THE METER READER

Coordinated by John Peirce

Glacial till: A troublesome source of near-surface magnetic

anomalies
S. PARKER GAY JR., Applied Geophysics, Salt Lake City, Utah, U.S.

otential-field geophysicists have been aware of the deleterious effects of glacial till on magnetic maps as far back as the
1940s. However, in the last decade or so, a new generation of
geologists and geophysicists entering the potential field arena
doesnt seem to have as solid a footing in geology as the earlier generation, and they are often unaware of the problems
that can arise when performing low-level aeromagnetic surveys over areas covered by glacial till. The present paper was
written to fill this information gap. In 1940, Nettleton in
Geophysical Prospecting for Oil referred to troublesome problems with magnetics in glacial areas; in 1946, Heiland in
Geophysical Prospecting stated: Magnetic interference may
occur... by magnetic rocks in glacial drift; and in 1951, Vacquier
et al. (Interpretation of Aeromagnetic Maps) stated (p. 1): There
is some benefit [to be] derived from removing the magnetometer a few hundred feet above glacial till and other nearsurface magnetic disturbances ... Today, however, one never
hears about problems resulting from low-level magnetic flying over areas of glacial deposits, although they commonly
occur without being recognized. Such problems can be a
strongly limiting factor in the accuracy of interpretations for
near-surface magnetic sources and for any attempt to map
sources within the sedimentary section via so-called HRAM
(high-resolution aeromagnetic) or depth slicing methods in
glacial till covered areas.
Brief tutorial on Pleistocene glaciation in North America.
Mountain glaciers, with their conspicuous lateral and terminal moraines, were observed from the earliest days of geology in the 1600s and 1700s, but continental glaciation was not
recognized until the early 1800s. The name most often associated with glaciation is that of Swiss geologist Louis Agassiz
(1807-1873), who became aware of its validity in 1836 after discussions with Johan von Charpentier, another Swiss geologist.
Agassiz was able to convince many of his reluctant colleagues,
both on the continent and in the British Isles, of the reality of
a gigantic ice cap covering all of northern Europe, similar to
the present-day ice cap on Greenland. In 1847, Agassiz left
Europe and accepted a professorship at Harvard, transferring
his knowledge of glaciation across the Atlantic. By the turn of
the 20th century, glacial studies by geologists on both continents had established the boundaries of glaciation in the northern hemisphere. In North America the presence of two
principal ice sheets was documented, the Labrador on the
Coordinators comments: This article is an excellent synopsis of a very real
problemthat of geologically derived noise contaminating a deeper signal. However, many designers of magnetic surveys would not agree with
Gays recommendation to fly higher above the ground to suppress the high
frequency contamination. They would argue that it is better to sample the
glacial signal accurately by flying low, and then to do ones best to eliminate the glacial signal with digital filtering, rather than using flying height
as a de facto filter. Furthermore, there may be unexpected value in the shallow signal. In Canada, such shallow signals have been used in sedimentary
basins to find kimberlites and gravel deposits, as well as to map shallow gas
reservoirs and drilling hazards caused by boulder filled channels.
542

THE LEADING EDGE

JUNE 2004

Figure 1. Extent of Pleistocene glacierization of North America. (From

von Engeln, 1949, after R. S. Tarr and L. Martin, College
Physiography, MacMillan, 1915.)

Figure 2. Photograph of glacial till in an esker at Hopeville, Ontario,

Canada. Note the large percentage of pebbles and boulders resulting from
reworking of the till by fluvial processes. (From Saunderson, Bedford
diagrams and the interpretation of eskers in Research in Glacial
Systems, 1982.)

east and the Keewatin in the west, that covered almost all
Canada and extended into the northern United States (Figure
1). Where the two sheets coalesced, as they did in the last glaciation, a single sheet, the Laurentide, resulted. Rough calculations of the thickness of ice in this sheet based on crustal
rebound in the Hudson Bay area are approximately 2800 m.
However, many geologists believe that figure should be much
higher. The two present icecaps, Greenland and Antarctica,
have, at the present timenot a glacial period but an interglacialapproximate maximum thicknesses of 4000 m and
3000 m, respectively. In the mid-1900s, four accepted glacial
JUNE 2004

THE LEADING EDGE

0000

Table 1. Areas containing documented magnetic anomalies over glacial till

Ex. No. Location
Description
1.
Northern New York
Lake Sylvia, a dammed river drainage, is filled with glacial
State
boulders. Documented by AMS*, GMS*, and a drill hole.
2.
Western New York
Cayuga and Seneca Lakes, two of the Finger Lakes, are
State
also filled with magnetic glacial materials. AMS
3.
Southern Michigan
MS ThesisGMS profiles over eskers.
4.
Southern Michigan
Glacial till anomalies caused problems in interpretation of
Albion-Scipio Oil Field
near-surface magnetic sources. AMS
5.
Northeast Wisconsin
Glacial till anomalies resulted in erroneous depth-to-basement calculations. AMS
6.
U.P. of Michigan
AMS anomalies over glacial till in the Ontonagon River
valley were misinterpreted.
7.
E. Minnesota
Many river valleys filled with glacial materials mapped by
AMS.
8.
E. Minnesota
Other river valleys in an adjacent county show well on
shade-relief magnetics. AMS
9.
N. Minnesota
Drumlins cause multiple anomalies on AMS.
10.
NE Montana
Both present-day and older, buried drainages with glacial
materials cause AMS anomalies.
11.
S. Montana
Lateral moraines of a mtn. glacier are mapped by AMS.
12.
NW Ontario, Canada
Glacial till anomalies mapped.
13.
N. Ontario, Canada
Mag. lows on AMS caused by removal of glacial till in river
beds.
14.
N. Saskatchewan,
Drumlin covered area is blanketed by magnetic anomalies.
Canada
15.
NW Alberta, Canada
River drainages coincide with mag. anomalies similar to
others shown here, although the author believed in a
nonglacial source.
16.
W. Alberta, Canada
Multiple short-wavelength anomalies drilled for kimberlites
proved to be glacial till.
17.
NE Alberta, Canada
Same results as 16.
18.

Central Alberta, Canada Same results as 16 and 17.

19.

N. Central Alberta,
Canada
SE Alberta, Canada

20.
21.
22.

NE British Columbia,
Canada
SE Alaska

Source
C. Ludwig, pers.comm. 2001
E. deRidder, pers.comm. 2001
Pub. by Onesti and Hinze, 1970
P. Millegan, pers.comm. 2003
pers.comm. 2002 Anon
pers.comm. 2002 Anon
Pub. by V. Chandler, 1985
Pub. by Boerboom, 2001
R. Ikola, pers.comm. 2002
S. P. Gay, 1990
T. Krebs, pers.comm. 2002
S. Hogg, pub.on Web site, 2002
S. Hogg, pers.comm. 2002
Pub. by Kornic, 1983
Pub. by T. Glenn, 1997

T. Rich, pers.comm. 2003

Shear Minerals, 2001 An.
Report
Montello Resources, AGS, 2000

Same results as 16, 17, and 18.

Keith Jones, pers.comm.2003

Magnetic anomalies along river systemsnot mentioned

by authors as due to glacial till.
...surficial magnetic layer [of]...glacial deposits.

Pub. by LeBlanc and Morris,

1997
Thurston et al., 1997

End(?) morraine from mtn. glacier causes very strong

anomaly, possibly from gabbro boulders.
23.
Alaska
Wide valleys infilled with glacial deposits. AMS
24.
Alaska
Multiple narrow river drainages are infilled with glacial
deposits. AMS
25.
Offshore U.K. in North Drainage system over glacially infilled rivers on formerly
Sea
high and dry British shelf, now inundated by North Sea.
AMS.
26-29. NE Montana-NW N.
Many river drainages, both old and present-day, are well
Dakota
mapped by properly processed AMS.

Schnepfe, pers. comm., 1990

J. Cady, pers.comm. 2002
J. Cady, pers.comm. 2002
J. Rowe, pers.comm. 2002

S. P. Gay, 2002

*Abbreviations: AMS = airborne magnetic survey; GMS = ground magnetic survey

JUNE 2004

THE LEADING EDGE

543

Figure 3. Glacial
drift sheets of the
central and
northern United
States. (From von
Engeln, 1949,
after Richard F.
Flint, 1947,
Glacial geology
and the
Pleistocene
epoch.) This 57year-old map
shows the age
divisions of the
glacial drift as
worked out in the
early to mid-20th
century.

periods had been recognized in North America (old divisions), but newer studies, based on modern age-dating techniques, reveal at least 10 different glacial pulses going back
approximately 800 000 years into the Pleistocene (Martini et
al., 2001).
Continental glaciers remove rock from the underlying
bedrock by a process called plucking, or quarrying. As
the glacier moves, these quarried blocks and boulders, as well
as soil and alluvial materials lying on the surface of the ground,
especially in river drainages, lakes, etc., including previous
glacial materials, become incorporated into the glacier. Since
the accumulation centers of ice in North America are underlain by igneous and metamorphic rocks of the Canadian Shield
(basement in the oil patch), all glacial materials contain a
component of these rocks and are therefore magnetic. Figure
2 shows a glacial esker with typical unsorted, unstratified
pebbles and boulders of these basement rocks in a matrix of
clay and silt.
Figure 3 shows the extent of glacial drift in the northern
United States as per the old division of glacial epochs. A comparison of this figure with a map of petroleum basins in North
America reveals that till covers the following basins (from east
to west):
1) the northern one-third of the Appalachian Basin
2) 100% of the Michigan Basin
3) the northern two-thirds of the Illinois Basin
4) the northern two-thirds of the Williston Basin
5) 100% of the Western Canada Basin, and, in fact, all of
Canada.
Types of glacial deposits and landforms and their magnetic
response. Glacial deposits take on many forms because of the
varied nature of glacial movement and because of later rework544

THE LEADING EDGE

JUNE 2004

ing of the glacial deposits by water. The materials deposited

by the glaciers are generally termed till and their landforms
are termed moraine, although some geologists use the terms
interchangeably. Ground moraine is a blanket of till of varying thickness laid down beneath a moving glacier. End moraine
is deposited at the end of a moving glacier where melting and
forward movement are in balance.
A knob and kettle topography is left by a glacier melting in place, where till sloughs off large blocks of unmelted
ice and is deposited on the sides in knobs, leaving a low
area, a kettle, where the block of ice once sat. The famous
cranberry bogs of New England are such glacial kettles.
Eskers are long, narrow, raised deposits left by streams
flowing beneath, through, or on top of glaciers. Drumlins are
deposits of bottom till in the form of long, low mounds
shaped by the glacier moving over them. A number of other
glacial landlords are recognized, but the ones Ive mentioned
will suffice for this brief discussion of typical magnetic anomalies. In Figure 4, I show the magnetic responses of seven of
the most common glacial landlords, or combinations thereof.
A brief perusal of this chart by the interested reader is recommended.
Examples of magnetic anomalies mapped over glacial
deposits. To demonstrate the common occurrence of magnetic
anomalies over glacial deposits, I thought it useful to gather
as many field examples of these types of anomalies as possible.
Some came from my own files and some from the literature, but the majority came by way of telephone calls from
geophysical colleagues in the United States and Canada. In
all, 29 examples, or example areas, have so far been obtained.
(Additional areas are solicited.) Their locations are shown in
Figure 5 and listed in Table 1. In 11 of the 29 examples I show,

Figure 6. Left: Aeromagnetic survey over Lake Sylvia, Balmat mine area,
St. Lawrence Co., New York. The drill hole encountered scattered igneous
and metamorphic boulders down to TD at 365 ft resulting from reworking of glacial till in this paleo-river channel. Right: Ground magnetic
profile and modeling results by Chris Ludwig, Denver, show that the
geometry of the interpreted source matches that of a paleo-channel under
the lake.

Figure 4. Types of magnetic anomalies over various glacial landforms.

Shapes of the anomalies vary not only with the type of landform but also
with its strike.

Figure 5. Locations of documented magnetic anomalies over glacial till

collected for this report. Numbers by each location may be used to reference the examples in Table 1 and in the text.

there was controversy as to the source of the magnetic anomalies; that is, the anomalies were at first incorrectly ascribed
to sources other than glacial till. Such controversies clearly
accentuate the need for an up-to-date paper on this subject,
with examples like the ones shown.
Space limitations permit discussion and illustration of
only five of the examples obtained. (The complete text of this
paper, with all examples, may be found at www.appliedgeophysics.com, as paper 47 under Papers and Publications.) The
first example I show here (Figure 6) is an aeromagnetic survey over a lake adjacent to the famed Balmat zinc mine in
upper New York state (example 1 in Table 1). Project geologists in the 1980s believed that the aeromagnetic high coinci-

dent with Lake Sylvia might be associated with an intrusive

body containing similar undiscovered deposits. However,
consulting geophysicist Chris Ludwig (who sent this material in 2002) pointed out that the magnetic high could be modeled by a magnetic source at shallow depth beneath the lake
bottom. The log of an earlier drill hole located in the mines
files proved this interpretation to be correct. The hole had penetrated a thick section of glacially deposited igneous and metamorphic boulders before being abandoned in these boulders.
Other lakes and river drainages in upper New York state,
for example the well-known Finger Lakes, show similar coincident magnetic anomalies, and comprise example 2 in Table
1 (provided by the late Ed deRidder). Both these correspond
to type example GL3 in Figure 4.
Figure 7 is an example (No. 10) from the authors files that
is similar to E1, except that it occurs in an area of high frequency noise (0.2-0.5 nT) over a ground moraine in NE
Montana. The anomaly over Big Muddy Creek clearly shows
through this noise, and can, in fact, be traced north into Canada
(on another survey) for a distance of 80 km more to (and
through) Big Muddy Lake in southern Saskatchewan (No. 26
in the appendix). This example corresponds to GL3 in Figure
4, and is highly relevant to the theme of this paper in that the
spectral signature of background glacial noise overlaps that
of magnetic anomalies that would arise in the underlying
shallow or mid-level sedimentary section. The inset in Figure
7 shows a ground magnetic profile in this same area that mimics the topographic profile, also illustrating the pervasive magnetic nature of the till and the pervasive magnetic anomalies
the till causes. As Thurston et al. (Structural and lithological
mapping of Archean and Helikian Basement underlying the
Liard Basin, northeastern British Columbia in Program with
Abstracts, CSEG HRAM Forum, 1997) state: The problem of
separating magnetic responses from [underlying rocks] is
complicated by the presence of a surficial magnetic layer ...
glacial deposits [that] have wavelengths on the order of 200
m ... these may be aliased into the frequency band dominated
by low-amplitude responses in the ... underlying [sedimentary section].
Just north of the area of Figure 7 is a magnetic anomaly
in the shape of an oxbow (not shown) that does not correspond
to a present-day drainage. It would be a buried or paleodrainage, GL5 in Figure 4, and would have resulted from an
earlier glaciation, not the last, or Wisconsin, glaciation. Such
ox-bows are common on the magnetic data in parts of
JUNE 2004

THE LEADING EDGE

545

Figure 7. Left: Short wavelength 2nd derivative magnetic contour map over the Early Wisconsin drift sheet in the Plentywood area of NE Montana.
Right: Note that the persistent magnetic high is located over the drainage valley of Big Muddy Creek due to deeper glacial till here, i.e. infilling of
reworked glacial boulders. (Thanks to Richard Hansen, of Pearson, deRidder, and Johnson, Denver, for the release of this data.) The inset shows ground
magnetic and topographic profiles of knob and kettle topography in a nearby area. Modeling gives a figure of 560
10-6 cgs for the magnetic susceptibility of the till here.

Alberta and Saskatchewan (J. Genereux, personal communication, 2002).

A different type of example (No. 14) is shown on the left
of Figure 8. This is from a magnetic survey over the
NEA/IAEA geophysical test area in Saskatchewan, Canada.
The published map shows myriad high frequency anomalies
trending northeast that are attributed to glacial sources. The
right side of the figure is an air photo from a Canadian textbook of a drumlinized area in Canada (no location given)
that would be similar to the area of aeromagnetic anomalies
546

THE LEADING EDGE

JUNE 2004

on the left.
Figure 9 shows another unpublished example (No. 13), this
one from the files of consulting geophysicist Scott Hogg of
Toronto. A series of succinct aeromagnetic profiles over a river
in the James Bay lowlands in northern Ontario, Canada, clearly
show magnetic lows coincident with a river. Here the glacial
till would have been removed by the river without any subsequent redeposition of boulders in the bottom, my example
GL4 in Figure 4.
Aunique and spectacular example (No. 25) of a glacial river

Figure 10. High-frequency residual magnetic map of an area in the North

Sea showing a drainage system infilled with glacial materials. (Data
provided by Jeff Rowe, Fugro Airborne Surveys, 2002.)
Figure 8. Left: A busy aeromagnetic map of short wavelength (i.e. narrow) anomalies from glacial deposits in the NEA/IAEA Athabasca test
area, Saskatchewan. (From Kornic, 1983, Geological Survey of Canada.)
Right: Air photo of a drumlinized area in Canada. The magnetic map
would have been flown over such an area. (From Thornbury, Principles
of Geomorphology, 1966, no location given.)

drainage system mapped by aeromagnetics is shown in Figure

10, unique because this area is located under the North Sea
and spectacular because of the detail obtained and the optimal processing employed by the contractor. Water depths
here are on the order of only 100 m, and the area was dry land
during the glacial epoch due to the accumulation of water from
the oceans on the earths various ice caps. Without knowledge
of glaciation and its geologic and geophysical effects, this
example would have been difficult to explain.
The remaining 24 examples of magnetic glacial till anomalies may be viewed on the Web site.
Conclusion. Multiple examples of magnetic anomalies over
glacial deposits of various kinds, and an understanding of the
causes and effects of continental glaciation, demonstrate the
widespread occurrence of glacial deposits in northern North
America and their deleterious effects on magnetic data. This
much was known to geophysicists in the 1940s and 1950s, but
it needs restating at the present time due to the recent emphasis on flying magnetic surveys in glacial till covered areas at
low ground clearances (HRAM), which causes the anomalies from till to be greatly enhanced (up to 10 times or more).
This problem occurs in all or parts of five major petroleum
basins in North America. In these basins, ground clearances
should be increased to 300-400 m or more to attenuate the
effects of the till, thus allowing more accurate magnetic interpretations to be made.
Suggested reading. Cenozoic history of northeastern Montana
and northwest North Dakota with emphasis on the Pleistocene
by Howard (USGS Professional Paper 326, 1960). Vertical magnetic gradiometer survey and interpretation,NEA/IAEA
Athabasca test area by Kornic (in Uranium Exploration in Athabasca
Basin, Saskatchewan, Canada, 1983). Principles of Glacial
Geomorphology by Martini et al. (Prentice-Hall, 2001). Glacial and
Fluvioglacial Landforms by Price (Oliver & Boyd, 1973). TLE

Figure 9. Stacked aeromagnetic profiles over river carved in glacial till,

James Bay lowlands, Ontario, Canada. (Data from Scott Hogg, consulting geophysicist, Toronto, Canada.)

Dedication: I dedicate this paper to two deceased colleagues who contributed

valuable material to the present study; both were excellent, skilled, knowledgeable geophysicists: Robert Schnepfe (1933-1991) and Edward de Ridder
(1939-2002).
Corresponding author: benagi@aol.com
JUNE 2004

THE LEADING EDGE

547