PLACE N RETURN BOXtonmmthbchockomfmn yourncotd. TO AVOID FINES Mum on Of More dd. duo. DATE DUE DATE DUE DATE DUE (- jl | | MSU I: An Affirmative Action/Ema! Oppommlly Inflation mm: Environmental Isotope (T :80) Studies of Storm Runoff in the Red Cedar Basin, Michigan By Cheol Woon Kim A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Geological Sciences 1995 ABSTRACT ENVIRONMENTAL ISOTOPE (T, 18O) STUDIES OF STORM RUNOFF IN THE RED CEDAR BASIN, MICHIGAN By Cheol Woon Kim The natural tritium and oxygen-18 composition of rainwater differs from that of the groundwater in the Red Cedar basin, Michigan. This isotopic difference was used to differentiate between groundwater and rain components in storm runoff in the Red Cedar basin for the period of September 24 to October 9, 1993. Before, during, and after the storm event of September 26 to September 30, 1993 samples of stream water, groundwater, and rain were analyzed for tritium and oxygen-18. Initial tritium and oxygen—18 results indicate that 69 % to 79 % of the storm runoff must have been groundwater during the storm runoff event which is somewhat greater than that obtained using a graphical technique (63 %) of streamflow separation. However, statistical analysis of the data shows that analysing tritium concentration in streamflow is not appropriate to separate flow components and that analysing oxygen-18 concentration in streamflow can be used to separate flow components. The results from these studies indicate that groundwater is an extremely important contributor to storm runoff events in the humid headwater catchment such as the Red Cedar basin. For the memory of my little brother, Cheol Woo ACKNOWLEDGMENTS Sincere appreciation and thanks to my advisor, Dr. Grahame J. Larson, for his thoughtful insight, intense interest, and willingness to offer guidance relating to both this thesis and to my education in general. I am grateful to my committee members Dr. David T Long and Dr. Nathaniel E. Ostrom for helpful suggestions and advice. I would like to thank Myeounghee Oh as my friend who was always willing to encourage me. Finally, special thanks go to my parents for their endless encouragement and support. Thank God. Iv TABLE OF CONTENTS List of Tables List of Figures 1. INTRODUCTION General Introduction Statement of Purposes Study Area Environmental Isot0pes (T, 18O) in Runoff Studies Tritium Oxygen-18 II. METHODS Sampling Stream and Rain Samples Analytical Methods Analyses of Water Samples for T and 18O Isotopic Hydrograph Separation Graphical Hydrograph Separation III. RESULTS Tritium Oxygen-18 Hydrograph Separation Using Tritium and Oxygen-18 Tritium Oxygen-18 IV. DISCUSSION V. CONCLUSIONS APPENDICES Appendix A - Sample Preparation Distillation Electrolysis Post-Distillation Appendix B - Tritium Activity Calculations V vii viii “5'5””.5 .D OOOOQNN 13 13 18 18 19 26 30 31 32 32 3‘2 33 REFERENCES VI 36 Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. LIST OF TABLES Tritium results. Oxygen-18 results. Storm hydrograph separation using tritium. Oxygen-18 data of groundwater in the Red Cedar Basin (Dannemiller et al., 1990). Storm hydrograph separation using oxygen-18. Result of hydrograph separations using graphical method, tritium, and oxygen-18. vii 14 16 21 22 23 24 Figure 1. Figure 2. Figure 3. Figure 4. Figure 5. Figure 6. Figure 7. Figure 8. Figure 9. Figure 10. Figure 11. LIST OF FIGURES The Red Cedar Basin in Michigan. The hydrograph of the Red Cedar River during Aug., Sep., and Oct. 1993. Precipitation in the Red Cedar Basin during Aug,. Sep., and Oct. 1993. Hydrograph separation using graphical method (semi-log scale). Hydrograph separation using graphical method (arithmetic scale). Variation of tritium in the Red Cedar River. Variation of oxygen-18 in the Red Cedar River. Relationship between tritium and oxygen-18. Hydrograph separations using tritium, oxygen-18, and graphical method. Variation of tritium in the Red Cedar River. Variation of oxygen-18 in the Red Cedar River. viii 11 12 15 17 25 29 I. INTRODUCTION General Introduction The term 'hydrograph separation' is normally associated with graphical hydrograph separation techniques which have been used for decades in predicting runoff components. For example, the graphical separation technique is commonly applied to storm hydrographs from forested catchments to separate 'quick flow' and 'delayed flow' contributions (Sklash, 1990). Another type of hydrograph separation, based on natural chemical tracers in water, attempts to be more physically based than the graphical technique in separating the rapid and slow delivery of water to the stream (Sklash, 1990). This separation technique apportions the storm hydrograph into contributing components based on the distinctive chemical signatures (TDS, electrical conductivity, and Cl) carried by each of the contributing components. The distinctive signature of each component is developed as the water passing through the catchment takes different flow paths and has different residence times. The tracer-based hydrograph separation technique normally involves a two- component mixing model for the stream. The model assumes that water in the stream at any time during a storm runoff event is a mixture of two components: 'new water', which is water from the current rain event, and 'old water', which is the subsurface water that existed in the catchment prior to the current rain event. This simple two-component mixing model approach allows the hydrologist to evaluate the importance of given conversion processes in a catchment. For example, if a rapid conversion mechanism (partial-area overland flow, saturation 1 overland flow or subsurface flow through macropores) is the dominant conversion process contributing to a storm runoff hydrograph, the tracer study should detect mostly 'new water' in the stream. Conversely, if a slow conversion mechanism (Darcian subsurface flow) is dominant in producing the storm runoff, the tracer study should find mostly 'old water' in the stream. There are some problems with separating hydrographs on the basis of chemical parameters such as TDS, electrical conductivity, and Cl because such parameters may have various sources or be affected by geochemical reactions (Sklash, 1990). For example, Nakamura (1971), Pilgrim et a1. (1979) and others have shown that the chemistry of the ‘new' water may vary as the rain water interacts with the catchment materials on the way to the stream. However, natural isotope tracers such as tritium, deuterium, and oxygen-18 may overcome these applied tracer problems mostly by the nature of their application. Statement of Purposes The purposes of this research are to analyze and understand the temporal variation of tritium and oxygen-18 in discharge of the Red Cedar River in southern lower Michigan and to test whether the concentration of tritium and oxygen-18 in discharge can be used to separate flow components, for example surface runoff and groundwater flow. Study Area The Red Cedar basin is located in Ingham County and western Livingston County, Michigan, and covers an area of 355 mi2 (Figure 1). The principal river in the basin is the Red Cedar River which is 42.2 miles long and has an average gradient of 2.7 ft/mi. OT .5322 E 53m 380 Ba 2:. ._ 23% mO—S OM £03 .033 o a: ”no r . N. W m m a . . cone; wean—=3 Saga comma mans» fig \. i / coma; mama—ES 8.95 53.5 I s / noun; gum 0928:. anbm .m.O.m.D +0 , DEED—30m \ / / I we 5:33 The basin is underlain by drift and bedrock. The drift is only a few feet thick in some areas, but is about 60 feet thick over most of the basin. It is composed mostly of till, but includes some fairly extensive beds of sand and gravel (Humphrys and Bradford, 1958). The bedrock includes the Saginaw Formation which is Paleozoic in age and consists mainly of sandstone and shale but includes some thin beds of limestone and coal (Humphrys and Bradford, 1958). The Saginaw Formation is a major source of groundwater for the municipal, industrial, and institutional water supplies in the basin (Humphrys and Bradford, 1958). Environmental Isotopes (T, "0) in Runoff Studies Tritium (T) and oxygen-18 are ideal tracers for runoff studies because of two attributes of these isotopes. Since tritium and oxygen-18 are constituent parts of some natural water molecules (e.g. HT“O, HZ'BO respectively), they travel at the same rate through the catchment as 'average' water (I-I,“O). Also, tritium and oxygen-18 are chemically conservative at low temperatures associated with most small watershed systems (Fritz er al., 1976; Kennedy et al., 1986; and others). This means that their concentrations in a volume of water do not change by reactions with catchment materials. Tritium Tritium is a radiogenic isotope of hydrogen whose half-life is in the order of 12.4 years. Tritium atoms represent an extremely small proportion of terrestrial hydrogen, about 10'14 - 10'“5 % of all hydrogen atoms. Concentrations of tritium are expressed in tritium units (TU), where 1 TU equals 1T/1018 atoms of protium (Fritz and Fontes, 1980). Tritium is produced naturally in the atmosphere by bombardment of nitrogen by solar radiation (Faure, 1986). It is also produced anthropogenically. Since the advent of atmospheric testing of thermonuclear devices in 1952, tritium as a by- product of thermonuclear testing ('bomb tritium') has been the dominant source of tritium in precipitation. After 1963, the year in which the atmospheric test ban treaty was signed, the tritium levels in precipitation began to decline gradually because of the cessation of atmospheric testing and radioactive decay (Gat, 1980). In addition to the order-of-magnitude variations in tritium in precipitation over the past few decades, tritium input also varies seasonally. For example, Chen (1993) reports that the tritium concentration of precipitation in East Lansing, Michigan is greatest during the months of June and July and least during the month of November due to the seasonal transference of tritium from the stratosphere to the troposphere. Distinct 'old' and 'new' water T values are required for storm runoff hydrograph separation. The required difference is possible because of the gradual decline in precipitation T values since 1963 and the even more gradual fall in groundwater T values. Oxygen-18 The important requirement in storm runoff hydrograph separation using oxygen-18 is that 'old' and 'new' water have distinct isotope values, like tritium. The 'new' water oxygen isotope values may vary considerably from event to event while the 'old' water oxygen isotope value remains fairly constant. Once rain passes into the saturated zone of shallow flow systems, the oxygen isotope values of subsurface water are changed only by mixing with waters with different isotopic contents. The homogenizing effects of recharge and dispersive processes produce groundwater with isotope values which approach uniformity in time and space and which approximate a damped reflection of the precipitation over a period of years (Brinkrnann et al., 1963). Evaporation and molecular exchange with ambient water vapor on the oxygen isotope content of water which infiltrates can lead to an enrichment in oxygen isotopes of the recharge as compared to precipitation (Gat and Tzur, 1967). H. METHODS Sampling Stream and Rain Samples The Red Cedar River was sampled on a daily basis for the months of August, September, and October of 1993 from under the Farm Lane Bridge located on the campus of Michigan State University, East Lansing (Figure 1). The apparatus for collecting the samples was 250 ml high-density polyethylene bottles. A subset of 16 samples (Figure 2) representing a single storm event (Sep. 24 ‘ Oct. 9) was selected based on 1) size of storm, 2) simplicity of hydrograph, and 3) clear definition of baseflow recession. This subset was then analyzed for T and 18O and used to separate 'new' and 'old' water for the storm event. Red Cedar River discharge data for the months of August, September, and October of 1993 were obtained from the U. S. Geological Survey gauging station located under the Farm Lane Bridge (Figure 1). Whole rain was also sampled for the months of August, September, and October of 1993 from behind the Natural Science Building on the campus of Michigan State University (Figure 1). From these samples, a subset of 4 samples associated with storm precipitation from September 26 to September 30 was also analyzed for T and‘sO (Figure 3). Precipitation data of the Red Cedar basin were obtained from the Michigan Meteorological Resources Station located at the Horticulture Research Center, on the campus of Michigan State University (Figure 1). 1000 m _ .............................................................. Red Cedar River sample subset N . .. 600 .J .................................................................................................. Discharge (cfS) 400 .. ......................................................................................................................................... 2m _. ............................................................................................................................. 14710131619222.528336912151821242? 9121518212A2730 Aug. Sep. Oct 0 IIIIIIIIIIIIIIIIIIITIIIIIIIIIIrITTITIIIIIIIIIIIlIIIIIIIFIFI%IIIIIJIIIIITITIITTIIIIIIIIIIIII 3 Figure 2. The hydrograph of the Red Cedar River during Aug, Sep., and Oct. 1993. 2 ml ........................................................................ Precrpitatron (m) Precipitationsample subset 1.. ...................................................................... L ....... I | 0.5" ......................................................... J ......................................... 0-1 - .1 WI 1 IIIIVII 14710131619222.528336912151321242? 36912151821242730 Oct. Aug. Figure 3. Precipitation in the Red Cedar Basin during Aug., Sep., and Oct. 1993. Analytical Methods Analyses of Water Samples for T and I‘0 Tritium analyses for the samples collected were done at the Michigan State University Tritium Laboratory and included electrolytic enrichment and scintillation counting (Ostlund and Werner, 1961, Taylor, 1981, and Wyerrnan, 1976, see Appendices). Results of the analyses are accurate to the 11.0 TU and are presented in Table 1. The 180 analyses for the samples were done at the University of Michigan Stable Isotopes Laboratory using the CO2 equilibration method (Roether, 1970) and mass spectrometry. The results of the analyses are shown in Table 2 and are reported as the relative ratio of 18O to 16O in the water sample expressed in parts per mil (960). This value,8‘8O, is defined by: 8‘80 = [(K - Rm...) / Rm...) x 1000 where R is the ratio of the heavy to light isotope (R = ‘80/“O), x is the unknown sample and SMOW (Standard Mean Ocean Water) is an international standard used for oxygen-18 analysis of water (Craig, 1961 a). Isotopic Hydrograph Separation Between storm events, stream flow reflects the isotopic composition mainly of 'old' water because mostly groundwater contributes to stream flow (Pearce et al. , 1986). During storm runoff events, however, the isotopic character of the stream may be altered by the addition of 'new' water from rainfall. The 'old' and 'new' water contributions can be calculated by solving the mass balance equations for the water and isotopic fluxes in the stream. These equations (Pearce et al., 1986) can be expressed as: Q, = Q, + Q. (1) 10 C,Q, = COO, + C,,Qn (2) These equations can be reduced to: Q. = [(C. - C.) / (Co - C.)] Q. (3) Q, = Q, - Q0 (4) where Q is discharge, C refers to tracer (isotOpe) concentration of a component, and the subscripts s, o, and n indicate stream, 'old' water component and 'new' water component, respectively. In order to test the validity of the above method of flow separation, each component of On and Q0 derived from tritium and oxygen-18 analyses of stream flow is compared to components of surface runoff and groundwater flow derived from hydrograph separation using graphical techniques. Graphical Hydrograph Separation Barnes' graphical method (Barnes, 1939) of hydrograph separation was used to separate the storm hydrograph of Sep. 24 ~ Oct. 9, 1993. The Red Cedar River discharge (Q) vs. time were plotted on a semi-log graph, the log axis for Q, (Figure 4). A straight line of the baseflow recession was extended from point 'C' back to point 'B' under point 'D', the inflection point. And then, a straight line 'A-B' was connected. The line 'A-B-C' represents baseflow during the storm event, likewise the area under the total hydrograph and above 'A-B-C' represents direct runoff during the event. Both the total hydrograph and the baseflow hydrograph are replotted on a arithmetic graph in Figure 5. According to graphical hydrograph separation, the mean discharge of the baseflow during the period of September 24 to October 9, 1993 was 230.52 cfs or 63 % of total runoff, and the mean discharge of the surface runoff was 133.08 cfs or 37 % of total runoff (Figure 4, Figure 5). 11 Inchs of precipitation Log scale 1133 i i “ 0.5 1 D l 0 Surface runoff, Q ‘= 133.08 cfs “" B Baseflow 0 go ’a? Q = 230.52 cfs 9-. o 8 .5 A C 152 i r i T I I T T i i r i T r 1 2 3 4 5 6 7 8 Oct. 24 25 26 27 28 29 30 Sep. Figure 4. Hydrograph separation using graphical method (semi-log scale). 12 Normal scale Inchs of precipitation 8°° L__] 4 0.5 700 .. ................................................................................................................................................ —< 1 500 J ....................................................................................................... 4 Surface runoff, Q = 133.08 cfs _ 0 500 - ............................................................................................................... GD _§ 1; a ‘30, 5 400 J ....................................................................................................................... Baseflow 100 T _ I i 1* 1 r T r m r r i I I 24 25 26 27 28 29 30 l 2 3 4 5 6 7 8 9 Sep. Oct. Figure 5. Hydrograph separation using graphical method (arithmetic scale). III. RESULTS Tritium The tritium data in Table 1 and Figure 6 indicate a high concentration of 19.45 TU in stream discharge before the storm of September 26 to September 30, 1993 and a low concentration of 14.94 TU during highest stream discharge associated with the storm event. The data do not show a smooth change in tritium concentration with time, however a smoothed curve for each data point can be approximated considering an error of 11 TU (Figure 6). The amount of tritium in precipitation during the storm period ranged from 14.99 TU to 16.70 TU (Figure 6). Oxygen The 8'80 data in Table 2 and Figure 7 indicate a high concentration of -8.43 %o in stream discharge before the storm of September 26 to September 30, 1993 and a low concentration of -11.20 960 during highest stream discharge associated with the storm event. The error bars of each data point were not constructed in Figure 7 because the amount of errors was too small to show in the figure. The amount of oxygen-18 in precipitation during the storm period ranged from -12.76 960 to - 12.87 %o. In Table 2 and Figure 7, sample number 18 and 19 of rain indicate -5.17 960 and ~10.69 960, respectively, but these values were excluded from further consideration because the standard deviations of these values were excessively high (i1.12 and i135, respectively) due to lack of sample volume. 13 Table l. Tritium results. SAMPLE CPM CE 6 EE DPM TU STD1 60.1 0.27 7.57 0.67 1.3623 191.67 BKG1 3.61 0.27 1 9 44 0.27 0.67 0.1361 19.45 2 9.1 0.27 0.67 0.1262 17.77 , 3 6. 01 0.27 0.67 0.1002 14.1 107.56) 4 6. 46 0.27 0.67 0.1155 16.27 5 6. 77 0.27 0.67 0.1202 16.93 6 9.17 0.27 0.67 0.1226 17.3 6 6.46 0.27 0.67 0.113 15.92 9 9 14 0.27 0.67 0.1221 17.2 10 9.1 0.27 0.67 0.121 17.04 STDZ 49. 02 0.26 6.65 0.66 1.3763 193.65 BKGZ 4.22 0.26 ' 11 6.06 0.26 0. 66 0.1124 15.64 12 7.47 0.26 0. 67 0.1119 15.77 13 7.79 0.26 0.67 0.1156 16.32 14 6 66 0.26 0.64 0.115 16.2 15 6 04 0.26 0. 66 0.1207 17.01 16 6. 05 0.26 0. 67 0.1248 17.56 rain 17 7. 69 0.26 0.66 0.1097 15.45 - rain 16 5. 69 0.26 0.65 0.0423 5.960506) rain19 6 23 0.26 0.65 0.1166 16.7 - rain 20 6.32 0.26 o. 66 0.0666 9.416499) STD1 56.43 0.2 16.66 0. 85 1.3526 190.54 BKG1 3.9 0.2 7. 6.1 0.2 0.65 0.1061 14.94 1 These numbers are from re- calculation considering the amount of dead distilled water (no tritium in it) added due to lack of sample volume. 2116s sample was re-counted. 15 Inchs of precipitation 16- 15‘- ! I I l l4 24252627262930i 2 3 4 5 Sep. Oct. I River 0 Precipitation Figure 6. Variation of tritium in the Red Cedar River. 16 Table 2. Oxygen-l 8 results. SAMPLE iDate run |8180,002|8180,C02|5180,H20[ std dev :onAs |17o corr [(SMOW) I(SMOW) | i #1 24-Oct; -1.579f 33.424? -6.47; 0.06 #2 24-Oct 4.533 33.464, -8.43= 0.06 #3 24-Oct 3.158: 36.299] 295:; 1.92 #4 24-Oct 4.6291 33.344 43.521 0.04 #5 24-Oct; 2.4361 32.509; -9.32;. 0.04 #6 24-Octl -2.415L 32.5311 -9.31 0.07 #7 24-Octi 4.316f 30.467] -11.2; 0.05 #6 24-Oct! 3905] 30.966 40.761 0.06 #15 24-Oct 2.1961 32.726 -9.06 0.02 #19 rain 24-Oct 4.6271 30.241 40.69 1.35 MDIW3 24-Oct 2.7541 32.16 -6.62 0.06 MDIW3 , 24-Oct. -2.691 § 32.245 376, 0.03 #9 1 25-Oct; 3724; 30.776 40.991 0.04 #10 e 25-Oct1 3905; 30.966 40.76 0.03 #11 25-Oct 2771 1 32.162 -9.66 0.04 #12 ’ 25-Oct 2751, 32.141 -9.66j 0.02 #13 25-Octn -2.598i 32.302 -9.53; 0.02 #14 25-Oct -2.426 32.517, -9.31 0.05 #16 25-Oct; -2.215 32.7031 -9.125 0.03 #17 rain 25-Oct 5991 26.629 42.871 0.03 #16 rain 25-Oct 1.7565 36.649 -5.171 1.12 #20 rain 25-Oct1 -5.892i 26.931 42.73 0.01 MDIW3 25-Octi -1.609l 33.157 -6.71 0.1 MDIW3 25-Oct: 2.0931 32.664 4.96 0.03 l5— 17 Inchs of precipitation oooooooooooooooooooooooooooooooooooooooo _14i1 1 1 1 1 1 2425262726293012 Sep. 3456 Oct. U River 0 Precipitation Figure 7. Variation of oxygen-l 8 in the Red Cedar River. 16 Hydrograph Separation Using Tritium and Oxygen-18 The relationship between tritium and oxygen-18 of the stream water samples analyzed is presented in Figure 8 and shows a general linear relationship between both values. The correlation coefficient for this relationship is 0.5293 and is statistically significant at confidence level 95 %. There is one anomalous point (- 2.95 %0) which is discounted because it has a high standard deviation ofi1.92 due to lack of sample volume. Tritium The weighted average tritium value of the rain water for the storm event of September 26 to September 30, 1993 is 15.73 TU with a range from 14.99 TU to 16.70 TU (Fable 1). The tritium concentration in prestorm water (groundwater) mixing with, or displaced by, the rainfall, however, is unknown since tritium data is generally not available for wells in the basin. Also, because mixing of groundwater has not been extensive since the introduction of "bomb" derived tritium, individual samples of groundwater from wells would not be representative for the entire basin. Information about tritium in prestorm water (groundwater), however, can be obtained from baseflow in the Red Cedar River on October 8 and October 9, 1993 and shows a range from 17.01 TU to 17.58 TU (Table 1 and Figure 6) and a weighted average of 17.30 TU which presumably represents an average of the water entering tributary channels from adjacent saturated zones. When the weighted average tritium content of rainfall is taken as Tn = 15 .73 TU, displaced prestorm water (groundwater) as To = 17.30 TU, and the tritium content for storm runoff (Ts) as in Table 1, application of equations (1) ~ (4) show that 69 % of the storm runoff during the period of September 24 to October 9, 1993 was prestorm water or groundwater and that 31 % of the storm runoff was 19 surface runoff (Table 3, Table 6, and Figure 9). The subscripts n, o, and s indicate the 'new' water component, 'old' water component, and the stream, respectively. Oxygen-18 The weighted average of the 8‘80 values for the rain samples from the storm event of September 26 to September 30, 1993 is -12.85 %o with a range from - 12.76 960 to -12.87 960 (Table 2), whereas that for baseflow is -9.09 %o with a range from -9.08 960 to -9.12 960. It is interesting to note that the baseflow values tend to match the 6180 values of groundwater samples collected from wells by the U. S. Geological Survey in the basin (Table 4). When the weighted average 8180 value of rainfall is taken as 8‘80ll = -12.85 960 , displaced prestorm water (groundwater) asfimOo = -9.09 %o, and theS‘SO values for storm runoff (880,) as in Table 2, application of equations (1) ~ (4) show that 79 % of the storm runoff during the period of September 24 to October 9, 1993 was prestorm water or groundwater and that 21 % of the storm runoff was surface runoff (Table 5, Table 6, and Figure 9). 20 -2 I .4 2.. .............................................................................................................................................. .6 J- .......................................................................................................... 6"O(%o) «- -3 4 -10 4 -12 . 1 - 1 - 1 1 . 1 14 15 16 17 18 19 20 TU Figure 8. Relationship between tritium and oxygen-l 8. 21 Table 3. Storm hydrograph separation using tritium. SAMPLE TS - Tn QS __ TS -Tn , Qn = QS - QO i To-Tn‘ Qo_To-Tn)QS 1 12' 243 243 0 2 1; 208 208 0 3 0.78 296 230. 88 65.12 4 0. 65 463 300. 95 162. 05 5 0.59 682 402.38 279.62 6 0.59 739 436.01 302.99 7 ~-————- 614 355" 259 8 0.55 473 260.15 212. 85 9 0.56 387 216.72 170.28 10 0.64 323 206.72 116.28 11 0.66 269 177.54 91.46 12 0.77 223 171.71 51.29 13 0.86 192 165.12 26.88 14 0.95 172 163.4 8.6 15 1’A 161 161 0 16 1‘ 157 157 0 Total 350.12 240.97 109.15 1 This number is not from the calculation, it is fiom the hydrograph separation curve, because the data are not valid for the calculation. 2 These number are slightly greater or less than 1, but they are adjusted to l in order to get a smoothed hydrograph separation curve. Table 4. Oxygen-l 8 data of groundwater in the Red Cedar Basin (Dannemiller et al., 1990). Well number 5‘80 (960) l -8.35 2 -8.80 3 -9. 10 4 -9.70 5 -9.20 6 -9.30 7 -9. 10 8 -8.90 9 -9.80 10 -9.95 l l -8.60 12 -9. 10 13 -9.00 Table 5. Storm hydrograph separation using oxygen-l 8. 23 SAMPLE 6‘°Os - 6'°On Qs QC = 8‘°Os - 6‘°On Qn = 06 - 00 6°00 - 6‘°On 6'°Oo - 6'°On 1 1‘ 243 243 0 2 1‘ 208 208 0 3 -— 296 296' 0 4 0.95 463 439.85 23.15 5 0.86 682 586.52 95.48 6 0.81 739 598.6 140.4 7 0.51 614 313.14 300.86 8 0.54 473 255.42 217.58 9 0.64 387 247.68 139.32 10 0.7 323 226.1 96.9 1 1 0.74 269 199.06 69.94 12 0.82 223 182.86 40.14 13 0.9 192 172.8 19.2 14 0.96 172 165.12 6.88 15 1" 161 161 0 16 1’ 157 157 0 .1.._IQI§i¥ 2 225-29 71 .86 “a. " 1 This number is not {Tom the calculation, it is fi'orn the hydrograph separation curve, because the data are not valid for the calculation. 2 These number are slightly greater or less than I, but they are adjusted to l in order to get a smoothed hydrograph separation curve. 24 Table 6. Result of hydrograph separations using graphical method, tritium, and oxygen-l 8 . Method Surface runoff Baseflow Total runofl' Graphical 133.08 cfs (37 %) 230.52 cfs (63 %) 363.60 cfs (100 %) Tritium 109.15 cfs (31 %) 240.97 cfs (69 %) 350.12 cfs (100 %) Oxygen-18 71.66 cfs (21 %) 276.26 cfs (79 %) 350.12 cfs (100 %) 2S D s 700 .1 .............................. Hydromh maximum-18 1 Discharge of groundwater = 278.26 cfs (79 °/o of total runofi) -. Hydrograph separation using tritium Discharge of grotmdwater - 240.97 cfs (69 % of total runoff) 6 5°“ 1 g A Hydrograph separation using graphical method 8 «3 Discharge of gromdwater =- 23052 cfs (63 % of total runofi) a 400 / .. K """""" 200 \r ‘1 100 I I 1 i 1 fl 1 r m I i f 1 j . 24252627282930123456789 Sep. ‘ Oct. Figure 9. Hydrograph separations using tritium, oxygen-l 8, and graphical method. IV. DISCUSSION The weighted average tritium and oxygen-18 concentration of baseflow and precipitation used to separate flow components include synthetic errors. For example, the tritium data has an error of :1:1 TU, and the oxygen-18 data has an error ranging from $0.01 %o to:t0.08 960 (Table 2). Also, the stream discharge and precipitation data each have errors of approximately 5 %. These errors are significant and clearly affect the analysis of flow components associated with the storm of September 26 to September 30,1993. The following equation of Lapin (1993) is used to calculate synthetic errors of the weighted average values. a = 1 - {(1 - otr)(1 - 09(1 - 06)"? where 01 indicates probabilities of synthetic errors, and 01,, 017, and 01, indicate errors probabilities of each data. The results show that the synthetic error for the weighted average tritium of baseflow (17.30 TU) and precipitation (15 .73 TU) is :1: 1.41 TU and i138 TU respectively, and that the synthetic error for the weighted average oxygen-18 of baseflow (-9.09 %o) and precipitation (-12.85 960) ist0.26 %o and 1:035 960 respectively (Figure 10 and Figure 11). Variation of tritium in the Red Cedar River together with the calculated synthetic error for the period September 24 to October 9, 1993 is shown in Figure 10. The data indicates an overlapping area between the weighted average tritium of baseflow with the synthetic error of 21:1.41 TU and the weighted average tritium of precipitation with the synthetic error of 1:138 TU. Clearly, the difference between the concentration of tritium in baseflow and precipitation during the period of 27 September 24 to October 9, 1993 is not enough to realistically separate flow components. Figure 11 likewise shows the variation of oxygen-18 in the Red Cedar River together with the calculated synthetic error. In this case, however, the difference between the concentration of oxygen-18 in baseflow and precipitation is enough to separate flow components. Inchs of precipitation 20 0.5 19 .1 18 2425262726293012345678‘9 Sep. 091- - River Weighted average ofbaseflow (17.30 TUilAl) O Precipitation Weighted average of precipitation (15.73 TU $1.38) Figure 10. Variation of tritium in the Red Cedar River. 29 Inchs of precipitation 2 L_l " - 0.5 -4 . ---------------------------------------------------------- _ 1 O ,6 .. ................................................. 8‘30 (960) .. -r V///////////X% 4H+ + 02 + 4e' Cathode : 2H" + 2e' --> H2 (Hoffman and Stewart, 1966) where H may be protium (I-I‘), deuterium (112), or tritium (H3). 33 During each electrolysis run, ten of the electrolysis cells were used for unknown samples, the remaining two cells however contained standards: one a known tritium standard and the other a blank. The blank is 'dead' or 'pre-bomb' distilled water which serves as a measure of background radiation. One g of sodium peroxide (N 202, electrolyte) was added to the cells. The cells were then connected electrically in series to the power supply and placed in the freezer at -9°C. A current of 5.8 A was maintained for approximately 1 10 hours or when a residual volume of 12-16 ml remained in each cell. The reasons for maintaining the freezer at cold temperature are: a) Evaporation loss is reduced. Evaporation can lower the recovery yield of the tritium during electrolysis (Ostlund and Werner, 1962). b) Warm N202 electrolyte can attack the anodes (Taylor, 1981). C) The cold temperature increases the separation efficiency of the hydrogen isotopes (Hoffman and Stewart, 1966). Post-Distillation In order to separate the electrolyte from the electrolyzed sample post- distillation was carried out after electrolysis was completed. The system for post- distillation consists of glass test tubes capped by modified distillation heads and encased in heating coils. All electrolyzed samples between 12-16 ml were transferred to the glass test tubes. Heat then was applied to the test tubes until dryness, which takes approximately 2-3 hours. Eleven ml of distillate was then pipetted into a plastic vial for liquid scintillation analysis. 34 APPENDIX B Tritium Activity Calculations The Packard Tri-Carb 1050 LS counter was used for scintillation counting. Prior to counting, 14 ml of cocktail was added to 11 ml of sample. The tritium activities of unknown samples were determined using the following equation (Wyerman 1975): DPM (s) / ml = [CPM (s) 1 ml - CPM (0) 1 ml] [{V. (S) / Vr (8)} "‘ CE * 13E(8)] where V0 (8) initial volume (ml) of sample Vf (s) final volume (ml) of sample CPM (s) / ml counts per minute of sample CPM (b) / ml counts per minute of background CE counting efficiency EB electrolytic efficiency of sample These values must also be known for the electrolytic standard (es). Conversion of DPM to tritium units is based on the relationship that 1 TU = 0.0071 DPM / ml. 1 TU = 1 tritium atom per 1018 hydrogen atoms (Faure, 1986). The counting efficiency of a standard is defined as: CE= |CPMZml - CPM(b)[ml| [DPMo/ml] where 35 DPM initial activity of standard Next the electrolytic efficiency of the standard, EE (es), is calculated from the 0 equation: EE (es) = |CPM (es) [ ml - CPM (b)[ m1] [{Vo (es) / Vf (es)} * CE * DPMo (es) / ml] This is used to obtain a fractionation factor B which relates the T / H ratio to that of the evolved hydrogen (Taylor, 1981). The equation for beta is: B = -ln |V (es) [ V,_(_e_s)1 ln EE (es) B is then substituted into the following equation to determine the electrolytic efficiency of the samples, EE (3). 1313(8) = W. (s) / V. (91‘ ’“ The final activity of the unknown samples is then determined: DPM (s)/ml = |CPM (s) 1 ml - CPM (b) [ ml| [{V0 (8) / Vr (8)} * CE * EE (8)] REFERENCES 37 REFERENCES Attanayake, P. M. (1983). A Field Investigation of the Groundwater Ridging Hypothesis, Unpublished MASc thesis, University of Windsor. Barnes, B. S. (1939). The structure of discharge-recession curves. Transactions, American Geophysical Union, 721-25. Barnes, B. S. (1940). Discussion of analysis of runoff characteristics. Trans. ASCE, 105, 106. Biggin, D. S. (1971). The use of natural tritium in hydrograph analysis. Subsurface Hydrology Rep. No. 26, Institute of Hydrology (UK). Brinkrnann, R., Eichler, R., Ehhalt, D., and Munnich, K. O. (1963). Uber den deuterium-gehalt von niederschlags- und grundwasser. Naturvissenschafren, 19. 611-12. Chen, M. (1993). Tritium in Precipitation from Lower Michigan, M. S. thesis, Michigan State University. Craig, H. (1961 a). Standard for reporting concentrations of deuterium and oxygen-18 in natural waters. Science, 133, 1833. Craig, H. (1961 b). Isotopic variations in meteoric waters. Science,l33, 1702-3. Crouzet, E., Hubert, P., Olive, P., Siwertz, E., and Marce, A. (1970). Le tritium dans les mesures d'hydrologie (16 surface. Determination expermentale du coefficient de missellement. J. Hydrol., 11,217-29. Dannemiller, G. T. and Baltusis, Jr., M. A. (1990). Physical and Chemical Data for Ground Water in the Michigan Basin, U. S. Geological Survey Open-File Report 90-368. Domenico, P. A. and Schwartz, F. W. (1990). Physical and Chemical Hydrogeology, John Wiley & Sons, Inc., New York. 36 Eriksson, E. (1967). Isotopes in hydrometeorology. In Isotopes in Hydrology, Vienna, IAEA, 21-33. Faure, G. (1986). Principles of Isotope Geology, 2nd edn, John Wiley & Sons, Inc., New York. Faure, G. (1991). Principles and Applications of Inorganic Geochemistry, Macmillan Publishing Company, New York. Fetter, C. W. (1988). Applied Hydrogeology, 2nd edn, Macmillan Publishing Company, New York. Fontes, J. Ch. (1980). Environmental isotopes in groundwater hydrology. In Fritz, P. and Fontes, J. Ch. (eds), Handbook of Environmental Isotope Geochemistry, Volume 1 The Terrestrial Environment, A., Elsevier, Amsterdam, 75-140. Friedman, I. (1953). Deuterium content of natural waters and other substances. Geochim. Cosmochim. Acta., 4, 89. Fritz, P., Cherry, J. A., Weyer, K. U., and Sklash, M. G. (1976). Runoff analyses using environmental isotopes and major ions. In Interpretation of Environmental Isotope and Hydrochemical Data in Groundwater Hydrology, IAEA, Vienna, 111-30. Fritz, P. and Fontes, J. Ch. (1980). Introduction. In Fritz, P. and Fontes, J. Ch. (eds), Handbook of Environmental Isotope Geochemistry, Volume 1 The Terrestrial Environment, A., Elsevier, Amsterdam, 1-19. Gat, J. R. (1980). The isotopes of hydrogen and oxygen in precipitation. In Fritz, P. and Fontes, J. Ch. (eds), Handbook of Environmental Isotope Geochemistry, Volume 1 The Terrestrial Environment, A., Elsevier, Amsterdam, 2147. Gat, J. R. and Gonfiantini, R. (eds., 1981). Stable Isotope Hydrology, IAEA, Vienna, Technical Reports Series No. 210. Gat, J. R. and Tzur, Y. (1967). Modification of the isotopic composition of rainwater by processes which occur before groundwater recharge. In Isotopes in Hydrology, Vienna, 49-60. Gillham, R. W. (1984). The effect of the capillary fringe on water table response. J. Hydrol., 67, 307-24. 39 Hoffman, C. and Stewart, G. (1966). Quantitative determination of tritium in natural waters, U. S. G. S. Water Supply Paper, 1696-D, D1-18. Hoefs, J. (1987). Stable Isotope Geochemistry, 2nd edn, Springer-Verlag, New York. Humphrys, C. R. and Bradford, W. F. (1958). Stratigraphic Study of the Red Cedar Watershed, Water Resource Survey, Red Cedar Watershed, Department of Resource Development, Agricultural Experiment Station, Michigan State University, East Lansing. IAEA (1969-71, 1973, 1975, 1979, 1990). Environmental Isotope Data. World Survey of Isotope Concentration in Precipitation (1953-1987), IAEA Tech. Rep. Ser., 96, 117, 129, 147, 165, 192, 311. Jacks, G., Olofsson, E., and Weme, G. (1986). An acid surge in a well-buffered stream. Ambio, 15, 282-5. J ansson, P. E. (1985). Flow paths and transit time distributions within small watersheds, model approaches and experimental evidence. In Johansson, I. (ed.), Hydrological and Hydrogeochemical Mechanisms and Model Approaches to the Acidification of Ecological Systems, Rep. 10, Nordic Hydrological Program, Oslo, 11-16. Kennedy, V. C., Kendall, C., Zellweger, G. W., Wyerrnan, T. A., and Avanzino, R. J. (1986). Determination of the components of stormflow using water chemistry and environmental isotopes, Mattole River Basin, California. J. Hydrol., 84, 107-40. Lapin, L. L. (1993) Statistics for Modern Business Decision, 6th edn, The Dryden Press, Orlando. Maulé, C. P. and Stein, J. (1990). Hydrologic flow path definition and partitioning of spring meltwater. Water Resour. Res., 26, 2959-70. McDonnell, J. J. (1990). A rationale for old water discharge through macropores in a steep, humid catchment. Water Resour. Res., 26, 2821-32. McDonnell, J. J., Bonell, M., Stewart, M. K., and Pearce, A. J. (1990). Deuterium variations in storm rainfall : Implications for stream hydrograph separation. Water Resour. Res., 26, 455-8. 40 McDonnel, J. J ., Stewart, M. K., and Owens, 1. F. (1991). Effect of catchment- scale subsurface mixing on stream isotopic response. Water Resour. Res., 27, 3065-73. Nakamura, R. (1971). Runoff analyses by electrical conductance of water. J. Hydrol., 14, 197-212. Novakowski, K. W. and Gillham, R. W. (1988). Field investigations of the nature of water-table response to precipitation in shallow water-table environments. J. Hydrol., 97, 23-32. O'Brien, A. L. (1980). The role of groundwater in stream discharge from two small wetland controlled basins in Eastern Massachusetts. Ground Water, 18, 359- 65. Ostlund, H. and Werner, E. (1961). The electrolytic enrichment of tritium and deuterium for natural tritium measurements. In Tritium in the Physical and Biological Sciences, IAEA, 1, 95-102. Packard Instrument Co. (1987). Chemicals, Supplies, and Instrumentation Catalog. Packard Instrument Co. (1990). T ri-Carb Liquid Scintillation Analyzer Operation Manual. Pearce, A. J ., Stewart, M. K., and Sklash, M. G. (1986). Storm runoff generation in humid headwater catchments, 1: Where does the water come from? Water Resour. Res., 22, 1263-72. Pilgrim, D. H., Huff, D. D. and Steele, T, D. (1979). Use of specific conductance and contact time relationships for separating flow components in storm runoff. Water Resour. Res., 15, 329-39. Rodhe, A. ( 1984). Groundwater contribution to streamflow in Swedish forested till soil as estimated by oxygen-18. In Isotope Hydrology 1983. IAEA, Vienna, 56-66. Rodhe, A. (1987). The Origin of Streamwater Traced by Oxygen-18, Uppsala Univ., Dept. Phys. Geogr. Div. Hydrol., Report Series A, No.41. Roether, W. (1970). Water CO2 exchange set-up for the routine 18Oxygen assay of natural waters, Int. J. Appl. Radiat. Isot. 21, 379. Rutledge, A. T. (1991). Methods of using streamflow records for estimating total 41 and effective recharge in the Appalachian Valley and Ridge, Pledmont, and Blue Ridge physiographic provinces. In Hotchkiss, W. R. and Johnson, A. I. (eds.), Regional Aquifer Systems of the United States, Aquifers of the Southeastern Area: 27th Annual Conference of American Water Resources Association, New Orleans, La., AWRA Monograph Series No. xxxx, p. xxx- xxx. Sklash, M. G. (1990). Environmental isotope studies of storm and snowmelt runoff generation. In Anderson, M. G. and Burt, T. P. (eds.), Process Studies in Hillslope Hydrology, John Wiley & Sons Ltd, New York, 401-35. Sklash, M. G. and Farvolden, R. N. (1979). The role of groundwater in storm runoff. J. Hydrol., 43, 45-65. Sklash, M. G. and Farvolden, R. N. (1980). Groundwater response during runoff events. In Proceedings of the Symposium on Watershed Management, American Society of Civil Engineers, Boise, Idaho, 743-54. Sklash, M. 6., Stewart, M. K, and Pearce, A. J. (1986). Storm runoff generation in humid headwater catchments, 2: A case study of hillslope and low-order stream response. Water Resour. Res., 22,1273-82. Sklash, M. G. and Wilson, B. A. (1982). An investigation of the groundwater ridging hypothesis for storm runoff generation. In Hydrological Processes of Forested Watersheds, Proceedings of the Canadian Hydrology Symposium '82, National Research Council of Canada, Fredericton, New Brunswick, 5 75 - 96. Stewart, M. K. and McDonnell, J. J. (1991). Modeling base flow soil water residence times from deuterium concentrations. Water Resour. Res. , 27,2681- 93. , Taylor, C. (1981). Present status and trends in electrolytic enrichment of low-level tritium in water. In Methods of Low Level Counting and Spectrometry, IAEA, Vienna, 303-23. Turner, J. V. and Macpherson, D. K (1990). Mechanims affecting streamflow and stream water quality : An approach via stable isotope, hydrogeochemical, and time series analysis. Water Resour. Res., 26, 3005-19. Vanlier, K E. and Wheeler, M. L. (1968). Analog Simulation of Groundwater Development of the Saginaw Formation Lansing Metropolitan area, 42 Michigan, U. S. Geological Survey in cooperation with Tri-County Planning Commission and Michigan Geological Survey. Wels, C., Taylor, C. H., Cornett, R. J ., and Lazerte, B. D. (1991). Streamflow generation in a headwater basin on the precambrian shield. Hydrological Processes, 5, 185-99. Wels, C., Cornett, R. J ., and Lazerte, B. D. (1991). Hydrograph separation : A comparison of geochemical and isotopic tracers. J. Hydrol.,122, 253-74. Wyerrnan, T. (1976). Laboratory Facility for the Analysis of Natural Levels of Tritium in Water, U. S. Geological Survey Open File Report. Yurtsever, Y. (1975). Worldwide survey of stable isotopes in precipitation. Rep. Sect. Isotope H ydrol., IAEA. "1111111111111