MICHIGA \ ll llllllll 3129 .‘I new LIBRARI ll \llllll ‘ llll 06“\350\8 This is to certify that the thesis entitled Radon-222 Distribution in the Aquifers of the Saginaw lowland presented by David W. Wills has been accepted towards fulfillment of the requirements for M.S. degree in Geological Sciences 9/4/41: / o M professor Date QM 5:, /.77/ / 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution F ‘1 LIBRARY Michigan State University L _,l _——i PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. t DATE DUE DATE DUE DATE DUE MSU Is An Affirmative Action/Equal Opportunity |nstttution c:\clrc\duodue. ems-p. 1 RADON-ZZZ DISTRIBUTION IN THE AQUIFERS OF THE SAGINAW LOWLAND by David W. Wills A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Geological Sciences 1991 ABSTRACT Three models are proposed to explain radon-222 distribution in groundwater aquifers of the Saginaw lowland. Random water samples obtained throughout the lowland indicate a high degree of variation and reveal no significant correlation with salinity, pH, groundwater temperature, depth of well, type of lithic source, or level of radium-226. Data does not support models for generation of radon-222 by radium-226 within the groundwater, or radon-222 derived from primary minerals as the source of radon-222 detected.in the groundwater of the Saginaw lowland. The high degree of areal variation supports the model of radon-222 generated by adsorbed and coprecipitated radium-226. ACKNOWLEDGEMENTS I would like to acknowledge the support of my family, who tolerated my educational pursuit, the influence of my father, who taught me perseverance, and the patience and persistence of my friend and advisor, Dr. Grahame Larson. Page ii Table of Contents LISTOFFIGURES........................................1 I. INTRODUCTION.. ...... ................... ......... .......2 II. OBJECTIVES........ ......... . ......... .. ..... ...........2 III.THEORY.. ..... .. ....... .................................3 3.1 General ............. . ..... ........................3 3.2 Models for Radon-222 Occurrence in Groundwater ...3 IV. LOCATION AND PHYSICAI.SETTING ........................12 4.1 General.........................................12 4.2 Geology.........................................12 4.3 Hydrology.......................................13 V. SAMPLINGSTRATEGYANDMETHODOLOGY....................15 Sampling Strategy and Methodology ...............15 Instrumentation.and Counting Procedure ..........18 Standards.......................................19 Determination of Radon Content . . . . . . . . . . . . . . . . . .20 SourcesofError................................22 ULTS..............................................22 Radon-222 Content of the Groundwater ....... .....22 Salinity.............................. ..... .....23 Temperature.....................................23 6.4 Radium-226Content..............................23 6.5 Alkalinityande...............................23 VII.DISCUSSION...........................................24 7.1 General.........................................24 7.2 Radon-222 Derived from Radium-226 in Solution ...26 7.3 Radon-222 Derived from Radium-226 in Primary Minerals and from Adsorption and Coprecipitated Minerals........................................26 VIII.SUMMARYANDCONCLUSION.......... ..... ................28 IX.BIBLIOGRAPHY....................... ...... .. ..... ......30' X.APPENDICES........................... ............... ..33 < H Mo 0 o o o UNHWU‘fiUNH mmmwmmmmm Appendix A: Table of Results Appendix B: Decay Spectrum of Radon-222 Appendix C: Liquid Scintillation Detection Efficiency Tests Page iii LIST OF FIGURES Figure 3.1 U-238 Decay Series......... .............. .........4 Figure 3.2 Radium-226 in solution decays to Radon-222........6 Figure 3.3 Radon-222 atoms ejected into intergranular space..8 . Figure 3.4 The recoil of Radon-222 atoms from secondary 5 mineraISOOOOO...OOOOOOOOOOOOO0.0.0.0000000000000011 Figure 5.1 Wells sampled in the Saginaw lowland.............16 Figure 7.1 Plot of Radon-222 vs. other parameters...........25 Page iv INTRODUCTION Page 1 I. Introduction Exposure to radon-222 has been determined by the National Academy of Science (1988) and the U. S. Environmental Protection Agency (1988) to be a health hazard primarily because when its short-lived daughters are inhaled as solid particles, they become lodged in the airway and continue to decay. This decay results in the release of alpha particles which damage lung tissue, causing lung cancer (Cothern and Smith, 1987). Groundwater is one of the primary pathways in which radon- 222 enters residences and.other structures, and.can contribute significantly to the level of radon-222 in a structure (EPA, 1986). It is therefore important to determine the level and source of radon-222 in groundwater. II. Obiectives The Saginaw lowland in southern Michigan (Figure 1.1) is an area of on-going research in hydrogeology and.geochemistry due to the anomalous levels of chlorides, nitrates, and other undesirable compounds detected in the groundwater (Long et a1, 1989: MDPH, 1986). To date, however, no significant survey has been conducted on levels of radon-222 in the groundwater. The objectives of this research are: 1) to determine radon- 222 levels in the groundwater within the shallow aquifers of Page 2 Page 3 the Saginaw lowland, 2) to evaluate the health risks to the lowland's community, and 3) to determine the source of radon- 222. III. Theory 3.1 General Radon-222 is a radioactive isotope in the U-238 decay series (Figure 3.1) and is the daughter product of radium-226. It has a half-life of 3.82 days and is highly soluble in water. Radon-222 decays to polonium-218 and a series of short-lived progeny through alpha and beta decay. The decay of radon-222 to polonium-218 involves the emission of an alpha particle containing two protons plus two neutrons from the radon-222 nucleus: Rn-222 --> Po-218 + He-4 The alpha particle, which is actually’a helium ion, is ejected at high speed (Friedlander et a1, 1955). It is alpha decay, chiefly from the short-lived radon-222 progeny, which is thought.to contribute:to the development of carcinoma in lungs and associated tissues through damage to cellular DNA (National Academy of Science, 1988; Cothern and Smith, 1987). 3.2 Models for Radon-222 Occurrence in Groundwater There are two possible sources for the origin of radon-222 in the groundwater within the shallow aquifers of the Saginaw lowland. Page 4 mwfluwm hwowo mMNID H.m wuamwm on. A||||\/>\.|| on mom erw er a... wmm 4: Cmmw e... La... wotom >mooo mmmj Page 5 Radon-222 Derived from Radium-226 in Solution Tanner (1964) describes conditions under which radium-226 can be leached from organic rich sediments high in uranium and thorium and in contact with brines, such as may exist in Michigan within the Antrim and Elsworth Shales (Long et a1, 1989). He states that positive ions, such as sodium, magnesium, and potassium associated with elevated chloride levels of brines found. at depth, combine with. reducing conditions in the shale to provide a situation which favors the displacement of radium-226. The positive ions also prevent the radium-226 in solution from subsequently adsorbing onto mineral surfaces due to competition with chloride ions (Tanner, 1964). This radium-226, while still in solution, may eventually diffuse upwards together with chloride ions and could be the source of the radon-222 detected in shallow aquifers of the Saginaw lowland. Page 6 GD .1 @ Figure 3.2 Radium-226 in solution decays to Radon-222 Page 7 Radon-222 Derived from Radium-226 in Primary Minerals As radium-226 decays and ejects an alpha particle, the resultant radon-222 atom can recoil in the opposite direction a distance of from 20 to 70 nm (Michel, 1987). If the original radium-226 atom is within a mineral and near a granular surface, then the radon-222 atom can be ejected into fluids within intergranular space (Figure 3.3). This process has been used to explain the occurrence of radon-222 in groundwater associated with granitic terraines, such as those found in Ontario or Utah (Lively and Morey, 1982; Tanner, 1964). It is also possible that radon-222 detected in the shallow aquifers of the Saginaw lowland originates from the same process 0 Page 8 Figure 3.3 Radon-222 atoms ejected into intergranular space Page 9 Radon-222 Derived from Adsorbed and Coprecipitated Minerals As previously mentioned, radium-226 can be leached from organic shales and can be associated with diffusing chloride ions. The solubility of both the radium-226 and chloride ions, however, would decrease upwards because of decreasing groundwater temperature towards the surface (Kirby and Salutsky, 1964) and may result in radium-226 adsorbing and coprecipitating onto primary and secondary mineral surfaces (Tanner, 1964). As previously mentioned, Tanner (1964) states that the presencezof positive ions associated with high.chloride levels in groundwater restricts the ability of radium-226 to adsorb onto mineral surfaces. However, he further states that since groundwater' near the surface is generally' a mixture of meteoric and subsurface waters, this results in decreased chloride levels and consequently reduces the competition for adsorption sites and also permits radium-226 to adsorb and coprecipitate onto primary and secondary mineral surfaces. The adsorbed and coprecipitated radium-226 resulting from a decrease in temperature and from lowered chloride levels subsequently decays, ejecting radon-222 into the groundwater within the intergranular space, similar to the process occurring within primary minerals. An alternate process for radium-226 enrichment on surfaces of primary minerals exists in the glacial deposits overlying Page 10 the Saginaw lowland. Prior to being deposited in Michigan, these glacial deposits were derived in part from the ‘weathering of granitic rocks in Canadan Michel (1987) describes a process whereby physical and chemical weathering in granitic rocks of uranium-rich feldspars form clays rich in uranium and progeny, including radium-226, which may intermix with primary rock formations or invade porous sediments or fractures and be deposited on rock surfaces. Eventually, these weathered sediments could be further eroded and transported by glaciers, and ultimately deposited in the Saginaw lowland. The radium-226 located in enriched grain surfaces in both the Saginaw Formation and the glacial drift can contribute significantly to the emanating power of a sediment as the resultant radon-222 atoms are more likely to be propelled into the intergranular pore space (Figure 3.4). Page 11 ENRICHED ACCRETIONARY LAYER Figure 3.4 The recoil of Radon—222 atoms from secondary minerals Page 12 IV. Location and Physical Setting 4.1 General The Saginaw lowland area of Michigan includes Bay, Saginaw, Midland, Gratiot, and portions of Gladwin, Tuscola, and Arenac counties (Figure 5.1), and is characterized by little relief. Surface elevations range from 176 meters along the shore of Saginaw Bay, to 282 meters in Gladwin County; the average elevation of the area is approximately 213 meters. The main rivers in the area are the Cass, Pine, Saginaw, and Tittibawassee. These drain generally eastward and discharge into Saginaw Bay. As of 1980, the population in the lowland was approximately 550,000, with approximately half, or 228,000, concentrated in Saginaw County. The vast majority of the land area is rural, with numerous small towns and villages, and one major metropolitan area: the Bay City, Midland, Saginaw tri-city area. The primary economic base is agriculture, although the cities of Midland and Saginaw are major industrial centers for the state. 4.2 Geology The lowland lies in the east central portion of the Michigan Basin and is underlain by the Saginaw Formation, which is Pennsylvanian in age and consists primarily of coal and limestone (Vugrinovich, 1984). The formation. ranges in Page 13 thickness from 0 meters in eastern portions of Arenac and Tuscola counties to approx1mately 163 meters in the western part of the lowland. In general, the formation decreases in thickness towards.the.east.andmdips slightly towards the*west. Overlying the Saginaw Formation is up to 30 meters of glacial drift (Rieck, 1980). The drift consists mainly of lacustrine clay and silt, till, and minor amounts of sand and gravel (Martin, 1955). Collectively, these sediments form a broad plain associated with glacial lakes Saginaw, Algonquin, and Stanley (Leverett and Taylor, 1915). Several morainic systems associated with the retreating Saginaw Ice Lobe cut.northwest to southeast across the lowland area. These include the Port Huron, Fort Worth, and Defiance moraines and are generally recognized by their gently rolling topography (Leverett and Taylor, 1915). They consist mainly of clay-rich till and minor amounts of sand and silt. Also occurring along the shoreline of Saginaw Bay and in scattered pockets inland from the Bay are ridges and dunes of well- sorted, fine-to-medium sand. 4.3 Hydrology The primary source of fresh water in the lewland is from bedrock aquifers, drift aquifers, and surface waters - chiefly Saginaw Bay waters. Most rock wells draw water mainly from sandstone units of the Saginaw Formation, although a feW”wells in the extreme northeastern and southeastern parts of the Page 14 sandstone unitslof the Saginaw Formation, although a feWHwells in the extreme northeastern and southeastern parts of the lowland draw from the Bayport Limestone or the Michigan Formation. In general the number of bedrock wells increases towards the west, where some are as much as 152 meters deep. Water from the Saginaw Formation does not exceed maximum contaminant levels (MCL's) for nitrate or fluoride, but some samples exceed secondary MCL's for chloride, iron, sulfate, total dissolved solids, and specific conductance (Long, et a1., 1985). Since clay-rich lake bed deposits are the dominant drift material in the lowland, wells are generally set to tap isolated deposits of sand and gravel beneath the surface. These deposits are usually thin and discontinuous, and recharge capacity is very limited. Approximately 42 percent of the region has greater than 10 percent bedrock wells as opposed to drift wells. This is primarily in the eastern portion of the lowland where the drift material is less than 30 meters thick. The percentage of bedrock wells decreases towards the west as the drift thickens; few bedrock wells occur in the western third of the lowland, which is underlain by 60 meters to 180 meters of drift. Water from drift aquifers is generally similar to that from the Saginaw Aquifer; nitrate and fluoride is below the primary MCL, but some wells contain water exceeding the secondary MCL Page 15 V. Sampling Strategy and Methodology 5.1 Sampling Strategy and Methodology A total of 56 wells were sampled for radon-222 within the Saginaw lowland (Figure 5.1). About half of these wells were from, Bay County, the rest were from the remaining six counties. Page 16 Tus Sagina Bay a, << 2f .0 3%.‘3. <0 o.°° E I n (I) 1180 0 V Q60? 00 o 588 ”m 990 Q C 02%) Fi ure 5.1 Wells sam led with the Sa inaw lowland Page 17 The following procedure (EPA, 1978) was used for collection of radon-222 samples: 1. A hose was attached to a household faucet after insuring that the water did not go through a softener system. The faucet was turned on full and allowed to run approximately 20 minutes to obtain a fresh sample from the well. The stabilization of temperature was used as an indication the sample was directly from the well shaft. 2. The flow of water was reduced to minimize turbulence during sample collection. 3. The end of the hose was placed in a funnel which filled with water, immersing the hose end. 4. The tip of a hypodermic needle was placed approximately 3 cm below the surface of the water, and 15 ml of water was slowly drawn into the syringe. This water was then ejected. This procedure was repeated twice to rinse the syringe. 5. Approximately 15 ml of water was again slowly drawn into the syringe. The syringe was inverted and any air bubbles and extra water was slowly ejected to leave exactly 10 ml. 6. The cap was removed from a scintillation vial containing a premeasured amount (10 ml) of N E N Products mineral oil-based high-efficiency scintillation solution. The Page 18 tip of the needle was placed near the bottom of the scintillation solution, and the water was slowly ejected into the solution. 7. The needle was slowly withdrawn and the cap of the vial was tightly replaced. 8. Each vial was identified by marking it with the time and date of collection, sample number, and field identification (site ID) number. Additional records were kept of the site location, well type and depth, water temperature, and conductivity. 9. Steps 4 through 7 were repeated to obtain an additional separate sample from the same source. 5.2 Instrumentation and Counting Procedure The radon-222 activity, as determined by disintegrations per minute, was ‘measured. with. a ZBeckman 8100 series Liquid Scintillation Detector; The following procedure (Gray, 1980) was used to determine the radon-222 activity: 1. Due to the short half-life of radon-222, all samples were measured for radon activity within 72 hours of the time of collection. Most of the samples were measured within 24 hours of collection. 2. The samples were allowed to equilibrate to room temperature for a minimum of three hours prior to counting. 3. The sample vials were shaken to insure equilibration, wiped with a clean damp cloth to insure the glass sides were clean, and placed in the Liquid Scintillation Detector in the Page 19 following order: background vials, standards, and samples. 4. The activity in each vial was measured for a period of twenty minutes. This step was repeated once. 5. At the end of the counting procedure, the time from the beginning of the count until each individual vial had been measured was added to the overall time since collection. The results of the two counts for each vial were averaged to give the overall activity, and.the ratios for the two channels were compared to insure they were similar to the ratios determined by the standards. The efficiency of the above procedure, as determined from the known radon-222 standards, is 97% (Appendix C). 5.3 Standard§ The following radon-222 standards, obtained from the Eastern Region U. S. Environmental Protection Agency Laboratory in Montgomery, Alabama, were used to calibrate the Liquid Scintillation Detector to establish efficiency and to determine the CPM/pCi conversion factor: 1650 pCi/l 2100 pCi/l 2400 pCi/l 4800 pCi/l 7500 pCi/l The 4800 pCi/l standard was also used to generate an energy spectrum for radon-222 decay by initially counting the standard with a "wide open" window (0-1000) on the detector. Page 20 The window'was then progressively closed from.the top limit of 1000 in intervals of 50 to generate the decay spectrum illustrated in Appendix B. Standards for C-14 and tritium were also used to generate decay spectrums in the same manner for their respective isotopes (Appendix B). During sample counting, one channel was devoted to measuring the activity above the C-14 and tritium decay spectrum limits, which includes the majority of the radon-222 decay spectrum. The activity of this channel was then compared to the wide-open channel to insure the resultant ratio was consistent with the ratio obtained for the same ranges for the radon-222 standards. This insured the activity being measured was radon-222 activity. 5.4 Determination of Radon Content CPM/pCi Conversion Factor Formula 5.1 (EPA, 1978) was used to convert CPM's to pCi/l. B=Sb-Rb/A 5.1 Where B= CPM/pCi conversion factor A= Activity of Standard (pCi) Sb= Count Rate of Standard (CPM) Rb= Background Count Rate (CPM) Calculation of Minimum Detectable Activity The minimum detectable activity, as determined for each water sample, is defined as the lowest ascertainable activity level for radon-222 (Gray, 1985): Page 21 AI = (1/13 x K1 Rb /T) x 100 5.2 Where AI = Minimum Detectable Activity in pCi/l B = Mean CPM/pCi conversion factor derived from five standards K1 = 1.65 Rb = Background count rate (CPM) E! II Count time in hours AI is the ability of the detector, within the limits of its efficiency and counting time, to reliably measure an activity above background. Calculation of Radon-222 Activity Formula 5.3 (EPA, 1978) was used to determine the radon-222 activity. pCi/l = (net CPM/B/decay) X 1000ml/liter/10 ml 5.3 Where net CPM gross CPM - background CPM B = CPM/pCi conversation factor decay = exp (-7.56 E3T ) T = time lapse from collection to counting in hours The two sigma counting error as given in percent is (EPA, 1978): §b + Eb T net CPM Where Sb:= Gross CPM Rb = Background CPM T Count time in hours Page 22 5.5 Sources of error To determine if there was a variation in radon-222 concentration in groundwater over time, five sites were selected.at.random, then sampled and analyzed three times from May, 1987 to August, 1987. The mean variation of the radon- 222 activity was 11.4%. A precaution for the possibility of errors in the sampling process included the repeated sampling idescribed above, and.the obtaining of two separate samples for each site. Errors due to machine counting were minimized by counting each sample twice, and by repeated counting of the standards to determine the amount of variation. Additional measurements included temperature and salinity, which were tested with a conductivity/temperature meter. Radiumr226 levels were also determined by MDPH for 26 wells in Bay County by measuring gross alpha radiation using an alpha track detector. Other parameters recorded included well type, drift or bedrock, and well depth which were obtained from well logs. The depth of the wells sampled ranged from 17 meters below ground level, to 171 meters. The mean depth was 57 meters, and the standard deviation was 35 meters. Of the wells sampled, 21 were drift wells, and.35 were bedrock wells. VI. Results 6.1 Radon-222 Content of the Groundwater The radon-222 concentration in the samples of groundwater obtained from the study area ranged from a low of 4.6 pCi/l, which is below the significance level for the analytical Page 23 method used (Gray, 1980), to a high of 562.3 pCi/l. The mean was 174.2 pCi/l, and the standard deviation was + or - 121.6. See appendix A for a complete list of values. 6.2 Salinity The salinity, or total dissolved salts, of the groundwater ranged from a low of 250 umhos to a high of .03 mhos. The mean was 3215 umhos with a standard deviation of + or - 4490 umhos. 6.3 Temperature of the groundwater The temperature of the groundwater samples obtained at the time of sampling was measured with an electronic thermometer previously calibrated with a laboratory thermometer. The resultant values ranged from 9.1 degrees to 15.7 degrees Centigrade. The mean temperature was 12.5 with a standard deviation of + or - 1.97 degrees. 6.4 Radium-226 Content Samples were obtained from Bay County approximately three to six weeks prior to radon-222 sampling. Samples were drawn directly from taps with unsoftened water, and poured into clean sample bottles“ These samples were analyzed for combined. levels of both radium-226 and radium-228, with radium-226 results ranging from 0.3iji/l to 108.7 pCi/la The mean was 12.96 with a standard deviation of + or - 4.10 pCi/l. 6.5 Alkalinity and pH The alkalinity of the groundwater samples ranged from 64 to 1250, with a mean of 373 and a standard deviation of 257.5. Page 24 The pH of the samples ranged from 6.8 to 8.5, with a mean of 7.6 and a standard deviation of .31. VII Discussion 7.1 General The levels of radon-222 measured in the groundwater samples from the Saginaw lowland were within anticipated values for groundwater from formations consisting of sandstones, gray shales, and carbonates that normally contain relatively low levels of uranium (Cothern and Smith, 1987), and are well below the 10,000 pCi/l proposed by the EPA as constituting a high-danger level. There was no discernable pattern to the areal distribution of the radon-222 in the groundwater (Figure _ 5.1). In fact, some of the lowest values recorded were from wells located within 4 or 6 km of those wells yielding relatively high values. The low levels measured should preclude the groundwater as a source for concern over radon hazards in normally constructed residences and other structures. With respect to the vertical distribution of radon-222, a plot of radon-222 vs. depth (Figure 7.1) shows little relationship. In fact, the correlation coefficient between the two parameters is 0.215. In addition, the data show that the highest and lowest concentration of radon-222 both occur at relatively shallow depth (<70m). Comparison of the radon-222 concentration recorded from wells open to the drift vs. those open to the Saginaw Radon 222 (pC/I) Page 25 LAIIILAI ILLIIII Rod§on 222 (pC/l) «HIONNQ 8 . . u 8 00:30:53. Ac330mv «No.3 0 . . » $8: 063: A3 «"0... Am §§§ Rodaon 222 (pC/l) 8 .nw. .me. . . .41.... . . .1813 man3 Nwm Abobv wfllohumm § Rodgon 222 (pC/l) o a q 1 all i. o .4 a 4 a 4 . m S S I 3 8 no 33.00338 30033 3 other arameters Plot of Radon-222 vs. 7.1 Fi ure Page 26 Formation also show no discernable relationship. 7.2 Radon-222 Derived from Radium-226 in Solution Generation of radon-222 by radium-226 within the groundwater should result in a high correlation between measured radon-222 and radium-226 values. A plot of the radon-222 and radium-226 for ‘wells in Bay County (Figure 7.1) however, yield a correlation coefficient of only 0.146. In fact, many of the lowest values for radon-222 obtained from Bay County also yield some of the highest values so far obtained for radium- 226~ These observations would suggest that radium-226 in solution is not a major source for the radon-222. This would be consistent with the findings of Lively and Morey (1982) for radon-222 in groundwater from east-central Minnesota and with the findings of Dyck (1980) from northeast Ontario. Furthermore, Michel (1987) and Tanner (1964) explain that radium—226 is primarily immobile and occurs in solution in only low concentration in the shallow (<50m) groundwater environment, and that the most likely source of radon-222 is radium-226 precipitated in the vicinity of the well, rather than radium-226 present in the groundwater. This process of radium-226 precipitation and immobility may also operate in the aquifers of the Saginaw lowland, and may explain the lack of correlation between radon-222 and radium-226. 7.3 Radon-222 Derived from Radium-226 in Primary Minerals and from Adsorption and Coprecipitated Minerals The results demonstrate that the radon-222 detected in the Page 27 groundwater of the aquifers of the Saginaw lowland is not derived from radium-226 in solution, therefore it must be originating from. either' the primary' minerals within ‘the Saginaw Formation and/or glacial drift, the secondary minerals, or both. The composition of the Saginaw Formation and glacial drift is known and includes the following primary minerals: quartz, calcite, dolomite, feldspars, "coal", clay' minerals, hornblende,l tremolite-actinolite, orthopyroxenes, clinopyroxenes, garnet, epidote, rutile, sphene, zircon, and tourmaline, as well as the secondary minerals Fe and Mn oxides, carbonates, hydroxides, and silicates (Vugrinovich, 1984, Dworkin, 1984). Of the above minerals; feldspars, "coal", zircon, clay minerals, Fe and Mn oxides, and silicates can contain significant quantities of uranium-238 and progeny, including radium-226 (Dyck, 1978, Tanner, 1964, Asikainen,. 1981, Lively and Morey, 1982). Whether the primary or secondary minerals containing radium- 226 are the source of the radon-222 in the groundwater can only be determined by direct measurement of the amount of radium-226 within the minerals. Such measurements are analytically very difficult and are beyond the scope of this research project. However, based on the extreme vertical and horizontal variability of the radon-222 measurements of the groundwater, it would appear that the probable source of radon-222 is from Page 28 secondary minerals. This is based on the likelyhood that, unlike primary minerals, secondary minerals are distributed within the Saginaw Formation and glacial drift non-uniformly, and would give rise to non-uniform values of radon-222 within the groundwater. The non-uniform distribution of secondary minerals is attributable to the high hydrologic variation within the Saginaw Formation and glacial drift; differences in permeability and flow rates leads to diversity in the availability of ions for adsorption and coprecipitation sites. Clay minerals, as compared to other minerals, also provide an effective adsorbent of radium-226 and other secondary minerals. VIII. Summary and Conclusion The data shows that radon-222 levels in the groundwater aquifers vary considerably, even over short distances, and that high and low values often occur in wells of close proximity and at equivalent depths. There is also a marked lack of correlation with any of the other parameters tested, i.e. well depth, conductivity, alkalinity, pH, and temperature. In particular, there is no correlation between radon-222 and radium-226 in solution which contradicts expectations if the model for radon-222 derived from radium- 226 in solution is used. Due to the general uniformity of distribution of primary minerals, as contrasted to the extreme variation of radon-222 values in the groundwater, there is no Page 29 discernable support for primary minerals as a major source of the radon-222 detected in the groundwater. The variation in permeability and flow rates within the Saginaw Formation and glacial drift leads to an expectation of variation in adsorbed and coprecipitated secondary minerals, including radium-226. This provides a plausible explanation of the variation in radon-222 levels within the groundwater, and identifies secondary minerals as the likely source of radon-222 in the groundwater. BIBLIOGRAPHY Page 30 BIBLIOGRAPHY Anderson, L., 1986. Partitioning of Trace Metals in Glacial Till in Southern Michigan. Masters Thesis. Asikainen, M., 1981. State of disequilibrium between U-238, U-234, Ra-226, and Rn-222 in groundwater from bedrock. Geochemica et Cosmochimica, Vol. 45, p 201-206. American Association of Petroleum Geologists, 1954. Geological Cross-Section of Paleozoic Rocks: Central Mississippi to Northern Michigan. Cercone, K., 1984. Thermal History of the Michigan Basin. American Association of Petroleum Geologists Bulletin, Vol. 68, NO. 2, p 130 - 136. Cothern, C. and Smith, J., 1987. Environmental Radon. Plenum Press. Dworkin, S., 1984. Late Wisconsinan Ice-Flow Reconstruction for the Central Great Lakes Region. Masters Thesis. Dyck, W., 1978. The Mobility and Concentration of Uranium and its Decay Products in Temperate Surficial Environments. in: A Short Course in Uranium Deposits: Their Mineralogy and Origin, University of Toronto Press. Dyck, W., 1980. Uranium, Radon, Helium, and Other Trace Elements and Gases in Well Waters of Parts of the St. Lawrence Lowlands, (Ottawa Region) Canada. Journal of Geochemical Exploration, V. 13, p 27 - 39. Environmental Protection Agency, 1978. Radon Sampling Procedures. EPA Manual EPA/EERF—MANUAL-78-1. Environmental Protection Agency, 1986. A Citizens Guide to Radon. Gray, D., 1985. Determination of Radon Gas Concentrations in Groundwater Using Liquid Scintillation Methodology. Masters Thesis. Page 31 Lane, A.C., 1899. Lower Michigan Mineral Waters. U.S.G.S. Water Supply Paper 31. Leverett and Taylor, 1915. The Pleistocene of Indiana and Michigan. U. S. Geological Survey Monograph #53. Lively, R.S. and Morey, G. B., 1982. Hydrogeochemical Distribution of Uranium and Radon in East-Central Minnesota. in: Isotope Studies of Hydrologic Processes. Long, D., Rezabek, D. H., Takacs, M. J., and Wilson, T., 1985. The Geochemistry of Groundwater of Bay CountyI Michigan. Report to Michigan Department of Public Health, MDPH: 0RD 38553, 1985. Makofske, W. and Edelstein, M., 1984. Radon and the Environment. Noyes Publications. Martin, 1955. Map of Surface Formations of the Southern Peninsula of Michigan. U. S. Geological Survey Publication 49. Michel, J., 1987. In Environmental Radon. Plenum Press. Michigan Department of Public Health, Division of Groundwater Quality, Open File Material, 1983. National Academy of Science, 1988. Health Risks of Radon and Other Internally Deposited Alpha-Emitters - BEIR IV. by the Committee on the Biological Effects of Ionizing Radiation. National Academy Press. Rieck, 1980. Bedrock Topography of Michigan‘s Southern Peninsula. Macomb, W. Illinois University. Unpublished Manuscript Map. Tanner, A., 1964. Physical and Chemical Controls on Distribution of Radium-226 and Radon-222 in Ground Water near Great Salt Lake, Utah. in: Natural Radiation Environment. Tanner, A., 1980. Radon Migration in the Ground: a Supplementary Review. in: Natural Radiation Environment. Page 32 Taylor, K. and Faure, G., 1981. Rb-Sr Dating of Detrital Feldspar: A New Method of Studying Till. In Journal of Geology, Vol. 89, p 97. Vugrinovich, R., 1984. Lithostratigraphy and Depositional Environments of the Pennsylvanian Rocks and the Bayport Formation of the Michigan Basin. Michigan Department of Natural Resources, Report of Investigation 27. APPENDICES Page 33 APPENDIx A ID TYPE DEPTH CORD TEMP RADON RADIUM Bayl 1 100 7200 13.7 33.0 —6._2 BayZ 2 260 30000 19.1 153.8 108.7 Bay3 2 92 4000 14.3 0 3.5 Bay4 2 116 3000 13.0 139.7 2.7 Bays 1 90 7800 13.4 110.6 4.1 Bay6 2 154 4000 12.7 165.6 2.4 Bay7 1 90 3000 14.8 11.4 19.6 Bay8 2 210 6000 13.3 263.7 13.2 Bay9 2 140 3000 14.4 8.4 15.8 Bale 2 192 4500 14.9 188.8 34.2 Bayll 1 111 2500 14.7 0 1.8 BaylZ 2 179 15000 14.4 83.0 38.9 « Bayl3 1 180 6600 15.4 106.3 16.1 Bayl4 1 190 650 12.5 133.3 14.0 BaylS 2 220 5500 13.9 0 15.8 Bayl6 2 202 3000 14.0 57.9 2.2 Bayl7 2 240 3500 13.2 237.1 14.0 Bay18 2 283 5600 13.4 125.5 5.5 Bayl9 1 147 1800 17.0 90.5 2.6 BayZO 2 380 2400 13.5 164.1 3.2 Bay2l 1 163 3000 12.3 111.5 3.6 Bay24 2 130 2500 17.5 83.0 17.0 Bay25 1 85 1300 13.2 163.1 0.8 Bay26 2 100 3300 12.4 27.1 15.7 Bay27 1 101 5000 13.1 84.3 2.1 Bay29 1 90 6000 12.0 94.5 2 5 Grtl 2 403 516 11.0 223.0 - 0:62 2 203 783 12.0 162.4 - 0:13 2 383 1911 11.1 71.0 - Grt4 1 114 463 10.6 167.2 - GrtS 2 565 1917 11.7 237.6 - Tusl 2 127 312 10.3 138.1 - Tus2 2 162 426 10.3 562.3 - Tus3 2 160 393 11.5 198.6 - Tus4 2 416 418 10.9 118.1 - TusS 1 68 275 10.9 165.4 - Gldl 1 79 250 9.1 314.0 - 0162 1 57 373 9.6 163.1 - 0163 1 95 433 10.7 211.1 - 0164 2 300 598 10.5 440.9 - 0165 2 202 820 11.0 223.2 - Arel 2 257 1713 11.6 333.5 - AreZ 2 133 2224 10.7 531.0 - Are2 2 300 1080 11.7 129.1 - Are4 2 432 1100 10.6 368.8 - AreS 1 90 493 10.3 245.9 - M161 1 80 383 12.1 124.3 - M162 2 470 2946 11.3 228.5 - M163 2 241 1128 10.8 346.6 - M164 1 70 4793 11.1 136.2 - M165 1 120 2160 11.8 273.3 - Sagl 2 144 2124 11.6 153.8 - Sagz 1 69 713 12.7 86.2 - 8ag3 2 135 2795 11.2 198.3 - Sag4 2 70 740 11.1 264.1 - SagS 2 226 5630 11.0 204.4 - Page 34 APPENDIX A Type: 1 = Drift well 2 = Bedrock well Depth: Measured in feet below surface Conductivity: Measured in micromhos Temperature: Measured in degrees Centigrade Radon: Measured in pC/l Radium: Measured in pC/l — Total Radium (Ra—226 + Ra-228) Page 35 .Flle >=>CU< «Na “.0 use ZDKFOmam >0¢m2m >