the cell and heat is applied. Heating is continued for an additional half hour he weighing bottle . is now measured to determine the exact amount of sample in it. Finally 8 ml of this enriched, post-distilled sample are placed in a borosilicate liquid scintillation vial in preparation for liquid scintillation counting. 27 Weighing V bulb acuum Distillation heed Liquid nitrogen Ost lund __ cell Elect rode Figure 7. Post-distillation apparatus. 28 I. ‘ . ‘1}; - ‘- APPENDIX II LIQUID SCINTILLATION COUNTING AND QUENCH CORRECTION. The tritium laboratory at M.S.U. employs a Beckman 8100 Liquid Scintillation Counter, LSC. The LSC does not directly measure the presence of a radionuclide but the interactions of the radionuclide with matter. The LSC is a passive observer which detects the light flashes or scintillations that are produced by the ionization of some of the molecules in the fluorescent medium (cocktail) as a result of the emission of beta particles from the decay of the radioisotope. The rate at which these flashes are produced within a finite time period is proportional to the rate of radioactive decay; furthermore the intensity of the light scintillation is proportional to the energy of the decaying beta particle (Beckman, 1978). The fluorescent medium used is Insta Gel Universal Scintillation Cocktail. 12 ml of the cocktail are mixed with 8 ml of sample achieving the most efficient 29 concentration of cocktail versus sample and safely avoiding the unstable zone (United Technologies Packard, 1984). The Beckman 8100 LSC is equipped with 2 channels both of which have adjustable lower and upper limits to maximize the efficiency and minimize background noise. Following the manufacturers instructions (Beckman, 1978) the Optimum window was determined to have a lower limit of 170 and an upper limit of 225. All samples, standards, and backgrounds were counted using this window. COUNTING EFFICIENCY AND QUENCHING. Any component in the vial which interferes with the scintillation process produces a "quenching" effect. Quenching lowers the observable counting rates and reduces the heights of the higher energy beta particles. By reducing the overall photon output, quenching makes the data obtained by the LSC ambiguous. To determine the amount of quenching in a sample the Beckman 8100 LSC utilizes an external standardization technique known as the H Number. Each sample is subjected to a high energy gamma source, such as Cesium 137, and the resulting Compton edge determined. If quenching is 30 present the Compton edge will be displaced relative to that of an unquenched sample. This displacement is the H number (Long, 1978). Samples labelled with an H number can be subjected to quench correction in order to determine the counting efficiency. Quench standards which contain the same type of isotope as the samples but whose activities are known are counted to determine their count rates and H numbers. A quench curve is generated which relates the counting efficiency of each standard to its corresponding H number (Figure 8). The counting efficiency of a sample of unknown activity is determined from the interpolation of its H number and the quench curve. 31 Counting Effic ienc y (96) 20‘ 10' O 0 00 0 ° ° e e l e e e M o - g A 4 100 120 no do 100 H-number Figure 8. 'Quench curve. 32 APPENDIX III TRITIUM CALCULATION AND TWO SIGMA SUBTRACTION To calculate the final activities of the samples the following data must be known (Hufen et al., 1969; Wyerman, 1976): Background count rate, BRC; Counting efficiency, CE; Electrolysis efficiency of standards, EEes; Electrolysis Fractionation Factor, B; Electrolysis efficiency of samples, EEs; Initial activity of electrolysis standard, Aoes; Final activity of electrolysis standard, Aes; Volume concentration of electrolysis standard, (Vo/Vf)es; Volume concentration of samples, (Vo/Vf)s; and, Sample count rate, CPM. Background count rate, BCR. A vial containing blank water, water containing 0 TU, was prepared in the same manner as the enriched samples. It was then counted for 16 hours and repeated 10 times to generate an average BCR of 4.73 t 0.12 counts per minute. 33 Counting Efficiency, CE. Based on the quench curve and the H number generated by the LSC, and discussed in appendix II, the counting efficiency of both the samples and the determined. Electrolysis efficiency of standards, EEes. EEes = (Vo/Vf)es x (Aes/Aoes). Electrolysis fractionation factor, B. B=ln (Vo/Vs)es / -ln EEes. Electrolysis efficiency of samples, EEs. EEs = (Vo/Vf)s1‘2 Final activity of samples, FAS. The final activity of the samples can background was be calculated using the above parameters and the following equation: FAS = (CPM—BCR) / CE x EEs x (Vo/Vf)s x V x K where V 8 volume of sample (8ml), and K=.0071 DPM/ml TU conversion factor. 34 TWO SIGMA ERROR DETERMINATION The LSC automatically determines the 2 sigma % error of all counts, based on the actual counting data (Beckman, 1978): H- 2o% = * 200 / (R x T) where R is the counting rate and T is the length of time the sample was counted. This 20 error represents the error due to both the sample activity (CPM) and the background (BCR). However what is desired is the 20 error of the sample activity alone. This is obtained by the following equation (Wyerman, 1978): 120% = t [(CPM x 20 )2 + (BCR x 20.... )2 1"2/ (CPM - BCR) cpm 35 BIBLIOGRAPHY BIBLIOGRAPHY Allen, W., Miller, J., and Wood, W., 1972. Availability of water in Kalamazoo County, South Western Michigan: United States Geological Survey Water Supply Paper 1973. Allison, G., and Holmes, J., 1973. The environmental tritium concentration of underground water and its hydrological interpretation: Journal of Hydrology, v. 19’ pe 131-143. Allison, G., and Hughes, M., 1974. Environmental tritium in the unsaturated zone: Estimation of recharge to an unconfined aquifer, in Isotope techniques in ground water hydrology; Proceedings of a symposium, Vienna, International Atomic Energy Agency, p.57—69. Allison, G., and Hughes, M., 1975. The use of environmental tritium to estimate recharge to a South-Australian aquifer: Journal of Hydrology, v. 26, p. 245-254. Allison, G., and Hughes, M., 1978. The use of environmental chloride and tritium to estimate total recharge to an unconfined aquifer in Australia: Australian Journal of Soil Research, v. 16, p. 40-54. Andersen, L., and Sevel, T., 1974. Six years' environmental tritium profiles in the unsaturated and saturated zones, Gronhoj, Denmark, in Isotope techniques in ground water hydrology; Proceedings of a symposium, Vienna, International Atomic Energy Agency, p. 3—17. Atakan Y., Roether, W., Munnich, K., and Matthess, G., 1974 The Sandhausen shallow ground water tritium experiment, in Isotope techniques in ground water hydrology; Proceedings of a symposium Vienna, International Atomic Energy Agency, p. 21—42. 36 Beckman, 1978. LS 8000 series liquid scintillation systems: Beckman Instruments Division. Irvine, California. Begemann F., and Libby, W., 1957. Continental water balance, ground water inventory and storage times, surface ocean mixing rates and world-wide water circulation patterns from cosmic ray and bomb tritium: Geochimica et Cosmochimica acta, v. 12, p. 277-296. Bredenkamp, D., and Vogel, J., 1970. Study of a Dolomitic aquifer with Carbon-14 and Tritium, in IsotOpe hydrology; Proceedings of a symposium, Vienna, International Atomic Energy Agency, p. 349-371. Bredenkamp, D., Schutte, J., and DuToit, G., 1974. Recharge of a dolomitic aquifer as determined from tritium profiles, in Isotope techniques in ground water hydrology; Proceedings of a symposium, Vienna, International Atomic Energy Agency, p. 73-94. Brown, T., and Taylor, 0., 1974. Geohydrology of the Kaikoura Plain Marlborough, New Zeland, in Isotope techniques in ground water hydrology; Proceedings of a symposium, Vienna, International Atomic Energy Agency, pe 169-1890 Buttlar, H. v., and Libby, W., 1955. Natural distribution of cosmic ray produced tritium II: Journal of Inorganic and Nuclear Chemistry, v. 1, p. 57-91. Buttlar, H. v., and Wendt, I., 1958. Ground water studies in New Mexico using tritium as a tracer: American Geophysical Union Transactions, v. 39, p. 660-668. Davis, 8., and DeWiest, R., 1966. Hydrogeology: New York, John Wiley and Sons Inc. Deutsch, M., Vanlier, K., and Giroux, P., 1960. Ground water hydrology and glacial geology of the Kalamazoo area, Michigan: State of Michigan Geological Survey Progress Report 23. Dincer, T., Al—Magrin, A., and Zimmerman, U., 1974. Study of the infiltration and recharge through the sand dunes in arid zones with special reference to the stable isotopes and thermonuclear tritium: Journal of Hydrology, v. 23, p. 79-109. Ehhalt, D., 1973. On the uptake of tritium by soil water and ground water: Water Resources Research, v. 9, p. 37 Eriksson, E., 1958. The possible use of tritium for istifiaging ground water storage: Tellus, v. 10, p. 72- 7 . Eriksson, E., 1964. An account of the major pulses of tritium and their effects in the atmosphere: Tellus, v. 17, p. 118-130. Eriksson, E., 1967. Atmospheric transport of tritium, in Stout, G., (ed.), IsotOpe techniques in the hydrologic cycle: Papers presented at a symposium at the University of Illinois, Washington D.C., American Geophysical Union Monograph Series No. 11, p. 56-58. Florkowski, T., 1981. Low level tritium assay in water samples by electrolytic enrichment and liquid scintillation counting in the International Atomic Energy Agency laboratory, in Methods of low level counting and spectrometry. Proceeding of a symposium, Vienna, International Atomic Energy Agency, p. 335-351. Forstat, D., 1982. Drift thickness of Kalamazoo county, in Selected geologic maps of Kalamazoo, Michigan, County Geologic Map Series, Lansing Michigan, Michigan Geological Survey Davison. Freeze, R., and Cherry, A., 1979. Groundwater, New Jersey, Prentice-Hall Inc. Hoffman, C., and Stewart, G., 1966. Quantitative determination of tritium in Natural Waters: United States Geological Survey Water Supply Paper 1696-D. Hubbert, M., 1940. The theory of groundwater motion: Journal of Geology, v. VXLVIII, p. 785-944. Hufen, T., Duce, R., and Lau, L., 1969. Some measurements of the tritium content in the natural waters of Southern Oahu, Hawaii: Water Resource Research Center, Technical report No. 34. Univ. of Hawaii. International Atomic Energy Agency, 1969 to 1975. Technical report series, No. 96, 117, 129, 147, 165, 192. International Atomic Energy Agency, Vienna. International Atomic Energy Agency, 1967. Radioactive dating and methods of low level counting, Vienna, International Atomic Energy Agency. Jens, 8., and McPherson, M., 1964. Hydrology of Urban Areas, in Chow,V.((ed.)), Handbook of Applied 38 hydrology, p. 20-3. Libby, W. 1953. The potential usefulness of natural tritium: National Academy of Science, v.39, p. 245-247. Long, E., 1978. Liquid scintillation counting theory and techniques: Beckman Instruments Davison, Irvine, California. Maloszewski, P., Rauert, W., Stichler, W., and Herrmann, A., 1983. Application of flow models in an alpine catchment area using tritium and deuterium data: Journal of Hydrology, v. 66, p. 319-330. Michigan Department of Natural Resources Water Quality Division, 1983. Groundwater management strategy for Michigan, Computer subsurface mapping and surficial textural mapping as assessment tools. Lansing, Michigan. Monaghan, G., 1984. Late Wisconsinan drift stratigraphy of the Saginaw and Lake Michigan ice lobes in Southwestern Michigan: Unpublished M.S. thesis, Michigan State University. Monaghan, G., and Larson, G., 1983. Selected geological maps of Kalamazoo County, Michigan. Michigan Dept. of Natural Resources Geological Survey Division, Lansing Michigan. Monaghan, G., and Ritter, M., 1982. » Identification and verification of natural recharge areas in glaciated terrains, in Abstracts with programs, Geological Society of America, vol. 7, p. 569. Munnich, K., Roether, W., and Thilo, L., 1967. Dating of ground water with tritium and C-14, in Isotopes in Hydrology: Proceedings of a symposium, Vienna, International Atomic Energy Agency, v. 1, p. 305-319. Nir, A., 1964. On the interpretation of tritium "age" measurements of ground water: Journal of Geophysical Research, v. 69, p. 2589-2595. Nurnberger, F., 1982. Summary of evaporation in Michigan, Michigan Department of Agriculture, Michigan weather service. Offer, 3., 1982. Determination of recharge rates to a drift aquifer using bomb tritium within the saturated zone. Unpublished M.S. thesis. Michigan State 39 University. Ogata, A., 1970. Theory of dispersion in a granular medium: United States Geological Survey Professional Paper 411 I. United States Geological Survey. Ogata, A., and Banks, R., A solution of the differential equation of longitudinal dispersion in porous media: United States Geological Survey Professional Paper 411 A. United States Geological Survey. Persson, G., 1974. Tritium data from ground water in the Kristiansted Plain, Southern Sweden, in IsotOpe techniques in ground water hydrology: Proceedings of a symposium, Vienna, International Atomic Energy Agency, v. 1, p. 45-55- Rabinowitz, D., Gross, G., and Holmes, 0., 1977a. Environmental tritium as a hydrometereologic tool in the Roswell basin, New Mexico I. Tritium input function and precipitation-recharge relation: Journal of Hydrology, v. 32, p. 3-17. Rabinowitz, D., Gross, G., and Holmes, 0., 1977b. Environmental tritium as a hydrometereologic tool in the Roswell basin, New Mexico III. Hydrologic parameters: Journal of Hydrology, v. 32, p. 35-46. Ritter, M., 1980. The use of tritium for confirming areas of ground water recharge. Meridian Township, Michigan: Unpublished M.S. thesis. Michigan State University. Roether, W., 1967. Estimating the tritium input to ground water from wine samples: Ground water and direct run-off contribution to Central European surface waters,in Isotopes in Hydrology: Proceedings of a symposium, Vienna, International Atomic Energy Agency, v. 1, p. 73-89- Smith, D., Wearn, P., Rowe, P., and Richards, H., 1970. Water movement in the unsaturated zone of high and low permeability strata by measuring natural tritium, in Isotope hydrology; Proceedings of a symposium, Vienna, International Atomic Energy Agency, p. 73-85. Stewart, G., 1967. Fractionation of tritium and deuterium in soil, water in Stout, G., (ed.), Isotope techniques in the hydrologic cycle: Papers presented at a symposium at the University of Illinois, Washington D.C., American Geophysical Union Monograph Series No. 11, p.97. 40 Stewart, M., and Downes, C., 1980. Isotope hydrology of Waikoropupu Springs, New Zeland, in Montgomery, 0., and Perry, E. eds., IsotOpe studies of hydrologic processes: Selected papers from a symposium, Dekalb, Illinois, Northern Illinois University Press, 1982. p. 15-23. Stronmen, N., 1971. Climatological Summary, Climatography of the United States No. 20-20. Sukhija, B., and Rao, A., 1983. Environmental tritium and radiocarbon studies in the Vedavati River Basin, Karnataka and Andhra Pradesh, India: Journal of Hydrology, v. 60, p. 185-196. SukhiJa, B., and Shah, 0., 1976. Conformity of ground water recharge rate by tritium method and mathematical modelling: Journal of Hydrology, v. 30, p. 167-178. Thatcher, L., 1967. Water tracing in the hydrologic cycle, in Stout, G., (ed.), IsotOpe techniques in the hydrologic cycle: Papers presented at a symposium at the University of Illinois, Washington D.C., American Geophysical Union Monograph Series No. 11, p.97-108. Todd, D., 1959. Ground water hydrology. New York, John Wiley & Sons, Inc. . Toth, G., 1962. A theory of ground water motion in a small drainage basin in central Alberta, Canada: Journal of Geophysical Research, v. 67, p. 4375-4387. Toth, G., 1963. A theoretical analysis of ground water flow in a small drainage basin: Journal of GeOphysical Research, v. 68, p. 4795-4812. Turner, J., Allison, G., and Holmes, J., 1983. The water balance of a small lake using stable isotopes and tritium: Journal of hydrology, v. 70, p. 199-220. U.S. Geological Survey, 1984. Computer print out. United Technologies Packard, 1984. Insta-Gel Universal Cocktail, Downers Grove, Illinois, United Technologies Packard. Verhagen, B., Sellchop, J., and Jennings, C., 1970. Contribution of environmental tritium measurements to some geohydrological problems in Southern Africa, in Isotope hydrology; Proceedings of a symposium, Vienna, International Atomic Energy Agency, p. 289-314. 41 Verhagen B., Smith, P., McGeorge, I., and Dziembowski, Z., 1978. Tritium profiles in Kalahari sands as a measure of rain water recharge, in Isotope hydrology: Proceedings of a symposium, Vienna, International Atomic Energy Agency, p. 733-748. Vogel, J., 1967. Investigation of ground water flow with radiocarbon in IsotOpe hydrology, International Atomic Energy Agency: Proceedings of a symposium, Vienna, International Atomic Energy Agency, p. 355-369. Vogel, J., Thilo, L., and Van DiJken, M., 1974. Determination of ground water recharge with tritium: Journal of hydrology, v. 23, p. 131-140. Walton, W., 1970. Groundwater resource evaluation, New York, McGraw Hill Book Co. Wyerman, T., 1976. Laboratory facility for the analysis of natural levels of tritium in water: United States Geological Survey Open File Report. Zimmerman, U., Munnich, K., and Roether, W., 1966. Tracers determine movement of soil moisture and evapotranspiration: Science, v. 152, p. 346-347. Zimmerman, U., Munnich, K., and Roether, W., 1967. Downward movement of soil moisture traced by means of hydrogen isotopes, in Stout, G., (ed.), Isotope techniques in the hydrologic cycle: Papers presented at a symposium at the University of Illinois, Washington D.C., American Geophysical Union Monograph Series No. 11, p. 28- 37. 42