Jökull - 01.01.2016, Qupperneq 79

Grain characteristics of silicic Katla tephra layers
geochemical composition seems an unlikely reason
for SILK-A11’s differences. The small age difference
between SILK-A11 and SILK-A12 eruptions (∼100
years) makes major changes in the glacier thickness
unlikely. A different source area with different ice
thickness and meltwater supply at the vents is a more
likely explanation.
Dellino and Volpe (1996) analyzed grain shapes
with the purpose of identifying the origin of the
grains, in order to establish if they were of hydro-
magmatic or magmatic origin. Their work suggested
that spherical shaped grains originate from hydro-
magmatic eruptions. However, Eiríksson and Wigum
(1989) state that grains from silicic magmas are more
readily elongated than those from basaltic ones and
magmatic grains tend to be more elongated than
the hydromagmatic grains. According to the elon-
gation values, the SILK grains are very elongated,
which agrees with Eiríksson and Wigum (1989). The
high proportion of fine grained material also supports
phreatomagmatic activity (Self and Sparks, 1978).
We believe that the elongated SILK grain shape is
connected to phreatomagmatic explosive activity. The
extreme elongate shape is the result of the break-up of
viscous silicic magma with abundant elongate vesi-
cles, as a result of a shock caused either by cooling
contraction, a sudden expansion of steam, or both. In
the case of the SILK-A11 tephra, the fragmentation
occurred before the vesicles acquired an elongated
form. We speculate that the ice at the eruption site
was thicker and degassing was arrested by abundant
meltwater before the magma froth was drawn out into
the elongated vesicles that characterize other SILK-
type tephras from this period.
SUMMARY
The younger SILK tephra layers appear to be coarser
grained, i.e. have larger maximum grain size, while
the older layers appear to be finer grained, except
for SILK-A7. These changes occur between SILK-
N1 (∼5800 year old) and SILK-A1 (∼6000 year old).
However, the mean grain size does not clearly follow
this trend.
Grain shape results for the SILK-LN, SILK-N1,
SILK-A8, SILK-A11 and SILK-A12 tephra layers
show that SILK-A11 is significantly different from
the other four layers. The main difference lies in the
elongation values, as no systematic changes with time
were observed in the shape parameters (elongation,
ruggedness and circularity). Omitting SILK-A11 and
adding the ∼3400 old SILK-LN to the measurements
does not change this conclusion.
The chemical composition of the SILK-A11 and
SILK-A12 show higher SiO2 values and lower TiO2,
MgO and CaO values compared to all other SILK lay-
ers.
Grain size and chemical composition appear to be
correlated, where the SILK tephra layers with the low-
est SiO2 and highest FeO and CaO tend to be coarser
grained than the others, but there is no correlation be-
tween grain shape (parameters measured in this study)
and chemical composition.
The results of this study do not support radical
changes in the eruption environment. On the contrary,
the results support a fairly stable eruption conditions
during the period under investigation, and presence
of ice cover throughout the Holocene as suggested by
Dugmore (1989) and Óladóttir et al. (2007).
Acknowledgements
This work was supported by the International Civil
Aviation Organization (ICAO) as a part of the Cata-
logue of Icelandic Volcanoes Project of the Icelandic
Meteorological Office and the Institute of Earth Sci-
ences, University of Iceland. We are grateful to
Bergrún A. Óladóttir, Johanne Schmith, Tinna Jóns-
dóttir and Agnes Ö. Magnúsdóttir for assistance and
good company in the field, and to Jón Eiríksson for
assistance with the grain shape analyses. We thank
the two reviewers for their constructive reviews.
ÁGRIP
Kornastærð og kornalögun súrrar til ísúrrar gjósku úr
12 Kötlugosum, sem urðu fyrir 2800 til 8100 árum,
var rannsökuð með það að markmiði að kanna hvort
kornaeinkenni gjóskunnar hefðu breyst með tíma og í
slíku tilfelli hvort breytingarnar endurspegluðu breyt-
ingar á umhverfi gosstöðvanna eða einhverjar aðrar
ástæður, t.d. breytta efnasamsetningu. Kornastærðar-
greiningarnar benda til að gjóska í yngri gosunum sé
JÖKULL No. 66, 2016 79