Jökull - 01.06.2000, Side 52

Aeromagnetic measurements over Mýrdalsjökull
due to its remanent and induced magnetization, un-
less its temperature is of the order of 500  C or above.
Superimposed on this gravity high there is a horse-
shoe shaped high along the south, west and east
rims of the caldera. It may originate in intrusions
concentrated at the border of the caldera above 2.5 km
depth. If these intrusions mark the lateral extent of the
magma chamber it would be circular and 6 km across.
Although acid rock occurs in most nunataks around
the caldera, the total volume of acid rocks is proba-
bly small as it would tend to cause a negative gravity
anomaly (Guðmundsson, 1994b).
GENERAL DESCRIPTION OF THE
RESIDUAL MAGNETIC FIELD
The magnetic field over the mountainous Mýrdals-
jökull and Eyjafjallajökull region appears to form a
part of the Brunhes age magnetic high of the ea-
stern volcanic zone. This is especially clear around
Eyjafjallajökull and the westernmost part of Mýr-
dalsjökull. The high is interrupted by a wide SE-
trending negative anomaly around the large Torfajök-
ull volcanic center (no. 79 in Figure 9 of Jónsson et
al., 1991), reaching the northern periphery of Mýr-
dalsjökull. The field is relatively smooth along the ea-
stern edge of Mýrdalsjökull.
The most noticeable feature of the residual field
is an NNW-SSE elongated depression of approxima-
tely 12 by 8 km in size, concentric to the subglacial
caldera. Its amplitude at our flight altitude is -1300 nT,
while the residual field surrounding it is of the order of
+600 nT. We shall refer to this 2000 nT magnetic bowl
and its rim as the Katla magnetic anomaly. Several
localized anomalies are seen at the rim or farther away
from the caldera.
A broad magnetic high extends south from the
eastern part of Mýrdalsjökull, to the coast. Its N-S
trend reflects the trend of the mountain ranges east of
the village of Vík but it could have a deeper source.
Some minor anomalies are found outside the glaciers,
most of unknown origin. A small negative anomaly is
clearly related to the Tindfjallajökull central volcano
(Figure 2, where lines 11 and iii cross).
THE KATLA ANOMALY
Data from several boreholes within or near the vol-
canic zones in Iceland show little systematic change
in magnetic susceptibility down to 2 or even 3 km
(Kristjánsson and Watkins, 1977). The maximum
depth of magnetic contrasts in the Icelandic crust must
be greater than this and may be set by the Curie point
for fairly pure magnetite (500-560  C). There is rea-
son to believe that a thermal anomaly is present under
the Katla caldera, associated with the magma cham-
ber inferred from seismic profiling. If this is the case,
a broad negative magnetic anomaly is to be expected,
in particular if the Curie point isotherm reaches close
to the surface (  1 km, say). However, such an anom-
aly would not be distinguishable from one due to
the presence of non-magnetic material such as tuffs
surrounded by crystalline rocks of normal magnetic
properties.
The maximum depth to the sources of the field
may be estimated using the half-slope depth method
(e.g. Sharma, 1986). In the central northern part of the
Katla anomaly (along line 8 of Figure 2) the source
appears to lie below sea level. In the southern part
of the anomaly the source would, on the other hand,
lie close to the bedrock surface. In addition to the
uncertainty of this method, the altitude of the aeropla-
ne in the older lines was not well known.
We model the field using the magnetic pole
concept (Telford et al., 1990). The generating bo-
dy is split into a number of square vertical columns,
all with the same uniform magnetization and their
contributions to the field are summed up. We assume
the columns to be sufficiently narrow to be replaced
with magnetic dipoles, with monopoles at the top
and bottom. The depth to the bottom monopoles is
held constant across the domain of the model but the
elevation of the top monopoles varies.
To model the field over Mýrdalsjökull we use for
the upper poles the two dimensional topographic grid
of the area provided by Björnsson et al. (2000). The
depth of the bottom poles (2000 m below sea level)
is not important as long as it is kept constant. In
order to eliminate edge effects, the working area is
extended in all directions by an 8 km wide tapering
zone (not included in the figures) where the topograp-
JÖKULL No. 49 51