V ‘h‘ V -fl-Qc-Qwo‘-..‘ .o-O .‘gf-fl‘of-O.‘W.Q ‘7 MAGNETIC MEASUREMENTS 0F 7 SAGINAW LOBE GIACI AL TILL- IN 'THE SOUTHERN PENINSULA 0F MIC. Q . :o . . HIGAN 1 : Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSITY . BRADLEY JAMES W 1968 , ‘ . ._ . .. _ a . . . . . ~ . . o ‘4 c v . o . . . u .‘n . . . p. .. . . . . . . . . .. c. . . r . . y . O ~ 0 . In: c I 4 . I . . . I .. . . . .. . . . . . I. C . . rpoauu...‘ ofuf flabe-Mo . .IanJa o‘- rv.. a O’~f ..Ia .. a.-I! a .... r..- ... Tut. .. . LIBRARY. " IIIIIIIIII I IIIIIIIIIIIIIIIIIIII . W” ,M _ 293008749545 I I . _ . f MAGNETIC MEASUREMENTS OF SAGINAW LOBE GLACIAL TILLS IN THE SOUTHERN PENINSULA OF MICHIGAN By 1 . . I (1" James w.'Bradley A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Geology 1968 ABSTRACT MAGNETIC MEASUREMENTS OF SAGINAW LOBE GLACIAL TILLS IN THE SOUTHERN PENINSULA OF MICHIGAN by James W. Bradley As the continental glaciers of Wisconsin Age covered the Southern Peninsula of Michigan, they carried and deposited the glacial till which now covers Michigan. This glacial till contains ferromagnetic material which past research has indicated causes changes in magnetic character from one glacial feature to another. Utilizing this information, vertical magnetic intensity readings and magnetic susceptibility readings of glacial soils were obtain- ed for several glacial features deposited by the Saginaw lobe. Among the features investigated were terminal moraines, ground moraines, a waterlaid moraine and its surrounding lake bed, and a glacial lake shore line. The vertical magnetic intensity data which were taken in groups of closely spaced stations indicates that the deviation from the representative readings for each small area varies predictively from one glacial feature to another as do the glacial soil magnetic susceptibility readings. Terminal moraine vertical magnetic intensi- ty readings show greater variations than the ground moraines. This is demonstrated by the 95.5 percent confidence interval of the standard deviation of the variations from a representative reading of the vertical magnetic intensity. The terminal moraines have a 95.5 percent confidence interval of 2.50 to 2.72 gammas while the ground moraines have a 95.5 percent confidence interval of 1.87 to 2.11 ii gammas. The individual morainic systems also show different variations in the vertical magnetic intensity. The other individual features investigated also demonstrate anamalous variations in vertical magnetic intensity. These data were analyzed in several ways which verify the previously mentioned relationships. Also a special study associates the strength of local vertical magnetic gradients with local variations of vertical magnetic intensity. The magnetic susceptibility measurements made on glacial soils collected from terminal and ground moraines also produce character- istic variations. The average magnetic susceptibility of terminal 6 moraine glacial soils is 63.96 x 10' 6 C.G.S. units while ground moraine soils average 27.5 x 10- C.G.S. units. It is also established that there is a direct correlation between local vertical magnetic intensity variations and magnetic susceptibilities of glacial soil. Finally, the data are evaluated to show the value of these measurements in geographically defining individual glacial features. 111 ACKNOWLEDGEMENTS The author wishes to express his sincere appreciation to Drs. William J. Hinze, James w. Trow and Hugh Bennett of Michigan State University for their cOOperation and assistance throughout this investigation. This study is an outgrowth of previous work carried out by Dr. Hinze and the Geology Department of Michigan State Univ- ersity (Hinze, 1961) and subsequent work by Thomas L. Lawler (Lawler, 1962). The author wishes to furthermore express gratitude to the Department of Geology of Michigan State University for the author's use of its instruments and facilities during this investigation. iv TABLE OF CONTENTS Page INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . 1 General Statement . . . . . . . . . . . . . . . . . . . . 1 Statement of Problem . . . . . . . . . . . . . . . . . . 2 Area of Investigation . . . . . . . . . . . . . . . . . . 2 Observation Sites . . . . . . . . . . . . . . . . . . . . 5 PREVIOUS INVESTIGATIONS . . . . . . . . . . . . . . . . . . . 16 Studies of Local Magnetic Disturbances . . . . . . . . . 16 Studies of Magnetic Susceptibility of Glacial Till . . . l7 LOCAL MAGNETIC DISTURBANCES . . . . . . . . . . . . . . . . . l9 Instrumentation . . . . . . . . . . . . . . . . . . . . . 19 Field Procedure . . . . . . . . . . . . . . . . . . . . . 19 Data Reduction and Analysis Procedure . . . . . . . . . . 22 Results: Percentage Evaluation . . . . . . . . . . . . . 23 Results: Standard Deviation and 95.5 Percent Confidence Interval O O O O O O O O O O O O O O O O O O O O I 0 314' Results: Significance Tests and Reliability Study . . . NO Results: Traverse Studies of a Glacial Lake Shore Line . A2 Vertical Gradients . . . . . . . . . . . . . . . . . . . h5 Evaluation of Field Procedure . . . . . . . . . . . . . . N6 MAGNETIC SUSCEPTIBILITY OF GLACIAL SOILS . . . . . . . . . . . 53 Measurements . . . . . . . . . . . . . . . . . . . . . . 53 Results: Average Magnetic Soil Susceptibility of Glacial Features O O O I O O O O O O O O O O O O O O O O O O 51" THE CORRELATION BETWEEN MAGNETIC SUSCEPTIBILITY AND THE STANDARD DEVIATION OF THE LOCAL VARIATIONS OF THE RELATIVE VERTICAL MAGNETIC INTENSITY . . . . . . . . . . . . . . . 56 III I A A A All: I 'l [I l [I [I I ll TABLE OF CONTENTS CONCLUSION . . . . . . . . SUGGESTIONS FOR FURTHER STUDY . BIBLIOGRAPHY . . . . . . vi II. N N l I [III I I‘ll... Ill fl 7. 10. ll. 13. 1’4. LIST OF FIGURES mpOfStUdyareao00.000000000000000 Map of area where the Charlotte terminal and ground moraines were studied . . . . . . . . . . . . . Locations at which vertical magnetic intensity measure- ments were taken on the Charlotte terminal and groundmoraineSooooooo00.00.0000. Locations at which glacial soil samples were taken for magnetic susceptibility measurements . . . . . Locations at which glacial soil samples and magnetic intensity measurements were taken together . . Locations at which glacial soil samples and magnetic intensity measurements were taken on the Kalamazoo terminal and ground moraines . . . . . . . . . Locations at which glacial soil samples and magnetic intensity measurements were taken on the plexus area terminal and ground moraines . . . . . . . . . Location of vertical magnetic intensity measurements on and around the waterlaid moraine and glacial lake ShorelinenearChapin............ Pattern of vertical magnetic intensity observations in anarray.o................. Deviation from the representative value curves for the Charlotte terminal and ground moraines . . . . . . Deviation from the representative value curves for the Kalamazoo terminal and ground moraines . . . . Deviation from the representative value curves for the first and second terminal moraines of the plexus area0.0000000000000000... Deviation from the representative value curves for the plexus area terminal moraines and the surrounding ground moraines . . . . . . . . . . . . . . . . . . Deviation from the representative value curves for the waterlaid moraine and its surrounding lake bed vii Page ll 13 1h 20 25 26 27 28 29 JIIIIIII‘III'I-Il'llllllul [II I I'll-It'll FIGURE 15. 16. 170 l8. 19. 20. 21. 22. 23. 2A. 25. 26. 27. LIST OF FIGURES Deviation from the representative value curves for the terminal moraines which were measured in the Saginaw 10b 6 O O O O O O O O O O O O O O O O O O O O O O 0 Deviation from the representative value curves for the ground moraines which were measured in the Saginaw lObe O O O O O O O O O O C I O O O O O O O O O O 0 Deviation from the representative value curves for the combined ground moraines and combined terminal moraines of the Saginaw lobe features measured in this study . . . . . . . . . . . . . . . . . . . . The 95.5 percent confidence intervals of the standard deviation of local magnetic intensity disturbances of the terminal and ground moraines of the Saginaw lObe O O O O O O O O O O O O O O O O O O O O O O O The 95.5 percent confidence intervals of the waterlaid moraine near Chapin and its surrounding lake bed . The 95.5 percent confidence intervals of the terminal moraines and ground moraines of the Saginaw lobe . Vertical magnetic intensity and standard deviation of Traverse I across the glacial lake shore line . . Vertical magnetic intensity and standard deviation of Traverse II across the glacial lake shore line . . The distribution of station vertical magnetic gradients fromarraySASEndA70000000000000. Three vertical magnetic intensity and vertical magnetic gradient models for a buried sphere . . . . . . . Curves showing the difference in average standard deviation varying the number of groups used to calculate the average . . . . . . . . . . . . . . Least squares curves depicting the correlation between magnetic susceptibility and standard deviation . . Standard deviation and magnetic susceptibility profiles across the Kalamazoo terminal and ground moraines viii Page 31 32 33 36 37 38 1+3 1m In I48 50 57 58 LIST OF FIGURES FIGURE Page 28. Standard deviation and magnetic susceptibility profiles across the Charlotte terminal and ground moraines . 6O 29. Standard deviation and magnetic susceptibility profiles across the Charlotte terminal and ground moraines using a grid system to average the values . . . . . 61 ix LIST OF TABLES TABLE Page 1. Probable Significance of Differences in Sample Variance by the Variance Ratio Test . . . . . . . . . . . . . N1 2. Groups Not Used in Finding Average Standard Deviation . 52 I II I I I'll l I [III I [I [III II [I [I III. till INTRODUCTION General Statement Glacial deposits have received only limited attention by geo- physicists. The primary concern of geophysical studies has been to delineate ground water reservoirs utilizing seismic, electrical, and gravity methods. This research uses magnetic measurements to investigate the magnetic character of glacial deposits. The magnetic character is a function of the ferromagnetic mineral content of the ultimate source of the deposits, the distance the deposited material has traveled from the source, and the depo- sitional process. This study is a continuation and further development of previous work which indicated a correlation between specific units of glacial deposits and amplitude variations of local magnetic disturbances and later work which confirmed the hypothesized relationship. In the latter work local magnetic amplitude variations were shown to have not only characteristic signatures over drift deposited by a specific glacial lobe but also over glaciomorphic features within each lobe. The present study further characterizes glacial features by their magnetic character- istics. Two types of magnetic measurements were utilized in this study. The first was obtained by observing the relative vertical intensity of the Earth's magnetic field. Within a rather small geographic area, the magnetic field is distorted abruptly by ferromagnetic minerals contained within the glacial drift. Occasionally extreme local magnetic disturbances occur from cultural features, but these are usually differentiated from 1 II I IIIII I II II I IA l l A l A [I I. I. III II I fl l I 2 natural variations by their extreme amplitude and thus can be eliminated. The second type of measurement consisted of observing the magnetic susceptibility of the glacial drift. This fundamental property is directly proportional to the ferromagnetic content of the drift material. Statement of Problem The objective of this study is to demonstrate the changes in local vertical magnetic intensity variations of the Earth's field associated with Saginaw lobe terminal moraines, ground moraines and other glaciomorphic features. The magnetic susceptibility of the glacial till was measured to determine the correlation of this parameter with known changes in glacial geology and local variations in the vertical magnetic intensity. The results of this study will be useful in predicting magnetic noise levels to be encountered in future magnetic surveys and they may suggest a method by which glacial features can be delineated from one another. Area of Investigation The observation sites are grouped along a diagonal line striking northeast through the center of the Southern Peninsula of Michigan. The area in which field work was conducted lies between longitudes 8&0 and 860 West and latitudes N20 and hho North and includes Calhoun, Barry, Eaton, Shiawassee and Saginaw Counties. The area roughly coincides with a perpendicular to the strike of the glacial features. The glacial till in this area is a result of continental glaciation during the Wisconsin Glacial Age. This 3 period of time occurred during the Pleistocene Epoch which embraced many periods of fluctuating glaciers covering the State of Michigan. The Wisconsin Age was the latest of these periods. During the Wisconsin Age four ice lobes protruded from their gathering grounds in Canada into what is presently the Southern Peninsula of Michigan. These four lobes were the Erie, Huron, Saginaw and Lake Michigan lobes. The general area in which features of these lobes occur are shown in Figure l. The surface t0pography of the study area ranges from hilly in the southwest to flat lowland type areas with occasional small ridges in the northeast. The glacial morphological features investigated within this area include terminal moraines, ground moraines (till plains), waterlaid moraines, glacial lake shore fines and lake bed deposits. The bedrock of the area under investigation is primarily Mississippian and Pennsylvanian limestones, dolomites, shales and sandstones. The glacial till consists of rock debris obtained from the bedrock which the glaciers have overridden. For this reason, the glacial till includes Precambrian igneous and metamorphic rocks some of which have high magnetic susceptibilities. Thwaites (19h6) explained that although the glacial till will contain primarily materials which have been carried only a moderate distance by the glacier, the till also will include a significant amount of materials which originated some distance from their present 10- cations. This explains the presence of large quantities of ferromagnetic material in glacial till deposited in an area where the bedrock consists of low magnetic susceptibility rocks. III 1‘51 l I lII'l I'lllllllllll lll'lll‘ll [III [I lI'lllll.‘ [. Huron Lobe FCOIures I D . .0 C) .1 SCQInOw Lobe Features I g l. A a 3: Michigan Lobe Features a , /‘ Efle Lobe Features. / / ' Figure 1. Map of study area. tures Fee 5 Anderson (1957) showed that there are significant lithologic variations in the material deposited by the major lobes of the Wisconsin Glacial Age in the central lowland of Michigan. Observation Sites The first glacialogical area upon which measurements were made was the Charlotte terminal and ground moraine of the Saginaw lobe. The general location of this area is shown as A in Figure 1. A remnant of the terminal moraine borders the western edge of Charlotte in Eaton County and extends west for approximately four miles, and the ground moraine occurs on the northern border of the terminal moraine in Figure 2. The sites at which vertical magnetic intensity observations were made appear as circles in Figure 3. Vertical magnetic intensity reading series on the Charlotte terminal moraine are designated by A while readings on the ground moraine are designated by B. The locations at which soil samples were collected for magnetic susceptibility study in the Charlotte area are shown in Figure A. Many of the magnetic susceptibility samples were taken in conjunction with vertical magnetic intensity reading arrays. The sites at which these measurements co-exist appear in Figure 5 as triangles within a circle. Figures 2, 3, h and 5 are the region defined as Insert I on the enclosure map. The Charlotte moraine has been correlated with the Cary substage or Middle Wisconsian, (Thwaites, l9h6). Leverett and Taylor (1915) describe the Charlotte moraine as being different from most morainic systems in that its outer border is vaguely defined. They state: "The nearly plane till tract on the south" rises "gradually in the space of one to two miles or more into a A ‘ll IIIII‘I‘II N . Ill r [([{l 85° EATON COUNTY I y fi F—u..- -... .. ............. an O | 2 3 Kg Charlotte Terminal Moraine § Miles 7 Charla? Is Ground Moraine % Figure 2. Map of area where the Charlotte terminal and ground moraines were studied. 85° N EATON COUNTY ? ,- ' 7 .. /’/.’ I ZE‘E%/Aé%¢f/z POTT R LI /11 1 BOW 0'37 may/[@V/vé/A Br¢/’7 / / 7/ a/ d 4 //. . v \CHESTERV/t WIS? ,/3/ a ‘ ‘ fi ngfi€%\ {1: : ye ' I I @§E%RLOTFE E Iayfix j\~\*K\\A (3 K a \s I. J [A Ff I5 w a . N V \ W T Jig-13 C“: O I 2 3 4 5 Chariot“ Terminal Moraine §\ Miles 1/ Charlotte Ground M Drains (CI/2.7:: @141 Figure 3. Locations at which vertical magnetic.intensity measurements were taken on the Charlotte terminal and ground moraines. 85° /L [Ir (f///3/T, :77“ X / | @/ 7/42/11,} POT‘IER LE % // “CHESTER //87% \\\ y“ \er \‘\l1 ‘ \ _ \ QR -—<>2! \ :r I I I j . \\§ 0 l 2 3 4 5 Charlotte Terminal Moraine \\ Miles “Charlotte Ground Moraine {,{f/ Figure A. Locations at which glacial soil samples were taken for magnetic susceptibility measurements. ‘ I. l l I I [III I I ‘ll [([fl 85° EATON COUNTY :5 POTTE : ILLE CHARLOTTE Scam: . §$§ O . 2 3 4 5 Charlotte Termlnal Moralne Q Mfles Charlotte Ground Moraine 4g Figure 5. Locations at which glacial soil samples and magnetic intensity measurements were taken together. ll l||l ‘II [I [l.[([l{l[([[ lO pronounced moraine." The structure of the moraine indicates that it was created by a brief halt in the recession of the Saginaw lobe. Leverett and Taylor feel that the large number of eskers leading into the moraine on its northern border, as well as the outwash and ice border drainage, verify this conclusion. The lithology of the drift is said to be exceedingly variable through- out the Charlotte morainic system. Only very minor amounts of stiff clayey material are to be found, although the main portion of the surface would be classified as till. This indicates that the moraine has been very highly waterworked. The ground moraine (till plain) which occurs to the north is for the most part very smooth. However, swampy depressions are encountered frequently. The Kalamazoo morainic system which was studied next (area B, Figure l) is of the same glacialogical age as the Charlotte moraine, but was deposited slightly earlier. The observation sites are illustrated in Figure 6 which is Insert II on the enclosure map. This particular portion of the Kalamazoo terminal moraine and ground moraine lies north-northeast of Battle Creek and is in Calhoun, Berry and Eaton Counties. The sites marked F in Figure 6 are measurements on the ground moraine and G are on the terminal moraine. A magnetic susceptibility sample was taken at each of these sites and was therefore identified by the same symbol. The moraine trends parallel to the Charlotte moraine previously discussed. In the area studied, the moraine is approximately ten miles in breadth. Leverett and Taylor observed that the outer border of the moraine is only slightly higher than its outwash .1“ . l 1' -. J ‘<. I?“ . «of. , 2T; ‘ Kal BATTLE CRE O|2345 Mlle: mazoo Terminal Moraine amazoo Ground Moraine IK 85° Figure 6. Locations at which glacial soil samples and magnetic intensity measurements were taken on the Kalamazoo terminal and ground moraines. llllll. llllltl‘, ll([{ I‘ll-‘(IIII [\(II. l2 plain, and on the inner border in Eaton and Berry Counties there is a gradual transition from the moraine to the bordering till plain. The ground moraine between the Kalamazoo and Charlotte moraines is undulating and of low relief. In general, they felt that the glacial drift in the Kalamazoo morainic system is rather sandy, but in Berry and Eaton Counties the glacial drift becomes very clayey toward the inner border. Leverett and Taylor also observed "great numbers of bowlders", the majority of which were granitic along the southern edge of the moraine as well as some distance back from the border in "belts" which are traceable for miles. The next area investigated is very interesting and somewhat more complex. The area is located in Figure l as C. This location is describable in glacialOgical terms as a plexus area. In this case the glacier progressed forward and then stopped leaving a terminal moraine. It then retreated for a period of time, pro- gressed forward once again and halted, depositing another terminal moraine contiguous with the first moraine. Figure 7 which is Insert III in the enclosure shows Just such an area at the Clinton- Shiawassee County border near Laingsburg. In this area, the first terminal moraine which predominates makes up the southern part of the morainal complex. The location of the vertical magnetic intensity measurements made in this area are shown in Figure 7 as small squares. Each array of readings is represented by a C on the first moraine, D on the second moraine, and an E on the surround- ing ground moraines. Again magnetic susceptibility soil samples were taken at all vertical magnetic intensity locations. The next glacial features studied are shown in Figure 8 which ill['|l[[[lllur|fllrl(ll 13 Z CLINTON CO —_— IN GHAM COU NTY Plexus Area Terminal Moraine: Plexus Area Ground Moraine: §HIA1IASSEE QOUNTY 2 3 4 5 Miles Figure 7. Locations at which glacial soil samples and magnetic intensity measurements were taken on the plexus area terminal and ground moraines. I III. 'I I\I|| I‘ I III I l I l f . lh ' GINAW comm! GRATLOT CO NT_Y___£ 6mm Lake 5m Line CLINTON CO TY I SHIAWASSEE COUNTY “\ Scale r71 ‘ ‘ ‘ j Wanlam 0'2345 r Moraine lflhs Figure 8. Location of vertical magnetic intensity measurements on and around the waterlaid moraine and glacial lake shore line near Chapin. Illl. . 'I.‘ I ‘ 'II [I r Ill \ \ \,“ C) :: : : e : T : ev, :vs 2 4 6 8 l0 EIQS Deviation From Representative Value in Gamma: Figure 12. Deviation from the representative value curves for the first and second terminal moraines of the plexus area. % at Stations r. :7) 70‘ 60.n- m C? 40‘ I l r -—-- PLEXUS AREA TERMINAL MC'RAEHES ""$URROUNDII‘IG GROUND MORAINES L .l \ \ ' \ \_ \ \ \ \ ‘.§_ ‘\ -r- L A I l ‘1‘ ——A l A . - r ‘r 1 r r w r T Figure 13. Devia the plexus area t z A 4 e 8 1682105 Devlction From RepresentativoVaIuo in Commas To the surrounding ground 901’ ‘36 at Stations 29 “*WATERLAIO MORAINE *SURROUNDING LAKE BED 30.. 20.. ‘ \ \\ \. '0" \\\ \ I \ .. \ \ ,a...‘ \\.”” “\‘~__ 0 : ¢ : s— : x. z 4 ¢ ¢——. 2 4 6 8 I0 I05 Deviation From Representative Value in Gammas Figure 1h. Deviation from the representative value curves for the waterlaid moraine and its surrounding lake bed. 30 The percentage of deviation curves for each terminal moraine investigated was plotted in Figure 15. It appears that there is a decrease in the deviation from the representative value as we progress north-northeast. That is to say, the earlier formed terminal moraines have a higher deviation from the representative values than those deposited later. Figure 16 demonstrates that the same relationship exists among the ground moraines of the Saginaw lobe. The Kalamazoo ground moraine has the largest deviation from the representative value, the Charlotte ground moraine has less, and the plexus area ground moraines have the smallest deviation. The last percentage evaluation graph presented in Figure 17 contrasts the combined terminal moraine percentages of deviation from the representative value with the ground moraine percentages of deviation for the areas investigated in the Saginaw lobe. As could be anticipated, the ground moraines show the lower per- centage of deviation from the representative value of the two. It should be noted that in this graph as well as the others, normal distribution of the values is demonstrated. It is of course also important to point out that since the least count of the instrument is 2.33 gammas, the assumption must be made that the values below the least count follow the normal distribution curve. The only way in which derived values below the least count can be regarded as valid is by accepting this assumption. % at Stations Figure 15. 31 ---- PLEXUS AREA TERMINAL MORAINES --'- CHARLOTTE TERMINAL MORAINE '----- KALAMAZOO TERMINAL MORAINE .\./‘ "m.—— s j Ems Deviation From Representative Value in Commas Deviation from the representative value curves for the terminal moraines which were measured in the Saginaw lobe. at Stations ' g 0 Figure 16. 804. 70- 60‘ MV- 32 ----- PLEXUS AREA GROUND MORAINES -—-—- CHARLOTTE GROUND MORAINE -——- KALAMAZOO GROUND MORAINE .>RT/”/” . '\\~ \\\\\\\:\\.\\. ‘\ ‘\\ \\ C\\\ :'\ - “ A l A 1 fi _ I I r : v 2 4 e_ s ‘ l02l0.5 Deviation From Representative Value in Gammas Deviation from the representative value curves for the ground moraines which were measured in the Saginaw lobe. [lllllll"..|l[lll|[i‘l 33 1 80‘ 50 \\ —— TERMINAL MORAINES °/o at Stations ’45:, 4 Qt. --- GROUND MORAINES 204i. IO.. é : 4 e 8 l0 2.10.5 Deviation From Representative Value in Gammon Figure 17. Deviation from the representative value curves for (the combined ground moraines and combined terminal moraines of the Saginaw lobe features measured in this study. 3h Results: Standard Deviation and 95.5 Percent Confidence Interval Since the last section indicates that the deviations from the representative value of the vertical magnetic intensity measurements have a unimodel, normal distribution, it will be possible to further examine the data with slightly more sophisticated statistical techniques. The standard deviation was calculated for each group according to the equation for standard deviation presented below, taken from Moroney, (1960, p. 61 and 62). 6‘ = \/ Inc-if N where 6 = standard deviation (x-x) = observation minus the representative observation in gammas N := number of observations These values for standard deviation were then averaged for the appropriate glacial feature. Using the average standard deviation the 95.5 percent confidence interval was calculated for each glacial feature using the following equation from Moroney (1960, p. 2A2). 95.5 percent confidence interval 6 i- 2 6 v5 is the standard error where 6 V2N Moroney (1960) states that 95.5 percent of "the pOpulation standard deviation will not differ from the large sample standard deviation by more than two standard errors." 35 Figures 18, 19, and 20 show the 95.5 percent confidence interval of the average standard deviation in gammas for the various glacial features. The Charlotte morainic relationship shown in Figure 18 demonstrates that the terminal moraine has a 95.5 percent confidence interval of 1.71 to 1.89 gammas and an average standard deviation of 1.80 gammas, while the ground moraine has a standard deviation of 1.h5 gammas and a 95.5 percent confidence interval of 1.59 to 1.31 gammas. The 95.5 percent confidence interval of the Charlotte terminal moraine and the Charlotte ground moraine are shown to be exclusive of each other. The average standard deviation of the Kalamazoo ground moraine is 3.0 gammas while the 95.5 percent confidence interval ranges from 3.72 to 3.30 gammas. The average standard deviation of the Kalamazoo terminal moraine is h.28 gammas with a 95.5 percent confidence interval of 3.27 to 5.29 gammas. This means the ground moraine and terminal moraine are not exclusive of each other as far as the 95.5 percent confidence interval is concerned; however, with more data it is probable that they would be. Also shown in Figure 18 are the 95.5 percent confidence intervals of the plexus area features. The average standard deviation of the latest terminal moraine in the plexus area is 1.07 gammas and the 95.5 percent confidence interval is 1.23 to 0.91 gammas. The earliest terminal moraine of the plexus area has an average standard deviation of 2.hl gammas and a 95.5 percent confidence interval of I 0.37 gammas. Also shown in Figure 18 are the 95.5 percent confidence intervals of the combined terminal moraines of the plexus area and the combined ground moraines which surround them. The terminal moraines have a standard deviation of 36 wz. b 0 E O 8 m c: a.) 0 '0 z: in ‘§ 3: < .J g (D 0.2-- 2 ‘° a 0’) 2 '0- 3 ° 0 a: '0- a: c D 0 en .- X h.) ‘_ ‘F—T Figure 19. The 95.5 percent confidence intervals of the waterlaid moraine near Chapin and its surrounding lake bed. 38 mac...— 3S.OZ Morainee 1.52 175 2.31 waterlaid Moraine 1.80 90 3.24 2.90 uo omtmzuh2. o_._.mzu<2 ._ "com- . V 05. H m» . a z \ Gum V sawuiae u! uouegnea piapuais hit .ooHH shone mama Homosam mop mmouoo HH omuo>mue mo sowpmw>oc duodenum was hpfiwmopafi capoawms HoOHpuo> .mm ouswfim .3... E one—.55 one cow can con emu CON On. 00. on O L. a .o. _ com . 23:3 139% o a: 026: 29:33 omgzfim « m ocm 1 n — m a G .m 2.0 T .m m. 29253 omgzfim .m 08 - .n .m a m. .m 0mm - - v m. H OOmT .m e V 05 r . w HH m» 2 \ Omm h sawuiog u! UOIiDIAGQ pmpuais 1+5 intensity change over the glacial shore line is shown in the upper curve of the two figures. It is apparent from the graphs that both the standard deviation and relative vertical magnetic intensity indicate a magnetic distortion associated with the shore line. Very likely the influences on the Earth's magnetic field are due to deposits sorted by water action at the shore line. Local variations in vertical intensity and similar evaluation techniques may prove fruitful in the following and in discovering ancient lake shore lines. Vertical Gradients A study was designed to determine if there are significant variations in the vertical magnetic gradient associated with high and low variations in vertical magnetic intensity. Two arrays of stations were reread utilizing this technique. One array has a large variation in the local vertical magnetic intensity and one has a smaller variation in the vertical magnetic intensity. Each station was first read in the normal procedure. After this, each station was reoccupied with a specially constructed nonmagnetic tripod which placed the instrument approximately five inches from the surface of the soil. The change in elevation of the instrument was 2.88 feet. The two arrays which were reread were A5 and A7 which are also used in the reliability analysis. Array A5 which has a greater variation in the intensity has an average difference in high and low readings of 9.6h gammas and an average vertical gradient of 3.3h gammas per foot. Array.A7 has an average difference in high and low readings of 2.89 gammas and an average vertical gradient of 1.00 gamma per foot. The percentage distribution of the vertical gradients for the stations in each array are shown in Figure 23. It can be clearly seen that #6 A5 has a larger percentage of higher vertical gradients. Three models were constructed in order to ascertain the nature of the variations in vertical gradient. The models appear in Figure 2A. The models are calculated for a sphere which is 10 percent magnetite and exists in a vertical magnetic field of 0.6 oersted. The graphs show the vertical magnetic intensity values for high and low instrument elevations on the left and the vertical gradients on the right. The first model is calculated for a sphere whose radius is 1 foot which is 5 feet below the surface. This model has a maximum difference of 66 gammas between high and low instrument readings and a maximum vertical magnetic gradient of 23 gammas/ft. The second model also was calculated for a sphere whose radius is 1 foot but it is buried 8 feet below the surface. This model has a maximum difference in vertical intensity of lh.28 gammas and a maximum vertical gradient of n.96 gammas/ft. The last model was computed for a Sphere which is buried 5 feet deep and has a radius of 6 inches. It has a maximum change in magnetic intensity of 8.29 gammas and a maximum vertical magnetic gradient of 2.88 gammas/ft. The observed values make all three models plausible. These models also demonstrate that 10 feet is a good station spacing. Features such as these spaced 10 feet apart have vertical magnetic intensity anomalies which are nearly independent. Evaluation 9: Field Procedure An evaluation of the field data was made to determine if the field procedure could be modified to make the time consuming task of gathering data more efficient. This could be accomplished by cutting down on the number of groups used in an array wherever possible, and A7 l00 ~ 90 - 80 - 70. g 60 b ‘5 A7 3; Average Standard Deviation I 4.3! gammas Average Standard Devitian- [70 gammas m 50 . Average Vertical Gradient-3.34qam/ft. Average Vertical Gradientswgam/ft. ‘6 ~2 ° 40 - 33. 20 - l0 - O O 174 3.48 5.22 6.98 870 [0.44 l2.l8 ‘ 0 I74 3.48 Difference in Gommas / Foot Figure 23. The distribution of station vertical magnetic gradients from arrays A5 and A7. 1+8 .oaosmm powuso_m how maovos psoaomew capoamma Hmoapuo> can hpflmoopcfl capoowms HmOHpNo> mouse .:m onsmah . 33.36;. coat...» Akinkihéhthkk A x x \\ \o\ x \N \.\\ /\..N.\ihk\q\ixihkki\ 1223.; 23:552.. 30.. I. a a 8:32. .5555 goalie .o... 5 8:30.. .2523: .9. A a a ON? ON. A. if LB 0' A.- 0' L ...a E t :3» + . boo: 9:23 z. 3.. of $99.5 0.523: 32935.5 .. 9:23 2. :6sz 43:5; 8.. cm 3. c 8; 8: co: 1+9 adding more groups wherever necessary. Six arrays of vertical magnetic intensity readings were used in this analysis. Three of these arrays have high average standard deviations of the variations of vertical magnetic intensity while three have low standard deviations. Beginning with nine group arrays, each array was decreased by one until only two groups remained in each array for purposes of calculating the average standard deviation. The specific number of groups removed from each array was repeated three times. Each time different groups were eliminated and an average standard deviation was calculated for that array. The difference in the averaged standard deviation for each reduced array and the original nine group array was calculated. The average difference in standard deviation between the reduced arrays and the nine group array was determined for each set of arrays. The results are presented in Figure 25. Arrays G3, F1 and Alh were considered high standard deviations. The average standard deviations for these three arrays is 2.60 gammas and the 95.5 percent confidence interval is 2.60 13.33 gammas. The average difference in standard deviation of these three arrays for each number of groups used is shown as curve A in Figure 25. As can be seen the difference in average standard deviation exceeded the 95.5 percent confidence interval of the standard deviation of the nine group array when the number of groups used fell below seven. The low value standard deviation arrays, A9, A15 and D1, were process- ed in the same manner. The average standard deviation for these arrays is 1.38 gammas while the 95.5 percent confidence interval was 1.38 i .17 gammas. Computed values for these arrays are represented as curve B in Figure 25. It is apparent that a minimum of five .50 .55 0 .50 4* 'A 0| if a. 9 hi 8' 95.5% CONFIDENCE __ INTERVAL LIMIT 0: 9 To 8" In Average Standard Deviation in Gammas .ZOT I 95.5% CONFIDENCE l __ INTERVAL LIMIT .lsT ‘———""“ 0 Difference .05-- G. A A I L_ I I V Y e 7 e A 3 2 Number Of Groups Used Figure 25. Curves showing the difference in average standard deviation varying the number of groups used to calculate the average 0 51 groups could be used and still remain within the 95.5 percent confi- dence interval for the standard deviation of the averaged arrays. The groups which were eliminated from each average standard deviation calculation are shown in Table 2. com: .o: 365 3.9.8 x 52 . 322.23 e. a o m o n n n v e e n n m N m N . . . TBS Goa... x x x .x x x x x m x x x x x x x x m. x x x x x x . x x . x N x x x x x x x x x x w ill x x x x x x x x x n x x x x x x x x a. x x x x x x x x x m x x x . x x x x x N x x x x x x x x . anomo zo_b<_>wo om< 0252.“. Z_ cum: #02 manomw "N .womdh MAGNETIC SUSCEPTIBILITY OF GLACIAL SOILS Measurements The measurement of the magnetic susceptibility of glacial soil had a twofold purpose. One purpose was to try to find evidence that the amount of local variation in the relative vertical intensity of the Earth's magnetic field discussed previously could be correlated with the magnetic susceptibility of the soil. The other purpose was to examine the possibility that magnetic susceptibility of glacial soil could be used to identify the extent if not the nature of a glacial feature. The instrument used to measure the magnetic susceptibility was the Magnetic Susceptibility Bridge, Model MS-3, made by GeOphysical Specialities, Division of Minnetech Labs. Most samples used in magnetic soil susceptibility measurements were taken coincident with vertical magnetic intensity observation locations. This permitted the direct correlation of these two types of observations. Extra magnetic susceptibility samples were taken independent of magnetic intensity observation locations on a profile x-x' (Figures h and 5) on the Charlotte terminal and ground moraines. Each sample was taken from a hole which was dug below organic soil into what is probably the Al soil zone. At each sample location a controlled volume of soil was taken by driving a pipe into the soil at the bottom of the hole to a constant depth and withdrawing the sample in the pipe. These samples were then transported to the geophysical laboratory where they were tested for their magnetic susceptibility. In each instance the measurements were repeated, repacking the soil and making sure that each time it was packed it had approximately the same volume as when it was extracted in the 53 5h field. It was found that the measurements had excellent repeatibility. The readings obtained were multiplied by a constant and were thus converted into magnetic susceptibility in C.G.S. units. These values have not necessarily been proven absolute though they are correct relative to each other. The magnetic susceptibility readings and converted values for each sample location are presented in Appendix B. Results: Average Magnetic Soil Susceptibility 9£_Glacia1 Features The three areas in which magnetic susceptibility samples were taken are: the Charlotte morainic system (susceptibility sample stations 1 to 27), the Kalamazoo morainic system, and the plexus area glacial features. The stations where samples were taken in the Charlotte moraine are shown in Figures 5 and 6. The stations in the Kalamazoo morainic system and the plexus area morainic features are in the same location as the relative vertical magnetic intensity arrays (see Figures 5 and 7). The average magnetic susceptibility value for the Charlotte terminal moraine is h3.h x 10"6 C.G.S. units. The average susceptibility for the Charlotte ground moraine is 17.75 x 10.6 C.G.S. units. The average susceptibility for the Kalamazoo morainic system is 91.79 x 10.6 and 31.81 x 10.6 C.G.S. units for the terminal and ground moraines respectively. The magnetic suscepti- bility of the soil in the plexus area terminal moraines is 56.7 x 10"6 C.G.S. units while the ground moraines have an average of 6 32.9 x 10' C.G.S. units. The average susceptibility for all ground moraine soils is 27.5 x 10- C.G.S. units. The average magnetic susceptibility for the terminal moraine soils measured is 63.96 x 10.6 C.G.S. units. It is obvious that the average magnetic 55 susceptibility of the terminal moraine soils is consistently larger than the ground moraines. ANDARD DEVIATION OF THE LOCAL VARIATIONS OF THE RELATIVE VERTICAL MAGNETIC INTENSITY Figure 26 relates magnetic susceptibility and standard deviations of local variations in vertical magnetic intensity for points at which both measurements were made. Each graph includes a first degree least squares computed line. This line was computed for the points plotted on the graph. Parts A, B and C demonstrate the correlation between the two magnetic properties of the soil for the Charlotte, Kalamazoo and plexus area morainic soils respectively. Part D is a composite of parts A, B and C. The similarity in slopes of the least squares lines depict the correlation between magnetic susceptibility and variations in vertical magnetic intensity. When the magnetic susceptibility goes to 0.0 there is still a standard deviation of the vertical magnetic intensity variation. This could be the result of natural phenomena such as remnant magnetism. It also could indicate that the correlation does not necessarily fit a first degree curve. There is also the possibility that it could indicate an absolute error in readings. In an effort to find more ways of evaluating the data, a series of profiles involving both magnetic susceptibility of the soil and the local variations in the relative vertical magnetic intensity were constructed from the data. Figure 27 is a profile drawn from north to south over the Kalamazoo moraine and its ground moraine; the measurement locations which are used in the graph are shown in Figure 6. The diagram contains profiles for both magnetic susceptibility and the standard deviation of the variations of the vertical intensity. As can be 56 57 . .eoHpmN>me assesses use apaaan IHpmoomsm capocmms soospmn coapmamahoo map weapowmmo mo>poo moamswm pmmog .mm shaman 0 5.155 t u hxda 2:5 woo e.e. x .333533 0:232 5.5 ”auto. x 2:33:39" 2.2.2.: ON. 0: OO- Om 8 ON 00 On 0* 0n CN 9 on. 0: 00. 0m 0.... 05 0m CO 0? On ON 0“ .i . 4 a I. a .lIIId .lliqll lHIlllgflliqullulu O .. I :J.III-. .IIIJIIIIaII,I_I..l .1 d 1.1 i 1 a O . a _ ~ . a. S S W i 0 NM ~ my D P n M an 0 o I V» cm m. t N w. o i D W n U . . m. w u w . l h w .m s is .o mm 1a u .52.. < hcca 2...: mouse. x b.3238.» £23: :2: moo 12 x b.3223...» 3.2.2. ON. 0: 00. cm 00 Du cm On 0* On ON 0. ON. 0: 09 om on 2 on On 0' on ON O. o a o J-III - 4 q i [114 a a 4 ~|ll1 J A q q u. a d 4‘ q 1 N .u a. _ e n A . . ‘ (of\ s O.‘ ‘eh‘lfld eel m NW. . on. an 1 N” P 6 D &. %@n 0‘. 0 cm new a m. w i 0“ A a“ m. g m. a we 0% m m . am I NM 9 I I n D . O L a O 58 ON sceptability X l0.SCGS Units 0 (0 Magnetic 5 u o o 3 2 g .mocwmuos unsoum pom assesses ooumemamm one access moafimoum hpfiafiofipaoomsm OHpoamma com ooflpmfi>o© pameomym .mm osswam w2_uo made 2 5b illii 40© vamdompm .mm ohswfih w2_w0 om mop ommuopm Op Bopmam deem u mean: moaamuos eosoum use Husfisuop oppoaumno on» mmouom moafimoum mpaafiofipooewsm oapmomms 6cm cOHpmH>oc unmecmpm .mm ousmfim O . _ . _ I _ o ,_ u. _ u _ 0 _ 0 _ m b < _ J i l i _ _ _ _ l _ mz.h.4.m