Jökull - 01.12.1978, Síða 104

America. The affinity of the Neogene Icelan-
dic flora with the Eastern Desiduous Forests of
North-America has been pointed out by
several authors. Our finding from the Hreda-
vatn area underlines this. A. askelssoni n. sp.
was found in interbasaltic limnic-fluviatile
sediments, the age of which has been deter-
mined by absolute dating (K-Ar-method) and
palaeomagnetic mapping. The age of the
Hredavatn sediments is upper Miocene (6—7
M.y.).
THE KRAFLA GEOTHERMAL SYSTEM
- CHEMICAL INTERPRETATION
Gestur Gíslason,
National Energy Authority
Temperature measurements in drillholes in
Krafla suggest a two layered hydrothermal
system and this interpretation is supported by
the chemistry of the fluids.
The upper system extends down to 1000—
1200 meters depth and has a temperature
around 200 °C and the chemical composition
of the fluids is a expected for that temperature.
The interpretation of the various gas ratios
from this system shows a convection system
with upwelling in Hveragil and downwelling
between Krafla and Leirhnjúkur.
The temperature of the lower system is in
excess of 300 °C and this system contains a
mixture of water and steam. The boreholes are
in the lower system fed by a mixture contain-
ing 40—90% steam (by weight).
CHEMICAL VARIATION WITHIN
SINGLE BASALTIC ERUPTIONS
Karl Grönvold,
Nordic Volcanological Institute
The chemical variation of some Icelandic
basaltic lavas is very large while others show
little or no variation.
Analyses of lava samples from the eruption
of the Laki craters in 1783 show no significant
variation with time in major elements.
Two shield volcanoes (Kjalhraun and
Borgarhólar) show highly variable com-
position. The phenocrysts of these lavas
appear out of equilibrium and movement of
phenocrysts appear unable to explain the
variation observed. Further the 87Sr/86Sr is
variable which suggests complicated origin.
Analyses of glass from the eruptions in the
Krafla 1975 and 1977 show variable chemistry
(Mg0 5.2—7.6%).
STATISTICAL ANALYSIS OF RECORDS
OF VOLCANIC ERUPTIONS AND
LARGE EARTHQUAKES IN ICELAND
Gudmundur Gudmundsson, Central Bank of Iceland,
Knstján Saemundsson, National Energy Authority
Volcanic eruptions in Iceland and on the
shelf recorded between 1550 and 1977 and
large earthquakes over the same pariod are
subjected to a statistical analysis. The histori-
cal record of both kinds of events is no doubt
incomplete. Only those earthquakes are in-
cluded that caused collapsing of buildings, or
earthquakes of magnitude 6 or larger after
measurements became available.
No significant trend in the number of
earthquakes and eruptions can be deduced
from the records. Separately neither record
departs significantly from a purely random
(Poisson) process. Several studies indicate that
earthquakes in restricted zones are not in-
dependent. There is also a significant depar-
ture from pure random (Poisson) processes in
the records of eruptions from separate central
volcanoes. But the records analysed here are
aggregates of more than one earthquake zone
and several volcanoes and neither of them is
significantly different from a Poisson process.
There is, however, a strong evidence of a
relationship between the two types of events.
The number of eruptions starting in the same
102 JÖKULL 28. ÁR