Potential-field Investigations of the Williston Basin Basement

Jiakang Li 1, Igor Morozov 1, and Glenn Chubak 1

Li, J., Morozov, I.B., and Chubak, G. (2005): Potential-field investigations of the Williston Basin basement; in Summary of
Investigations 2005, Volume 1, Saskatchewan Geological Survey, Sask. Industry Resources, Misc. Rep. 2005-4.1, CD-ROM,
Paper A-5, 11p.

Abstract
Simultaneous interpretation of gravity and aeromagnetic data from the Canadian part of the Williston Basin, taking
into account other available information (geology, borehole, and seismic), was carried out as part of the Williston
Basin Architecture and Hydrocarbon Potential Project. This study focuses on seamless and uniform processing and
interpretation of the potential-field data in eastern Saskatchewan and western Manitoba, with an emphasis on high-
resolution imaging of the Precambrian basement. Processing methods include gridding, reduction to pole,
downward continuation, horizontal and vertical derivatives, analytic signal, local wavenumber, and Euler
deconvolution. The interpretation based on these new attribute maps provides a characterization of structural
domains and potential basement faulting.

Major structural contrasts and domains are identified within the Precambrian basement of the Williston Basin. The
domains are determined primarily by magnetic characteristics of crystalline basement blocks combined with their
gravity attributes. The boundaries between Precambrian blocks are commonly associated with high-gradient zones
between the major high- and low-field anomalies. Within these domain blocks, major cross-cutting fractures can
also be mapped from the trend patterns of gradient maps.
Three major domains are recognized within the Williston Basin basement and analyzed for their internal structures:
1) the Sask Craton and Reindeer Zone, with its major north-south structural trend, 2) the Churchill-Superior
Boundary Zone (CSBZ), and 3) the Superior Province, which is characterized by a number of east-trending sub-
parallel domains. On top of these major structures, an additional group of linear features with weaker fabric is
identified in the newly processed data maps. These northwest-trending features cross nearly the entire study area
and are sub-parallel to the margin of the basin, suggesting that the deposition of Phanerozoic sediments could be
genetically related to them.

Keywords: gravity, magnetic, derivative, trend, boundary, fault, domain, basement, Williston Basin.

1. Introduction
This study is part of the Williston Basin Architecture and Hydrocarbon Potential Project (Phase 1) undertaken in
partnership with Industry, Economic Development and Mines (IEDM) of Manitoba, and Natural Resources Canada
(NRCan) as a component of NRCan’s second round of Targeted Geoscience Initiatives (TGIs). The project area
extends from W96º to W106º, N49º to N56º, and covers much of eastern Saskatchewan and western Manitoba. The
overall objective of the project is to produce a geological model of Phanerozoic strata of the Williston Basin over a
significant portion of Saskatchewan and Manitoba to enhance our understanding of the hydrocarbon and mineral
potential of these strata. This goal is being approached through subsurface geophysical, geological and
hydrogeological mapping, and remotely sensed imagery analysis. More details on the project are available at
http://www.gov.mb.ca/iedm/mrd/geo/willistontgi/. The specific objective of the project’s geophysical study
presented here is to incorporate seismic, aeromagnetic, gravity, and remotely sensed imagery into a seamless (in
particular, across the SK-MB border) 3-D regional geological model of the subsurface to improve our understanding
of the Precambrian basement and its possible tectonic relationship with the overlying Phanerozoic rocks.
For the purposes of this study, the Williston Basin was defined as the entire area south of the Precambrian Canadian
Shield edge in Saskatchewan and Manitoba. With the available new data compilations, it becomes possible to
achieve a seamless and consistent coverage across this entire area, link the existing isolated interpretations, provide
basin-scale perspectives on geological structure and evolution, and extend geological mapping from the exposed
regions into sediment-covered areas. A fundamental building block in these interpretations is the geophysical
domain, distinguished on the basis of its potential-field anomaly trend, texture, and amplitude. Where the basement

1
Department of Geological Sciences, University of Saskatchewan, 114 Science Place, Saskatoon, SK S7N 5E2.

Saskatchewan Geological Survey 1 Summary of Investigations 2005, Volume 1

is exposed, these domains often coincide with lithotectonic domains, which in turn depend on the scale of
investigation. Structural trends identified from the geophysical anomalies may indicate the types of crustal and basin
deformations in the area.

In order to obtain an integrated characterization of the area, it is important to utilize a combination of geophysical
attributes. Gravity and magnetic signatures most closely reflect the tectonic and structural characters of the area, and
potential-field data also provide the necessary spatial continuity of coverage. However, due to its inherent
limitations, potential-field mapping needs to be further calibrated and constrained by the available well logs and
seismic data. Gravity data have been used successfully to map crustal-scale and basin-scale structural domains. Due
to steeper distance dependence and higher susceptibility to the metamorphic and deformational processes within the
crystalline rock, magnetic anomaly patterns are likely to be the primary indicators of the character of the basement
in the covered regions.

2. Data
The datasets consist of 235,413 gravity and 15,737,915 aeromagnetic survey points. Gravity data were collected
from the 1950s to 1990s at various survey scales. In Saskatchewan and Manitoba, data sampling is highly variable,
ranging in density from <1 km in the areas of oil and mineral exploration to 10 km or even more than 20 km in the
east and northeast parts of the basin. For improved resolution, magnetic datasets interpolated to 200 m spacing were
used in this work.

3. Data Processing and Inversion Methods

To improve the resolution of gravity and magnetic images, and also to emphasize the effects of the geological
contacts critical for the structural framework of the area, our data processing generally focused on accurate
enhancement of the short-wavelength and linear features in the data. Processing began with interpolation and
downward continuation of the point-gravity data using the Continuous Equivalent Source technique and derivation
of the horizontal- and vertical-field gradients. Based on the results of these primary operations, several advanced
attributes were also extracted. The key processing steps thus included:

1) Interpolation and gridding of the gravity data – locally, the interpolation is performed into a rectangular grid,
with the final results merged and stored in the form of a global latitude-longitude grid (see the following
section). Where necessary, the interpolation can also include downward continuation in order to bring the data
to a common datum.
2) Reduction to the pole (RTP) of the aeromagnetic data – the reduction to pole can reduce the effect of Earth
magnetic field and provide a more accurate determination of the position of source bodies.
3) First and second derivatives – the maxima of the first horizontal derivative, and first and second vertical
derivatives are commonly used to delineate the boundaries of source bodies and local anomalies.
4) 3-D analytic signal – the analytic signal could be useful to reveal the basement structure. The maxima of
analytic signal are located close to the outlines of the source (i.e., gravity or magnetic) bodies. The analytic
signal (AS) is defined from the field derivatives as follows (Roest et al., 1992):
2
⎛ ∂P ⎞ ⎛ ∂P ⎞ ⎛ ∂P ⎞
2 2

AS = ⎜ ⎟ + ⎜⎜ ⎟⎟ + ⎜ ⎟
⎝ ∂x ⎠ ⎝ ∂y ⎠ ⎝ ∂z ⎠
5) Local wavenumber attributes called TDR and THDR – these operations suppress the longer wavelength
anomalies and emphasize the effects of the sedimentary cover and of the basement. Local wavenumber
attributes are defined from the vertical derivative (VDR) and the total horizontal derivative of the gridded field
(THDR) as (Fairhead et al., 2004; Verduzco et al., 2004):

Tilt derivative:
VDR
TDR = tan −1 ( );
THDR
Total horizontal derivative of TDR:
2
⎛ ∂TDR ⎞ ⎛ ∂TDR ⎞
2

TDR _ THDR = ⎜ ⎟ + ⎜⎜ ⎟⎟ .
⎝ ∂x ⎠ ⎝ ∂y ⎠

Saskatchewan Geological Survey 2 Summary of Investigations 2005, Volume 1

6) 3-D Euler deconvolution – this attribute provides estimation of source location and depth, including the faults
and other sharp structural contrasts affecting magnetic strata. 3-D Euler deconvolution is performed through
solving the equation (Thompson, 1982; Reid et al., 1990, Silva and Barbosa, 2003):
∂ ∂ ∂
( x − x0 ) T ( x, y , z ) + ( y − y0 ) T ( x, y , z ) + ( z − z0 ) T ( x, y , z ) = − N ⋅ T ( x, y , z ).
∂x ∂y ∂z
Here, T is the magnetic or gravity field and N is the structural index corresponding to the structural type (e.g.,
distinguishing point or linear sources) of interest.

For gridded-data displays, it is important to utilize colour schemes that allow the best visual contrast and depth
while offering a natural perception of the polarity and amplitude of the anomalies. The traditional “rainbow”
palettes (used in Figure 1) may not always be the optimal choice. We experimented with a number of colour palettes

Amgal

Figure 1 - Interpolated Bouguer gravity anomaly map of study area. The white line indicates the edge of the Canadian
Shield; dashed black line represents the Saskatchewan-Manitoba border.

Saskatchewan Geological Survey 3 Summary of Investigations 2005, Volume 1

and preferred using bi-colour (blue-to-orange) palettes to display magnetic data. To improve the depth of colour
rendering, we applied histogram balancing to the palettes. Such histogram-balanced palettes are derived from the
actual data being displayed, and are designed so that each colour occupies approximately the same area of the
resulting image. This is achieved at the expense of making the colour axis non-linearly dependent on the anomaly
magnitudes, which is insignificant for visual perception. An example of such a histogram-balanced magnetic image
is shown in Figure 2.

I
NE
Z

KISSEYNEW O
SB

GLENNIE IT
KW
HR

PI
GODS nT
AWEKKA LAKE
WAP
TOBIN
ON
MOLS
FLIN FLON ISLAND
SMEATON BZ LAKE
CS

SASK
? BERENS
CRATON RIVER
HU

RZ

SUPERIOR
SR B Z
MB
OL

UCHI
DT

PROVINCE
ENGLIS
H RIVER
HR
SB MIN N

BIRD RIVER
W RA

Z
YO TO
C

ER
I NN IP EG RIV
W
? OON
?
G

WABIG

Figure 2 - Map of aeromagnetic anomaly reduced to pole and interpolated. For display clarity, a histogram-balanced blue-to-
orange colour palette is used. With histogram balancing, every colour level occupies approximately the same area of the
image. The white line indicates the edge of the Canadian Shield; green lines represent boundary-zone margins; black solid
lines represent inter-domain boundaries; and dashed grey lines represent intra-domain boundaries between possible sub-
domain blocks. Green labels indicate major tectonic regions: CSBZ, Churchill-Superior Boundary Zone; HRSBZ, Hearne-
Reindeer-Sask Boundary Zone; RZ, Reindeer Zone; and SRBZ, Sask-Reindeer Boundary Zone. Structural domains are
labelled in black.

Saskatchewan Geological Survey 4 Summary of Investigations 2005, Volume 1

4. Parallel Processing
Analysis of large geophysical datasets, such as the gravity and aeromagnetic datasets of this project, poses stringent
requirements to the organization and their efficient handling of the potential-field data. The sheer volumes of the
datasets and their complexity (grids of variable densities, point and flight-line readings; seismic lines and well logs)
first require careful pre-processing (data management, interpolation, and filtering) that cannot be carried out in a
single process. The area thus has to be subdivided into smaller blocks that can be processed either sequentially or in
parallel, using a multi-processor computer. Sequential data processing is most common, and was used in the
processing examples presented below. However, in fulfillment of the project objective of making a seamless and
accurate interpretation of very large datasets spanning a broad geographical area, we are also developing a
framework for uniform, accurate, fast large-scale processing of potential-field datasets. This objective is achieved
by utilizing Beowulf clusters as data-analysis workstations.

Apart from the volume of the data, the need for subdividing the project’s potential-field datasets into smaller blocks
is dictated by the area of data coverage spanning two UTM zones. Potential-field analysis and inversion is normally
done on a Cartesian grid, and only smaller blocks can be mapped onto such a grid with negligible distortions
(Figure 3). The extracted rectangular data blocks then undergo a series of transformations (see above), after which
they can be transformed into the latitude-longitude coordinate system and merged to form a continuous and
distortion-free attribute grid.

The novelty of our approach to potential-field data processing is in utilizing the data-management techniques that
are normally used in seismic processing. The data blocks (Figure 3) are treated as multilayered 2-D seismic traces
that are passed from one tool to another in the processing sequence, in a way that is routinely done in seismic
processing. This allows combining multiple tools into complex and well documented processing sequences

W106º W96º
N56º N56º

Y Y
Single Data Block

X X

N49º N49º
W106º W96º
Figure 3 - Schematic subdivision of the study area into potential-field processing blocks (blue lines). Red dots are gravity
stations and black contours represent the results of gravity interpolation. Details of data distribution are not essential here;
however, note the large size of the study area and non-uniform data coverage, with most observations in the areas of oil
exploration. Processing, such as interpolation from irregularly spaced survey points into a grid, is performed in local
Cartesian coordinates (panel on the right, corresponding to the shaded blue block on the left from and into which the data are
transformed using local stereographic projections. The resulting (e.g., filtered) grids are finally transformed back into
latitudes and longitudes, and merged into the original global grid. Overlapping margins of the processing blocks are averaged
in order to reduce the edge effects. Processing can be conducted concurrently using a parallel computer system.

Saskatchewan Geological Survey 5 Summary of Investigations 2005, Volume 1

that could be executed concurrently. This use of modular seismic processing enriches potential-field data analysis
with numerous additional tools (such as filtering, plotting, and potentially visualization) and provides a natural way
for integration of potential-field and seismic data.

Implementation of such a serialized and parallelized potential-field analysis is only possible with the use of a
seismic-processing package that is capable of handling data types that are significantly more general than the
traditional seismic records. Such a package was recently developed by Morozov (Morozov and Smithson, 1997).
Recently, we built into this package an integrated Graphical User Interface and implemented extensive parallel
processing functionality using the Parallel Virtual Machine. With potential-field processing and inversion tools
integrated into the system, we expect to obtain a gravity- and aeromagnetic-data processing package of unusual
pass-through, modularity, flexibility, and scalability.

5. Preliminary Results: Delineation of Boundaries and Structural Domains

Identification of crustal blocks and domain boundaries is the key application of potential fields to geological
interpretation. Previous gravity and magnetic mapping and domain analysis of Saskatchewan and Manitoba were
carried out by Miles et al. (1997), Kreis et al. (2000), and Pilkington and Thomas (2001) respectively. In the present
study, we focus on an integrated interpretation of the Williston Basin.

Major basement domains should have pronounced expressions in aeromagnetic and gravity data. Domain
boundaries cannot, however, be distinguished from other linear structural contrasts (e.g., created by stacking
accreted terranes) on the basis of potential-field interpretation alone. Thus, we prefer a conservative term “structural
contrasts” for the extended linear features observed in the maps, although many of these features could still be
associated with significant basement blocks. Only seismic profiling across such features could provide definitive
constraints of the degree of basement displacement associated with these contrasts.

In our interpretation, major basement blocks and structural contrasts are identified mainly from the horizontal and
vertical gradients and attribute maps. Domain boundaries are then roughly delineated by the areas of parallel or sub-
parallel trends that have comparable amplitudes of the anomalies. This method was applied to both gravity and
magnetic data. From a combination of gravity and magnetic results, we present the interpreted structural boundary
and domain definition based on the additional and improved potential-field attributes as well as on seismic and well-
log data and geological information.
The various magnetic and gravity attributes describe the features and patterns of the basement blocks relative to
their surrounding medium. For virtually any attribute, we look for spatial trends corresponding to the geologically
known or inferred structures. The Bouguer gravity attribute map (Figure 1) outlines the lighter and heavier masses
within the crust. Because the structural features within the crystalline basement have significantly stronger magnetic
than density expressions, and also because magnetic-data spaces are much closer, the resulting magnetic attribute
maps (Figures 2 and 4) are considerably more detailed than the Bouguer gravity map.
The Williston Basin can be subdivided into three main regions with distinct structural patterns: 1) the central region,
represented by the north-northeast–trending Churchill-Superior Boundary Zone; 2) the western region, including
Sask Craton and Reindeer Zone, with numerous northwest-, northeast- and near north-trending structural
boundaries; and 3) the eastern region (Superior Province), which is mainly composed of a north-south stack of
nearly east-trending boundaries and domains. These regions are outlined on Figures 2, 4, and 5.

a) Churchill-Superior Boundary Zone

The Churchill-Superior Boundary Zone (CSBZ), which crosses the entire Williston Basin within the study area and
is marked by north-northeast–trending linear anomalies with strong gravity highs (Figure 1); this trend changes to a
north-south orientation near the Canada–US border.
The north-northeast–trending magnetic fabric of the CSBZ contrasts strongly with the near east-trending magnetic
anomaly patterns of the Superior Province and with the more complex magnetic patterns of the Reindeer Zone
(Figure 2). Bleeker (1990) suggested that this change from magnetic low to high was related to Hudsonian
metamorphic overprinting of granulites in the CSBZ, which destroyed magnetite through biotite-, hornblende-,
and/or garnet-forming reactions. The low magnetic field over the northern margin of the CSBZ contrasts strongly
with the eastern part of the Superior Province showing several large, strongly positive magnetic basement blocks
(Figure 2). At the same time, the magnetic field character also changes significantly from north to south within the
CSBZ.
The western and eastern boundaries of the CSBZ are clearly defined by their gravity magnetic characteristics
(Figures 2, 4, and 5). The western boundary corresponds to the boundary of north-northeast–trending linear

Saskatchewan Geological Survey 6 Summary of Investigations 2005, Volume 1

structure of the Flin Flon Domain in its northern part and extends southward to the edge of the study area. Its
eastern boundary against the Superior Province is defined by a high magnetic gradient (Figure 4).

b) Sask Craton and Reindeer Zone

In the western region, we recognize eight sub-zones: the Reindeer Zone, the Sask-Reindeer Boundary Zone, the
Wapawekka Domain, the Tobin Domain, the Smeaton Domain, the Humboldt Domain, the Hearne-Reindeer-Sask
Boundary Zone, and the Wyoming Craton. The major structural boundaries and domains are shown in Figures 2 and
4. Note that from its magnetic signature, the unnamed area located south-southeast of the Humboldt Domain and
extending from Regina south-southeast to Weyburn could be a part of the Hearne-Reindeer-Sask Boundary Zone

nT/m

Figure 4 - Vertical gradient of magnetic field reduced to the pole. Tectonic zones, domain boundaries, and the exposed
Precambrian Shield boundary are indicated as in Figure 2. The white line indicates the edge of the Canadian Shield; green
lines represent boundary-zone margins; and black solid lines represent inter-domain boundaries.

Saskatchewan Geological Survey 7 Summary of Investigations 2005, Volume 1

 mgal/km
8.0

4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
-2.5
-3.0
-3.5
-4.0

-8.0

Figure 5 - Vertical gradient of gravity. This attribute is complementary to the reduced-to-the-pole magnetic field and can be
used for additional constraining the domain and zone boundaries. Note how the contrasts between the Churchill-Superior
Boundary Zone and Reindeer Zone, and also between the domains in the southwest part of the study area stand out better in
this display. Lines defined as on Figure 4.
(Figure 2). However, vertical gradient of the gravity field indicates that this block is clearly different from the
adjacent boundary zones.
From the magnetic and gravity anomaly and their vertical derivative maps, the amplitudes are clearly reduced in the
Weyburn area, in which the Phanerozoic sedimentary thickness determined from well and seismic data gradually
increases to 2600 m. The vertical gradients of potential field gradually decrease with increasing depth of the
Precambrian sources, which facilitates understanding of the depth to the basement in any area. Along the western
margin of the CSBZ, the Reindeer Zone manifests as a broad belt of low to moderate magnetic field containing
small isolated magnetic highs and short quasi-linear magnetic highs.
The Sask-Reindeer Boundary Zone, which cuts the Saskatchewan portion of the Williston Basin into two parts, is
well expressed in negative narrow linear north-south trends on magnetic and gravity gradient maps (Figures 2 and

Saskatchewan Geological Survey 8 Summary of Investigations 2005, Volume 1

5). On both sides, it has clear boundaries defined from the amplitudes and gradients of magnetic anomalies. This
zone extends from the Precambrian margin in the north, continues straight south to approximately N50º30’, and
then turns south-southeast, adjacent to the Wyoming Craton. The zone exhibits no clear gravity boundary (Figure
1), although some gravity anomalies are located within it.

The Tobin Domain is widest in the north and narrows towards the south, where overlapping magnetic and gravity
highs are observed (Figure 2). South of about N53º30’, it becomes a narrow tract with strong magnetic contrast with
the adjacent areas. The domain is characterized by strong positive magnetic anomalies and moderate-intensity
gravity anomalies.

The Smeaton Domain between the Tobin Domain and the Humboldt Domain is a low negative magnetic anomaly
area that corresponds to a gravity high. It appears as an independent weakly magnetic block placed between two
high-intensity magnetic blocks.
The Humboldt Domain shows similar characteristics to those of the Tobin Domain: high-intensity magnetic
anomaly with some interior boundaries subdividing it into sub-domain blocks. In its northern part, the trend of its
magnetic fabric is north-northeast, but towards the south, its magnetic trend is oriented north-northwest. Thus, this
domain may potentially be considered as consisting of two distinct sub-domain blocks.
The Hearne-Reindeer-Sask Boundary Zone is a northwest-trending, narrow belt with a negative magnetic anomaly
located between the Sask Craton and the Wyoming Craton.

Finally, in the southwestern corner of the study area, part of the Wyoming Craton is represented by a zone of
gravity low (Figure 1) and magnetic high (Figure 2). The Wyoming Craton is characterized by its strongly magnetic
crystalline basement and topographic high (not shown here).

c) Superior Province
From their magnetic-field character, the Superior Province domains mapped within the Precambrian Shield are seen
to extend into the eastern part of the Williston Basin. The Superior Province within the study area has been
subdivided into east-trending parallel domains (Card and Poulsen, 1998; Pilkington and Thomas, 2001). From north
to south, these domains are: Gods Lake, Molson, Island Lake, Berens River, Uchi, English River, Bird River,
Winnipeg River, and Wabigoon (Figure 2). Each of these domains has a clear margin separating it from other
domains (the separation between the Bird River and Winnipeg River domains is, however, less distinct). Most of
these domain boundaries are gradient zones with negative magnetic anomalies (Figures 2 and 4). The intensities and
trends of magnetic anomalies within the domains can commonly be used for defining the interior sub-domain
blocks.

The common characteristics of the Superior Province domains within the study area are their high-intensity
magnetic fields and clear boundaries with the CSBZ, where gravity-anomaly trends essentially coincide with the
magnetic trends. The Berens River Domain can also be subdivided into two sub-domain blocks based mainly on the
presence of distinctive linear magnetic features that may signify major faults rather than boundaries between
compositionally different blocks (grey dashed line in Figure 2; Bleeker, 1990; Pilkington and Thomas, 2001). The
southwestern sub-domain block of the Berens River Domain is cross-cut by a northwest-trending feature that
appears to also overprint the Churchill-Superior Boundary and Reindeer Zones (Figure 6). This and other similar
features are discussed below.

d) Northwest-trending Magnetic Contrasts

Along with the main structural-domain subdivisions discussed above, additional weaker linear magnetic contrasts
were also found (Figure 6). These features have short spatial wavelengths and consequently they could be related to
the upper part of the basement or associated with magnetized sediments. The northwest-trending magnetic contrasts
extend from the Sask Craton to the Superior Province, crossing nearly the entire study area and overprinting the
basement blocks with different potential-field signatures. They are located close to the Precambrian margin of the
basin, and deposition of Phanerozoic sediments may thus be genetically related to them. Similar structures may also
exist in the deeper parts of the basin, as indicated by identification of major basement faults in several seismic
studies (Hajnal et al., 1996; Hamid et al., this volume). However, observation of these structures at depth using
magnetic data is difficult because of the loss of high-frequency resolution with depth.

6. Conclusions
Using conventional and recently developed 3-D potential-field data processing methods, structural domains and
boundaries within the Williston Basin TGI study area are identified. Combinations of the horizontal and total

Saskatchewan Geological Survey 9 Summary of Investigations 2005, Volume 1

 nT

?
Figure 6 - Weak, linear, narrow northwest-trending basement contrasts (dashed black lines) presented on magnetic
background. Note that these features overprint the boundaries of tectonic zones and domains (Figure 2).
gradient, analytic signal, Euler deconvolution, and local wavenumber attributes maps offer the best horizontal
resolution, and also provide ways for estimating the depths to the magnetic sources.

The Williston Basin in Canada is subdivided into three structural regions: the Churchill-Superior Boundary Zone
characterized by gravity highs, clear boundaries, and north-northeast–trending structural trends; the Sask Craton and
Reindeer Zone with numerous near-north-south magnetic trends; and the Superior Province, a north-south stack of
structural domains with predominantly west-east trends.
On top of these regional domain patterns, weaker, linear magnetic contrasts are found close to the margin of the
Precambrian Shield. These features cut across the domain and structural boundaries and may possibly be related to
faulting genetically associated with sediment deposition in the basin. Similar features might also exist within the
deeper parts of the basin where their detection using surface data is difficult because of the loss of resolution with
depth.

Saskatchewan Geological Survey 10 Summary of Investigations 2005, Volume 1

7. Acknowledgments
Many researchers from: Saskatchewan Industry and Resources; Manitoba Industry, Economic Development and
Mines; the Potash Corporation of Saskatchewan; and the Geological Survey of Canada contributed to this project.
Special thanks go to W. Miles (Geological Survey of Canada) who provided the potential field datasets, A. Costa
(Petroleum Geology Branch, Saskatchewan Industry and Resources) for providing the well data, and S. Sule
(University of Saskatchewan) for his preliminary seismic interpretations. Olympic Seismic donated seismic data
used to calibrate the potential-field results. Critical reviews by Z. Hajnal and B. Pandit (University of
Saskatchewan), and F. Haidl and C. Gilboy (Saskatchewan Industry and Resources) have greatly helped in
improving this manuscript. GMT programs (Wessel and Smith, 1995) were used in preparation of the illustrations.

8. References
Bleeker, W. (1990): New structural-metamorphic constraints on Early Proterozoic oblique collision along the
Thompson Nickel Belt, Manitoba, Canada; in Lewry, J.F. and Stauffer, M.R. (eds.), The Early Proterozoic
Trans-Hudson Orogen of North America, Geol. Assoc. Can., Spec. Pap. 37, p57-73.

Card, K.D. and Poulsen, K.H. (1998): Geology and mineral deposits of the Superior Province of the Canadian
Shield; in Lucas, S.B., St. Onge, M.R., and Percival, J.A. (comps.), Geology of the Precambrian Superior and
Grenville provinces and Precambrian fossils in North America, Geology of Canada Series, No. 7, Geol. Surv.
Can., p15-204.

Fairhead, J.D., William, S.E., and Flanagan, G. (2004): Testing magnetic local wavenumber depth estimation
methods using a complex 3-D model; Soc. Expl. Geophys., 2004 Annual Meeting, Denver, exp. abstr., p0742-
0745.
Hajnal, Z., Lucas, S., White, D., Lewry, J., Bezdan, S., Stauffer, M.R., and Thomas, M.D. (1996): Seismic
reflection images of high-angle faults and linked detachments in the Trans-Hudson Orogen; Tectonics, v15,
p427-439.

Kreis, L.K., Ashton, K.E., and Maxeiner, R.O. (2000): Interpretive geophysical maps of Saskatchewan, Sask.
Energy Mines, Open File Rep. 2000-2, Sheet 1 of 8.
Miles, W., Stone, P.E., and Thomas, M.D. (1997): Magnetic and gravity maps with interpreted Precambrian
basement, Saskatchewan; Geol. Surv. Can., Open File 3488, maps at 1:1 500 000 scale.

Morozov, I.B. and Smithson, S.B. (1997): A new system for multicomponent seismic processing; Comp. Geosci.,
v23, p689-696.
Reid, A.B., Allsop, J.M., Granser, H., Millete, A.J., and Somerton I.W. (1990): Magnetic interpretation in three
dimensions using Euler deconvolution; Geophys., v55, p80-91.

Roest, W.R., Verhoef, J., and Pilkington, M. (1992): Magnetic interpretation using the 3-D analytic signal;
Geophys., v57, p116-125.

Pilkington, M. and Thomas, M.D. (2001): Magnetic and gravity maps with interpreted Precambrian Basement,
Manitoba; Geol. Surv. Can., Open File 3739, maps at 1:1 500 000 scale.

Silva, J.B. and Barbosa, V.C. (2003): 3-D Euler deconvolution: Theoretical basis for automatically selecting good
solutions; Geophys., v68, p1962-1968.
Thompson, D.T. (1982): EULDPH: A new technique for making computer-assisted depth estimates from magnetic
data, Geophys., v47, p31-37.
Verduzco, B., Fairhead, J.D., and MacKenzie, C. (2004): New insights into magnetic derivatives for structural
mapping; The Leading Edge, v24, p116-119.

Wessel, P. and Smith, W.H.F. (1995): New version of the Generic Mapping Tools released; EOS Trans. Amer.
Geophys. Union, v76, p329.

Saskatchewan Geological Survey 11 Summary of Investigations 2005, Volume 1