TheNorth American Central Plains conductivity anomaly and ...world (Van Schmus et al., 1987). In central North on the southernmost profile, and the correlation of America, it separates

Physicsof theEarth andPlanetary Interiors, 60 (1990)169—194 169ElsevierSciencePublishersB.V., Amsterdam— Printed in TheNetherlands

TheNorth AmericanCentralPlainsconductivityanomalyandits correlationwith gravity, magnetic,seismic,

andheatflow datain Saskatchewan,Canada

Alan G. JonesandJamesA. Craven

GeologicalSurveyof Canada,1 ObservatoryCrescent,Ottawa Ont. K1A 0Y3(Canada)

(ReceivedJanuary13, 1988; revisionacceptedJune6, 1989)

Jones,A.G. andCraven,J.A., 1990. The NorthAmericanCentralPlainsconductivityanomalyandits correlationwithgravity, magnetic,seismic, andheatflow datain Saskatchewan,Canada.Phys.Earth Planet.Inter., 60: 169—194.

The North American Central Plains conductivity anomaly (NACP) lies, virtually in its entirety, within theTrans-HudsonOrogen.Accordingly, should thesetwo featuresprove to be contemporaneous,then the geometricalrelationshipbetweenthesetwo is of foremostimportanceto anyevolutionarytectonic model proposedto explainthecollisionof theSuperiorandChurchill Provincesin theHudsonian,anda modelthat doesnot includea mechanismforthegenerationof this anomalyis obviouslyuntenable.

In orderto mapbetterthetrendof theNACP in the Provinceof Saskatchewan,Canada,two magnetotellunc(Ml’)profiles wereconductedover theNACP previouslydefinedby magnetometerarraysandprofiles.Data from thesetwoprofiles, alongwith anearlierMT profile just north of the U.S./Canadianborder,suggestthat the NACP is not acontinuousfeature,but ratherthat it exhibits a definitebreakat latitude51°N. Othergeophysicalevidenceexaminedhereinis concordantwith thischaracteristic.If a secondMT anomalymappedto thenorthwestof thisbreakis, in fact,a manifestationof the same geological structure,then one possibleinterpretationis of a major NW—SEtrendingsimstralfault in thedeep crust,previouslyundetected,with a movementof some100—150km alongstrike.

TheMT datafrom thesouthernmostprofile, aftercorrectionfor staticshift, aremodelledin a 2D manner,andit isshownthat the NACP is consistentwith an arcuatestructurein verticalsectionof high conductivity (>2 S rn

1)beginningata depthof 10 km centredon 103°W dippingdownto thewestreachingpossibly thebaseof thecrust. Itis also shownthat suchan arcuateshapein sectioncan explain an observedgravity high of 40 mgal. Such highconductivitiescannotbe explainedin terms of connectedfluids, as it would require implausibly high porosities(12—20%). A comprehensiveinterpretationof all the geophysicaldata would have to accountfor the observedcharacteristicsof theanomalousregion;whichare,high electricalconductivity,positivedensitycontrast,no magnetiza-tion, andhigh heatflow in thebasement,in termsof a singlecausativebody. Onepossibleexplanationis presentedintermsof a zone of lithosphericweaknessalong which was emplacedlow-densitydifferentia from a mantle-derivedbody. An alternativeexplanationcouldbethephasetransformationof obductedmaterialinto eclogiteor serpentinite.However,difficulties existwith both of theseexplanationsand furtherdata arerequired.

1. Infroduction Gough andCamfield, 1972;Alabi, 1974;Alabi eta!., 1975;CamfieldandGough,1975,1977; Handa

The enigmaticNorth AmericanCentral Plains andCamfield,1984; Guptaet a!., 1985) suggestedconductivity anomaly (known by the acronym that the NACP is a linear feature,extendingforNACP), was first discoveredin the late 1960s over 2000 km. The NACP has beeninterpreted(Reitzel et al., 1970) by a geomagnetic depth variouslyas: (i) associatedwith conductivemner-sounding(GDS) array. SubsequentGDS studies als, such as graphitein schistoserocks in a belt(Camfield et al., 1970; Porath et aL, 1970, 1971; mappedby Lidiak (1971), that lies within the

0031-9201/90/$03.50 © 1990 Elsevier SciencePublishersB.V.

170 AG. JONES AND J.A. CRAVEN

Precambrianbasementrocks beneath the Great sensing the presenceof porous (7—10% porosity)Plains (Camfield et al., 1970; Gough and Cam- obductedoceanic crustalmaterialsaturatedwithfield, 1972); (ii) dueto the presenceof salinewater highly saline fluid (sea water obductedwith thein fractured rocks (Handa and Camfield, 1984); oceanic sediments).This ,suggestionappearstoand(iii) dueto partial serpentinizationof oceanic havebeengiven support by Kurtz et aL’s (1986)maficandultramafic rocksat theridge crestof an interpretationof their MT data on Vancouverancientformer oceaniccrust (Guptaet a!., 1985; Island that a conductingzone,coincidentwith aseealso Greenet a!., 1985). seismicreflection horizon (Green et a!., 1987),is

Figure 1 illustrates the location of the NACP due to the presenceof trappedsaline fluids with

from the GDS experiments.Its location, and the porosity in the range 1.6—3.6%. (An alternativecorrelationof its ‘ends’ with major Early Protero- explanation for the conducting zone/seismiczoic zones of shearingand folding in the Pre- reflectivity hasbeenoffered by Hyndman(1988),cambiancrust, suggestedto Camfieldand Gough who appliesthe suggestionof Jones(1987a)of the(1977) that it may be the electromagnetic(EM) growth of an impermeablelayer at the 400°Cexpressionof a geosuturethat formedduring Early isotherm.)Proterozoic time due to the collision of two The suggestionby Camfield and Gough(1977)Archeanplates.That deep-probingnatural-source of aplateboundarybeneaththe thick sedimentaryEM techniquescanlocatethe boundariesof pres- Phanerozoiccover of the mid-North Americanent,or former,plateswaspreviously illustratedby continent waslaterunited with the proposalsforLaw andRiddihough(1971),which led Drury and Proterozoic plate boundaries in the exposedNiblett (1980) to proposethat EM methodswere CanadianShield(Gibb andWalcott, 1971),andin

6 1l~o 1OO~

~ *~++j

:~‘~ ~‘‘ : 0 ~0

(~~o o,op c~O’o,oooo ++

00 ~ ~ 0 0 00, * *÷:.N÷÷*+~4 ++ * + **** * +*+* *

* o ~ ~ $UPERi~~:~:~:÷~* + * + * + + + + + *• ~ ~ + ~j7* +,• ~ * + • + + ** * *

0000 1+ + + •* • * + + + ** * + +* + + * 50

+ ÷::: 0 ~00 ++ + * ** + + *+** + + *+ *

0;

NACP From GDS+ — MT Profiles

Tectonic Provinces* :~~~Jo + •~ * + +

0 * Deformed Continental~++ ~+ El Margins

+ .0 •~ Continental Magnetic4Qo * + ~ + + +÷ +~ ++ El Arcs

llOo CeN~~~—, ~ ~ Juvenile Crust

1000 900

Fig. 1. TheNACPstructureastracedby theGDSexperimentstogetherwith therelevanttectonicprovinces;Rae,Hearne,(thesetwowere formerly theChurchill province)Wyoming and Superior(from Hoffman 1988).It should benotedthat throughoutits entirelength, theNACPresideswithin thepostulatedlocationof theTrans-HudsonOrogen.

THE NORTH AMERICAN PLAINS CONDUCTIVITY ANOMALY 171

southern Wyoming (Hills et al., 1975), into a portant tectonic feature and usedit to correlate

singleNorth AmericanHudsonianagesuturezone, the buried part of the THO in the Dakotaswithsince termed the Trans-HudsonOrogen (THO: the exposedPrecambrianterraneof the cratons.Hoffman, 1981). Whereexposedin the Canadian They interpret their agedataas concurringwithShield, the THO consistsof alternatingbelts of thoseof Peterman(1981) in revealingEarly Pro-volcanicrockswith associatedgranodiorite—tona- terozoicrocks,mainlygneissesof unknownorigin,lite intrusionsandvolcanogenicsedimentaryrocks in thewesternDakotasbut that, as yet, thereis no(T.M. GordonandE. Froese,personalcommuni- evidencefor oceanic crust there. In their litho-cation). tectonicsubterraneIVA (seeFig. 3 andTable1 of

Extensionof the Early ProterozoicTHO, and Klasner and King (1986)), defined by Pre-associatedsuture zones, beneath the Palaeozoic cambrianchip samplesanddrill corein thenorth-sedimentsof the Williston Basin(WB) has since em part of North Dakotatakenmostly at a longi-been confirmed by other evidence, principally tude of 103° W (i.e. the ‘new’ location of theinterpretationsof theBougergravity andmagnetic NACP as delineatedby JonesandSavage(1986)compilations(Green et al., 1979, 1985; Dutch, and as modelledbelow), the rocks are predomi-1983; Klasner and King, 1986; Thomas et a!., nantlygranodionticto tonalitic gneiss.1987),but alsoby geochrono!ogicalevidencefrom In this manuscript, we will illustrate the resultsbasement-reachingboreholes(Peterman,1981). obtained from three MT profiles recently con-

The THO is interpretedby someto extendfor ductedin the Provinceof Saskatchewan,Canada,over 5000 km, from central U.S. in the south, and correlatethis information in both a qualita.northward into Canadaand there eastwardsto tive and a quantitative mannerwith other geo-Greenland.It is the largest coherentlypreserved physicaldatafor this region. A 2D model will beandexposedEarly Proterozoicorogemcbelt in the presentedthat satisfiesthe observedMT responsesworld (Van Schmuset al., 1987).In centralNorth on the southernmostprofile, andthecorrelationofAmerica, it separatesthe Wyoming—Churchill(or this anomalywith gravity, aeromagneticand geo-‘Wyoming—Heame’accordingto Hoffman(1988)) thermaldatawill be shown.Considerationwill beArcheanprovincefrom theSuperiorprovince(Fig. given to the NACP at the Black Hills, and the1). Formationof magmaticrockswithin the THO GDS data of Porathet a!. (1971), modelled inoccurredbetween1910 Ma and 1830 Ma (Van terms of a shallow highly conducting arcuateSchmuset a!., 1987). structure,will be presented.Finally, two possible

The NACP shouldbe considereda significant tectonic interpretationsare given for the NACPand important element of any evolutionary anomaly— onein termsof materialderivedfromtectonicmodel for formationof the THO. Green the mantle since the onsetof subsidenceof theet al. (1985)suggestedthat the anomaly‘is caused Williston Basin(400 Ma), and the other in termsby buried slices of hydratedoceanic-typecrustal of phase-transformedobducted material duringmaterial’ which ‘might also explain the spatially the THO (1.8 Ga).However,difficulties exist withrelated seismic low-velocity zones’. However, both of theseexplanationsand further data aredetailedanalysesof the refractiondata(Morel-a. required to elucidate the reasonfor the NACPl’Huissier et al., 1987) were inconclusiveconcern- anomaly.ing the relationship of the seismic low velocityzonesto theNACP mappedby Jonesand Savage 2. Previouselectromagneticresults(1986). It shouldbenotedthat hydratedmineralsare not intrinsically conducting(Olhoeft, 1981); it Figure 1 illustratesthe location of the NACPis the fluids releasedby dehydrationreactionsthat determinedfrom the GDS experiments,togethercausethe highconductivity, with the recenttectonic map of North America

Klasner andKing (1986), in their studyof the presentedby Hoffman (1988). It shouldbe notednatureof the Precambrianbasementrocksof the that the Churchill Provincehas beensubdividedDakotas, consideredthe NACP to be an im- into the Heameand Rae Provincesby Hoffman.

172 AG. JONESAND iA. CRAVEN

Primarily becauseof the large separationof the east),is maximally enhanced.Thus, close to anGDS stations,typically 150 kin, this initial loca- anomalyit is extremelydifficult to determineon

tion of the NACP is crude.An MT survey(Jones which side of the station the anomalylies, espe-and Savage, 1986) conductedjust north of the cially when thedataused,aswere thoseof ACG,U.S./Canadianborder(Profile S in Figs.1 and2) are contaminatedby source-fieldeffects. Further-providesmoredetailedcontrol and indicatesthat more, close examinationof the dataof ACG re-the initial locationwas in error at that latitudeby vealsthat the anomalyis not well locatedin thisapproximately75 km, or half aGDS stationspac- region. Figure 3 (reproducedin part from Fig. 8 ofing. Sucha locationerror would appearto imply ACG) clearly indicates that the maximumof thethat the NACP structureshouldpassto the east realpart of the H,, componentlies to the east ofof GDS stationQUA (Fig. 2), not to the west of GDS stationQUA (Fig. 2). Moreover,the ENE-it, as interpretedby Alabi (1974)andAlabi, Cam- pointing quadratureinduction arrow, derivedbyfield and Gough (1975; referred to hereafteras ACG (1975, Fig. 9), for stationQUA at a periodACG). This apparent discrepancyis easily ex- of 68.3 mmclearlyindicatesthat theNACP passesplainablewhenonerealizesthataboveananomaly to the east of that station. For stations southofcausedby a zoneof higherelectricalconductivity, QUA, ACG (1975, p. 828) comment that ‘Thethe vertical componentof the time-varyingmag- quadrature-phaseinduction arrows proved unex-neticfield, H~,is zeroor closeto zero,whereasthe pectedlysuccessfulin showingthe presenceof theperpendicularhorizontal component,H,, (the co- NACP conductoralong the whole length of theordinatesystemusedthroughoutis x north and y array,by pointing to it from both sides’ — why

520j~_~~j~r~ 00 NjB~c / R~Y

~ * * + ~f.J11__15I0

MO~ ~ua IN ~o ~L~—15°°

II *++**+***4_~ ***t**1+ •000 * * **t+ 0

____________________________________ I 490

108 060 040 02’ 00’ 98

Fig. 2. Positionsof thethreeMT profiles; S: South;M: Mid; and N: North Saskatchewanrespectively.NandT referto thelocationof theNACP andTOBE anomaliesrespectivelyasidentified from the MT data.Also shownarethelocationsof someof theGDSsites of Alabi et al. (1975)referredto in thetext; BRC, RAY, MOR, QUA, and MOO.


1150 105° 950 1150 105° 95°____________________________________________ - I

Alto. Sask. Man. Alta. Sask. Mon.

io x- I i . ca~- - I

Mont. ~---“:‘~~T ;. ~ N.

/w 40 ~ ~~~~j:SD 40

Fig. 3. Maps of theresidualFouriercosinecoefficientsof theeastwardhorizontal component,H~,at periodsof 68 and 170 mmobservedby Alabietal. (1975)for oneparticularevent.It shouldbenotedthatthemaximumlies to theeastof stationQUA (Fig. 2),not to the westasconcludedby Alabi et al. (1975).

this shouldnotbeacceptedfor QUA is notstated. continuousfeaturebut ratherthat it is discontinu-Oneothernoteworthyfeaturein the GDSwork ous, and that the contoursof the map of Fig. 4

of Alabi (1974) andACG is that the characterof illustrate one of the breaks.This latter explana.the NACP changesmarkedly north of approxi- tion, that the NACP is a discontinuousstructure,mately 50°N. Figure 4 (reproducedfrom Fig. 6 was suggestedrecentlyby Thomaset a!. (1987)inof ACG) shows a concentrationof contour lines their interpretationof thehorizontalgravity gradi-south of the U.S./Canadianborder which is in ent mapof centralNorth America. Theyproposestark contrast to the dispersedattitude of the that theNACP‘could be the expressionof a seriescontoursnorthof the border.Thischangein char- of discontinuous,perhapsen echelon,conductorsacter is also evidentin the Fourier transformmaps that havenot beenresolvedby the coarsespacingdeterminedfrom a magneticeventwith a more of magnetometerstations’. We concur with thisuniform source-field (Fig. 7 of ACG), although conclusionandwill reinforceit by the argumentsthe very existence of the NACP conductor is presentedin this manuscript.exhibitedby the reversalin the vertical magneticfield componentsbetweenGDS stationsRAY andBRC and stations MOO and MOR of Fig. 2 3. Magnetotelluricobservations(Alabi, 1974). Accordingly, source-effectscannotexplain this along-strikevariation. All MT responsesreported herein were ob-

This variation in charactermay indicate that tamed using PHOENIXE Real-Time MT datathe NACP dips more steeply to greaterdepth to acquisition systems. Such a system records thethe north,or it may infer that the NACP is not a time-varying MT fields in the period range

174 AG. JONES AND iA. CRAVEN

115’ 105° 95~ GeologicalSurveyof Canada’sPHOENIX system.Alto, / Sank. Man. At all thirty-four sites,sixteenalongprofile N and

34 ~ eighteen along profile M, ten-componentMTmeasurementswere made. This two station re-

38 34 dundancyat eachlocation enablesus either(i) tochoosethoseestimatesof the responsesthat were

42 38 so’ leastnoisecontaminatedandexhibitedleastscatter

do for interpretationand modeffing, or (ii) to com-

—- - .._~— ~ bine the estimatesfrom the two stations in aND. weighted manner(see, for example,Craven and

Mont. Jones,1989). In this report, we adoptthe former44 . of thesetechniquesfor profiles M and N.

~--- SD. Data recordedon all three profiles were re-/ 48 ~ edited to reject cross.spectralaveragesthat con-

Wyo taminatedthe cumulative average.The data for‘i~~— 34° Neb profile S havebeenstatic shift corrected(Jones,

48 40~ 1988a),but thosefor profiles M and N haveyet

to be correctedin this manner.Accordingly, in- . this manuscriptin order to avoid static shift ef-

Fig. 4. Map of the Founer transform phaseof the vertical .

magneticcomponent,H, ata period of 68.3 nun observedby iects, tor tnesepronieswe win consiuertne qua -

Alabi et al. (1975) for oneparticularevent.The concentration tative variation of phasesof the MT impedanceof the contoursto the south of theU.S./Canadianborder in off-diagonal elementsZ~~ and Z,,~alone, andcontrastto their dispersednatureto thenorthshouldbenoted. quantitativelyinterpretthemagnitudesandphases

of thoseelementsfor profile S only. Thedatawere0.0026—1820 s (frequency range of 384—0.0055 all rotatedinto the geographicalcoordinatesys-Hz), and derives and displays the MT response tem,with the x-directionbeing true north andthefunctionsin real-timein the field. All dataareof y-direction true east, which is particularly ap-exceptionalquality with standarderrors generally propriate for theseprofiles in that the dominantless thana few percent. structuraland geophysicalstrike of the region is

Datafor the southernprofile (profile S, Figs.1 north—southandthat the dominantrotationdirec-

and 2) were recordedduring 1984 and 1985 by tion of the MT impedances(definedby the usualPHOENIX Geophysics of Toronto for Pan- Swift (1967)criterion) is within 5°of N—S. In thisCanadianPetroleumLimited of Calgary. At all coordinatesystem,the Z~,,estimatesrepresentthethirty-five, sites, seven-componentMT measure- E-polarization modeof induction(i.e. north—southments were madethus facilitating remote-refer- electriccurrentflow) in the 2D anomalies,whereasenceprocessing(Gambleet a!., 1978) of the data. the Z,,~ones are the B-polarization modeJonesand Savage(1986) discussedthe main fea- (east—westcurrentflow).tures of thesedata,andillustrated that the NACP Pseudo-sectionsof thephaseresponsesobservedstructureis some75 km eastof the location shown along the threeprofiles are illustrated in Fig. 5.in Fig. 1 at the latitude of the U.S./Canadian The phasesrange from 10° to 70°,and the J’~border.Jones(1988a),analysingthehighfrequency phaseshavebeenrotatedinto the first quadrant,portions(frequenciesabove0.1 Hz) of thesedata, from the third quadrant, for direct comparisonshowedthe correspondencebetweenthe 1D mod- with ~ On these figures blue signifies a lowels and the well-log information, andpresenteda phaselead of the horizontal electric field compo-techniquefor static shift correctionof MT data nent over the perpendicularhorizontal magneticfrom asedimentary-basinenvironment, field component,which is indicative of a transi-

Data for the two other profiles (M and N, tion from shallow conductiveto deeperresistiveFigs. 1 and2) were recordedduring 1987 with the zones,whereasredsignifiesa high phaselead,and


-.~ 0

~ U

i Z .EI-

I—

z2-z

C0 za

za

—

0.0

~ .~~D1~-—I ~

—~—

I I ~

~

~~ .~vu

—— cO~~.Zo

— ~ 0,0= t~a ——

I ~• ~I— ___________ ~0,0) zC0 zF..J

~ao z-ö

0 0 0 I0 — 0

~w 0 — 6~ .-(s)aoI~3d —‘~‘~ ~

(S) 00I~d 0“

6... ~‘0 co .~0.~~ EL)

I 20<— -~ — 0 0~ 0 — 0 OO~ C)0

(S) ~OI~3d


accordinglythe reversesituation.The 45°contour on top of the conducting shalesof the Upperis illustratedin dark-yellow.Theconjecturedloca- CretaceousFaval Formation. For the basementtions of the NACP and the TOBE (ThOmpson structure, thereis no strong indication in eitherBElt, seeJonesand Savage(1986)) anomalieson the jx~,,or the 4c,,~phasesfor eitherthe NACP orthe profiles are indicatedby the boldfacedletters theTOBE features.By extrapolatingthe trendsofN andT respectively.Noteworthyfeaturesof these the potential field data(seeSection4), it is possi-pseudo-sectionsare discussedbelow. ble to infer that theminor phaseanomaliesin the

~,, data, indicatedby N, maybe associatedwiththis feature.However,thisphaseanomalyis small,

3.1. Profile 5: SouthSaskatchewan of less than — 10°,comparedto over 30°on

profiles S and N. We interpret this as an ‘edgeThe depthof the baseof a conductingzonecan effect’ dueto non-continuityof the NACP struc-

be given approximatelyby the Niblett—Bostick ture. Onefeaturethat is discerniblein thecontourdepth (Niblett and Sayn-Wittgenstein, 1960; plotsis the lateralvariationin theresistivityof theBostick, 1977; Jones,1983a)for the apparentre- basement;to the east,the basementis moreresis-sistivity at the period where the phaseresponse tive than to the west, as demonstratedby thecrossesfrom above450 to below 45°.For profile higher phases(darkershadesof blue).5, this phasecrossingoccursat a periodof — 5 sto the west, and shallows to a shorterperiod of— 3.5 s to the east(Fig. 5, bottom). Accordingly, 3.3. Profile N: North Saskatchewanthe baseof the upperconductivelayer — associ-atedmainly with the top of the Lower Palaeozoic On the N-Saskatchewanprofile, the conductingsequencesof the WB (Jones,1988a) — occursat sedimentsare well mappedas they thin from westdepths of approximately1450 m and 1250 m (7 s) to east (0.05 s); Niblett—Bostick depths ofrespectively.TheNACP andTOBE anomaliesare approximately1200 m and 400 m respectively.clearlyobservablein the 4~phases,but notin the Also apparentin the contour plots is the lateral

f~phases,which is a manifestationof the hori- variation in the resistivity of the basement,withzontalEM ‘thinness’ of the two anomalies;that is, the moreresistivestrataoccurring to the east,ascurrentsflowing along the y axis (i.e. east—west) profile M. On thesepseudo-sectionsit is againare not affectedsufficiently for theseN—S trend. possibleto discern two anomaliesin the phases,ing anomaliesto be seenat the ground surface. labelledN andT. As with profile S, the anomalyTheseMT responsesare modelledquantitatively, that can be associatedwith the NACP (N) has awith a 2D model, below, and a combined geo- phaseindicator in 4~,,only, not in 4,,~and thusphysicalinterpretationof theNACP structurewith the NACP is herealso electromagnetically‘thin’.the gravity, magneticsand heat flow observedin For the TOBE feature (T) however, there is athis regionis discussed. strong responsein both phaseswhich indicates

that, at this latitude, the TOBE structure cansufficiently affect both the parallel (E-polariza-

3.2. Profile M: Mid-Saskatchewan tion) and perpendicular(B-polarization) currentsto anextentthat is observableon the surface.This

Along thisprofile (Fig. 5, middle) the conduct- variationalong strikeof the TOBE is also evidenting sediments (not all sediments but those of in the gravity data (see below and Fig. 6). Pre-higher electrical conductivity — see above) are liminary 1 D modelling of the MT responseob-shallower, as evidencedby the shorterperiod of servedat this locationindicatesthat the top of the2 s for their ‘base’; which correspondsto a TOBE anomalyis at a depth of some 13.5±0.5Niblett—Bostick depth of approximately750 m. km which clearly cannotbe associatedwith theThephaseanomalyjust west of longitude100°W exposedrocks of the ThompsonNickel Belt butis dueto that particularsitebeinglocateddirectly mustbe a deeperfeature.


2:9’U

zC)

.0

0‘I,a0CO

00000000000 0000000 ‘0oo,o..oooo ~ 9

CIII III 111111 I .c

_________________________ ‘0aCO

0

0 Q-. to 0.C’.J to ,

°~r ~°—~0 to to _—~~

(0 to .0It)

°0 I2COC

00

—

0(0 °toto If) ON, o~ °

~78 AG. JONESAND J.A. CRAVEN

C

V

V

0

0000000000000000000UlOlflOlfl 0 0tflOit0IflOIfl000000W C

2aIm 110

~

‘-I- IllIllIluIluIs

V

C

0

0

0 it) U~°cr

(0 Iit)

L)

V.0

00 ~H ~ QJ

____________________ I_ I II


4. Qualitative correlation with other geophysical latedto the exposedrocksof the CanadianShield,observations but rather to its eastbetweentwo magnetichigh

zones(red contours).Although the NACPis very

In this sectionwe describeother geophysical crudelylocatedby the widely-spacedmagnetome-observations,i.e. gravity, magnetics,heatflow and ter sites some 400 km further north of profile Nseismic along the THO, and comparethem with on the exposedpart of the Trans-Hudsonorogenthe MT datain a qualitativesense.A quantitative by HandaandCamfield (1984), thereis herealso2D modellingof theMT responsesobservedalong the indication that the NACP correlateswith aprofile S, anda combinedgeophysicalinterpreta- magneticquiet zone, in this case the Reindeer-tion of the availablegeophysicaldata,is given in South Indian Lakes belt; a correlationthat wasthe following section. earliernotedby Greenet a!. (1985). It shouldbe

noted that also along profile S the NACP corre-4.1. Gravity data lateswith a magneticquietzone(yellow andgreen

contours).Figure 6 is aBouguergravity anomalymapof

the region with the three MT profiles marked. 4.3. HeatflowApparent in the gravity field is the high thatcoincidesexactlywith the NACP structureat the The heatflow below the Palaeozoicsurfaceforlatitude of the S profile. This high, of approxi- the region is illustrated in Fig. 8 (taken frommately 40 mgals relative to the ‘regional’, trends Majorowicz et al. (1986)). Strong anomalies, ofslightly west of duenorth, and terminatesat the greaterthan 100 mW m 2 are derivedfor loca-latitude of profile M. Along profile N thereis no tionseastof Weyburn(to the southeastof Regina),distinctiveanomalypatternthat canbe associated and to the north-westand west of Regina, fromwith the q~ phase anomaly identified as the the strong geothermalgradients,of greaterthanNACP. Thereis abroadhighin approximatelythe 25°Ckm— ~. Elevatedtemperatures,of 70°Catcorrectlocation,but nothing as definitive as for the 2 km depthlevelandup to 100°C at the 3 kmprofile S. For the TOBE structure, there is acoincidentgravity high on profile S that can betracednorthwardsacrossprofile M, at the loca- ~

tion of the minor ~ phaseanomaly,and then I I I

northeastwardsacrossprofile N at its easternmost 54~ SASKATCHEWAN MANITOBA

extremity.At this northernlocation,both the EMand gravity signaturesaremuch more enhanced (than thoseto the south. I g ~ N

I ,- ~o_-.~ I I

52° I ( ~ SASKATOON4.2. Magneticdata I ‘~ (~

50 /a variety of sourcesof varying quality, is given in oyC~) ~~6O

Greenet a!. (1985).The contourinterval of Green ~c~DA - — - / — - I~ - — - — -

et al.’s map is 200 mgals, and thus some of the USA MONTANA NORTH

fine structuresdiscerniblein Fig. 7 arenot visible 48’ I I I I DAKOTA I

on Green et al.’s map. It is apparent,however, . . -Fig. 8. Heat flow pattern,in mW m 2 observedin southern

from both mapsthat the NACP of profile N does and central Saskatchewan(redrawn from Majorowicz et al.,

not correlatewith the magnetic low (dark blue 1986).Also shownarethelocationsof theMT profiles andthe

contour at — 106.5°W) which can be extrapo- identifiedlocationsof theNACP(N) andTOBE(T) structures.

180 AG. JONESAND iA. CRAVEN

depthlevel,are extrapolatedfrom thesedata.This the earthquakesrecordedup to the end of 1988makesthe topof the Precambrianbasementhotter for this region are illustrated in Fig. 9. The twohere than anywhere else in Saskatchewan. largestwerehistoricaleventsfrom the lastcenturyMajorowicz et al. (1986)concludedthat the blan. on the order of magnitude5 (stars),both locatedketing effect of the sedimentscan only partially approximatelywithin the WB but with very largeexplain the Weyburn temperaturehigh, and that uncertaintyas to their actualposition. All otheras well as a hydrodynamiceffect, there may also recordedeventsoccurredafter1968 andarelocatedbe a contributionfrom a high-temperatureregion to within a precisionof 0.2°(J.A. Drysdale,per-in the crystallinecrust. This is discussedfurther sonal communication).ACG and Camfield andbelow. . Gough (1977) refer to earthquakesof magnitude

It is apparentfrom Fig. 8 that a north—south 3—4 in the zone49—49.5°N, 104—106°W. How-trending geothermalanomalyoccursat the same ever,this areais the location of the Regina-Hum-location of the NACPat the latitudeof profile S. mingbird Troughand thereis the possibility thatThis trendin the anomalyendsratherabruptly at theseevents are all associatedwith s!umping/— 50.50 N, at which the trend in the anomaly collapsingof solutioncavities.Homeret a!. (1973),becomesnorthwest acrossprofile M. There are however,argueagainst slumping for at least oneinsufficient data to know if thereis a geothermal of these,andprefer a deepcrusta!depth.The twoanomalyon profile N associatedwith the NACP. tight clusters of magnitude3—4, one of thirteen

earthquakesat 50.75°N, 102°W and the other4.4. Seismology of six earthquakesat 52.1° N, 106.9°W, are

believed to be induced by potash miningA comprehensiveseismic refractionstudy was (Gendzwill et a!., 1982; Drysdale et al., 1985;

carriedout acrossthe THO/WB by the Canadian Drysdale and Homer, 1986). For most of theCOCRUST (Consortiumfor Crustal Reconnais- other seismic events in Saskatchewan,there issanceUsing Seismic Techniques)group, and all insufficient depthcontrol to be certainof whetherthe data were presentedand comprehensivelymodelledby Morel-á-l’Huissieret a!. (1987). TheNACP was one of the primary targetsof thisexperiment,but as a consequenceof the unfor-tunateerroneousprior location of the NACP from I 106° I I I

the GDS work (see above) at the latitude of theseismic profile, there was no sampling of the . 52°

NACP anomalyalongits strike.The seismicprop- SASKATCHEWANA MANITOBA

ertiesof the crust beneaththe neareststrike line —M °

(their line D) some 75 km to the west of theNACP were modelled as exhibiting virtually no - N .

lateral variation, and no unusualvelocity effects. I I TIn contrasta strike line just 50 km further to the —_~ I L.

CANADAwest (line E), which was intendedto samplethe .

NACP, exhibits two crustal low-velocity zones, MONTA * 4one at 10 km and the other at 22 km, at its - - 48’

southernextremity,but thesepinch out some120 M <3

km northof the U.S./Canadianborder,nearlati- M ~ 3 A

tude51.5°N. M �4 • IIn their paper, ACG consideredthe known - M ~ ~ 1* I I I 46°

earthquakesfor the region with referenceto the . .Fig. 9. Earthquakeepicentresfor centralNorth Amenca.Also

NACP, and concluded that therewas a correla- shown are the locationsof two of the MT profiles and the

tion, albeit tentative.The epicentrallocations of identified locationsof theNACP (N) andTOBE(T) structures.


they occurredin the basementor the sediments. and are within the crust. Given the paucity ofHowever, the smallevents,of less thanmagnitude eventsin this region one must concludethat the3, closeto Saskatoonare well locatedin depthby earthquakedatabaseis far too sparse for anythe Universityof Saskatchewanseismicarray,and quantitative,or evenqualitative, inferencesto besome of theseare definitely below the sediments made.

105.66° 103.190102 I I I I 102 I I I I

I US U

•5 U

I U1. U U

Pa I I10’ I••:.. I • 10

(fim) •.:~‘, .:‘ ‘11151•• N

0

10 I I I I 100 I I I

90°

~ 450~-- ~ ---~-“1~’si~ ‘. 45°1.___ hh,u,s,suIh’0°~ 1~2 1Ib~’ ib° ,ibl 1b2 ~ 0°3 i0~ id~’ 100 10’ 102 io~

PERIOD CS) PERIOD (Si

100.770 100.31°102 I I I I 102 I I I I

.•:~lI .‘ :S I

I .‘ IPa • H ,1

10’ 10’ III

(em) •• I~ ~ ~

I I I I 100 I I I I

90° • 1 90°

~ 45° ‘!!s’.s..,0 ~ ~ i&~ th° IbI ic~ io

3 0 ~ l&2 i~’ ib° ibl 1? io3

PERIOD CS) PERIOD (Si

Fig. 10. Static shift correctedapparentresistivitiesof theE-polarization (dots)and B-polarization (squares)apparentresistivitiesandphasesobservedat four locationsalongthesouthernmost(S)profile. At eitherendsof theprofile, i.e. at thesitesat105.66°and at100.31°,the data arevirtually 1D. The NACP structureis below thesite at103.19°,and theTOBE structureis below thesite at100.77°.It shouldbenotedthat theB-polarizationdata areinsensitiveto both theNACP and TOBEstructures.


C)

°~

C’J~C)G0

.0

0iiiiiinul:~ CO.0a) 000Wc’J~0WN~0 0.d1~ ~ I-

A‘0CCOI’

.9:~2

2CO0.0.CO

E 0

02C)”

1

-c‘SI-’~0.00~

‘0

0Q

~ .9

0 00U-)

0000.0

0

0.-S

9.-,

EEOC)0.0

c-~C

~IIIIIIII III III Will IIIIi~ 1111111 IIIIIIIII ‘0 CO

~ .90 2 0 0CII t’)

(wj) H.Ld~a


5. Modelling of MT responsesfrom profile S (20 km for NACP, 5 km for TOBE), below whichthereis a decreasein conductivity.

Thissectionpresentsthe 1D and2D modellingundertakenon the static shift correctedresponses(seeJones,1988a).The phasesof the E-polariza- 5.2. 2D forward modellingnon and B-polarization responsesare illustratedinFig. 5, and the static shift correctedapparentresistivitiesat four locationsalong the profile are Two-dimensionaltrial-and-error forward mod-illustratedin Fig. 10. Two of thesefour locations elling of the MT responseswasundertakenusingare at either extremity of profile 5, and the data Madden’s EMCDC code (Madden, 1973) mod-from bothare 1D in that the two modesof polari- ified for useon a CONVEX mini-supercomputer.zation give the same responseat each site. In E-polarization and B-polarization responseswerecontrast,the stationat 103.19°is directly overthe calculatedat sevenperiodsover the two decadesNACP, and 100.77°over the TOBE, anomalies, iO—iO~s (threepoints/decade),andthen interpo-and the datafrom bothexhibit 2D effects. How- lation to six points/decadefor contouringpur-ever,notethat the B-polarization apparentresistiv- poses.The interpretationwas thus by total re-ities andphasesexhibit no laterally varying struc- sponsematching of the apparent resistivity andture. phasepseudo-sectionsfor the data and model

response,andnot just at chosenperiods.5.1. ID inversion Owing to their lateral separationof over 100

km. modelling of the two anomalies,NACP andUsing the ‘Occam’ algorithm that inverts the TOBE, canbeundertakenindependently.A model

responsesto a least structuremodel in a 8a/8z that explainedthe anomalousMT responsesabovesense(Constableet al., 1987), the E-polarization the TOBE structurewas foundvery quickly duetoresponsesfrom the 35 locationson profile S were the simpleresponseandbecausethe anomalywasinverted. These models are compiled and pre- insufficiently sampledspatially, i.e. thereareonlysentedin pseudo2D section format, with linear somethreeor four siteswith data that are indica-depthon the ordinate,in Fig. 11. Theseinversions tive of the structure,and they have stationsep-are leaststructure in a gradientsense,and there- arationsof over 10 km. The TOBE structuremustfore any structurethat is apparentin them must be thin, of less than 5 km laterally, with anexist. However, one must bewareof interpreting integrated lateral conductance(thickness timesthe E-polarization responsesoff the flank of a 2D conductivity) or around 10000 S. Its top is rea-body because of ‘false conducting layers’ sonablywell locatedat the baseof the Palaeozoic(BerdichevskyandDmitriev, 1976; Jones,1988b). sedimentsat 2 km (see Jones,1988a). Its baseisAlso, it mustbe rememberedthat the B-polariza- not well resolved,but it mustexceed4 km in totaltion dataare totally insensitive to the structure vertical dimension. More exact featuresof thegeneratingthe anomalousresponses,and so this body mustawait a moredetailedMT surveyof theinversionmustbe treatedwith caution, feature.

Notwithstandingthesecaveats,the NACP and The NACP can be much better resolvedfromTOBE featuresare well imagedby the Occam1D the data as it is deeperand has a far greaterinversions.The top surfaceof theNACP structure spatial extent.UsingFig. 11 as a basis,an initialis arcuatewith a minimum depth of approxi. model was constructedto describethe observa-mately 10 km at longitude103°W. It dips sharply tions from the twenty-two sitescentredabovetheto the west at an angle of some60°.In contrast, bodyat periodsgreaterthan 10 s. The E-polariza-the TOBE feature is vertical with its top surface tion apparentresistivity andphasedatafrom thesevirtually at the baseof the Palaeozoicsediments, sitesare illustratedin Fig. 12 (left side).For both structures,thereis the inferenceof in- Critical to the 2D modelling of the NACPcreasingconductivity with depth to a maximum featurefrom thesedataare the following points.

184 AG. JONES AND LA. CRAVEN

U

>

Uo o I.CII - -2_U

- . 0.0.CO

- I..II.l..IIl..lII.~.I ‘ •..i.~~•u••iiii• •.0.

II _-_~_ ——- __v

:

II

111111 111111 IllIllIlIlIll

0 0 0 0 0 0rO 2 ~ r’_) 0 0

(s) PO!~13d (S) PO~J~d


V V V ‘Vto) ~

resistivity(Am)

0 200 crust

_________ Liii 25 ied iments

U ______________ VE 1:1 ~ :.:

-4o -10 -10 II~0 b 0 io ~o ~o so 02DISTANCE 1Km)

Fig. 13a.A single-body2D model derivedto explaintheMT observationsalongprofile S of theNACP feature.

1. TheB-polarizationresponses(Figs.6 and10) body model wasfound which fit well the E-polari-are insensitiveto thepresenceof the NACP struc- zation and B-polarization data (Fig. 13a). Theture. This dictated that the anomalouszone of E-polarization responseof this made!, and itsenhancedelectricalconductivity mustbe‘thin’ in comparisonwith the observeddata, areillustratedanelectromagneticsense,bothvertically andlater- as contouredapparentresistivities and phasesinally, such that deviations of current flowing Fig. 12. Note the arcuatenatureof the body inthroughit arenot observableon the surface(due verticalsectionview,as predictedby the 1D inver-to the thick sequenceof conductingsedimentsin sions, and the asymmetry.The body must be

the WB). vertically thin and of very high conductivity —

2. The zoneof enhancedconductivity is closest from 2 to 5 5 m1. Although the model found isto the surface, along profile S, at longitude by no meansunique(seebelow), it doeshavethe103.25°±0.10. property that thereis no B-polarization response

3. The structuremust increaseits conductance to it detectableon the surface. If the anomalyiswith depth.The top layer has a vertical conduc- modelledwith greatervertical extent but smallertance (vertically-integratedconductivity) of the conductivity, i.e. keep the depth-integratedcon-orderof 2000 S,which increasesby a factorof 2—3 ductivity constantbut ‘spread’ it out in depth,at a depthof 16—18 km. then the E-polarization responsesfit the observed

4. There must be a zoneof increasedvertical data,butaneffect in the theoreticalB-polarizationconductancesome10 km to the west of longitude becomesevident,which is unacceptable.For the103.25°.Thiszone,of some100005, is evidentin B-polarization mode of induction, obviously thethe 1D Occaminversion(Fig. 11). effectsof the chargeson the boundariesare at-

By trial-and-errorforward modelling, a single- tenuatedby the thick conductingsedimentsof the

Ib) 1. •‘~‘ •. . ~...~“ .~. “ iiiiiiiiiiiiIIIIIIIIIIIIIII.IIIiiiiIIIiIII:II.:::.: :11i1

111

resistivity(Am)

ID 200 crust

, ~ __________ 25 sediments

20

________ VE 1:1 03

30 _________________________________________________________________________ 02-50 -40 -30 -20 ~I0 0 ID 20 30 40 50

DISTANCE (Km)

Fig. 13b. A multi-body 2D model derivedto explaintheMT observationsalongprofile S of theNACP feature.

186 AG. JONES AND J.A. CRAVEN

Williston Basin, whereas for the E-polarization a multi-bodymodel that fit thedataalmostequallymodethe currentsflowing along the body are not well is illustrated in Fig. 13b. The discontinuousso strongly attenuated. nature of the model ensuresthat the currents

Alternatively, the datacould equally be mod- flowing east—westare not affectedsufficiently forelled by morethanonebody. An exampleof such the structureto be detectablein the B-polarization

06

-04 910s

I I I I I I I I I I I I I I I I I I I I

~

-200 -150 -100 -50 0 50

distance (km)Fig. 14. The GDS transferfunctions,with 68% confidenceintervals, for currentflowing north—southobservedat 910 s at thesitesonprofile S. The solid line is thetheoreticalresponseto themodel illustratedin Fig. 13a.

THE NORTH AMERICAN PLAINS CONDUCTIVITY ANOMALY S 187

responses.The arcuatenatureof the top of the sent the ratio of the total horizontalmagneticfieldmodel is still required,as is the high conductivity, to the verticalmagneticfield, themaximumof theandthesetwo mustbe a featureof anysingle-body realGDS transferfunction responseto this modelor multi-body model that satisfiesthe data. is not at this period, but is at 1000 s —

although the differencein the transferfunctions

5.3. GDSresponses between1000 s and1 h is within usualdataerrors(±0.05) for theseestimations.

Although the 2D trial-and-errormodellingcon-centratedsolely on the MT data, the theoretical 5.4. Gravity data interpretationGDS responseof the resulting single-bodymodel(Fig. 13a) compareswell with the observations The gravity dataalongprofile S are illustrated(Fig. 14) at a period of 910 s. The data are in Fig. 15 (solid line). Owing to the well-knownscatteredat theselong periods,but the shapeof non-uniquenessassociatedwith modellinggravitythe responsesagreewell. Thesestatementsare data,we decidedto usethe dataonly to determineequally true for the multi-body model (Fig. 13b) if anomaliesin the data could be modelled byalso, bodiesof thesamegeneralcharacterasthatimaged

Both models(Figs. 13a,b) exhibit a maximum by the MT data.responsein the real parts of both the anomalous The generaleastwardincreasinggravity trendeast—westhorizontal magneticfield and the verti- in Fig. 15 can be attributed to the thinning sedi-cal magneticfield at a periodof — 2000 s, which mentsof the Williston Basinwith a densitycon-is in reasonableagreementwith the commentby trast (comparedto the ‘host’ backgroundrock) ofAlabi that the anomaly‘has a peakresponsenear —180 kg m3. Thesesedimentscan be modelleda periodof onehour’ (Alabi, 1974,p. 202). How- by a thinning wedge as illustrated in Fig. 15.ever, becausethe GDS transfer functions repre- There are two positive anomaliesin the gravity

Gravity (observed) Gravity (modelled) Aeromagnetic-~:~0 __________ ~

TOBE

I ANACP I I

0 100 200 300 400DISTANCE)km)

Fig. 15. Upperpart.Thegravity(solid line) andaeromagnetic(dottedline)dataobservedalongprofile S.Thedashedline representstheresponseto thegravity model illustrated in thelower part of thefigure. Lowerpart. A model that explainstwo of theobservedanomaliesin thegravity data.Thesedimentsaremodelledwith a densitycontrastof —180 kg m ~, theNACPwith + 270 kg mand theTOBE with +180kg m3.


data that lie directly abovethe NACP and TOBE 5.5. Aeromagneticdataconductivity anomalies — theseare at distancesof 180 km and350 km on Fig. 15. The anomaly The aeromagneticdataabovethe NACPin thiscentredon 180 km is of 50 km in spatial wave- regionare anomalouslyfeatureless(Fig. 15, dottedlength,whereasthe anomalycentredon 350 km is line; also Fig. 7). This easternedgeof this mag-much narrower,of 6 km spatial wavelength. netic quiet zone appears to correlatewith the

By placing an arcuatestructurewith its top easternedgeof the NACP structureas modelledsurfaceat 10 km andwith a shapesimilar to that by the MT data (Fig. 13). The western edgesofof the NACP anomaly(Fig. 13a),but of slightly the two do nothavethe samecorrelation.smallerspatialwavelength,and a positivedensitycontrastof 270 kg m3, the gravity anomalycan 5.6. Geothermaldatabe explainedalmost totally (Fig. 15, dashedline).The differencesin spatialwavelengthbetweenthe The heatflow dataalongprofile S is plotted inmodels from thesetwo techniquesis not signifi- Fig. 16, togetherwith two 2D modelsof the crustalcant given the inherentnon-uniquenessesof the structureof geothermaleffects to explain the ob-two methods.The actualdensitydifferenceis, of servations(Majorowicz et al., 1989).The heatflowcourse,dependenton the thicknessof the limbs of datahavebeennormalizedto an assumedregionalthe structure.However, a lower bound is givenby heat flow of 60 mW m 2, and the shapeof themodelling the structureas‘filled-in’, which gives a anomalousbodieshavebeendictatedby the MTdensitycontrastof 110 kg m~. resultsof bothourselvesandof RankinandPascal

For the TOBE anomaly,a thin vertical struc- (1990) (see Majorowicz et al., 1989). The twoture, of dimensionssimilar to thoseof the body models representeither anomalousheat genera-that describethe MT observations,with a positive tion of ~ A = 4 ~tWm ~ (Fig. 16A), or anomalousdensitycontrastof 180 kg m3 can explain quite thermal conductivity causing a rise in the iso-well the gravity data(Fig. 15, dashedline). therms(Fig. 16B). The interpretationfavouredby

Q (mWm2)~Q(mWm2) -4i--welI data estimates 20

~:~ __________________________

0 20 40 60 80 00 120 40 60 4C I I I I I I I I

Distance (km) 0 20 40 60 80 100 120 140 160.~ C Distance (km)

4 . 1 _/oo~K=2.5 ~—2oo

~~.24A=4.O~~ 2C~IIiiI

0 20 40 60 80 100 120 140 160 0 20 40 60 80 100 120 140 I60

Distance (km) Distance (km)

Fig. 16. The heatflow data acrossthe profile S andits interpretationin termsof (A) anomalousheat generationmodel, and (B)anomalousthermalconductivitystructure(takenfrom Majorowiczet a)., 1989).

THE NORTH AMERICAN PLAINS CONDUCTIVITY ANOMALY 5 189

Majorowicz et at. (1989) is of the former, with not be explainedby induction alone, and accord-mineralization and redistribution of radiogenic ingly thesedatawereincompatiblewith eachotherelementsin the crustduring the THO collision, in that no 2D inductionmodel could be foundto

explain both responses.Forward modelling wasattempted,however, using Wright’s (1969) pro-

6. Porath’sBlack Hills data gramme but with derivation of the anomaloushorizontal field throughHilbert transformationof

It has long beencontendedby Gough and his the vertical field (Siebertand Kertz, 1957). There-colleaguesthat the NACP structure‘defies’ 2D fore they concludedthat the spectreof currentmodelling. Given that we are able to model our channelling had to be invoked to explain thedata from profile S by a 2D model, we wish to observations.examinethe basisof that statementwith particu- However, it is now known that the anomalouslar referenceto the GDS data from across the horizontal field can be larger than the normalBlack Hills. horizontal field above a 2D body by induction

Illustrated in Fig. 17 are the vertical-to-normal alone(see, for example,Dupis and Thera,1982;east—westhorizontal(Z/D~,crosses)and anoma- Jones,1983b).bus-to-normal east—west horizontal (D11 / D5, In order to test the statementby PGCconcern-pluses)magnetic field ratios observedby Porath ing theapparentincompatibilityof thesedata,thisand co-workers (Porath, Gough and Camfield, Hilbert transformation coupling between the1971; hereaftercalledPGC)at a periodof 3000 s anomalousfields was investigated.The Z/ D~datafor five locationscrossingtheBlack Hills of South were interpolatedonto a regular spacing usingDakota.Theseare amplitudedataonly, andit was quasi-Hermitesplines (IMSL routine IQHSCU),noted by PGC that the phasesof the fields are whichhavethe propertyof being continuouswithvirtually zero,i.e. thesedataare equal to the real a continuousfirst derivative,but not a continuouspartsof the ratios.Note that the Da/Dn response second derivative. This interpolation procedurereachesa maximumof — 2.35, i.e. the anomalous was found to result in a curve which passedhorizontalmagneticfield is 2.35 timesas large as through all the datapoints but which had littlethe normal horizontalmagneticfield. apparentextraneousstructure(Fig. 17, full line).

PGC commented that this situation, of the Theinterpolatedregularly-spacedZ/D~datawereanomaloushorizontal magnetic field above an then Hilbert transformedby Fourier transforma-anomalybeing larger than the normal field, can- tion, exchangingtherealandimaginaryparts(with

the necessarysign convention),andinverseFour-ier transformation.The resulting Hilbert-trans-

4 I I I formed curve is illustrated in Fig. 17 (long and— Z/Dn HIL short dashedline), and, to within observational— — — 3t{Z/Dn}

z,’o accuracy, it passesthrough the DU/D~ data.2 1 F~D Accordingly,onecanbeconfidentthat thesedata

/.k,~,- are, in fact, compatible,and that a model whichD~C U:~T/ I, ~ DRA can explain one set of responsesmust,by induc-

o —..es”’~”~ .1 -~+-—___—— tion, alsoexplain the other.Also illustratedin Fig.

7 17 is the GDS transferfunction Z/D calculatedfrom the derivedresponses.The maximumin the

I S I 200 I 400 transferfunction is — 0.9, which is large.Distance 1km) The data publishedby PGC were scaledand

Fig. 17. The GDS Z/D,, (crosses) and D0/D0 (pluses) re- are illustrated at three periods in Fig. 18. Thesponsesobservedat 3000 s at stationsacross theBlack Hills of error bars on the data at penodsof 1500 s andSouth Dakota by Porath et al. (1971). The full line ts aninterpolationof the Z/D,, dataonto aregularspacing,and the 3000 s representthe scatterfrom the two eventslong andshort dashedline is its Hilbert transformation. analysed(their Fig. 2). The dataare not of suffi-


-300 -200 -100 0 100 200 300 in Fig. 18 are the realpartsof the ratios only, the0 .~ I I ~___~SS

1IS~L~ 5Am’ 5O0n~ imaginary parts were less than 10% of the real0.8flin 5Qfl

E100Am accord with the observationalevidenceof zero—.2 j; parts over the whole period range, which is in

phase for these anomalous fields above theI-

S is Isw anomaly (see above). This in-phaserelationshipl000nm betweenthe anomalousfields is not surprising

8 given Summers’(1981, 1982)discussionof currentS channellingwhich demonstratedhow this occurs

10 I I I I for highly conductiveanomalies.1500 It mustbe emphasizedthat thereis virtually no

2resolution whatsoeverof the shapeof this highly

1-conducting body beneath the Black Hills from

‘4 thesedata— a block of 0.1 ~ m, 80 km wide and4 km thick would also adequatelyexplain theobservations.Such a block body would not have-2 I I I I the extreme variations in the ratios Z/D~and3000

2 r~,~, D~/D~ for distancesclose to 0 km as for thearcuateshape. However, an arcuateshape waschosen to be consistent with the body imaged

beneathprofile S. Obviously, further MT profilesneedto be undertakenin the Dakotasto improveour image of the NACP at theselatitudes.—2 I I I

2 •Z/Dn ~‘ 7200oDo/Dn HIL/~’

DRA 7. Conclusions

i the MT phaseresponsesobserved along threeWe havepresenteda qualitative summaryof

profiles acrossthe NACP and TOBE structures.I I I I Theseresponses,displayed as contouredpseudo-

-300 -200 -100 0 100 200 300DISTANCE (km) sections,indicatethe following features.

Fig. 18.A 2D model of theGDS responsesobservedby Porath (1) The lateralvariatonof the sediments.etal. (1971). Themodel consistsof anarcuatefeature,of 0.1 l~ The baseof the conductingzoneon profile S wasm, buried within a hostwith a sedimentary structure as in- shown to be the baseof the Upper Palaeozoicterpreted by RankinandReddy(1973). Thedata arethe Z/DC

strata (the Ashern Dolomite) by Jones (1988a).(solid circles) and Da/DO (open circles) responsesobserved at

three periods at stations across the Black Hills of South Well revealedin the pseudo-sectionsis the thin-Dakota. The responsesto themodel aregivenby thefull and ning of the sedimentsfrom west to east.dashedlines for Z/D,, and Da/D~respectively. (2) The lateralvariationof the basement.

On the profiles M and N the basementis ap-parently more resistive to the east than to the

cient quality to warrant extensive2D modelling, west. Sucha featuremaynotbe resolveablein theHowever, it is of interest to note that a feature data from profile S due to the maskingeffect ofwith the sameshapeas under profile S, i.e. an the TOBE anomaly,although thereis the indica-arcuateshapeof 80 km width of 0.1 ~ m, buried tion in the ~ phasesfrom the easternmostsitein a hostwith a sedimentarystructureas interpre- that this is also true at the latitude of this profile.ted by RankinandReddy(1973),canexp!ainwell Thus,from the tectonicmapof Fig. 1 we caninferthe observationsat all periods(Fig. 18). Illustrated that the mid-crust of the Superior province is


more resistive than the mid-crust of the Trans- that an anomalyexists within the crust directlyHudsonOrogen. below 103° W in southernSaskatchewan.The

(3) Thevariation alongstrike of the NACP. anomalyhasthe following characteristics.The MT datapresentedhereinshowsconclusively (1) Its top surfaceis at adepthof — 10 km.that theNACPis not thelinearcontinuousfeature (2) It hasawidth of some80 km.inferredfrom the GDS work (Fig. 1). In contrast, (3) It appearsto be arcuatein section.from thesethreeMT profiles in Saskatchewanit is (4) It is asymmetricwith ahigherconductanceapparentthat thereis abreakat — 510 N, with a to the west comparedto theeast.shift to the North and West for a possiblecon- (5) If it is a singlebody, then it mustbe thin,tinuation of the samestructure. If the features not morethana few kilometresin thickness.Anyobservedon profiles S at — 103°W andon N at greaterthicknessextentis not penrnssible.— 105.15°W, but which are absentfrom profile (6) It mustbehighly conductive,of conductiv-M, areindeedmanifestationsof the samegeologi- ity greaterthan1 Smcal structure, then one possible interpretationis (7) An explanationof such high conductivitiesfor a massiveNW—SE trending sinistral fault in in terms of conducting fluids, such as Archeanthe deepcrust,with a movementalong strike of brines with a 20 wt% NaC1 solution at 3750 (seesome 100—150 km, cutting the NACP at 51°N. Kurtz et al., 1986), requires implausibly highDextral NE—SW trendingfaults in the basement, porositiesof the order of 12—20%.with a small movementalong strike of some 20 (8) If the positive gravity high is due to thekm, are interpretedto existbeneaththe sedimen- samefeature, then the anomaly has a densitytary cover of the Dakotas (Kiasner and King, contrast(comparedto the hostrock) of approxi-1986) from displacementsof a magnetic low mately 270 kg m3.boundaryzone. An alternative geometryfor the (9) There appearsto be no magnetichigh orNACP could include a ‘bend’ to the northwest low associatedwith the anomaly.On the contrary,(see Fig. 1) suchthat it wasmissedby profile M. the NACP body is in a magneticallyquiet region.If this is the casehowever,the NACPmustbe at Nevertheless,the apparentlack of information is,beast50 km west of the westernend of profile M in itself, information.as thereare no phaseeffectsevident in the data (10) If the samestructureis causing the MTfrom profile M. anomalyas well as the thermalanomaly,then the

(4) The correlation of the NACP with other featureis eitherdue to enhancedheatgenerationgeophysicaldata. or to enhancedthermalconductivity.In this paper,we haveshownthat the location of Various possible explanations could be pro-the NACP on profile S correlateswith a gravity posed to explain the MT, the gravity, thehigh, a magneticquiet zone, and a geothermal aeromagnetic,andthe thermaldataindependentlyanomaly. In the MT, gravity, magneticandheat of each other. However, there is a reasonableflow data thereis concurringevidencefor a non- probability that the collocationof all four of thesecontinuousgeophysically-observablefeature that anomaliesis dueto a singlecause.Also, thissingleappearsto terminateat — 51°N. Furthermore, causemust explain the very long extent of thefrom these other geophysicaldatasetswe have NACP andits associationwith the Trans-Hudsonconjectureda NW—SE trending sinistral fault in Orogen.A completeinterpretationof thesedata-the deepcrust,which displacesthe NACP feature sets is out of the boundsof this paperandwill besome100—150km alongstrike, from 51°N, 102° the subject of further study, but two possibleW to 52° N, 106°W. This feature is also ap- ‘end-member’ explanations can be examined;parentin the tectonicboundarylineationsdrawn where the term end-memberis used becauseinon the compiledaeromagneticand Bouguergray- onescenariothe conductingmaterialcomesfromity anomalymapsby Greenet al. (1985). deepbelow, i.e. the mantle,whereasin the other

From analysis and modelling of the available the materialcomesfrom thesurface.geophysicaldataalong profile S, we canconclude One possible explanation for the NACP

192 A.G. JONES AND iA. CRAVEN

anomaly, originally suggestedby Fowler and lead to removalof the magneticgrains (this ex-Nisbet (1985) to explain the subsidenceof the plains the aeromagneticdata). The body couldWB, is of a slab of oceanic,mafic continental,or havea concentrationof radiogemcelements,lead-continentalmargincrust that was overthrustdur- ing to a higher than backgroundheatgenerationing the Hudsonianorogenyandthat,at deepcrust (not likely, but explains the geothermaldata).ambient F—T conditions,recrystallizedto eclogite Finally, the body could havea concentrationofover a very prolongedtime scale.Independently, highly conductivemineralsfrom the mantle,withGuptaet al. (1985) and Greenet al. (1985) also carbonbeing onepotential candidate(Duba andconsideredthe role of obductedoceanicrocks, in Shankland,1982).To explainconductivitiesof thetheir casesuggestingthat theserocks underwent order of 2—5 S m ~ only requires a graphiticserpentinization.However, experimentaldata on carbonfractionof theorder of 0.1%(extrapolatingeclogite suggestthat at mid-crustal temperatures from Fig. 3 of DubaandShankland(1982)).How-its conductivity is two order of magnitude too ever, although this scenariomay well explain asmall(LastovickovaandParchomenko,1976),and body beneaththe centreof the Williston Basin, itOlhoeft (1981), in avery carefullaboratorystudy, is not likely that such an emplacedbody wouldillustrated that hydratedminerals are not intrin- extendfrom the Black Hills to northernCanadasically conductingbut that it is fluids releasedby which is the tracedNACP from GDS studiesdehydrationreactions that are the causeof the (Fig. 1).high conductivities measuredin laboratories.As Finally, a moreexotictheory for the generationproposedby Jones(1987a),fluids releasedby such of the WB is by meteoriticimpact (SearsandAlt,reactionscouldbecometrappedbeneathanimper- 1989),andJones(1987b)hasexaminedthe poten-meablebarrier which forms at temperaturesof tial for the generationof conductinganomaliesinaround400°C.Suchtemperaturesarein reasona- the lowercrust from suchenergeticimpacts.How-ble agreementwith the geothermsfrom the model ever, such a theory raisesmore questionsthan itof anomalousthermal structure (Fig. 16B) pre- answerswith regardto the NACP structure.sentedby Majorowicz et al. (1989). Additionally, Obviously,manyquestionsneedtobe answeredif theserocks are abovetheir Curie temperature, before the tectonichistory of the region, and inthen therewould exist a magneticquiet zone as particular the role of the NACP, can be un-observed. However, the porosities required for ravelled.conductivitiesgreaterthan 2 S m~are implausi- Is there a geothermal high along the wholebly high (12—20%, dependingon the poregeome- length of the NACP?try). Also, this would not explain the present-day Is therea gravity high alongthe wholelengthofheatflow anomaly. theNACP? Certainly the highin Fig. 6 on profile

Another possibleexplanation for the NACP S is not readily traceableon profile N.structure could be the fairly recent (compared Is therean aeromagneticquiet zonealong thewith the THO at 1.8 Ga)intrusionof a magmatic wholelength of the NACP?body from the mantleinto the crustalonga zone To what extentdoesthe Williston Basinmodifyof previous weakness,i.e. the THO. Ahern and andalter the NACP structure— if at all?Mrkvicka (1984) proposedthe emplacementof a It is hoped that thesequestionswill becomehot body into the subcrustallithosphereat 400 addressedas both LITHOPROBE (Hajnal andMa to explain the subsidenceof the Williston Lewry, 1988)and COCORP(Brown et al., 1989)Basin. One possiblescenariothen for the crustal direct their attention to the Williston Basin andbody is for low densitydifferentiafrom this sub- the Trans-HudsonOrogenoverthe nextfew years.crustalbodyto havebecomeemplacedin the midcrust. As the body cooled to ambient tempera- Acknowledgementstures,it would havea higherdensity(this explainsthe gravity observations).Vigorous fluid circula- We wish to acknowledgePete Savageof Pan-tion subsequentto the injection of the bodycould CanadianPetroleumLtd., Calgary, for the MT


datafromprofile 5, andwe thankall thosefarmers associatedfault systemsbelowPrinceEdwardIslandimaged

who kindly let us perform our ‘magic’ on their by the magnetotellurictechnique. Can. J. Earth Sci., inpreparation.

land durmg the recordmgof profiles M and N inDrysdale,J.A. andHomer R.B. 1986. CanadianEarthquakes

thesummerof 1987.Thefield crew for theselatter — 1984. Seism.Ser. Earth Phys.Branch, No. 93.

profiles included three summer students, Rob Drysdale, J.A., Homer, R.B., Wetmiller, RJ., Stevens,A.E.,Stevens,Blair Walker and John Glasko, and a Rogers, G.C. and Basham,P.W., 1985. CanadianEarth-

Chinesevisitor NongWu, andwe are grateful to quakes— 1983. Seism.Ser. Earth Phys.Branch,No. 92.Drury, M.J. and Niblett, E.R., 1980. Buried oceancrust andall of them for their hard work and pleasant . . .

continentalcrustgeomagneticinductionanomalies:a possi-dispositions. S ble association.Can.J. EarthSci., 17: 961—967.

The potential field mapsand the earthquake Duba, A.G. and Shankland,TJ., 1982. Freecarbonanddee-epicentresof the regionwereprovided by staffof trical conductivity in the Earth’s mantle. Geophys.Res.

the GeophysicsDivision, GSC,GeologicalSurvey Lett., 9: 1271—1274.1~f’ .1 Ti 11’ ‘ Dupis, A. and Thera,A.L., 1982. Naturalelectromagnetismin01 ~.anauaruojication i~O~~ theRhineGraben.Geophys.J.R.Astron.Soc.,68: 545—557.

Dutch, S.!., 1983. Proterozoic structural provinces in the

north-centralUnitedStates.Geology,11: 478—481.References Fowler,C.M.R. andNisbet,E.G., 1985. Thesubsidenceof the

Williston Basin. Can.J. Earth Sci., 22: 408—415.Ahern, J.L. and Mrkvicka, SR., 1984. A mechanicaland Gamble,T.D., Goubau,W.M. and Clarke, J., 1978. Magneto-

thermal model for the evolution of the Wiffiston Basin. tellurics with a remotereference.Geophysics,44: 53—68.Tectonics,3: 79—102. Gendzwill, D.J., Homer, R.B. and Hasegawa,H.S., 1982. In-

Alabi, A.O., 1974. A Study of the North AmericanCentral duced earthquakesat a potash mine near Saskatoon,Plains Conductivity Anomaly. PhD Thesis, University of Saskatchewan.Can.J. Earth. Sci., 19: 466—475.Alberta. Gibb, R.A. and Walcott,R.I., 1971. A Precambriansuturein

Alabi, A.O., Camfield, P.A. and Gough,D.I., 1975. The North the CanadianShield. Earth Planet.Sci. Lett., 10: 417—422.AmericanCentralPlains anomaly.Geophys. J.R. Astron. Gough,D.I. andCamfield,P.A., 1972. ConvergentgeophysicalSoc.,43: 815—834. evidenceof a metamorphicbelt throughtheBlack Hills of

Berdichevsky,M.N. and Dmitriev, V.1., 1976. Basic principles SouthDakota.J. Geophys.Res.,77: 3168—3170.of interpretationof magnetotelluricsoundingcurves.In: A. Green,A.G., Cumming,G.L. and Cedarwell,D., 1979. Exten-Adam (Editor), GeoelectricandGeothermalStudies.KAPG sion of theSuperior-Churchillboundaryzoneinto southernGeophysicalMonograph,Akadémiai Kiadô, pp. 165—221. Canada.Can.J. EarthSci., 16: 1691—1701.

Bostick, F.X., 1977. A simple almost exact method of MT Green,AG., Hajnal,Z. andWeber,W., 1985. An evolutionaryanalysis. In: Workshop on Electrical Methodsin Geother- modelof thewesternChurchillProvinceandwesternmarginmal Exploration. U.S. Geol. Survey, Contract No. of the SuperiorProvince in Canadaand the north-central14080001-8-359. UnitedStates.Tectonophysics,116: 281—322.

Brown,L.D., Kaufman,S., Nelson,K.D. andOliver, J., 1989. Green,A.G., Milkereit, B., Mayrand, L., Spencer,C., Kurtz,Intracratonic basins, Proterozoic layered complexes, and R.D. andClowes,R.M., 1987. Lithoprobeseismicreflectionfluids in thecrust:A threecomponentprogramof COCORP profiling acrossVancouverIsland. Geophys. J.R. Astron.

deepseismicreflectionresearch.NSFGrantApplication. Soc., 89: 85—90.Camfield, P.A. and Gougii, D.I., 1975. Anomalies in daily Gupta,J.C., Kurtz, RD., Camfield, P.A. and Niblett, E.R.,

variationmagneticfields andstructureundernorth-western 1985. A geomagneticinduction anomaly from IMS dataUnited Statesand south-westernCanada.Geophys.J.R. near Hudson Bay, and its relation to crustal electricalAstron. Soc., 41: 193—218. conductivity in centralNorth America. Geophys.J.R. As-

Camfield,P.A. and Gough,D.I., 1977. A possibleProterozoic tron. Soc., 81: 33—46.plate boundaryin North America. Can.J. Earth. Sci., 14: Hajnal, Z. and Lewry, J.F., 1988. The Trans-HudsonOrogen1229—1238. and theIntracratomcWilliston Basin LITHOPROBEPro-

Camfield, P.A., Gough, D.I. and Porath,H., 1970. Magne- ject. LITHOPROBEGrant Proposal.tometerarray studies in the northwesternUnited States Handa,S. andCamfield,P.A., 1984. Crustal electricalconduc-and southwesternCanada.Geophys.J.R.Astron. Soc., 22: tivity in north-central Saskatchewan:the North American201—222. CentralPlains anomaly and its relation to a Proterozoic

Constable,S.C., Parker,R.L. and Constable,C.G., 1987. Oc- plate margin.Can.J. EarthSci., 21: 533—543.cam’sinversion:apracticalalgorithmfor generatingsmooth Hills, F.A., Houston,R.S.andSubbarayudu,G.V., 1975.Possi-models from electromagneticsounding data. Geophysics, bleProterozoicplateboundaryin southernWyoming. Geol.52: 289—300. Soc. Am. (Abstracts),7: 614.

Craven,J.A. andJones,A.G., 1990. The MagdalenBasinand Hoffman,P., 1981. Autopsyof Athapuscowaulacogen:a failed

194 AG. JONES AND JA. CRAVEN

armaffectedby threecollisions.In: F.H.A. Campbell(Edi- Morel-à-l’Huissier, P., Green, A. and Pike, C., 1987. Crustaltor), ProterozoicBasinsof Canada.GeologicalSurveyof refraction surveys across the Trans-Hudson orogen/Canada,Paper81-10,pp. 97—102. Williston Basinof southcentralCanada.J. Geophys.Res.,

Hoffman, P., 1988. United platesof America, the birth of a 92: 6403—6420.craton:earlyProterozoicassemblyandgrowthof Laurentia. Niblett,E.R. and Sayn-Wittgenstein,C., 1960. Variation of theAnn. Rev. Earth. Planet. Sci., 16: 543—603. electrical conductivity with depthby the magnetotellunc

Homer, R.B., Stevens,A.E. and Hasegawa,H.S., 1973. The method.Geophysics,25: 998—1008.Bengough,Saskatchewan,earthquakeof July26, 1972. Can. Olhoeft,G.R., 1981. Electricalpropertiesof granitewith impli-J. Earth Sci., 10: 1805—1821. cations for thelower crust.J. Geophys.Res., 86, 931—936.

Hyndman,R.D., 1988. Dipping seismic reflectors,electrically Peterman,Z.E., 1981. Dating of Archeanbasementin north-conducting zones and trappedwater in the crust over a easternWyoming and southernMontana.Geol. Soc. Am.subductingplate. J. Geophys.Res., 93: 13,391—13,405. Bull., 92, 139—146.

Jones,A.G., 1983a. On theequivalenceof the “Niblett” and Porath, H., Oldenburg, D.W. and Gough, D.I., 1970. Sep-“Bostick” transformationsin themagnetotelluricmethod. aration of magneticvariation fields and conductivestruc-J. Geophys.,53: 72—73. tures in the westernUnited States.Geophys.JR. Astron.

Jones,AG., 1983b. The problem of “current channelling”: a Soc., 19, 237—260.critical review. Geophys.Surv., 6: 79—122. Porath,H., Gough,D.I. andCamfield, P.A., 1971. Conductive

Jones,A.G., 1987a.MT and reflection: anessentialcombina- structuresin thenorthwesternUnitedStatesandsouth-westtion. Geophys.JR.Astron. Soc.,89: 7—18. Canada.Geophys.J.R. Astron. Soc.,23, 387—398.

Jones,A.G., 1987b. Are impact-inducedlower crustal faults Rankin, D. and Reddy, I.K., 1973. Crustal conductivityobservable?Earth Planet. Sci. Lett., 85: 248—252. anomalyunder the Black Hills: a magnetotelluricstudy.

Jones,A.G., 1988a.Static-shiftof magnetotelluricdata andits EarthPlanet. Sci. Lett., 20: 275—279.removal in a sedimentarybasin environment.Geophysics, Rankin,D. andPascal,F., 1990. A gapin theNorth American53: 967—978. Central Plainsconductivity anomaly. Phys. Earth Planet.

Jones,A.G., 1988b. Discussionof “A magnetotelluricinvesti- Inter., 60: 132—137.gation under the Williston Basin of south-eastern Reitzel, J.S., Gougli, D.I., Porath, D.I. and Anderson, III,Saskatchewan”by J.M. MaidensandK.V. Paulson.Can.J. C.W., 1970. GeomagneticdeepsoundinganduppermantleEarth Sci., 25: 1132—1139. structure in thewesternUnited States.Geophys.J.R. As-

Jones,A.G. and Savage,P.J., 1986. North AmericanCentral tron.Soc., 19: 213—235.Plainsconductivityanomalygoeseast.Geophys.Res.Lett., Sears,J.W. and Alt, D., 1989. Impact origin of large intra-13: 685—688. cratonic basins, the stationary Proterozoiccrust, and the

Kiasner, J.S. and King, ER., 1986. Precambianbasement transition to modem plate tectonics.Contributedpaperatgeologyof North and SouthDakota.Can.J. EarthSci., 23: ‘8th Intemat. Conf. BasementTectonics(in press)’.1083—1102. Siebert, M. and Kertz, W., 1957. Zur Zerlegungeineslokalen

Kurtz, R.D., DeLauner,J.M. andGupta,J.C., 1986. A magne- erdmagnetischenFeldesin äusserenund inneren Anteil.totelluric soundingacrossVancouver Islandseesthe sub- Nachr.Akad.Wiss. Gottingen,II, Math.Phys.Ki., 87—112.ductingJuandeFucaplate. Nature,321: 596—599. Summers,D.M., 1981. Interpretingthe magneticfields associ-

Lastovickova, M. and Parchomenko,E.I., 1976. The electric ated with two-dimensionalinduction anomalies.Geophys.propertiesof eclogitesfrom the BohemianMassif under J.R.Astron. Soc.,65: 535—552.high temperaturesandpressures.Pageoph.,114: 451—460. Summers,D.M., 1982. On thefrequencyresponseof induction

Law, L.K. andRiddihough,R.P.,1971. A geographicalrelation anomalies.Geophys.J.R.Astron. Soc.,70: 487—502.between geomagneticvariation anomaliesand tectonics. Swift, C.M., 1967. A magnetotelluricinvestigationof anelectri-Can.J. Earth Sci., 8: 1094—1106. cal conductivity anomaly in the south-westernUnited

Lidiak, E.G.,1971. BuriedPrecambrianrocksof SouthDakota. States. PhD Thesis, Departmentof Geology and Geo-Geol. Soc. Am. Bull., 82: 1411—1420. physicsMIT, Cambridge,Mass.

Madden, T.R., 1973. Instruction Manual for EMCDC and Thomas,M.D., Sharpton,V.L. andGrieve,R.A.F., 1987.Gray-EMUVC (EMCAL). Exploration Aids Inc., Needham, ity patterns and Precambrian structure in the NorthMass.,31 pp. AmericanCentralPlains.Geology,15: 489—492.

Majorowicz, J.A., Jones,F.W. and Jessop,AM., 1986. Geo- Van Schinus, W.R., Bickford, M.E., Lewry, J.F. and Mac-thermics of the Williston Basin in Canadain relation to Donald, R., 1987. U-Pb geochronologyin theTrans-Hud-hydrodynamicsand hydrocarbonoccurrances.Geophysics, sonOrogen,northernSaskatchewan,Canada,Can.J. Earth51: 767—779. Sci., 24: 407—424.

Majorowicz,J.A., Jones,F.W. andErtman,M.E., 1989. Model Wright, J.A., 1969. The magnetotelluricand geomagneticre-of terrestrial thermal effects in the area of the crustai sponseof two-dimensionalstructures.GAMMA 7, Inst.electricalconductorsof theWilliston Basin. Can.J. Explor. Geophys. Met., Technical University of Braunschweig,p.

Geophys.(in press). 102.

Documents

TheNorth American Central Plains conductivity anomaly and ...world (Van Schmus et al., 1987). In central North on the southernmost profile, and the correlation of America, it separates