Jökull - 01.01.2012, Page 29

Deciphering eruption history and magmatic processes from tephra in Iceland
A tephra layer that has reached the level of being
preserved in soil will with time be affected by chem-
ical weathering. Tephra grains are altered by hydra-
tion and element redistribution and thus produce alter-
ation products, such as palagonites from basaltic glass
(Peacock, 1926; Moore, 1966; Thorseth et al., 1991;
Stroncik and Schmincke, 2002). Physical factors that
are likely to control the rate of palagonitisation are:
temperature, structure of the primary material, reac-
tive surface area of the primary material, structures of
the precipitation secondary phases, the growth rates of
the secondary phases, time and different fluid proper-
ties (e.g. pH; Stroncik and Schmincke, 2002; Gíslason
and Oelkers, 2003). These factors interact in a com-
plex manner, a few thousand years old tephra may not
become palagonized at all whereas it took less than
10 years to create palagonite rims and compact tuff
close to heat sources after the Surtsey eruption in 1963
(Jakobsson, 1978). Therefore, some thin (and proba-
bly the oldest) tephra layers may have been weathered
out and lost from the tephrostratigraphy.
ERUPTION FREQUENCY VS. TEPHRA
LAYER FREQUENCY
The compilation of a complete eruption history of a
volcanic system, even over a relatively short period
like the Holocene in Iceland (∼10 ka), will prob-
ably never be achieved as very small tephra layers
are sometimes not preserved in the geological record.
The best approximation of the true frequency of ex-
plosive eruptions in a particular volcano is obtained
from mapping of all tephra layers at several localities
around it. Each locality will have its own local tephra
layer frequency (number of tephra layers divided by
time) as the number of layers will vary between out-
crops. An example of this is the fact that due to pre-
vailing weather conditions, tephra deposition in Ice-
land is more common towards east than towards west
(e.g. Jónsson, 1990). Good tephra preservation lo-
calities will, however, have similar local tephra layer
frequencies, given that the tephra distribution was the
same in both places. By combining and correlating
the records from all outcrops, a best estimate of the
regional tephra layer frequency for the period under
study is obtained.
The ideal time period to develop a high resolu-
tion, nearly complete record of eruption history is the
historical time due to the combination of two inde-
pendent data sets, (1) tephra layers and (2) written
sources. It is possible to maximise the tephra layer
frequency by looking at tephra layers in different en-
vironments, especially the glacial ice but as most of
the Icelandic volcanoes are located under glaciers this
environment preserves evidence of eruptions that only
deposited tephra close to source. It is then possible to
correlate the glacier tephrostratigraphy to layers in the
soil around the glaciers and with written documenta-
tions.
Unfortunately the prehistoric time period is not as
well documented, due to lack of glacial tephra record
as the oldest ice in the largest Icelandic glacier is only
800-1000 years old (Larsen et al., 1998; Thordarson
and Larsen 2007). One way of estimating the erup-
tion frequency of a volcanic system during the prehis-
toric time is to use a ratio between the tephra layers
found in the soil and all known tephra forming erup-
tions during selected periods (Larsen and Eiríksson,
2008a; Thordarson and Larsen, 2007; Óladóttir et al.,
2011a). For example, in 7 sections around Vatnajökull
16 of the 64 known Grímsvötn tephra layers erupted
from AD 1200–2010 were found. This ratio of all
known explosive eruptions and the observed layers in
the studied sections is then used as a correction fac-
tor, termed preservation ratio. Thus, the eruption fre-
quency for the prehistoric period in Grímsvötn was
assessed by multiplying number of Grímsvötn tephra
layers with the factor of 4 (64/16; further information
see Óladóttir et al., 2011a). In short, the tephra layer
frequency is used with preservation ratio to account
for unpreserved and undetected tephra layers in order
to estimate the eruption frequency of a given volcanic
system.
What influences eruption frequency?
Tephra studies allow us to read the eruption history
that different volcanic systems have left in soils and
other sediments. Individual volcanic systems seem to
follow similar trends in volcanic activity, i.e. periods
of high and low activity seems to take place in simi-
lar time periods (Jakobsson, 1979; Larsen and Eiríks-
son, 2008a; Óladóttir, 2011a; Guðmundsdóttir et al.,
JÖKULL No. 62, 2012 27