WWWMIMI \ 1 NM 1 ‘ A CORRELATION STUDY OF ENVIRONMENTAL RADEOACTIVETX’ W “19:13 for the Degree 0; M. S. MECHIGAN STATE UNZKZEESETY ‘1"hozmas Richard Colpetzer 1962 THESIS This is to certify that the thesis entitled smut" WW presented by “WWW has been accepted towards fulfillment of the requirements for LdWw tum fix; Z4474, J Major prdiessor Date 26 .377” ///Z’ 0-169 LIBRARY 1 Michigan State 31 University JWNHJMII.JIU A CORRELATION STUDY OF ENVIRONMENTAL RADIOACTIVITY By Thames Richard Colpetzor AN ABSTRAOT Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER 0! 8013303 Department of Civil and Sanitary Engineering 1962 Approved: THOMAS RICHARD COLPETZER ABSTRACT this thesis reports an investigation of the relation. ship between air partieulate radioactivity and rainfall ‘ radioactivity. The effects of the amount of rainfall upon rainfall activity levels were also studied. Statistical tests were made to determine whether any association or correlation exists between air particulate activity and rainfall activity. In addition, inferences were made about how rainfall activity levels vary with amount of rainfall. The analysis of data collected during periods of low activity levels indicated that rainfall activity in units of uc/nl tended to increase as the amount of rainfall increased. For the same period no relationship was found between rainfall and rain activity in units of pc/cme/day, and the correlation coefficient for these two variables was not significantly larger than zero. Rainfall activity in units of pc/ml and pc/cmz/day was found to be correlated with air particulate activity when the radioactivity levels were relatively high. No correlation was found between the two variables when activity levels were low. . It was found that when activity data were analyzed on.a basis of menthly averages rather than weekly averages, correlation was found between air particulate activity and total deposition of rainfall activity (Po/ena/day) when the activity levels were low. Statistical tests not used in this study were proposed for use in future investigations. A CORRELATION secs: or snvzaosnznrit asnzciorxvirr By Thomas Richard Oolpetser A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Madras or scissor Department of Civil and Sanitary Engineering 1962 ' ACKN OWLEDGEMEN T8 The author would like to express his sincere thanks to his major professor, Dr. Shosei Serata, Assistant Professor of Civil Engineering, Michigan State University, for his assistance in the study and for his checking of the manuscript. Gratitude is also expressed to the National Science Foundation for their support, to Dr. James Staplcton, Assistant Professor of Statistics, Michigan State University, for his advice, to Dr. Conrad P. Straub, Director of the Division of RadiolOgical Health, R. i. Taft Sanitary Engineering Center, for his very helpful comments and discussion, and to D.E. van Parcwe, Seed of the Division of Radiological Health, Michigan Department of Health, for his kind assistance. In addition the author wishes to thank the Michigan Department of Health and the U. 8. Public Health Service for the use of their data in the analysis. ll TABLE OF CONTENTS Page AOKNOVLEDGEMENTS. . . . . . . . . . ll LIST OPPIGURES . . . . . . . . . . iv LIST OF TABLES . . . . . . . . . . V1 Chapter I. INTRODUCTION. . . . . . . . . 1 II. LITERATURE REVIEW . . . . . . . 2 III. APPROACH TO PROBLDI . . . . . . 8 IV. “ALISIS OF DATA . . . . . . . ‘4 v. assume in immune . . . . . so VI. CONCLUSIONS . . . . . . . . . 27 VII. FUTURE INVESTIGATIONS. . . . . . 29 APPENDIX-8ample Calculations . . . . . 33 Figures . . . . . . . . . 35 Tables. . . . . . . . . . 50 BIBLIOGRAPHI . . . . . . . . . . . 55 ill Figure l. 2. 3. 7. 8. 9. 10. 11. LIST OF PIGURBB Formulation of Attack to Problem. . . . Radioactivity of rainfall in uo/cma/day versus rainfall for Station 2P at Monroe,Michigan............ Radioactivity of rainfall in go/ on2 /day versus rainfall for Station at Monroe, Michigan . . . . . . . . . Radioactivity of rainfall in uo/ml versus rainfall for Station 2 at Monroe, Michigan. . . . . . . . . . . . Radioactivity of rainfall in uc/ml versus rainfall for Station 2 at M0m0., MIChlgan. e e e e e e e e e e e Radioactivity of rainfall in {o/m versus the product of rainfal and rain activit for Station 2 at mom..n1°EEneeeeeeeeeeee Variation of rain and air activity with time for Station 2 at Monroe, Mlohlgln................ Variation of rain and air activity with time for Station 2 at Monroe, Hichlgan................ Scatter diagram of air articulate activity versus rainfal activity for Station t at Carleton, Michigan . . . . Scatter diagram of air particulate activity versus rainfall activity for Station 2 at Monroe, Michigan. . . . . Scatter diagram of air particulate activity versus rainfall activity for Station 1 at Carleton, Michigan . . . . iv Page 36 37 38 39 #0 41 42 43 45 #6 Figure 12. 13. 14. Page Scatter diagram of air articulate activity versus rainfal activity for Station 2 at Monroe, Michigan. . . . 47 Scatter diagram of air articulate activity versus rainfal activity for Lansing, Michigan. . . . . . . . . . 48 Scatter diagram of air articulate activity versus rainfal activity for Lansing, Michigan. e e e e e e e e e 49 Table 1. LIST OF TABLES Page Radioactivity data used in analisis from Station I at Carleton, Mic igan. . . 50 Radioactivity data used in analysis from Station 2 at Nonroe, Michigan. e e e 52 Rainfall data used in analysis from Monroe,Mlchlgan............. 54 Radioactivity levels reported by U. S. Public Health Service for Lansing, Michigan.................55 I . IHTRODUOTIOH The increase in the background level of environmen- tal radioactivity in recent years has prompted many investigators to study the sources and characteristics of radioactivity in our environment. Monitoring of the radioactivity in rain water, surface water, food, soil, and the atmosphere is being carried on continuously. It has been the purpose of this study to investigate air particulate activity and rainfall activity to deter- mine if any relationship exists between these two vari- ables. Sone degree of correlation was expected since the radioactive matter in rainfall originates from air particulate activity and high activity levels in the stmcs- phere would be espected to cause correspondingly high levels of rainfall activity. The data used in this study were from three sources. The Division of Occupational Health of the Michigan Depart- mont of Health provided records of radioactivity levels at various stations near the nuclear reactor site at Hearse, Michigan. Precipitation records for the same peg-10¢ of time were obtained from climatological reports published by the U. 8. Government. The final source of data was the U. 8. Public Health Service which reported atmos- Phoric and rainfall radioactivity levels for Lansing, Michigan. II. LITERATURE REVIEW r es f viro e t Rad a t v Better and Russel (1)* discuss radioactive contami- nation cf the environment in the Cincinnati, Ohio area. They indicate that the more important sources of man ‘made radioactive wastes are from: use of radioisotopes in industry, medicine, and research institutions; mining and chemical processing of uranium ore; nuclear reactors for power production or research: chemical processing of spent reactor elements for the recovery of nuclear fuel: and fallout due to the use of nuclear testing devises. In their study of radioactivity in the surface waters of Texas: Gloyna. Drynan, and Smallhorst (2) state that the major potential sources of surface-water activity are fall- out and washout, natural deposits of radioactive materials, and radioactive wastes. e c Rad t t Recently, emphasis has been placed on the study of nuclear detonation fallout; whereas, it has long been established that significant amounts of radon (an???) and thcrcn (anao) in equilibrium with their daughter products are found existing naturally in the atmosphere. IDavis. et al. (3) found that decay curves for air samples ‘Ihich they collected indicated an apparent half-life of —_ * Numbers in parenthesis are references listed at the end of this paper. 2 about 30 minutes. Since radon in equilibrium with its daughter products exhibits an apparent half-life of approximately 30 minutes, it was assumed that radon was the primary active element in the air samples. On this basis, Davis. et al. (3) performed a correlation analysis on the relationship between weather variables and atmos- pheric radioactivity. They found statistically signifi- cant correlations between wind speed at the time of sampling and temperature, dew point, visibility, and the northpsouth component of the wind vector prior to the time of sampling. Calculations were made by Dunning (4) on the contri- bution of shorter lived fission products in radioactive fallout to the world-wide external gamma levels. These shorter-lived isotopes represent a mixture of fallout debris from many nuclear detonations in the past and have been stored for different lengths of time in the strato- sphere and troposphere. Radioactivity in the atmosphere is associated with particulate matter to a greater or lesser degree. Setter and Russell (1) report that in 1957. the annual fallout in the Cincinnati, Ohio area amounted to 15,200 muc/m2 of which 56 per cent was attached to suspended matter. .a large proportion of the fallout was young fission products (2 - 5 days old). The 1-yearbold portion of this fallout was estimated to be 200 capo/ma. R r t without precipitation there is little fallout even though the particulate activity of surface air is abnormally high. Fallout measurements, largely as pre- cipitation, have been made on a continuous basis at Cincinnati, Ohio since March i953. Better and Straub (5) state that a correlation study is now possible between air-particle concentration and raincut (washout by precip- itation). Analysis of data by List (6) shows that a pre- pcnderence of the fallout occurs during precipitation. The activity which is measured in rainfall is that of the long-lived fission products which are potentially the most hazardous. Morgan and Stanbury (7) state that the fission products derived from the testing of atomic and thermonuclear weapons are normally washed out of the lower atmosphere by rain after hold-up periods in the stratosphere during which time the shorter-lived nuclides decay. Beta activity in rainfall reflects nuclear weapons testing. In a study of the radioactive fallout in the Cincinnati, Ohio area by Nader, et al. (8), it was found that a sharp rise in the beta activity of rain occurred twa days after the first Atomic Energy commission's nuclear-weapons test in March 1953. In addition, the radio- activity of subsequent rains during the test period was usually of greater intensity and varied widely. It was found that the activity of rains can generally be attributed to particular bomb bursts, on the basis of a comparison of the half-lives of rain activities with the theoretical decay of the fission product. The units in which the activity of rainfall is reported are of two general ferns. Many researchers (S.7,8,9) prefer to report rainfall activity in terms of activity per unit volume of rainfall such as micro-curies per liter (us/l). Others (i,10,11) prefer to report rain- fell activity in terms of activity per unit area of sur- face on men the rainfall is collected. Typical units ' would be milli-micrc-curies per square meter per day (mus/mQ/day). The first method of reporting is useful when one is interested in the variation in rainout of activity with time. When total radioactive raincut is being compared with radioactivity which appears in surface waters, the latter method appears more appropriate. According to the Division of Radiological Health, Public Health Service (12), measurements have indicated that the ball: of deposited activity occurs through precip- itation, but concentrations in surface air are not directly related to the amount deposited through precipitation. Precipitation of radioactivity may be in the form of snow as well as rain. Observations on radioactive snows at Ann Arbor, Michigan (‘3) definitely established the presence of radioactive rare earth isotopes. The energy and half-life for these isotopes were such that it could be stated that the activities undoubtedly originated in the Las Vegas atomic test explosions. Activities as high as 100 times background were obtained in these samples. R act r W are Increasing attention has recently been paid to the possible contamination of drinking water supplies with” radioactive materials. The measurements of Kahn and Reynolds (14) have shown that water from rivers and lakes contains detectable amounts of 83:90 and 03137 which are potentially the most hazardous of the long-lived fission products. The quantities cf‘pollution by natural radios activity and from nuclearbenergy operations are so smell quantitatively that chemical separations, energy determina- tions, and half-life measurements are difficult (15). Better, at al. (10) found a linear relationship to exist between radioactivity and concentration of suspended solids in stream surface water. These findings provide circumstantial evidence that radioactive runoff is proportional to rain intensity, because the concentration of suspended matter is roughly proportional to rain inten- sity. Because radioactivity is associated with suspended matter, impounded waters have a high ratio of dissolved to total activity and a low concentration of suspended matter (10). Thomas, et al. (‘6) found significant increases in :activity in the surface waters in.Massaohusetts which were apparently due tcfallout of fission products from llevada nuclear-weapons tests. Previous testing did ncteause large increases in the activity of surface waters in Massachusetts, but meteorological conditions were such that a portion of the air mass over the Nevada test site on June 1st moved rapidly eastward with only a small amount of rainout before reaching the Massachusetts area. 't—x‘. III. APPROACH TO PROBLEM b so The primary purpose of this study has been to deter- mine if there is any measurable amount of correlation between radioactivity in rainfall and the level of radio- activity in the atmOSphere. It would be desirable to ascer- tain what, if any, effect a rise in the atmospheric radio- activity level would have upon the amount of radioactivity which reaches the earth in the form of precipitation. There is also interest in the levels of radioactivity in other media such as surface waters, soil, and biological specimens. It was hoped that on the basis of this statis- tical study, new light would be shed on the factors which influence these radioactivity levels. The results of this study will provide a basis for re-evaluation of the present method and extent of data collection. A considerable amount of time, effort, and money has been, and will be, spent on the monitoring of radioactivity levels in various media. If some part of this data collection is found to be unnecessary, a consid- erable saving in effort and expenditure could be attained. in ideal development would be the elimination of the necessity for monitoring of rainfall radioactivity. This could be realized by finding a high degree of correlation between the levels of radioactivity in the atmosphere and in rainfall. An additional objective has been to provide a some- what systematic procedure which can be followed in the future study of correlation between radioactivity levels in other media. This thesis reports the first extensive study of correlation between levels of gross beta radio- activity in the atmosphere and in rainfall, and the statistical procedure presented will be found useful in future studies. On the basis of the results of this study, it was hpped that some answers could be given as to the method of transfer of the radioactive particulate matter from the atmosphere to the earth in the form of rainfall. It was realized that this goal might be severely limited by the method of attack used in this study and the amount of data used in the analysis. Finally, the purpose of this study was to provide an insight into the difficult problems involved and to provide a basis for further research in the field of environmental radioactivity. W One of the most serious problems encountered in this study was the limited amount of data which was available for analysis. Some of the possible methods of analysis which might have been used were precluded by the fact that, for various reasons, the weekly rainfall activity levels 'were not all reported. If a continuous series of weekly IO observations had been available, it would have been possible to investigate the activity levels for the existence of a time lag between the atmospheric radioactivity and radioactivity of rainfall. In any future study of this type where data are being collected by the investigator, an effort should be made to assure that all the possible data are taken and that no unnecessary omissions are a“- allowed. Radioactivity levels are often reported on a weekly basis and the data used in this study were weekly averages. Analysis of this data did not permit insight into how air particulate activity is affected by daily rainfall. If records of daily activity levels had been available, it would have been possible to determine how efficiently vari- ous types of rains remove radioactive particles from the atmosphere. Future analysis should be based on daily activity levels if at all possible. in inherent difficulty of the statistical analysis . of data of this sort is that it is not possible to prove beyond doubt that correlation between weather variables does not exist. If correlation is found to exist between two variables, this result may be stated with a well defined degree of certainty and no special problems arise. 0n the ~ pcther hand, if no correlation is found using a certain type of analysis, this does not preclude the possibity that the two variables being studied might be found to be highly correlated if some other type of analysis were ‘1 used. This problem cannot be avoided, but an attempt has been made to conduct a thorough analysis reporting the procedures used so that future efforts by other analysts will not be wasted. Low levels of radioactivity are a problem to a correlation study of this type. such low levels are difficult to determine accurately and the standard errors of the results tend to be quite high. This normal variation of counting makes a correlation analysis very difficult and a large amount of data is required to show correlation even.if the variables are indeed related. Many weather variables have been found to influence the atmospheric radioactivity levels to a greater or lesser extent. in important factor is the movement of air due to wind from one area to another. Air moving from an area where a rain has Just occurred is likely to have a low concentration of radioactive particulate matter. It is obvious that the activity of a given rainfall might be quite low if the air activity is low on that particular day while at the same time, the average air particulate activity for the week of record has been very high. If situations of this sort occur often, it will appear that there is no correlation between air and rain activity. IPcssibly the activity levels in rainfall are also affected by a number of undetermined weather parameters. It is obviously impossible to recognise and take into account all the possible variables which exert their influence A. . ..,ts......sa 1m I2 and this limitation should be recognized when evaluating the results of this study. - ‘ The radioactivity levels being analysed in this study are composed of contributions from various sources. In addition to the nuclear testing debris in the atmosphere, there is secondary radiation from cosmic rays, activity initiated at nuclear power reactors, and natural backgreund radiation. For a strictly accurate correlation analysis . it is necessary that there be comparability among data for different periods. If any of the variables are varying with time the data are said to be non-homogeneous. This was actually one problem faced in this study, and the assump- tion was made that this non-homogeneity would have no adverse effect upon the comparability of the data for different periods. The validity of this assumption deter- mines the validity of the results of this study to some extent. Air particulate activity as reported in the data used in this study is actually the activity of air which is near the surface of the earth. Thunder storms which occur during the summer months originate as ice crystals at alti- tudes ranging from 30,000 to 50,000 feet (21). These ice crystals are formed around particulate matter came of which is radioactive. Therefore, particulate activity from very high altitudes as well as from.surface air is rained out during the summer months. Particulate activity at the higher altitudes is not necessarily the same as surface 13 activity and it is possible that no correlation exists be- tween surface air activity and the rainfall activity. Winter rainfall is of a different nature than that which occurs during the summer months. Winter rains are caused by the movement of a warm front into a lower temperature area. The warm air is forced to rise and at altitudes of 3,000 to 8,000 feet the temperature falls below' .iq the dew point of the air and precipitation results. There- fore, during the winter months the activity in rainfall may be quite different than the activity found during the summer, and comparison of the data without regard to the seasons is not entirely justified. Seasonal variation in activity was not taken into account in this study because of the limited amount of data available for analysis. Eu ture Study For the purpose of future study a block diagram is presented in Figure 1. The steps shown in the diagram are essentially the ones followed in the approach to the problem of this study. Future correlation studies of environmental radioactivity may use this approach with some modifications. IV. ANALYSIS OF DATA W The data used for analysis in this study were obtained from the Division of Occupational Health of the Michigan Department of Health (11) and from the 0. 8. Public Health Service (1?). Supplementary data on precip- itation at Monroe, Michigan were obtained from U. 3. weather Bureau publications (18). The data supplied by the Michigan Department of Health were observations of radioactivity levels in rainfall and air particulate matter. These observations were originally made to determine any changes in background radiation which might have occurred as a result of activation of the Enrico Permi nuclear power reactor near Monroe, Michigan. This analysis used data which the Michigan Public Health Depart- ment collected in the period from June 15, 1959 to Decem- ber 19, 1960. The U. 8. Public Health Service data covered the period from October 29, 1961 through May 20, i962 and 'was collected by the Lansing, Michigan station of the Radiation Surveillance Network. Radioactivity of air particulate matter waszrepoptgd as gross beta activity one day after’ccllecticn. Weekly average values of activity were reported in units of micro-micrc-curies per cubic meter of air sampled. _ The radioactivity of rain samples was also expressed as gross beta activity which did not include the 1A ‘5 contribution of naturally occurring radon and thcrcn.and their daughter products. The data were tabulated in two separate units; micro-caries per'milliliter, and micro- curies per square centimeter per day. is with the air particulate activity, the results were reported in terms of weekly averages. Certain weekly data were missing where no samples were collected and where the net count rate of the samples was less than or equal to the backgrpund count rate of the counting equipment, the results were reported as, ”Trace”. The precipitation data obtained from U. 3. Weather Bureau publications were reported in inches of rainfall. Rainfall data were only available at the Monroe station and daily rainfall records were condensed to weekly totals corresponding to the weekly sampling periods for radioac- tivity of the rain samples. WW Data from the hichigan_Department of Health were used to investigate how the radioactivity of rain varies with rainfall. Rainfall activity expressed in units of micro- curies per square centimeter per day (no/cmz/day) was plot- ted against the corresponding volume of rainfall in inches. These plots are shown in rigures 2 and 3. It was espected that some trend would be observed in these plots if average activity precipitated per day was: 16 (1) independent of the atmospheric activity levels, and (2) independent or the frequency, intensity, and duration or the various rains contributing to the total weekly raintall. It was further expected that the trend would be for an increase in average rain activity to occur it the rain- fall increased. Another approach used in the investigation of rain activity was to plot activity, expressed in units or micro- curios per milliliter (us/ml), versus rainfall in inches. It is to be noted that each value plotted was a paired observation of rainfall and rain activity for a certain week. A trend in the plots was expected if the average activity per unit volume of rainfall was: . (1) dependent upon the total amount of rain which fell during the week, and (2) independent of the atmospheric activity levels. These plots appear in Figures 4 and 5. The third relationship studied was a plot of rainfall activity in uc/ml versus a measure of the total deposition of activity. The measure of total deposition was obtained 'by multiplying the rainfall activity expressed as uo/ml ‘by the corresponding total weekly rainfall in inches. .Again, each or the points plotted represented a paired observation. For convenience, the plot was made on 17 semi-logarithmic paper and is shown in Figure 6. tmcs h ric Acti v on t t In an effort to determine if there was any relation- ship between atmospheric radioactivity and radioactivity of rainfall, several different approaches were taken. Figures 7 and 8 show air particulate activity and rainfall activity plotted on the same time scale. These figures show data reported by the Michigan Department of Health. The time scale is not continuous because of the disconti- nuity of the data, and therefore the rain activity and air particulate activity can only be compared on a same week basis. Figure 7 shows rain activity in units of uc/ml . and Figure 8 shows rain activity in units of uc/cmZ/day. Using the data reported by the Michigan Department of health, air particulate activity was plotted versus rainfall activity in Figures 9 through 12. The points on these scatter diagrams each represent air particulate activity and rainfall activity for a given week. t O T For preliminary investigation for correlation, Figures 9 through 12 were used to make a statistical test called the Corner rest (19). The Corner Test was made to test the hypothesis that the two variables were independent. the non-parametric test statistic was computed in the following manner. A vertical line was passed through each 18 scatter diagram which divided the data with 50 per cent of the observations on the right and 50 per cent on the left. A similar horizontal line divided the observations with 50 per cent above and 50 per cent below. Positive and negative signs were assigned to the various quadrants as follows: upper right positive, upper left negative, lower left positive, and lower right negative. Starting ’““' at the extreme right and moving toward the vertical median line, the number of observations encountered before an observation was found across the horizontal median were counted. In Figure 9 this number was +1. This number was given the sign of the quadrant in which the extreme observations were found. Proceeding similarly from the top, left, and bottom, four numbers were obtained. the algebraic sum of these four numbers (+6 in Figure 9) was taken as the test statistic. If this sum was found to be too large at the 5 per cent level of significance, the hypothesis that the variables were unrelated was rejected. The test statistic was Judged to be too large when its value exceeded a computed value found in an appropriate table (19). The chance that this computed value would be exceeded was only 5 per cent if the variables were not associated. $22l3223$2§ g; z-Test Air and rain activity data from the Michigan Depart- ment.of Health and from the U. 3. Public Health Service ‘were used to test for the existence of correlation. the 19 Department of Health data were analysed on a monthly and weekly basis to see if the same results would be obtained in both cases. The Public Health Service Data were analysed only on a weekly basis. All rainfall activity was expressed in units of uc/ml and uc/cmz/day. To simplify the analysis, the rain activity and air activity observations were classified into various categories according to the magni- in- tude of the activity. The r-test was applied by comparing, (1) the total variation in the rainfall activity for spe- cified levels of air activity with, (2) the total variation between the mean rainfall activities for the various cate- gories and the overall mean rainfall activity. The test statistic, P, was the ratio of the variation between the categories to the variation within categories. If P was found to be larger than the critical P value at the 5 per cent level of significance, the hypothesis that there was no correlation between the variables was rejected. The critical P value was selected so that the chance of it being exceeded was only 5 per cent if there was truly no correlation between the variables. A sample calcula- tion for the test statistic, F, is shown in the appendix. V. RESULTS AND EVALUATION Radioactivity of rainfall, in units of uc/cmZ/day, was plotted versus rainfall in inches in Figures 2 and 3 to determine if there was any relationship between the two variables. It was predicted that radioactivity would increase with rainfall if the rain activity is independent of the corresponding air particulate activity and indepen- dent of the frequency, intensity, and duration of the various rains. By examining figures 2 and 3 it was seen that no such trend was present. This result does not ~»- indicate that rainfall activity is dependent upon the charu acteristics of the various rains and upon the air partic- ulate activity. However, it has been shown that the possibility of dependence of the two variables does exist. A similar approach was made in Figures 4 and 5 where rain activity, expressed in units of uc/ml was plotted versus rainfall. A downward trend in rain activity was expected if the rain activity in these units is independent of air particulate activity levels and dependent upon the total amount of rain which falls. This trend was observed to some degree in Figure 5, although the trend was not definite and unquestionable. because of the dispersion of the observations in these figures, the strict indepenp dance of rainfall activity and air particulate activity was not verified. The fact that rainfall activity in units 20 21 of pc/ml will decrease with increasing rainfall if air particulate activity remains essentially constant was not definitely proven and further investigation of these variables was made. In Figure 6, radioactivity of rainfall in uc/ml was plotted versus a measure of the total deposition of activity. The measure of total deposition was obtained ro+ by taking the product of rain activity in pc/ml and total . weekly rainfall in inches. It was expected that the con- centration of activity in rainfall would decrease as the total amount of deposition increased. The result, however, was in direct disagreement with this prediction. The obser- vations, when plotted on a semi-logarithmic graph, indi— cated an increase of the intensity with an increase of total deposition. The data used in this particular analysis were those collected by the Michigan Department of Health and the activity levels in rainfall at the time of collection were very low. These low activity levels 'which are very difficult to measure may be the reason that the prediction was not verified. In very small amounts of rainfall where concentrations of activity are expected to be high, low activity levels would be more difficult than usual to measure and counting errors as high as so or 100 Per cent of the observed count rate might be present. a Similar'situation might be found when the total precipitation 18 high, The rained-out activity would be very highly diluted in this case and again difficult.to measure. It 22 is not impossible that a complete reversal of the results could occur within the limits of the counting errors. It is concluded that investigation of the variation of activ- ity concentration with total deposition of activity should be made when activity is at a much higher level. No definite statement can be made about the results of this particular analysis. A statistical test called the Corner Test was applied to the scatter diagrams of air particulate activity and rainfall activity shown in Figures 9 through 12. The data were those taken by the Michigan Department of Health at two separate stations. The purpose was to test the hypoth- esis that the two variables were not associated. The results of the analysis were as follows: m. this greases: sass. “151°“ 9- 1 uc/cmz/day 6 10 1O 2 uc/cmz/day 5 1o 11 1 uc/ml 13 1O 12 2 uc/ml 10 to With rain activity orprcssed in uc/cme/day, the hypothesis that there was no association between the variables was not rejected at the 5 per cent level of significance for either Station 1 (Carleton) or Station 2 (Monroe). However, when the rain activity was expressed in units of 23 pc/ml, the hypothesis of no association could be rejected for the data from Station 1 and Station 2. This means that the test statistic, S, could have taken values as high as those observed no more than 5 per cent of the time if the variables were truly not associated. In summary, it has been found that weekly rainfall activity in units of uo/ml was associated with air particulate activity for the same all? week. Weekly rainfall activity in units of pc/cmz/day was not associated with air particulate activity. The F-test was used to test the hypothesis that there was no correlation between air particulate activity and rainfall activity. The Michigan Department of Health data were tested both on a monthly and weekly basis to determine if results would be influenced by the period of record. The results are shown below. (1) Monthly averages of radioactivity levels reported by the Michigan Department of Health gave the following results: gtagign Units 3: Raine_ __ F Critical uc/ch/day 5.87 __ v 3.48 2 uc/ch/day 3.69 V. 3.58 1 pc/ml 0.03 . 3.53. 2 Pc/ml 0.80 3.58 (2) Weekly averages of radioectivity_levels reported by the Michigan Department of Health yielded: 24 Station Units of Rain- Critical ° t al: E___. 1 pc/cmg/day 1.03 2.54 2 uc/cmZ/day 0.73 2.64 1 uc/ml 2.39 2.36 2 uc/ml 0.52 2.37 Using the weekly averages it is seen that the F-ratics were less than the critical values with only one exception and in this case the P-ratio was only very slightly larger than the critical level. In general it could be stated that there was not sufficient evidence to reject the hypothesis of no correlation between air particulate activity and rainfall activity in uc/ml or uc/cma/day. However, when the monthly averages were tested for corre- lation, it was found that rainfall activity in units of uc/cme/day appeared to be correlated with the average air particulate activity for the same month. This result can be explained by the fact that total deposition of radio- activity as uc/cma/day depends upon the amount of rain which falls. Even if air particulate activity remained at a constant level for a series of weeks, the total depo- siticn would be highly variable because of the large vari- ation in weekly rainfall. When monthly averages are con- . sidered, the fluctuation in amount of rainfall is much smaller. is a consequence, a correlation between total deposition of activity and air particulate activity is more easily discovered when monthly averages are considered. 25 The results of the F-tests applied to the data ' collected at Lansing, Michigan are shown below. ‘cni ts of Rain- Critical iifllaluflfldfldiaa» E :Z.__. uc/ch/day 6.72 2.83 uc/ml 3.47 2.83 It is possible to reject the hypothesis that there is no correlation between air particulate activity and rainfall activity in units of uc/cma/day and uc/ml. This disagrees with the results of analysis of the data collected by the Michigan Department of Health where it was concluded that no correlation existed between air particulate activity and rainfall activity. The activity levels reported by the U. 8. Public Health Service were approximately ten times as high as those reported by the Michigan Department of Health. The higher activity levels were the result of recent nuclear testing by the USSR. The higher levels swore determined more accurately because the counting errors were greatly reduced, and the results of analysis of the higher activity levels are more-reliable. In addition, it is obvious that the components of gross beta activity are different for fission products of different ages. It is quite possible, although not verified by experimental results, that correlation exists between air and rain ac- tivity for some nuclides and not others. it present it is only possible to state that there appears to be definite correlation between rainfall activity and air particulate 26' activity when the activity levels are high and consist of young fission products, and no correlation is found for the low levels of activity. Because of the dispersion of the observations on the scatter siagrams, it was decided that development of prediction equations for the data was unwarranted. For the same reasons, correlation coefficients were not computed. If the correlation coefficients had been com- puted they would have been quite small and would not have been sepecially useful in this study. VI. CONCLUSIONS From the results of this study theiollowing general conclusions may be drawn: . . (1) Radioactivity of rainfall in units of uc/cma/day is apparently not related directly to the amount of rainfall in inches. Two conditions would have to be satisfied for for these variables to be so related that an increase in rainfall activity as uc/cmz/day would result from an increase in rainfall. First, it would be necessary that the level of air particulate activity remain constant or have no effect upon rainfall activity. Second, the concen- tration of rainfall activity (pa/bl) would have to remain nearly constant for any amount of rainfall. Since concen- tration of activity varies independently from amount of rainfall then no relation would be eXpected between rainfall and rainfall activity as Pc/cma/day. (2) The average weekly concentration of rainfall activity as uc/nl has been found to increase as the total amount of rainfall increases. This result disagrees with original predictions that rain activity concentrations would decrease with increasing rainfall. These results were interpreted as being inconclusive because the low levels of activity being analyzed were highly variable. It would be necessary to analyze data taken at a time when the activity levels were much higher to obtain reliable and conclusive results. (3) When environmental radioactivity levels were low, 27 rainfall activity in units of po/cme/day appeared to be unrelated to air particulate activity when average weekly levels were considered. However, the monthly average depo- sition of activity appeared to be highly correlated with the corresponding monthly average air particulate activity. rhe explanation for the monthly correlation and not weekly correlation was that total deposition of activity depends upon the number of rains which occur during the period of record. Since weekly rainfall is much more variable in intensity, frequency, and duration than monthly rainfall. a better correlation is expected if monthly averages are considered rather than weekly averages. . With high environmental radioactivity levels, there appeared to be highly significant correlation between air particulate activity and rainfall activity as uo/cmg/day. Analysis of the high activity levels was made only on a 'weekly basis. A (4) The concentration of gross beta activity in rainfall was found to be unrelated to air particulate activity levels when the activity levels were generally low.‘\ both weekly 'and monthly averages were considered. Significant corre- lation was found between air particulate activity and rain- fell activity as uc/ml when the activity levels were high. It was suggested that different fission products were conn tributing to the gross beta activity depending upon the relative ages of the activity. Further study of specifia nuclides and their relation to gross beta activity will be ‘neoeesary to provide reliable answers to these questions. VII. FUTURE INVESTIGATIONS The Public Health Service Radiation Surveillance Het- ‘work was established in 1956 in cooperation with the Atomic Energy Commission to provide a means of promptly determin- ing increases in levels of radioactivity in air and precip- itation due to fallout from nuclear weapons tests. Since the resumption of nuclear testing in September 1961, the network has been expanded to 62 stations. Gross beta activ- ity of air and rain are reported daily by these stations. It is suggested that the data collected by the Radiation Surveillance Network stations should be used in the future study of gross beta radioactivity levels in air and rain. The advantages of using these data for analysis are: (i) the data are readily available, (2) the activity levels observed since September 1961 are relatively high, and (3) the results of such an analysis would be applicable to the entire continental United States. In addition to the statistical procedures used in this study, more refined and extensive analysis can be made on the Public Health service data. The following systematic approach is suggested and is fully described by Mills in W (20). To provide a rational attack ‘upon the problem the following questions are to be asked: (I) Do the available observations provide sufficient evidence that the two variables being studied are related? 29 3O (2) If the existence of true correlation is assumed, will a straight line acceptably define the regression? (3) If correlation exists and a straight line is not appropriate as a regression function, will a given second degree function provide an acceptable measure of regression? If such a function is not suitable, will a different function with the same number of constants, or a polynomial of higher degree, give an acceptable fit? To obtain the answer to the first question, the procedure followed in this study is adopted. If the answer to the first question is no, the investigator should go no further, If it is yes, the testing of regression func- tions should be carried out until an acceptable function is found. In the second step a straight line is fitted to obser- vations by the method of least squares. Computed values of the dependent variable are compared to the observed values by the technique of analysis of variance. the var- iation of the dependent variable due to the influence of random forces is compared to the variation between computed and observed values by the r-test. If the latter variation is significantly larger than the random.variation, it is concluded that a linear relationship does not adequately predict the results. , a search is made for a polynomial of higher order 31 which defines the relation between the two variables if a straight line is not acceptable. The procedure is the same as described above where the method of least squares is used to determine the best values of the constants in an equation of the desired form. To make such an extensive investigation feasible, it is suggested that the analysis be done on an electronic computer, The application of computer analysis is~desirable to save time and labor and to obtain accurate results. After the analysis of gross beta activity levels in air and rain, an investigation should be made of the correo laticn between specific nuclide concentrations in air and rain. Potential health hazards such as strontium-90, iodine-131, and cesium~i37 contribute to gross beta activity and research is required to determine if any correlation exists between concentrations of these isotopes in air and rain. Study is also required to determine how various individual nuclides are related to gross beta activity levels. it the present time no relationship has been discovered between gross beta activity and concentrations of various radioactive isotopes, and great deal of study is required in this particular area. APPENDIX 32 SAMPLE CALCULATION the formulas to compute the r-ratic used in this study are shown below. in example of the computation of the F-ratio for monthly averages of activity levels at Monroe, Michigan is also included. Agglzgis 2f laziance fable gatggggz gum g; Squares 5;; .M E 2 3.1739011 fn1(z.1e - 2..)2 k a ‘ Stan 8 2113(Ia. - Xe.) - T p “a 2 Within 2}}:(113 - 21.)2 N - k 32 3 E2281: - 11-) J N - 1: Pass/8% where, k a number of categories into which the air particulate activity levels are divided. n1 a number of rain activity observations in the ith category of air activity. 3 a total number of rainfall activity observations. :13 z the measurement of the 3th rainfall activity observation found in the ith category of air activity. 21. a the average level of rainfall activity within the ith category of air activity. 2.. a the overall mean rainfall activity obtained by summing all the rainfall activity observe- ticns and dividing by the number of total observations. 33 I I 2.20 l l l l I _L_- _-J H- ._.l__ I.0————L-———l——-l—h°5-f——-—I'— . I . ' ' __ __I_____I____I.- 0.9 —-‘—l———‘——~rv-T ‘T‘ l I I {m8”“—V“—T'“T——r— r I I i ‘6. 6___I..._I_:_I_"haven"?-221_ am5~——.ffi.’];g ‘I T“ 1; .us I“ _K__.__L_J_. ._ ___________ | HmI_—' ' fl '1 I I I z : ' 43.1! I ——l— __-f_——l————l—— — _.L._ 30.3——‘T"é—6 T I l 27I l l I ° I l .23 ___ ._ -—-——l—-—--—--——l—-— m2———f-—fl-——T- 1 r I i J I l ____ ————-L __-_ _- c$30.1 —--—I-—7f—.10+——T——-II- 4|— 1 I L“): { 00310-0 4' L 1 L 00 0.1 0.2 0.3 0.4 0.5 0.6 1.38 3.53 Air Particulate Activity (ups/m5) 1 n1: 3 6 3 1 1 Ex”: .80 2.12 1.28 .27 .46 2.2 r' 43 27 .A6 2.2 X1. 3 .27 e35 e e N a in; z.- ‘5 ‘ J 2.. 2 Grand Mean = 7.13/15 2.: 0.48 Within Sum of Squares: 2 " e O (.46 - .27 g a .036 .?g - :3; 2 : .822 .26 - .27 2 s.- .002 .03 - .35 2 g .102 '08 ' '27 9 a '034 (I'os - .43 2 = .384 '63 - '33 2 : '3?) 323 - .43 g c .040 :23 : :35 2 a :005 .00 - .43) a .185 0.988 35 Between Sun of Squares: 3 027 " 0‘8 2 3 00‘32 6. - .aa2ao.1o1 3 oz; ' 0‘8 g I 0.008 .aazzwkiaaa 3.233 Summary: WW: Category Sun. or Squares a: 1mm aqum Botvoon 3.233 5 0.647 Within 3.988 9 0.110 F.95‘509) 3 3048 Oonclunicnt Since tho r-ratio 1' greater than the critical P-ratio for 5 and 9 degrees or freedom. no reject the hypothesis that that. is no corrolcticn bctwocn air particulate activity and ruin activity in units of no/cnz/day. The value of 2.20 for rain activity contributed heavily to the high value obtained for the F-ratio. It is quite probable that the F-ratio would be significantly reduced if this observation were not used in the analysis, and the strength of the conclusion obtained is not great. 36 FORMULATION OF ATTACK TO PROBLEM Basic investigation Literature review of data - charts. to determine the diagrams, calculations work already done i l J l Formulation of basic problems to be solved F l Recognize and enumerate Recognize all P0335“? the main difficulties to ways to analyze the be overcome data I l 0 Limited Inherent data problems I I Limited Decay, wea- applica- ther, var- tion of iation of results counltim Form of Dependence data on unmea- sured var- iables Apply statistical tools to the anal sis. Rationalize statis- tica approach and make method general enough to be useful to others i l 1 Draw conclusions from Sug est_possible work the results of the whic might be .done in analysis future studies Figure l r l ' I ”A .'.' | g l " If ( Pc/cmz/ day ) Rain of Radioactivity 37 5.0 Weekly Averages Date: 0-59 to 12-60 2.0«- O O 1.04- O O O O 0 O O O O O.S"'o O O o 5. O o O o O 0 0° 0 O 024* ° 0 O O 0 0 O O O 0.1-r° ° ° 0 O O 0.05 t 5 + t 0 0.4 0.8 l.2 LG 2.0 Rainfall (inches) Figure 2. Radioactivit'y of rainfall in pc/cmZ/day versus rain all for Station 2 at Monroe , Michigan . 38 .4 {Fiji I £03522 .00.:52 *0 N c0205 3» 2350.. 39-2, >00\~Eo\u1 E. :359 *0 33300050". .m 330C 7.3.0:: :otzom 9n o.~ .9. one and 03 no.0 . ~90 66 .r 4 w w a. a a 00.0 0 O O O 0 O O 11—.0 o ....N.o o o O o .0 O O O o o c tn.O O O .0 o O O O O ..o._ . 3.2 0+ amé "8.00 . neonae>< 3x003 .. o.~ (flap/zmo/orl) ugoa 50 Minuooogpoa 39 7 Weekly Average: 0 Da+e: 6-59 +0 I2-60 0 an 6‘ o '2 X E o \ ._ u 5 :L V O c U o I: 4__ 4 O a... o a a O .0- .2 34v- 0 0 * u 0 a .2 '3 c: o° ° z-II- O O 0 O O O "b O o a o Do a o a o o 9 0 ° 0 o o : : : : : 0 0.4 0.8 LZ L6 2.0 2.2 Rainfall (inches) . . . . . . -8 Figure 4. Radmacl’wnly of rainfall In pc/Inlauo versus rainfall for Sfa’rian 2 al Monroe, Michigan. 7. .l‘). 40 20 °° Weekly Averages Dale: 6-59 +o l2-60 [0" «no 0 IIF o u.- 0 O 54- . . O A "' o O Q o '2 .0. o a K E 00 0 \ 2-- o. 0 o o o .3" ° 3 a- O a 4!- a: '4_ o o u- -» o ‘P 9 o r «l- 0 0° 0 '; 0.5" ° 2: o 0 4’ o a O .o '0 +- o a O! 0.2" o O 0.! : J, : l 0 0.4 0.8 LZ l.6 Rainfall (inches) Figure 5. Radiaacfivil'y of rainfall in pc/Inlzilo"B versus rainfall for Slalion 2 al Monroe, Michigan. 2.0 Radiaacfivify of Rain (pc/mlxlo'a) 41 I0 0 o 0 5" +- o o o «i- a a 1 o a 2-0- 0 O o o 0 Id. ‘b o o a o o o a 0.5+ ° 0 db 0 a . Weekly Averages dh O 0'2 ° Dale: 6-59 fa I2-60 o o i 00'L # % + + 0 0.4 0.8 LZ LG Rainfall (inches) It Rain Acfivify Figure 6. Radioacfivify of rainfall 2.0 in Pc/ ml versus fhe producf of rainfall and rain acfivify for Sfafian 2 af Monroe, Michigan. 42 £007.32 .00.:32 +0 N c0205 Law 25+ f; 3.230 to tag 50.. mo co_+o_._a>. .5. 9:49.“. Jw Kiwwv 3l°ln3ElJ°d (em/odd) n: . amass: x~o>> nno o . \ . i) \‘X . ./(. , . . /\ /.- . / ua . I ’ I." In. NOL: \ . _ O /.l.)/ /’ s\ /i. ‘\ . ADJ IOI < \ \ , a .A . . :o i / . , .3 “a m. u . #2 .m: no: a w oo-~_ 2. mmé H. ua+ao I 1.0. O. 0.0:: .24 iiiii 1W c.0m vceoou ON 0.. 43 4w mev ammuwd (cm/odd) 53222 .0923. +0 N cor—3m .5... 05.... p.33 3.3.60 to tea Eu. .0 co..o.ua> 335:: :33 . m 0.50.... No: 0.0.“ 0.0.. \. f» , .. . .\ ,/.\\/ _S / \. \/ y \ \-}-/, l I o. O l N o lo. 0 0.. 0.. K+w+°°°9p°a i 'f. o (KOPéIflO/Dfi) ‘1an 50 44 . 000.52.). 59.0.00 +0 _ co.+0+m 3.. 3.3.00 :0...c_0._ ”33> 3.3.30 30.002000 :0 ..o 0.0.30.0 aa++0um .0 0.30.... .>00\~Eu\u1. 3310.0. 50¢ N. 0.. 0.0 0.0 0.0 N0 0 w r w w n no. .. v iiiiiiiiiiii iillliiiilliiia 5. M. 1 ... u H d — W 0 — 11—.0 D _ a 0 u- o m. _ . . . .. m. _ . . . _ m. _ 0 .F o o 00 M .N.O a ll. - a In 0 . m a 4 v .. _ _ p a o o I. + m. _ . _ ... _ c o + (A _ o o m: _ a _ ...m.O d . N _ 9 .. l _. .......................... ._ . m e O .... l.\ . Lf a 4.0.. e oo-~_ 0. am... «too «000.330. 3x003 o L o.~ Parficulal‘e Acfivify (ppm/ma) Air 45 2.0 0 ___.___._________L.____ Io«- [—0 1 4» : g 0 I 4y- | I ... I l 05« + . ° o I I “7" .. o I . . I T | . I I ° . .. . I O . ' lo 0 o—4L. o o I O'Z‘i i ' o o [ O l . I O I O o O l 0 0 I O O 0 0.14 o I .. I . . I ...- I I ‘F' l O i I ‘i L-._____'.__% ___________ LJ .05‘l" o 0 Weekly Averages Dale : 6-59 f0 I2-60 .02 i i 4i ‘i 0 0.2 0.4 0.6 0.8 I.0 Rain Acfivify (pc/cmz/day) Figure l0. Scalfer diagram of air parlicula+e acfivily versus rainfall aclivi+y for Sfafion .2 of Monroe, Michigan. 46 .c00....u..2 .co..0...0U .0 _ co.+0.m .0. 3.3.00 ...0..r_.0.. ”at”; 3.330 30.00....00 ...0 .o 50200.0 20+.0um ... 0.39.... “0:0.x.E\u_... 3.3.00. __0.c.0m o~ o. on cm 9. Go Nd .d .. -J...r. 4+ . .+..T++fi. . ..o 0 L 0 ..I llllll I VIIIIJIIIIIIJJ o 0 _ 0 o .0 _ _ o a _ - o o o . o w o o .1 ..md _ 0 .. o _.. 0 _ o 00 _ L _ . _ 0 n iiiiiiiii Lipiiiililiiliu [I O . ......o f . .r . .3 Jr 0 110.. O O a . .54... oo-~_ 0. am-o ”n+oo .. 0 «000223 3.003 .. Qm J!V K+w+9v 3+°|“°.'H°d (em/odd) 47 ON £007.22 .00202 +0 N 00:05 La”. 3.3.00 20.50.. manuo> 3.3.00 2030.200 ...0 “.0 5000.0 00+.00m .N. 0.59... 3-0:. E}... 3.52 :33: o. 3 . o.~ 0.. n .o u .o ..o . J." n «J. . . . + w I +1 . w w . w .o w IIIIIIII ii:|+..||...:i.i-.i..:il. _ 0 . O O _ _ o _ . . o o _ . .. . . o _ ..~.o W a .0 I I IllOblI-ITII a a 60 o _ _ . .. _ . _ o 0 _ _ o _ .. riiiiiiiiiii.iiiiiioi iiiiiii 0.:— ...“.O L. 1. .. a H.0.. 0 sfo.~ oo-~. 0. 3-0 ”2.5 . o «000...;4 3x00? 0.... 3l°|“°!l-‘°d 4w? MMHDV (Em/add) 48 I50 Weakly Averages Data: IO-6l to 5-62 0 100‘- O 70-»- ; 00 I o - O ’t 504' . 0 )5 o 0 v ~\ 40 E u 0 \ 0 db 1 30 o o >~ .‘1’. a > 20-0- .— O ... O o < O [5* '6 G.- .E 0 a no» 00 o 0 7+ o O 0 5-4- 0 4 : 41 'r % : 1‘ 1r i r O 2 4 6 8 IO l2 l4 l6 IO 20 Figure l3. ScaHer diagram of air par+icu|a+a aciiviiy Air Par+icula+e Aciivh‘y (ppc/m3) versus rainfal! aciivify for Lansing, Michigan. I”; L'fl" 49 I400 WOO“)! Averages Date: IO-GI *0 5-62 [200m 0 O IOOO—w- 5.‘ 2 " 800... a E \ u :1. >~ :0; 600» > 22 a u < a O '6 40m- 9— o 9 o .5 . ° . o O O m 0 O O ZOO‘ii' 00 O O 00 O % % + : s 4 % 4t #1 O 2 4 6 8 IO 12 l4 l6 IE 20 Air Pariiculafe Aciiviiy (ppc/ma) Figure I4. ScaHer diagram of air parficu|a+e aciiviiy versus rainfall acfiviiy for Lansing. Michigan. ‘.'_' -'"'.'. ‘. “305‘.¥l95 [Ci-”M’n. ‘ .‘V".-9-‘etm 50 Table l. Radioactivity data used in analysis Iran Station 1 at Carleton. Michigan. Sampling Air Particulate Rainfall Rainfall Week Activigi Lotizity Activity (nun/h (no/on /day) (80/01) 39 3.6 2.7 4.10 40 2.3 53.0 41 1.6 8.0 #2 1.6 0.32 1.8 #3 1.8 0. 3.1 44 1.2 0.69 2.2 #5 0.94 46 1.0 0.42 2.2 4 0.81 0.58 2.3 4 0.06 0.10 3.3 49 0.20 0.09 0.16 50 0.23 0.18 0.56 51 0.56 52 0.36 0.20 6.9 53 0.31 54 0.2 1.9 3.8 55 0.16 0.12 0.14 56 0.11 0.10 0.11 57 58 0.12 0.10 0.59 59 0.19 0.41 2.0 so 0.18 0.28 0.72 61 0.11 0.02 0.73 62 0.19 63 0.13 0.71 7.0 61 0.0 0.11 0.85 65 0.1 0.26 0.65 66 0.08 0.57 4.0 67 0.10 0.12 0.39 82 0.21 0.27 1.9 83 0.15 1.10 .5 80 0.89 0.38 8.3 85 0.20 0.57 1.7 86 0.27 0.14 1.1 8 0.20 0.33 1.5 8 0.16 0.78 2.7 89 0.28 0.30 1.6 90 0.21 0.39 1.4 I? . T.""‘L . D” was: 0 ..~.\r..l.1 ‘Abfli .51 .. .mm-u 51 Table 1. (continued) Sampling Air Particulate Rainfall Rainfall Week Activigy iotig}ty Lot vity (into/m (pt/om day) (1:0 ml) 91 0.27 0.64 1.10 92 0.1 1.10 1.40 93 0.4 ,O.43 1.90 94 0.13 0.11 0.17 95 0.41 96 0.38 0.09 0.15 9 0.35 0.09 0.90 9 0.37 99 0.18 100 0.18 0.16 0.59 101 0.19 0.25 0.80 102 0.25 10 0.19 . 2.7 1 0.37 O. 1.2 105 0.26 106 0.25 1.20 1.9 107 0.20 108 0.45 109 0.21 0.12 0.90 110 0.09 111 0.19 0.15 0.55 hi. lime. ”(Lsm -....l‘mnflsm-Dor“ 52' Table 2. Radioactivity data used in analysis from Station 2 at.Monroe, Michigan. Sampling Air Particulate Rainfall Rainfall Week Activigy Actigity Activity (mic/m (uO/cm /day) (us/m1) 39 3.8 1.5 5.1 40 2.5 41 1.5 2.2 7.2 42 1.6 43 1.8 'O.31 3.8 44 1.1 0.66 3.4 45 1.1 “6 '00 00‘? 2.2 47 0.8 0.37 1.8 48 0.5 0.63 0.16 #9 0.24 0.10 0.20 50 0.28 0.12 0.39 51 0.113 52 0.90 53 0.35 51,- 00‘9 ‘08 6" 55 0.15 0.11 0.12 56 0.10 0.23 0.23 57 0.18 0.08 17.0 58 0.12 59 0.14 0.82 0.04 60 0.14 0.20 . 0.61 g1 0.09 0.28 0.66 2 63 0.11 0.49 3.9 64 0.06 0.08 0.58 55 0.05 0.24 0.52 66 0.05 0.26 9.1 67 0.13 0.10 0. 68 0.06 0.82 1.8 82 0.21 0.10 3.0 83 0.16 0.82 0.2 84 0.50 0.60 17.0 85 0.21 0.80 2.8 86 0.25 0.89 507 87 0.26 0.40 1.6 88 0.17 0.79 2.2 89 0.20 0. 1 1.2 90 0.26 0. 8 1.7 53 Table 2. (continued) Sampling Air Particulate Rainfall Rainfall Week Activi Actigity Activity (mm/m (us/cm /day) (pa/ml) 91 0.26 0.53 1.2 92 0.18 0.61 0.65 93 0.39 0.73 2.0 94 0.12 0.21 0.29 95 0.44 96 0.42 97 0.34 0.29 2.1 98 0.40 99 0.22 0.08 0.19 100 0.11 0.10 0.36 101 0.11 0.16 0.24 102 0.13 0.37 6. 103 0.07 0.15 1.8 104 0.16 105 0.26 0.47 4.7 106 0.23 0.13 3.0 107 0.23 108 0.13 8.9 109 0.27 110 0.10 111 0.23 0.22 0.93 112 0.08 0.24 2.0 113 0.96 0.08 0.55 114 0.34 0.11 0.58 .Ilnv 54 Table 3. Rainfall data used in the analysis from MOnrO. , Michigan e Sampling Rainfall Banyling Rainfall Week (inches) Heck (inches) 39 0.75 83 0.52 :0 0.02 34 0.10 1 2.97 0. 22 0.35 87 0.56 0.39 88 0.99 45 0.33 89 0.64 47 0.64 91 "‘ 48 0.78 92 .. 49 2.24 93 -- 50 0.72 94 -- 51 2.22 95 Trace 52 Trans 95 0.36 53 0.04 97 0.09 54 0.84 98 0. 55 2.02 99 0.98 56 2.60 100 0.28 57 0.07 101 0.50 58 0.50 102 0.93 59 0.52 182 0.03 60 0.Z9 1 Trace 61 1. 0 105 1.64 62 0.04 106 0.43 63 0.51 107 0.15 65 1.13 109 0.05 66 0.17 110 0.16 67 0.58 111 0.68 68 0.27 112 0.12 113 0.40 82 0.07 114 0.39 Tabl. 40 Sampling Week Radicactivi Lansing, Mic igan. fl \ 2 \\\\§SS<° W {\ 2‘92 dd‘d“‘d“ m“ UUNNNMd-‘d-‘NMMMN D MN-fi \\\\\

>\§>> \ 0\ n) m...- NU" \\ \ O\O\O\O’10\ 101010101X) n- {\ U'IUI vP >\§ U! ‘0 Air Particulate Activi (1810/ Rainfall data used in analysis from Rainfall A ti Activity (1107013531) (lac/m1) M f “ ‘ 4 «FM 479113-0401 00-4 0110 03410 d‘O‘O 01W 0‘0 ‘U‘I (DO 0.10-4n)mxooekmo~coaomomouooamuommucou 90.2 116:8 64.2 19.0 .0 ..e ImngN .eeeIe IOU #‘OOfiOO’t e e e e e e e e e e “MkmUOO-P’Ot U170“) 0 [DUI ‘0de 55 ’0 . 2. 3. 1.. 5. 5. 7. s. 9. 10. 56 W Setter, L.R. and Russell. 8.3.. “Radioactive Contami- nation of the Environment in the Cincinnati Area," _ a. - v... V: ; '11:: 4.43-2 1”, 511449 P Clgzna, 3.3.. Drynan, W.R., Smallhoret‘ D., I'Radicac-- ti ty in the Surface Waters of Tunas. Paper Presented at Texas Water and Sewage Work Short School, A and N College, (March 1961). Davis, J.A.‘ Logic, L.C.. Robinson, 3.8., and Heathing- ton. 3.3.. A Correlation Analysis of the Existionehip Between Weather Variables and Atmospheric R icactivity,' W, Vol. 4, Pergamon Press, (October 1960). Dunning, G.M. "ShorterbLived Fission Products in Fallout," W, 701. 4, Pergamon Press, (OOtOber 1 e Setter. 1.3., and Straub‘ C.P.. "The Distribution of Radioactivity tram Rain. saotio s or can Ge - mam. 01.39. W . um- - - List, R., ”Radioactive Fallout in North America from Operation Teazot ” U t er esion Qggggegt7319= 626, (1356?. it"; 3 Morgan, A. and Stanbury, 0.0., ”The Contamination of Rivers With Fission Products Prom Fallout ".523l32 ‘zgygigg. Vol. 5. Psrganon Press, (June 1961). Nader, J.S., Goldin, A.8.. and Setter, L.R., "Radioactive Fallout in Cincinnati Area.” W. 4611096 1 er 9 . Xilcawley, E.J., et al., "Measurement of Radioactive Fallout in Reservoirs ' a or o W ter W s W, 4611101 Haven er . Setter, L.R.. Regnier. J.B., and Diephaus, E.A.. ”Radio- active Surface Waters in the United States " 0%1959) . eric I e s , 5111377 N ..- u. n- uns“ an yv "sen-r... 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 57 W ( continued) ”Monroe Monitor Program,” A Report by the Division of Occupational Health of the Michigan Department of 0.8. Department of Health, Education, and Welfare, Public Health Service, W, 701. 2, N0. 4 (April 1961). Meinke, 11.11., "Observations on Radioactive Snove at Am; )Arbor, Michigan,” m, 701. 113. 545. (May 19 1 . Kahn, B. and Reynolds, 8.1., "Detemination of ladie- nuelides in Low Concentrations in Haters'(18urn§ W122. 50161 Goldin, 1.3., Nader, J.8., and Better, 1.3., ”The Detectability of Low-Level Radioactivity in Water,” figfil American Water Works Association, 45:73 Thomas, H.A., et al., "Radioactive Fallout in Mass- achusetts Surface Waters," 0 erican Water nggs Assgcigfiigg, 451552 (i933;- Unpublished Data, 0.8. Public H.610. Service, R. .1. Taft Sanitary Engineering Center, Cincinnati, Ohio, (1961). 0.8. Department of Comoros, Weather Bureau, W9 “959: 195°)- Dixon, W.J. and Massey, F.J., troduction To Stat s- W 2nd ed., McGraé-liflifl BEof'Com" 'pm','£L' nc., e1r o , ' (1957). Mills, F.0., Stat}et%cg% Methods, 3rd ed., Henry Holt and Company, ew or , .2 . Personal conversation with A.H. sickneier, Climatol- ogist for the State of Michigan. ‘. IA .‘fflu'u 83313? ESE. (33.1.1 Milli lleiliflilliilii Iiiiiliiiiiiiilllililiililfis 3 1293 03046 6530