Jökull - 01.12.1985, Qupperneq 59

but the distribution of thermal manifestations seems not
to be affected.
THE GROUND MAGNETIC SURVEY
The aim of the ground magnetic survey was to reveal
the structural features of the basement, where covered
by soil or sediments of small or móderate thickness. The
total magnetic field was measured with a proton preces-
sion magnetometer along lines approximately per-
pendicular to the main line of hot springs. Readings
were made every 5 m along lines spaced 20 m apart.
The magnetic data was processed in a computer to
facilitate interpretation. Short wavelength anomalies
were filtered out with upward continuation to enhance
anomalies of a deeper origin. The weaker anomalies of
shallow depth were enhanced by taking the vertical
gradient.
The magnetic contour map of the Vellir thermal field
is shown in Fig. 5 with a schematic interpretation in the
lower left corner. The magnetic anomalies can be classi-
fied into three groups based on their origin. Firstly,
those caused by underlying dykes, secondly, those
caused by faults and fractures and thirdly, anomalies
caused by late magnetization of the Finiglacial sedi-
ments.
At least three anomalies were correlated with dykes.
Two of them are exposed at the surface (Fig. 4), sr
which is normally magnetized and 1 which is reversely
magnetized. The third, dl, which is also reversely
magnetized, may possibly be either ss or k. No obvious
correlation is observed between these dykes and the
distribution of hot springs. One additional anomaly, d2,
may also be caused by a normally magnetized dyke but
the data are not conclusive. It transects the Vellir line at
the Snaeldubeinsstadahver hot spring. If this is a dyke,
it may affect the upflow of thermal water.
The magnetic anomalies indicate two faults with a
trend close to N20°E. One of these anomalies corre-
sponds to the Snaeldubeinsstadir fault (SSF). The other
is more complicated but corresponds to the Logaland
fault (LF). Both these faults may influence the ascent of
hot water. Three additional anomaiies are interpreted
as faults. Two of them are just west of Sturlu-Reykir; F1
has a trend N20°E and can not be correlated with any
fault observed on the surface; the other corresponds to
the Sturlu-Reykir fault (SRF) and has a northwesterly
trend. Its southwards continuation is observed in the
southern slope as suggested in Fig. 4. Finally, the north-
westerly trending F2 seems to have some bearing on the
ascent of hot water along the Hagahús line.
A weak positive linear anomaly was recorded above
the hot springs. It is most pronounced in the central part
of the thermal field. It is of a shallow origin and is due to
some sort of secondary magnetization of the Finiglacial
sediments along the thermal anomaly. Similar anoma-
lies have been observed elsewhere in the Reykholts-
dalur valley (Georgsson et al. 1978). The anomaly and
the linear distribution of hot springs supports the exist-
ence of a fracture trending close to N3°E controlling the
upflow of hot water along the central segment of the
Vellir line.
CONCLUSIONS
Geological and geophysical evidence indicates that
the ascent of hot water at the Vellir thermal field is
controlled by fractures, faults and dykes and may be
considered as typical for the Reykholt thermal system.
Two northeasterly trending faults, the Snaeldubeins-
stadir and Logaland faults seem to control the flow of
water towards the Vellir line. A northerly trending
fracture intersects the faults and the two main hot
springs, Badlaugahver and Vellir, are located at the
intersections. The fracture is exposed in the unconsoli-
dated marine sediments in the river bed. The central
segment of the Vellir line coincides with the fracture but
the southern segment coincides with the Logaland fault.
The interpretation of the magnetic data (Fig. 5) suggests
that Snaeldubeinsstadahver may be located where the
the Logaland fault is intersected by a northeasterly
trending dyke.
Tectonics and topography (i.e. higher elevation) indi-
cate that the Sturlu-Reykir line has a different aquifer
from other parts of the Vellir field. The hot springs
break out on a northwesterly trending fracture associ-
ated with the Sturlu-Reykir fault. The fracture is
intersected by a northeasterly trending fault 100 m
northwest of the hot springs and a dyke 50 m further
northwest.
The Hagahús warm springs are on a northwesterly
trending line, probably a fracture, which may be the
same as controls the Sturlu-Reykir line.
We suggest that the main aquifers of the Vellir ther-
mal field are associated with northeasterly faults.
Where they are intersected by open fractures the water
flows to the surface. This is in agreement with studies of
other thermal fields of the Reykholt thermal system as
well as regional studies of the Reykholt thermal system
(Georgsson et al. 1984), which have emphasized the
importance of northeasterly structures for regional flow
at depth towards the thermal fields. Fig. 6 shows a
simplified model of the Vellir thermal field. The base
temperature of the thermal field is about 130°C.
The structure of the Vellir thermal field is quite
similar to the Baer thermal field. Experience from the
JÖKULL 35. ÁR 57