Magnetic surveys of Iceland

Magnetic surveys of Iceland

Tecfonophvsics, Elsevier 229 189 (1991) 229-247 Science Publishers B.V.. Amsterdam Magnetic surveys of Iceland Geirfinnur Jonsson, Science Inst...

2MB Sizes 2 Downloads 11 Views

Tecfonophvsics, Elsevier

229

189 (1991) 229-247

Science Publishers

B.V.. Amsterdam

Magnetic surveys of Iceland Geirfinnur

Jonsson,

Science Institute, (Received

Leo Kristjansson

and Marteinn

iJnwersi
February

2. 1989: revised version accepted

November

Sverrisson Iceland 15. 1989)

ABSTRACT Jonsson. G.. Kristjansson, Husebye, H. Korhonen Iceland. Tectonophysics,

L. and Sverrisson, and C.-E. Lund 189: 229-247.

M., 1991. Magnetic surveys of Iceland. In: S. BjGmsson. S. Gregersen. (Editors). Imaging and Understanding the Lithosphere of Scandinavia

E.S. and

Data from a 1968-1980 aeromagnetic survey of Iceland and a 1972-1973 marine magnetic survey of a part of the shelf around Iceland have been digitized and upgraded. Additional aeromagnetic survey lines were flown in 1985- 1986. After appropriate low-pass filtering and regridding. a new multicolor map of geomagnetic field intensity anomalies over the island and its surroundings was produced. The main features of the anomaly field are of two types: The first (a) comprises broad lineations subparallel to currently active or extinct spreading zones. These are believed to be mostly ,due to tilled basalt lava flows of alternating remanence polarity; in Quaternary regions, other extrusives, of subglacial origin, also contribute to the lineations. The second (b) comprises localized anomalies of the order of 10 km or less at individual volcan c centers. We describe some of these features in terms of current knowledge of the age, structure and magnetic properties of the uppermost crust in Iceland

Introduction

ponent measurements, although in 1c:eland the field is quite steeply inclined (I = 76” now, 77” for

Magnetic

The sources of the geomagnetic anomaly field, which occur on scales reaching hundreds of kilometers, lie in lateral contrasts in the magneti-

axial dipole) and the total field anomalies closely approximate those from vertical component measurements. Mathematical downward continuation does not appear to be practical in i lterpreting the present data, partly due to limitations of the data

zation vector between different media in the Earth’s crust. The main reason for using magnetic survey results as a contribution to mapping geo-

set and partly because this technique is not well suited to dealing with the sloping boundaries between layers of normally and reversely magnetized

logical features structural/tectonic

rocks

anoma!): interpretation

has traditionally trends and

been to (a) find age relationships,

in the lava pile. We prefer

a forward

ap-

factors), and (c) to estimate the size etc.) of selected rock formations or

proach, where the measurements are compared with the fields generated by plausible models of the source. The models must be consistent with the results of stratigraphic mapping and ground geophysical measurements. but unfortunately such

Even with complete coverage of all field components. aeromagnetic data cannot be interpreted unambiguously in terms of subsurface structure. Scalar measurements, such as those obtained from proton precession magnetometers, have further disadvantages in interpretation compared to com-

observations are still scarce in most parts of Iceland. Interpretation of aeromagnetic data in terms of structural models, even if only qualitative, can help in understanding results from measurements with other geophysical techniques, and it can point the way to further research. A major advantage of

(b) locate regions with unusual magnetic ties (these properties can often be linked geological (thickness.

properto other

overburden.

0040.1951/91/$03.50

‘1; 1991 - Elsevier Science Publishers

B.V

230 the

magnetic

survey

method

is the fact

that

all

sediments are essentially transp~ent to the magnetic field, as are seawater and glaciers. This allows tracing of basement ages and identification of basement structures in regions where other methods are not applicable. The magnetization vector in a rock is composed of its remanence plus an induced magnetization proportional to its susceptibility. In many continental areas, such as Precambrian shields, the latter component dominates, whereas in areas of oceanic crust, including Iceland, remanence is most important. In a large number of measurements on samples from upper Tertiary basalt lava flows, which are the chief constituent of the uppermost crust in Iceland, Kristjansson (1984) reported an average primary (thermal) remanence intensity of 3.5-4 A/m, decreasing slightly with decreasing

altitude.

In two collections

from

altogether

657

lava flows in Western Icetand, the mean primary remanence intensity in 13X nlivine-rich flows is only a few percent lower than that in the more common tholeiite flows. The induced magnetization in the Tertiary Lava pile is generally ahout I A/m. The geological structure of Icelund

The rock formations of Iceland are almost entirely igneous, having been generated in a central zone of spreading which is part of the mid-Atlantic ridge system. This zone includes several individual volcanic systems of about 10 km in width. The main active elements of each system are a fissure (or dike) swarm and a central volcano (Walker, 1963); their products are mostly subaerial lava

Fig. 1. Map showing the approximate boundaries of the present zones of volcanism and crustal spreading in Iceland as inferred from the distribution of Brunhes age rocks in outcrops. Various geographical names referred to in the text are also included.

MAGNETIC

SURVEYS

231

OF ICELAND

flows. The lava pile partly systems zone. pile

degrees.

15 and

towards

The oldest old,

and

erosion,

older

volcanic

of the central it by a few to above

sea level

in Northwestern

sections

and

(McDougall

Walker,

lava pile can be observed erally comprise

lavas

respectively

Watkins

Pleistocene

out

near this zone, the lava

‘tilted

13 m.y.

Iceland

buries

drifts

Due to subsidence

Eastern 1984;

gradually

is generally

several are

and

1977).

et al., Due

to

of the top 1 km of the directly.

only a few percent

the crust, but in the swarms 8% at sea level.

Intrusives

gen-

Matuyama

reversed

geomagnetic

m.y. ago). Other anomalies be traced

coherently

The most U.S.

Navy

survey

Project

covered

seafloor

Magnet

most

adjacent

profiles.

spreading

in

survey

out by the

197331974.

of the seas around

itself. The part north

described

(0.7-2.5 could not

aeromagnetic

area to date was carried

but not the island has been

between

comprehensive

of the Iceland

chron

over Iceland

and

interpreted

This Iceland

of Iceland in terms

of

by Vogt et al. (19bO) and others.

of this part of

their volume

exceeds Sigurgeirsson ‘s survey

In a very simplified view, in surface exposures the edges of the present central zone of volcanism are commonly set at the boundary between mally magnetized extrusives of the Brunhes

norgeo-

magnetic chron (last 0.7 m.y.) and reversely magnetized rocks of the Matuyama chron. This is indicated in Fig. 1, which also serves as an index map for the various geographical locations mentioned in the text. In the area of the volcanic zone and at some distance beyond it the influence of glacial climates on the character of the volcanism is evident, as the regularly stratified lava pile is interrupted by structures such as hyaloclastite ridges and table mountains.

In 1968, the late Th. Sigurgeirsson of the University of Iceland began a series of measurements of the geomagnetic field over Iceland from a small single-engine aeroplane (Sigurgeirssc n, 1970a). The flight altitude was generally 900 m above sea level over low areas of Iceland, 1200 m ever areas with moderate relief, and 2100 m over parts Vatnajiikull glacier. The field was measured

of the with a

novel proton precession magnetometera continuous AC signal was recorded on magnetic tape, the frequency of the signal (2-2.5 kHz) being proportional to the field intensity. During the first four years of the survey. the flight routes chosen were approximately straight

Magnetic surveys in the Iceland area to 1980

lines spaced 4 km apart and passing over recognizable features such as farms or road junctions.

Eurly U.S. and Canudian

During

work

The history of magnetic surveys in the Iceland area has been reviewed by Kristjansson (1987). Of

the

flight

an operator

viewed

vertically

down from the plane and marked these “events” on a separate channel of the tape which also recorded the field signal. The velocity of the plane was assumed to be constant between the event points, and the magnetic field was plotted by hand from average signal readings over intervals of 10 s

the early regional surveys, the best known is probably that of Heirtzler et al. (1966) over the Reykjanes Ridge, which demonstrates convincingly the presence of magnetic lineations parallel to the ridge axis. Their results were important in promoting the genera1 acceptance of seafloor spreading

(generally about 500 m of track; closer spacing was used where rapid changes in the field were taking place). The scale of the plotting sheets was

in the late 1960s. A three-component survey at an altitude of 3-4 km and a spacing of 35 km over the northern Reykjanes ridge and Iceland was carried out by the Dominion Observatory of Canada in 1965 (Haines et al., 1970). It showed positive lineations over the active volcanic zones, flanked by negative anomalies presumed to originate in rocks from the

the same as that of the main topographic map series of the Geodetic Survey of Iceland. i.e. 1 : 100,000. with 2.5 cm/1000 nT for the field. The above methods were used for surveying about 40% of the country. In the Reykjanes-Langjiikull volcanic zone, the flight lines were oriented at right angles to the dominant trend, but elsewhere the lines trended E-W or N-S.

In

mid-1972

changes

Sigurgeirsson

in his techniques.

was now recorded and a stable distance

digitally

oscillator Iceland.

on board transmitter

Position

the transmitter, along

sults of Haines et al. (1970): this pol~n~~rnia~ diifers from the values derived from the Interna-

signal

tional

circular

the tracks

was at first found

or from photographs,

distance

the

U.K.

Rugby

from arcs

I

I

I

values

variation

to a common

have been

iceland).

corrected

time (late

All the

for secular

1967) by using Observatory

in

was In

1972-1973,

the

U.S.

Defense

Mapping

Agency and several Icelandic institutions carried out a marine survey on the Icelandic shelf. Included were bathymetric, gravity, magnetic and sediment

thickness

ship’s position

I

‘.._’

(sparker)

within

measurements.

t

I

16'

/,

I

The

each survey area of the shelf

I

I

I

20"

-

Fig. 2. Survey tracks of Sjgurgeirs~n

more off Northeastern

magnetic

by up to IO0

The marine magnetic data from I972 to 1923

lished on nine sheets (Sigurgeirsson, 1970b-1985). A “regional” or baseline field was calculated as a third degree polynomial of position from the re-

I

Field

visu-

measured for this purpose. Sigurgeirsson~s maps were then redrafted on a scale of 1:250,000 (and 1 cm/1000 nT) and pub-

28"

Reference

annual means from the Leirvogur Southwestern Iceland.

but later the

transmitter

Geomagnetic

nT (and

at Sandur

spaced 3 km fl0 ps) apart.

ally, as before, from

major

the plane.

The survey lines flown

this time on (Fig. 2) are accordingly around

two

magnetic

was used to keep constant

from the Loran-C

in Western

made The

121

km I

I

and other regional magnetic surveys in Iceland. The straight-line tracks were flown between 1968 and 1972 and the curved ones between 1972 and 1980. with the exception of supplementary airborne measurements in 1984-1986 and tracks (widely spaced) of the 1972-1973 marine magnetic survey.

MAGNETI<‘

was

SURVEYS

determined

mitters

on

by

shore.

perpendicular spacing

two

The

mobile

of lo-12 with

Magnetic

by

in digital

redundant

a nominal

a Varian

V-75 proton between

field was precession

field record-

to an average

140 m in the track

dis-

line direction.

data were obtained

only off the south

(widely spaced lines of Fig. 2). for temporal field changes and

Kristjansson

(1976,

and

references

overlaps

served from values and

on

as drawings these

we digitized

1987. This

of large Tertiary

and Quaternary

volcanic

centers,

and (b) a scarp-type anomaly occurring off South and Southeast Iceland, well inside the shelf edge as defined

from the bathymetry.

as part of the et al. (1983). This

map clearly shows the reduced regularity in magnetic anomalies on a transverse zone through Iceland and the shelf, as compared with the general anomaly pattern over the Mid-Atlantic Ridge. We felt that it might be possible to discern additional information on regional trends from this extensive set of measurements by making it

field

and

included

pre-

position

and

1985

some

we also

previ-

corrected

in the published

maps.

data set from 197221973.

Furthermore,

where

we discarded

previous

the‘

where the temporal

of the field were unusually correction

attenpts

large

and

(using

Leirvogur Observatory recordings and cross line checks) appeared to have been insufficiently constrained.

In 1985 aeromagnetic

overlay to indicate trends, 1:2,000.000 scale map of Nunns

from

only

500 m between

data from a few line segments

Reprocessing of Sigurgeirsson’s aeromagnetic und of the 1972- I973 marine magnetic data

surements have been very helpful in various local geological studies. The aeromagnetic profiles were later reproduced, with a manually drawn two-color

data

which was preserved on tape, has been thoroughly inspected by us using modern comsuter graphic techniques. A number of minor accidental or systematic errors in field and position values, which were largely due to a faulty on board data logger, have been eliminated, and the regional field has

New aeromagnetic

The aeromagnetic measurements of Sigurgeirsson covered essentially all of Iceland, with the exception of part of Northeastern Iceland, and the aeromagnetic maps which arose from these mea-

the

data,

magnetic

Preparation of data base for the new magnetic map

data

were

several minor errors present

variations

to be the result

His

hand,

of about

unpublished

The marine

a degree of

noise.

processing

central

believed

and

original

1972 and 1980 were

on a scale of 1 :lOO,OOO, but

at intervals

been recalculated.

Iceland

Sigurgeirsson’s

the other

therein). The two outstanding features of the maps were (a) a concentration of localized anomalies off Western

format.

to us on tape, but contained

1968-1972,

ously

every 10 minutes.

low-pass filtering to reduce topographic effects in shallow-water areas, the data were plotted by hand at a scale of 1:250,000. These magnetic profile maps were published in reduced scale and discussed

available

data from his surveys between

km, but a few cross lines were

were recorded

and west coasts After correction

mostly

accessible

the interval

of about

trans-

ran

line

with

ings was 30 s, corresponding Positions

lines

in each area. The magnetic

magnetometer: tance

Raydist

survey

to the shore

also surveyed measured

233

OF ICELAND

survqv profiles

the University of Iceland resumed surveying in order to eliminate some

of the major gaps work. A Geometries

in the coverage of previous G-856X digitally recording

magnetometer was installed in a twin-engine plane together with a programmable Loran-C receiver and portable computer for data surveys in 198551986 concentrated

storage. Our on three criti-

cal areas, two in Faxafloi and Hunafloi bays off Western Iceland and one in the northeast. The Faxafloi survey overlaps with that of Sigurgeirsson on the Reykjanes peninsula. The Ilight altitude was generally 900 m. 1.-ith measurements made at 4 s intervals corresponding to an average distance of 280 m. Line spacing was about 3 km. Details of the procedures used in these surveys (navigation, field corrections, etc.) and of the software written for reduction of the data are described by Kristjansson et al. (1989).

234

Main charucteristics

and 1imitation.s oj’ the rnugnetrc,

duta Some main parameters in the magnetic

of the data base acquired

surveys are listed in Table

the table, Fig. 2 and the description be seen that the amount

of quantitative

tion which may be carried limited point

by several spacing

combined

survey,

or above

basement

ways known large

is quite

for tracing

features

(a) the line and

in various

(b) the altitude

accurately,

stages of the

above

variable

sea level

and

(c) the line spacing

of short-wavelength

between

adjacent

it may

interpreta-

out done on this data is

imperfections:

is different

1. From

above

profiles,

not alFig. 3. Histogram

is too

field

(< 5 km)

intensity

surveys (1968-1980

(d) the data

of anomalous

values

(i.e. observed

in nano-Teslas

and 1985-1986)

minus regional)

for the aeromagnetic of Fig. 2. Average:

nT. standard deviation:

+ 17

435 nT.

were acquired over a period of 18 years, during which time the secular variation at points distant from Leirvogur

Observatory

may not be quite the

same as that at Leirvogur, (e) the polynomial used for the 1967 regional field correction may diverge from the actual regional field, especially in offshore areas, (f) the angles between the survey lines and the dominant regional trends are variable, and (g) some minor

gaps in the coverage

magnitude caused by the procedure of first plotting the field manually and then digitizing it. Fortunately, the observed magnetic anomalies have sufficiently large amplitudes (Fig. 3) that these errors are of Little importance. For the reasons listed above, most suited for studying features

still remain.

Due to incomplete correction for short-period temporal variations and disturbances caused by the surveying

vehicle,

errors in the recorded

wavelengths, i.e. 5-100 km. However, several distinctive short-wavelength anomalies which occur

mag-

netic field values themselves may amount to a few tens of nano-Teslas. In the case of Sigurgeirsson’s 1968-1972

TABLE

our data set is of intermediate

in the surveys

survey, there are also errors of similar

will

be discussed

in a separate

section.

1

Acquisition

parameters,

survey coverage and filter characteristics

for data included in this work

Survey

Nominal

Sampling

Average

Number

Number

Distance

Area

Low-pass

Spacing

and

line

interval

point

of

Of

covered

covered

filter

after

spacing

(s)

spacing

lines

points

(km)

tkm2)

width

decimation

(km)

(km)

year

(km)

(km) Aeromag. 1968-1972

4

10

0.498

172

25,691

12,700

50,800

4.5

0.498

3

10

0.593

237

34,203

20,159

60,500

5.3

0.593

IO

30

0.142

193

70,138

9898

-

4.3

0.568

3

4

0.284

73

16,181

4576

13,700

5.1

0.568

675

146,213

47,333

125,000

Aeromag. 1972-1980 Marine t972-i973 Aeromag. 1985-1986 Total

_

-

MAGNETIC

SURVEYS

OF ICELAND

0

E

a 0 --

735

appears

on

aer(~magnetic The residual ble

magnetic

data are now all avaiia-

in computer-accessible

temporal

variations

intensity

over Iceland

to represent

conveniently

computer,

a survey

{ X, Y, field

line consists

value},

the distance

for

the residual

sized ASCII

and a number

map-projected

corrected

field

to represent

of data records X,Y

being

coordinates.

the spherical

Because

or

spacing

versus line spacing),

marine

surveys,

These data were kindly

each survey group

in terms of average

plotter

ing of that group

and was in all cases about

spac-

gridding of continuity

data

from the

4). There

for the (1987).

by the U.S. NaHard copies of with an ink-jet

are eight

colors

for low

features of the composite

map Linear features (isochrons) In the foltowing

it may be assumed

as a first

approximation that positive and negative anomalies indicate the presence of substantial bodies of

of weighted means was employed for the gridding. All survey points within a 5 km radius of the grid point were found and the field value calculated with inverse

(Fig.

provided

Data Center. were produced

D~cripti~n of ~~vidual

of of

Our magnetic data were transformed to a 2 x 2 km grid within the map projection used by Sigurgeirsson-the conical (Lambert) projection of the Geodetic Survey of Iceland. A simple method

weighted

of the marine

fields and eight for high fields; the field change between adjacent color bands is 125 nT.

4-5

km (Table 1). In order to save on computer storage and to improve the uniformity of the data set,

using these points,

by up to 5 km.

line spacing

we have displayed

tional Geophysical the screen picture

point point

outwards

the nominal

Gaussian filter window was run along each survey line. The width of the filter was determined for

after filtering we used only every fourth the 1972-l 973 survey and every second the 1985-1986 survey.

field ix believed

the real field in the two dimensions.

set making up the map of the Committee Magnetic Anomaly Map of North America

of the

a low-pass

point

hlf the

in Fig. 21

only at points where survey lines cross each other. Offshore, in areas not covered by the 1972.-1973

the digitai data and detailed documentation are available from the Science Institute of the Univer-

data (point

;tre;t

survey was twice the gridding radius, these data improves the 2-dimensional

charge,

sity of Iceland on request. in order to counteract the biased density

the

line spacing

to the closest survey point is gcnerall>

lies may be extended

(one

of the form

For a nominal

Withm

except at the edge of the area where local anoma-

files. In the

of a header

map.

survey (dense

less than 2 km and the calculated

in 1967. The data are stored

in several record)

form,

the

normal and reverse magnetization respectively. We assume, following Piper (1973, fig. 14) that in areas aider than the Quaternary, the boundaries of

distance.

The disadvantage of this method is the nonuniform low-pass filtering of the field, but in terms of

these bodies are tilted towards the volcanic zones by several degrees. It may then be calculated that,

gridding artifacts it is probably the safest way of gridding this type of data. For each grid point, records were kept on the number of survey points found within the 5 km radius and on the distance from the grid point to the closest survey point. If there is no survey point within the 5 km, a blank

for example, the anomaly corresponding to any particular major geomagnetic chron does not appear directly above the surface outcrops of lavas of that age, but will be displaced by some kilometers towards the volcanic zone. Those magnetic polarity zones in the lava pile which correspond to

___ Fig. 6. contour

Contour map of geomagnetic intensity in the region of the Reykjanes peninsula. Only positive anomalies are represents

the

+600

nT

SU = Sveifluhals

hyaloclastite

ridge;

volcanic center:

LJ = Lyngdalsheidi

anomaly

and

the contour

Br = Brennisteinsfj&ll

interval

Mountains:

volcanic shield; &J = Kalfstindar hyaloclastite

hills. Numbers

is 200

nT.

01 = Blafjiili hyaloclastite

Gr = Grindavik ridge:

village:

Hp = Hellisheidl

ridge: Ef=

refer IO local anomalies in Fig. 9.

Efstadalsfjall:

---

shown. the Iowest

Kr = Krisuvik plateau:

farm;

He = Hengill

Ru = Raudafell.

both

MAGNETIC

SURVEYS

,

OF ICELAND

237 I

, 18’

I

22”

Fig. 5. Sketch of main areas of positive

anomaties,

with numbering

to indicate

, 14”

I

particular

features

discussed

km 0

lo

20

1

30

in the text.

2.38

short subchrons tions

due

will not give rise to distinct

to

volcanism,

the

imperfect

faulting

Identification

and

numbers

linea-

continuity

other

of

filtering

the

effects.

in the text are found

in

Figs. 5 and 6.

area are quite strongly magnetized compared to the upper Tertiary lava flows elsewhere in Iceland. Their

natural

pletely dominates zation

age magnetic

anomaly

of the Reykjanes generated

continues

the central

Ridge.

It is understood

by rocks

erupted

axial high to have

during the last

0.7 m.y., or 1.0 m.y. if the Jaramillo

subchron

is

published) samples should there

which

or viscous 13-16

100 samples

is about 7 A/m,

(L. Kristjansson,

unin 20

from seven pillow lava sites is 12 A/m. also be kept in mind are abundant

of breccia

than 1.5 km, there is no clear correlation lithology and the west to east amplitude

material.

A good example

( > 600 m) series of reversely magnetized lavas below sea level underlies a thin discontinuous cover of uppermost In the southern tive magnetic

Quaternary volcanics. Reykjanes peninsula, the posi-

anomaly

is on closer inspection

to consist of several elongated westernmost of these coincides

seen

highs (al-a5). The with the Reykjanes

and Grindavik fissure swarms (Fig. 6) (Jakobsson et al., 1978), and the next with the Krisuvik swarm including the Sveifluhals ridge. Their trend is N30”-45”E. Less regular positive anomalies (a3 and

a4) occur

over

the Brennisteinsfjoll

Moun-

tains and the Blafjall ridge respectively. It is noteworthy that high-temperature geothermal areas occur close to many of the peaks of the anomalous Reykjanes peninsula field; in contrast, the large geothermal region of the Hengill volcanic center has a relatively flat magnetic field. A detailed map of this anomaly has been presented by Palmason (1987) who suggests that at least some localized lows in the Hengill area are due to the destruction of magnetic minerals by hydrothermal activity. The Hellisheidi plateau, in contrast, which lies along strike to the south-southwest and forms part of the fissure swarm of the Hengill area, has a pronounced broad positive anomaly (a5). The lavas and pillow basalts in the Reykjanes

It

that on the peninsula

occurrences

older,

reversed,

(mea-

The inten-

and the intensity

palagonite remanence.

mostly

magneti-

from prehistoric

included. At the boundary of this region we expect to have a cover of Brunhes lava flows on top of is the northern coast of the Reykjanes peninsula, which is well outside the central positive anomaly but which is entirely covered by normally magnetized rocks. Drill core data for testing this conjecture are, however, not available, except for the Reykjavik area (Sigurgeirsson, 1967) where a thick

com-

A/m

1970,1972).

lava flows on the peninsula

Iceland (a) This anomaly been

averages

by Kristjansson,

sity in an additional

in Southwestern

intensity,

any induced

components,

surements Brunhes

remanence

material which carry essentially In drill holes reaching depths

and

no net greater between variation

(see Fig. 6) (H. Franzson, pers. commun., 1989); however, the proportion of elastic material is high in drill holes on the southwestern tip of the Reykjanes peninsula, where the field amplitude is relatively low. Kristjansson’s for the effective crust

(1970, 1972) estimate of 6 A/m magnetization for the uppermost

in the peninsula

and

elsewhere

in Upper

Quatemary formations in Iceland appears to still be valid as a general average; a lower value may apply in the Hengill volcanic center. The topographic expressions of the volcanic lineaments on the Reykjanes peninsula are far too small to account for the amplitude of the anomalies. Simple calculations indicate that magnetization contrasts of the order of 6 A/m must persist to at least a depth of l-l.5 km. By heating twelve samples of Reykjanes peninsula surface pillow lavas and Reykjanes Ridge dredge material (Curie points mostly between 100’ and 2OO’C) in air we have tested the possibility of secondary heating and cooling in the Earth’s field creating a new induced or remanent magnetization of high intensity in these rocks. Our results indicate that these parameters (measured at room temperature) either remain fairly constant or decrease, after heating to 100°, 200” and 300°C. However, it is possible that other changes occur in the magnetic minerals if the heating takes place in the presence of geothermal fluids. Farther east, a fairly continuous anomaly band (a6) stretches from east of lake Thingvallavatn in

MAGNETIC

SURVEYS

239

OF ICELAND

a northeasterly

direction

through

of the LangjSkull

glacier.

(Fig. 6) this band

is composed

some

of which

correlate

activity

northeast

volcanic The main and

anomaly,

some

graphic

peaks

dalsheidi, (1982)

such

Efstadalsfjall

has described

and breccia probable netization

material than

typical

the

of the active

Thingvallavatn.

with

the

A lateral

at 64”N

(Einarsson

east,

ground

composed

farther

Lagarfljot

and

Raudafell.

Barker

tive conclusion

magnetized

pillow

South

ridge, and it is

has a higher net magflows

which

running offset

are more

through

and

lake

of possibly

Bjornsson,

as

a5 and seismic 1979)

over the eastern

active

age rocks

above

et al., 1982; Kristjans-

son et al., 1988). An E-W trend in the tectonics of these mountains (Saemundsson, 1978) is not clear in the magnetic data; instead, the anomaly to continue south, with an offshore outlier

seems to the

west past the Vestmannaeyjar Islands. The large Torfajijkull volcanic center and high-temperature area is the site of the main part of a barrier-like NW-trending low separating segments bl and b2 of the Brunhes anomaly. Segment b2, which is unusually large in terms of width and amplitude, coincides with productive fissure swarms that disappear under the western part of the Vatnajiikull glacier. The next segment (b3) is rather irregular in shape but covers northwestern Vatnajiikull including the Kverkfjoll volcanic complex, w&h has been suggested as being located over the center of current hotspot activity in Iceland. The segment b3 also reaches the mountainous areas of Dyngjufjoll and Herdubreid to the north, and has an

Iceland

end

with

landscape

of Brunhes

level (Johannesson

together

the volcanic

of the anomaly

from Dyngjufjbll,

northeastern

topo-

The Brunhes anomaly over this zone seems to be composed of four distinct segments. In the south (bl) a broad high occurs over the mountainous Eyjafjoll-Myrdalsjiikull area, which is mostly

to the northeast This,

tail towards

In Fig. 4 there are some indica-

that this segment

present

but otherwise it is difficult to correlate this zone with any particular features of the anomaly field. Brunhes age anomalies volcanic zone fb,)

tions

Lyng-

with

much as 20 km occurs between anomalies a6 at the western end of the South Iceland zone

SE-trending

Kalfstindar,

it coincide as

on Kalfstindar

area

glacial

is still farther

and

lava

late

unexpected

center of EsjufjMl.

of lake Thingvallavatn.

strongly

that such material

part

ex~nation

of several sections,

however, within

features

the eastern

On closer

i.e. beyond

of the

lineations

volcanic

to the northeast, that

the eastern

1 or 2 m.y.) farther

of

6xarfjordur

Lake

zone of the last

northeast

than it doe:; now and activity.

Brunhes

anomaly

Myvatn

and

rather

to the northwesterly

trend of the Tjarnej

Zone,

southern

the

segment

areas to the

continues

Bay, where it changes

towards

Lake

our tenta-

volcanic

(b4) covers the highly active volcanic east

the

(some time during

that it still shows some residual The most northerly

in

between

support

the zone.

evident

Vopnafj~rdur, reached

continues

end

into abruptly Fracture

of Kolbeinsey

Ridge. A similar trend is also evident ir the magnetic data of McMaster et al. (1977). and as suggested by them, this trend may consist of en echelon steps; see fig. 2 of Vogt et al. (1980) for details. In Gxarfjiirdur Bay, b4 is enveloped between the Kopasker and the Husavik faults. From this area it seems evident that the magnetic field is reformed by crustal strain. We explain pected NE-SW trending negative traversing the central anomaly Dyn~ufj~ll as a recent strain with the previously

mentioned

the unexanomaly

in the area north of zone in conjunction eastwarcl

extent

of

the neovolcanic zone south of it. This is supported by recent earthquake activity with NE:-SW epicenter trends (Einarsson, this issue) in this area. An unexpected kink into b4 from t?e east is assumed to be related to a transform fault which is recognized in Triillaskagi further to the west. The average width of the central positive anomaly (Fig. 7A) is about 20 km, whereas the “ volcanic zone”, as commonly delineated on maps of Iceland, is closer to 50 km in width. We have constructed several source models in an attempt to account for the field over the Brunhes, Matuyama, Gauss and Gilbert age formations in Northeastern Iceland, and we have come to the conclusion that sources with at least moderately tiltmg (2 7”) boundaries (cf. fig. 13a of Piper, 1973) would be necessary to represent the observed field. This model, however, does not correspond well to the

240

Matuyama

age (0.7-2.5

East of Reykjavik,

Ma) anomalies (c)

the lava pile includes 800 m

of mostly reversely magnetized rocks which have been assigned to the Matuyama

chron by Krist-

jansson

et al. (1980);

their conclusions

ported

by K-Ar

sponds

to the R2 and R3 series

dating.

This

are sup-

sequence

corre-

of fig. 3 in

Einarsson (1957)

the outcrops of which continue

to the northeast

along the coast of Hvalfjordur

Fjord. The anomaly lineation cl in Southwestern Iceland is clearly caused by this sequence. Farther northeast, towards the Langjijkull glacier, there is a surface cover of Brunhes age rocks (Piper, 1971) Mv.5

4

3

2

10

I part of Northeastern the accretion

the Reykjanes ridge to the east just south of the Reykjanes peninsula. An offshoot (~2) turns to-

I anomaly

lineations

over a

Iceland. Diagonal shading = positive field. magnetic

zone, showing

(white) magnetization. polarities

of smoothed

but this cover is probably thin. A small-amplitude negative anomaly runs from

0

I

Fig. 7. (A) Drawing (B) A hypothetical

1

survey line running transverse

Inclination

over a square-block

to

only normal (black) and reverse assumed

W”. (C) Anomaly

crustal model, assuming

an ac-

wards the north and terminates at Lake Thingvallavatn. Kristjansson et al. (1988) have located the Brunhes-Matuyama

boundary within this anoma-

ly in Mount Ingolfsfjall and it is reasonable to assume that the lowlands from the lake to the

m.y. (see text).

south coast are underlain by reversely magnetized rocks at shallow depths. The same may apply around the River Hvita (c3), but no paleomagnetic results or dates have been published from that

geology of the volcanic zone, and compared to a model with vertical boundaries does not give significantly better representation of the field (see

area. The negative anomalies (c4 and westernmost part of c5) flanking the eastern volcanic zone seem to be due to Matuyama age lava flows generated in that zone. This is consistent with the field

Fig. 7C).

mapping of polarities

tive spreading M = Matuyama; (C) in millions

rate of 9 mm per ridge flank.

B = Brunhes;

G,G = Gauss and Gilbert. (D) Time scale for of years. An unconformity

occurs beyond

5

by Vilmundardottir

et al.

(1984) around the Thjorsa River in the area of c4, Trends of the neouolcanic zones in South Iceland Anomaly trends over the two neovolcanic zones in South Iceland, referred to as Bnmhes lineations in the above, are relatively consistent within each zone. It is frequently indicated in diagrams in the literature that these zones are parallel, but our results confirm the recent observation of Eirrarsson (1988) that there is a noticeable difference in their trends. The western anomaly, inchtding the segment over the Blafjoll ridge, strikes between N30” and 35”E, in good agreement with the dominant direction of structural features in the Hengill area. The magnetic and tectonic line&ions in the eastern volcanic zone, on the other hand, strike about N50”E.

and with K-Ar

dates quoted by these authors. A

small positive magnetic lineation within the low at Kaldakvisl River may be due to formations of Jaramillo or Brunhes age. In the north, Matuyama lineations occur on both sides (c6 and c7) of the volcanic zone. The western lineation is better developed, but it is seen to terminate or bend abruptly at the NW-trending Husavik fault system in Skjalfandi Bay. Gauss (2.5-3.4 Ma) and Gilbert (3.4-5.3 Ma) age magnetic lineations (d) The Gauss (dl) and Gilbert anomalies in Southwest Iceland correspond well to the age and polarity of the outcrops mapped by Kristjansson

MAGNETIC

SURVEYS

241

OF ICELAND

et al. (1980) and Piper (1971, fig. 2). The former

units.

anomaly,

cal faults with a throw of up to 200 m or more are

north

however,

is absent

of the Reykjanes

over the anticlinal normally

peninsula.

Hreppar

magnetized

in the offshore

Lineation

area must

d2

be due to

lavas (see Fridleifsson

19580); the oldest outcrops

region

et al,,

in the area are assumed

This is at least partly

known

to occur in units as young

Hvalfjordur Two

thickness Iceland,

also several

which

assign to normal More extensive

flows which subchrons stratigraphic

in the area of the anticline

Fridleifsson

within

et al.

the Matuyama.

mapping

and dating

A Gauss age for the Adaldalur valley area (d4) and a Gilbert age for the hills west of Adaldalur is supported by K-Ar dates obtained by Jancin et al. (1985). These authors find that an unconformity still farther west probably represents the period between 9.5 continuity between d3 in Fig. 5 because to be east-west Johannesson, pers.

and 6.5 Ma. We do not assume the lineations marked d2 and the geological strike is known in the Hofsjbkull area (H. commun., 1989).

so it may

difficult to discern individual anomaly lineations and to correlate them with distinct stratigraphic

I

I

22"

I

as 3 Ma south of commun.,

1988).

approximate

mean

zone

in the lava

pile of

be expected

are considerably

formations

pers.

verti-

is the

of a polarity

thicker

continuous

that

cnly

than linear

zones

200 m give anomalies

in

older than 3 Ma.

unusually

long

period

of mostly

normal

polarity ranging in age from 9 to 10 Ma has created a very distinct positive signature known as “Anomaly 5” over the ocean ridges, including both the Reykjanes and crops of normal polarity

Kolbeinsey ridges. Outlava flows known to us

which most probably correspond to this period are indicated in Fig. 8 . Off Western Iceland, the lineations marked el in Faxafloi Bay and e2 in Breidafjordur Bay may represent Anomaly 5 stripes on both sides of a spreading axis crossing the present

Anomaly 5 (approximately 9-10 Ma) (e) With increasing age, it becomes more and more

meters

rise to reasonably The

is required.

(P. Imsland,

hundred

to date from late in the Gauss chron, but there are lava

due to faulting:

Snaefellsnes

the spreading extinct

zone

6 m.y. ago (Johannesson,

ity shifted to the current and Langjiikull.

I 18"

peninsula.

is believed

I

This

to have

part

of

become

1980) when activ-

zone between

I 14O

56"

64"

Fig. 8. Lava outcrops of Anomaly 5 age in Iceland, simplified from various references mentioned in the text.

Reykjanes

Anomaly lineation el becomes wide and indistinct over the onshore areas in the eastern part of

history of this area as well as that of the Iceland shelf in general cannot be worked out in detail

the Snaefellsnes

peninsula,

unless more geophysical

is not far from

localities

quences

of normally

but this extension

of thick (> 600 m) se-

magnetized

lur valley which are presumed age (Johannesson tion

to be of Anomaly

indistinct

at lo-11

of normal

for the small dimensions

not

be correlated

with

peninsula any

prediction

onshore.

of W-dipping

the Trijllaskagi

peninsula

magnetic

along

the

with a

Saemundsson

and dated

to correspond

(see Fig. 5) can-

particular

continuation

thick sequence

and Helgason lineation pass-

data 5 of

by Bott (1985, fig. 4), but it does not

lavas believed

could be

unbroken

Bay (e4), in agreement

et al.,

probably continues farther to the southwest than is shown in our Fig. 8 (fig. 7 of Johannesson,

Northwest

continues

Fjord on

the presence of a thick reversely magnetized lava series below (see McDougall et al., 1984, figs. 1 and 5). The presence of variable dip vectors and faulting in the area may also cause a reduction in the positive anomaly. The normal lava sequence

ing through

Ridge

et al. (1980) have described

of the anomaly

1980). As pointed out by Kristjansson (1988) the main positive anomaly

Kolbeinsey

east side of Hunafloi

magnetization,

that the reason

available.

In the north, our 1986 airborne magnetic have shown for the first time that Anomaly

have a direct

(McDougall

indicate

data become

in the vicinity

Ma, in Steingrimsfjordur

the east side of the peninsula 1984). Model calculations

5

1991). Linea-

over Northwestern

but a small high occurs

of a thick lava sequence dated

lavas in Hitarda-

and Kristjansson,

e2 is similarly

peninsula,

of it

normally

an 1800 m magnetized

to Anomaly

5 from

at e5. A pair of interest-

ing N-S lineations offshore (Fig. 4) appears to connect this anomaly with the prominent Anomaly 5 lineation farther west. On the eastern side of Eyjafjiirdur, a smaller anomaly corresponds to a normally magnetized series (> 1 km?) dated at about Anomaly 5 time by Hebeda et al. (1974) and Jancin et al. (1985) (see Fig. 8). It should however be noted that the magnetic polarity determinations of the latter authors are only based on measurements entirely reliable (Kristjansson, In Eastern

on hand samples, which are not in lava sequences of this age

1984). Iceland

(Fig. 8) anomaly

e7 may be

group in the local stratigraphy. It is possible that a major (known but so far unmapped) volcanic system north of Arnarfjijrdur was very active in lava production during a period of predominantly normal polarity. The strike of lineations in Northwestern peninsula is to some extent different from

caused by an 800 m thick series of mostly normal polarity lavas; these have been mapped by G.P.L. Walker (Watkins and Walker, 1977), and dated by

that assumed by McDougall et al. (1984), and continued geological mapping on the peninsula (H. Johannesson, pers. commun., 1985) has shown that the stratigraphic correlation proposed by these authors between profiles on its south and east

north

Bakkafloi

Bay.

coasts must be substantially revised. Along with other linear anomalies east of the Reykjanes Ridge, the Anomaly 5 Iineation (e3) peters out not far north of 63”N on the shelf,

Localized

anomalies

south of the Reykjanes peninsula. It appears from the shape of the anomaly that the trend of underlying rock structures turns towards the east closer to shore, but other possibilities, such as their continuation at depth under Iceland or the presence of a sediment-covered fault zone south of Reykjanes, must also be considered. The tectonic

McDougall et al. (1976) to correspond to Anomaly 5. No published geological mapping results are available from the northern part of the anomaly, of Mjoifjordur

Fjord,

but

Saemundsson

(1986, fig. 1) has drawn a 10 m.y. isochron from Mjoifjardur to the north-northwest,

onshore towards

In addition to the magnetic lineation patterns which can be explained by ridge-controlled aceretion and tectonic activity, numerous localized anomalies of geological interest are seen in the magnetic field over Iceland. The typical axnplitudes which are observed are of the order of 500-1000 nT. The anomalies are typically 5-10 km in size and appear to be mostly equidimensional rather than elongate. These parameters are,

MAGNETIC

SURVEYS

however,

not

anomalies

243

OF ICELAND

well constrained,

as many

of the

are only seen in one flight line or they

may be affected the Snaefellsnes

by topographic and Trollaskagi

noise

(e.g., over

peninsulas).

Fig-

ure 9 shows the distribution of 82 selected anomalies. We shall describe some of them qualitatively below, classified

according

a more complete

listing is given by Kristjansson

al. (1989). These anomalies Quatemary central

volcanoes

volcanoes.

to geographical occur mostly

of subglacial

In the latter,

region; et

at either

origin

or at

the source

rocks

low in the Krafla-Namafjall refer

to

designations

(Schonharting,

area

in

1969) to have been

due to the geometrical

effects of the caldera

but

the available

region cient.

and

data

their

The other

amplitude,

(46), Herdubreid

contain

reversely

shape, in the

are insuffiare smaller

to the young

most likely explanation

et al.,

mountains

types

properties

three lows, which

are related

tains Baejarfjall

on the rock

magnetic

of intermediate

(Kristjansson

caused

alteration of young basalts. caldera (60) may be partly

(44). While it is not impossible

composition

mostly

by local hydrothermal The low at the Askja

may be gabbro stocks, cone sheets, or lavas (Fridleifsson and Kristjansson, 1972) or even rocks 1977).

(45; numbers 9) is known

Fig.

table

in

moun-

(59) and Blafjall that some of these

magnetized

rocks, the

for these lows is that the

flight line may have passed to the side of a main mass of normally magnetized rocks. Other mountains

The spreading zone in North Iceland Within the Brunhes age positive anomaly in North Iceland at least five distinct lows occur. The ,

I

22”

I

in the

area

show significant in the Matuyama

I

18"

around anomalies. (negative)

I

Lake

Myvatn

do not

Just west of Blafjall, lineation,

a positive

1

l4O

P

Fig. 9. Index map of the main localized magnetic anomalies in Iceland. Shaded circles = positive anomalies; unshaded circles = negative anomalies; circles surrounded b> shading = negative anomalies surrounded by strong positive field.

244

localized magnetic anomaly occurs over Mt Sellandafjall

(43). The Triilladyngja

on the Brunhes-Matuyama

The volcanic zone of Southwestern Iceland The magnetic field of the Reykjanes peninsula

lava shield (61)

anomaly

boundary

causes a pronounced peak in the magnetic field.

has already been discussed. In particular, anomaly a5 in Fig. 6 has a very high peak (2500 nT) over Skalafell hill (5) on the Hellisheidi plateau; this is caused by volcanics

Southeastern

anomalies in a positive field region: highs over the

The field in western Vatnajiikull with

relatively

long-wave

normal

is irregular and

reverse

anomalies,

reflecting different stages of subglacial

volcanism

and metamorphism.

Skjaldbreidur Hlijdufell

of the zone

origin. In the

northern

Iceland

part

of subglacial

The Vatnajiikuli glacier and the volcanic zone of

interglacial

table mountain

we see three local

lava

shield

(11)

and a low over

(12)

and

another table mountain, Skridan (10).

A magnetic high

(73) is related to the Kverkfjiill center; southwest of it, on the line connecting Kverkfjiill and Grimsvotn and south of the southern Kverkfjiill

Pre-Brunhes anonuxlies in Western Iceland Some localized anomalies are situated in the Matuyama lineation on the west side of the main

caldera (Thorarinsson

et al., 1973), there is a major

volcanic zone. The best known of these is at the

positive anomaly (74) in an area which has a thick glacial cover and no known volcanism. The ice-

eroded Stardalur volcanic center (7) and appears

filled Bardarbunga magnetic

map

by

caldera is represented a

strong,

arcuate,

on the positive

anomaly (75). Radio echosoundings (Bjomsson, 1988) have revealed the subglacial caldera rim, which has shape similar to that of the anomaly, but on closer inspection the southern end of the magnetic anomaly lies farther south than the topographic rim. A kind of “barrier” in the field, a transverse (E-W) negative anomaly extending from Mt. Hamarinn to the subglacial lake Grimsviitn (76) is clearly related to a transverse zone of seismic activity (Einarsson, 1989) and to the high crustal heat flow observed at Skaftarkatlar (Bjiimsson, 1988). Another barrier-like anoma-

to be the result of normal polarity remanent magnetization of Olduvai or Reunion age (about 2 Ma) in lavas and intrusions within the caldera of the center (Fridleifsson and Kristjansson, 1972; Kristjansson

et al., 1980).

Positive anomalies

at

the Ferstikla (15) and Ok (13) volcanic centers are farther away from the spreading zone, contributing to the Gauss anomaly lineation (dl of Fig. 5). The Husafell central volcano (14) within the Gilbert age anomaly lineation has a large magnetic low. K-Ar dating of this center (Saemundsson and Noll, 1974; McDougall et al.,1977) indicates that its activity was mostly in the later part of the Gauss chron, but the negative anomaly may

may reflect a

be caused by the last phase of the activity of the

buried nonmagnetic hyaloclastite ridge. A positive anomaly (72) occurs over the active stratovolcano of Graefajiikull, Iceland’s highest mountain; a chain of localized positive anomalies follows the

volcano in the lower Matuyama. The negative anomaly at Hafnarfjall (16) in Borgarfjardur is of Gilbert age or older, corresponding to reversely

ly in the vicinity of Mt. Palsfjall(77)

mountainous southeastern edge of the glacier from there, but the magnetic anomalies do not reflect a proposed lineation of volcanic centers (figs. 2 and 4 of Saemundsson, 1986) towards the northeast. The very active subglacial Katla caldera in Myrdalsjbkull glacier creates a deep magnetic depression (80) probably due to high-temperature alteration, whereas the stratovolcano Eyjafjallajokull, which has had a relatively low level of activity (Kristjansson et al., 1988) generates a positive anomaly (81).

magnetized gabbro outcrops in the core of a volcanic center. The nearby positive magnetic anomaly at Hvanneyri (17) occurs over flat farmland, and its source may possibly be found below the major unconformity which is seen in outcrops in that area (Johannesson, 1980). Other localized pre-Brunhes anomalies Many of these coincide with the locations of the main late Tertiary volcanic centers of Iceland (cf. fig. 2 of Kristjansson and He&son, 1988). Examples include the Setberg center (sigurdsson,

MAGNETIC

1966)

SURVEYS

245

OF ICELAND

in the Snaefellsnes

peninsula

northeast, and Eystra-Horn several central netic

volcanoes Published

centers,

can be of considerable

use in leading

(63) in the southeast.

new ideas on the tectonic development of the area.

there are

(e.g., in Northwestern

which do not generate distinct

anomalies.

Pre-

geologists to interesting sites for mapping and to

However, at the 1 km flight altitude peninsula)

(22),

(53) in the

stbakki (26) in the north, Vopnafjordur

geological

mag-

maps are

The anomaly lineations

are most distinct

have highest amplitudes over and around the active zones. We expect that the reason for this lies partly in a higher effective

remanent

available for much less than half of these volcanic

tion in upper Quatemary (O-l.5

systems, and the presence of some is only inferred

other formations,

from casual observations. We anticipate that com-

faulting with age. A more speculative

bined

concerns

interpretation

of detailed

geological

and

and

magnetiza-

Ma) rocks than in

and partly in an increase

of

possibility

slow changes in the magnitude

of the

geophysical maps, which has so far only been carried out for very few of the volcanic centers,

geomagnetic dipole moment and in the rate of geomagnetic reversals. The thickness of The crustal

will in the future enable researchers to delineate different categories of volcanic centers and to

layer which contributes significantly to the magnetic anomalies is probably at least 1.5 km, and

suggest plausible ment.

more beneath the central volcanoes. Improved understanding of the main sources of

Discussion

mechanisms

for their develop-

and conclusion

magnetic anomalies in Iceland through detailed mapping will be invaluable in the extrapolation of geological

The magnetic anomaly lineations over Iceland,

models to detritus-covered,

glaciated,

and offshore areas. With our present state of knowledge we do not consider it advantageous to

in conjunction with comprehensive mapping on the ground, can be of considerable use in unravell-

interpret

ing the complex history of the volcanic zones. A

downward continuation;

good example concerns the two active zones in South Iceland: how has the total amount of crustal

servative method of making comparisons between the observed anomalies and those to be expected

spreading been divided between them in the last few million years? It seems from the magnetic field that the western zone has not been as dominant in this respect as is commonly thought, and

from the local geology and rock magnetic properties. Offshore, the results of gravity and multichannel seismic observations are needed in order to constrain the possible models of magnetic anomaly sources; thus, a simple model of the pronounced step-type anomaly around Southeast-

also that the difference in trends for the two zones has persisted at least for much of the Brunhes epoch. Another example is in the trend of Anomaly 5 in Faxafloi Bay (part of el in Fig. 5) and of older anomalies on Northwest peninsula, which have more northerly directions than expected from pre-

the

anomaly

field

by

computing

its

we prefer the more con-

em Iceland has already been proposed by Kristjansson (1976) but the limited geophysical data which have become available since have not permitted any more realistic modelling of the basement step or of its causes.

vious geological studies; these studies should be carefully re-evaluated.

Acknowledgements

The lineations are quite irregular compared to those observed over the ocean ridges, and they do in many cases not correspond well to the known

We wish to pay particular tribute to the skill and perseverance of the late Th. Sigurgeirsson in

magnetic polarities or strikes of surface rocks. For example, there is no obvious E-W trend to be seen in the anomaly field over the Quaternary volcanism in the Snaefellsnes peninsula or across the central highIands. However, some of the irregularities, such as those associated with major volcanic

his aeromagnetic surveys. He designed both the magnetometers and navigation equipment for these surveys, piloted the survey plane for much of the time, and supervised all data processing for the map series published in 1970-1985. Sigurgeirsson was ably assisted in his work by a large number of

dedicated

students

ence Institute.

and technicians

We thank

International

Atomic

ing this work workstation

Energy

by giving

Agency

Department

for support-

us access

to a graphic

from the computer

of Physics

lent us an ink-jet

to Kjartan

Emilsson

plotter;

who wrote

For their help in the present processing,

and

Stefan

Kristjan

the

of Ice-

thanks

the driver

surveying

we thank

Kaiser

because

of the University

land kindly

data

and the

free of charge. We were able to obtain

color hard copies

Isburg

from the Sci-

Sigfus J. Johnsen

also for it.

phase and

Saemundsson,

Leosson.

This

work

was supported by the Icelandic Science Foundation and the Research Fund of the University of Iceland.

anomalies

over the Reykjanes

Helgason,

J., 1984.

Iceland. Jakobsson,

M.,

Young,

Saemundsson,

volcanoes

rock

in Iceland.

magnetic

M.S. Thesis,

study

of

subglacial

Univ. Georgia,

Athens,

Bjljmsson,

H., 1988. Hydrology

Sot. Sci. Iceland Bott,

M.H.P.,

1985. Plate

transverse

of ice caps in volcanic

regions.

tectonic

ridge and adjacent

evolution

regions.

of the Icelandic

J. Geophys.

Res., 90:

for the Magnetic

1987. Magnetic sheets:

scale 1:

Einarsson,

Anomaly

anomaly

map

Map of North

of North

rift zone in Iceland. Geosci.

Einarsson,

velocity

In: Symp. Volcanic

Sot., Reykjavik

P. and Bj&rnsson,

Jokull,

America

in four

5,000,OOO. Geol. Sot. Am., Boulder, Colo.

P., 1988. Gn the propagation

Iceland

America,

of the eastern

Activity

in Iceland.

(in Icelandic).

S., 1979. Earthquakes

in Iceland.

T., 1957. Magnetogeological

with the use of a wmpass. Fridleifsson, magnetic

I.B. and

Kristjansson, SW Iceland.

I.B.,

nlaugsson,

Haraldsson,

Natl. Energy

L., 1972. The J&t&

G.I.,

E. and Bjomsson,

verjahreppi.

in Iceland Stardalur L.S.,

B.J., 1980. Jardhiti

Auth.,

Reykjavik,

Gun-

i Gnup-

Rep. OSOOlO/

JHDO6 (in Icelandic). G.W.,

Johannesson,

with English

land and Norwegian

Geod.

E.H.,

Int. Meet. Heirtzler,

Earth Planet.

inve&gations

side of Eyjafjllrdur,

C&chronology,

H.N.A.,

1974.

of a series of lava Northern

Cosmochronology

Iceland.

and Isotope

(Paris) (Abstr.). J.R., LePichon,

J.G.,

sequences

Gwl.

of the magnetic

magnetic

survey

crustal

off southern

on palwmagnetic

sampling

in

fra fhrgvelum

og

and

(Editor),

pp. 209-225. Some properties

volcanic

centres

In: A.C. Morton

L., Thors,

Volcanism

K. and

I Hlutatins

J., 1988.

Sot. London

of central

Kristjansson,

of

in a plate-

and L.M. Parson

and the Opening

of the

Spec. Publ., 39: 147-155.

Karlsson,

volcanoes

L., Fridleifsson,

and

H.R.,

1977. Con-

off the Icelandic

coast.

Na-

1%6. Magnetic

and

Akrafjall

I.B. and

palwmagnetism

mountains,

Watkins,

N.D.,

of the Esja,

SW Iceland.

1980.

Eyrarfjall

J. Geophys.,

47:

31-42. Kristjansson,

L., Johannesson, lava

sections

L., Jonsson,

surveys

Iceland,

H.,

Eiriksson,

J. and

A.I., 1988. Brunhes-Matuyama

at the

in Iceland.

Can.

Institute.

pp. and wlour

Gud-

paleomagnetism J. Earth

G. and Svenisson, Science

Rep. RHO1.89.40

Sci., 25:

M., 1989. MagSci. Inst.,

Univ.

map at a scale of

1:1,000,000. McDougall,

I., Watkins,

N.D.,

Walker,

son, L., 1976. Potassium-argon of Icelandic

X. and Baron,

surveys

Sci. Lett., 16: 237-244.

Reykjavik,

envionment.

Kristjansson,

Priem,

and magnetic

Planet.

Helgason,

Early Tertiary

Obs., 39: 123-

and

sheet 6, 2nd ed.

In: Th. I. Sigfusson

boundary

firmation

K., 1982.

34: 67-76.

L. and

NE Atlantic.

(In prep.).

Res., 2: 315-326.

(Editors),

Seas. Publ. Dominion J.J.

Earth

Edli. Menningarsjodur, lava

and

in the lsafjar-

Mus. Nat. Hist. and

L., 1987. Segulsvidsmaelingar

Kristjansson,

a

Sci. Lett., 8: 101-108.

L., 1984. Notes

vid Island.

215-225.

and paleomagnetic

flows on the eastern Geology

Surv., Reykjavik,

Jokull,

in three

Hantelmann,

sequences

Icelandic

L., 1976. A marine

skipum

rekbelta

L., 1991. Stratigraphic

L., 1972. On the thickness

Iceland.

of

(in Icelandic,

areas, W Iceland.

Mar. Geophys.

basalt

og throun

L., 1970. Paleomagnetism

Kristjansson,

reorganization

S.P. and Saemundsson,

map of Iceland.

Kristjansson,

Axial Rift Zone, in

50: 13-31

of lava

H., Jakobsson,

Kristjansson,

netic

K-Ar

mapping

in Iceland.

and

ages across

Abstr.).

and Hnappadalur

1970. Mag-

149. Hebeda,

of

Res., 90: 9961-9985.

H. and Kristjansson,

palwmagnetic

and the Green-

P.H.,

Iceland

Natturufraedingurinn,

Johannesson,

countries

W. and Serson,

maps of the Nordic

in

19:

J.L.

and K-Ar

H., 1980. Jardlagaskipan

Vesturlandi.

mundsson,

Hannaford,

netic anomaly

J. Petrol.,

B., Aronson,

J. Gwphys.

Stratigraphy

22: 69-78.

Georgsson,

zone

ture, 268: 325-326.

Adv. Phys., 6: 232-239.

anomaly,

Fridleifsson,

mapping

Voight,

Iceland.

Kristjansson,

29: 37-43.

Einarsson,

Iceland.

to the 7 Ma volcano-tectonic

Kristjansson,

9953-9960. Committee

Haines,

K.D.,

the west flank of the Northeast

Iceland.

Publ., 45: 130 pp. and 21 maps.

1.t:

F., 1978. Petrology

peninsula,

K., 1985. Stratigraphy

layer in SW Iceland.

138 pp.

of the volcanic

relation

Iceland A

shifts

J. and Shido,

Reykjanes

Geological

1982.

Res..

669-705. Jancin,

dardjup

G.S.,

Deep-Sea

12: 212-216.

S.P., Jonsson,

the western

Kristjansson, Barker,

Frequent

Geology,

Johannesson,

References

Ridge.

427-443.

Gwphys.

G.P.L.

and Kristjans-

and paleomagnetic

analysis

lava flows: limits on the age of Anomaly Res., 81: 1505-1512.

5. J.

MAGNETIC

SURVEYS

McDougall,

I., Saemundsson,

N.D. and Kristjansson, netic polarity Western

time scale to 6.5 m.y.: K-Ar

Iceland.

land. J. Geophys. R.L.,

boundary A.G.,

Schilling,

J.-G.

Tjomes

margin.

M.,

in

Ice-

(Editors),

IceP.R.,

Vogt,

H.C. and Vopand

surand

Ridge.

over Iceland

and

Development

of the

Plenum,

New York,

pp. 661-

G.,

orkulindum Hlutarins

Gildi

landsins.

jardedlisfraedi In:

Th.

i rannsoknum

I. Sigfusson

(Editor),

Edli. Men~ngarsj~ur,

Reykjavik,

pp. 239-249

1971. Ground

studies of crustal

Earth

Saemundsson,

Planet.

K., 1978. Fissure

Saemundsson,

K., 1986.

Atlantic.

The Geology

of some magnetic swarms

Subaerial America.

Husafell,

and central

central

Iceland.

J.

Iceland.

des

erdmagnetischen

in Nord-Island,

Doct. Dissert.,

deren

of the Setberg

Sot. Sci. Iceland

Auswer-

Univ. Munich, area,

121

Snaefel-

Misc. Pap., IV (2):

Th., 1967. Aeromagnetic

anomalies

surveys

Mid-Ocean

Sot. Sei. Iceland

Sigurgeirsson,

Ridges.

and

Iceland

and

Publ., 3X: 91-96.

Th., 1970a. Aeromagnetic

Sci. Iceland,

of Iceland

(Editor)

survey of SW Iceland.

2: 13-20. Th., 1970b.

Aeromagnetic

1979, 1980a, profile

maps

b, c. 1981, 1984a,

b,

of

1:

Iceland.

Sheets l-9. Sci. Inst., Univ. Iceland, S., Saemundsson,

K. and

1973. ERTS-1 image of Vatnaj~kull: Vilmundardottir.

and volcanic

Jardfraedi

dingurinn,

Burfells

54: 97-113

og

Reykjavik.

Williams,

Analysis

features.

E., Gudmundsson,

RX,

Jr.,

of glaciologi-

Jokull,

A. and

Scale

23: 7-17. Snorrason,

nagrennis.

S.P.,

Natturufrae-

and map (in Icelandic.

with English

volcanoes

in the western (Editors),

Vogt,

P.R., Johnson, G.L. and Kristjansson,

L., 1980.

phology and magnetic anomalies north of I&and.

Mor-

J. Geo-

phys., 47: 67-80. Walker, G.P.L., 1963. The Breiddalur central vclcano, eastern Iceland. Q. J. Geol. Sac. London, 119: 29-K;. Watkins, N.D. and Walker, G.P.L., 1977. Magnetostratigraphy

Iceland, volcano.

Vermessung

Profile

In: S. Bjomsson

1984.

Geol. Sot. Am., Boulder,

K. and NolI, H., 1974. K-Ar western

paleomagnetic

Abstr.).

In: P.R. Vogt and B.E. Tucholke of North

growth

Geol. J., 10: 415-432.

volcanism

I. and Watkins,

and

in northern

its neighbourhood.

Thorarinsson,

Vol. M, pp. 69-86.

Saemundsson,

Sigurgeirsson,

I

16: 163-187.

zones of Iceland.

succession

H., 1966. Geology

lsnes, Western

250,000.

Sci. Lett., 12: 199-207.

Tectonophysics,

of the neovolcanic

Husafell

magnetic

1973. Interpretation

over Iceland.

Colo.,

PP. Sigurdsson,

Cal, structural,

in Iceland.

North

1969.

tung und Interpretation.

a

and map. Piper, J.D.A., Piper. J.D.A.,

G.,

Feldes langs einiger

1985. 1987.

lava

geological

Res., 85: 36283646.

Sigurgeirsson,

678 and map. Palmason,

L., McDougall,

dating,

53-125.

G.R.,

Bott, S. Saxov, M. Talwani

Structure

Greenland-Scotland

1977. Plate

on northern

L., Larsen,

Geophys.

K-At

Schonharting, K., 1984.

269: 663-668.

anomalies

seas. In: M.H.P.

J. T’hiede

of a 5-km

P.R.,

Zone

Lorentzen,

Th.. Kristjansson,

pel. D., 1983. Magnetic rounding

study

geologi-

of Northwest

and Pinet,

Fracture

Nature,

Talwani,

Sigurgeirsson,

dating,

Res., 89: 7029-7060.

within

land’s insular

1980.

L. and Saemundsson,

and geochronology

K., Kristjansson,

N.D.,

Bull. Geol. Sot. Am., 88: l-15.

I., Kristjansson,

&Master,

Saemundsson,

H., Watkins, of the geomag-

study of a 3,500 m lava succession

Magnetostratigraphy

Nunns,

K., Johannesson, L., 1977. Extension

cal and paleomagnetic McDougall,

247

OF ICELAND

and

the

ages of rocks from development

Jiikull, 24: 40-59.

of the

of eastern Iceland. Am. J. Sci., 277: 513-584.