Jökull - 01.01.2020, Síða 44

The 1845–46 and 1766–68 eruptions at Hekla volcano
written descriptions by J. C. Schythe and H. Finnsen,
we interpret them ourselves instead of relying solely
on Thórarinsson (1967). Because the present study
focuses on the effusive activity, descriptions of lava-
flows in the literary sources (Finnsen, 1767; Schythe,
1847; Thórarinsson, 1967) are mainly taken into ac-
count. This entails specific time/periods and locations
of the effusive activity. Furthermore, information re-
garding the openings of new vents was noticed and
used here to reconstruct the emplacement of the lava-
flow fields. Qualitatively assessed time-lines of the
lava emplacement during the 1766–68 and 1845–46
eruptions are created similar to those of the 1947–48
eruption by Thórarinsson (1976). It is important to
underline that the flow-lines are approximate, but the
best possible from combining the historical sources
and our analyses. The flow-lines show dates from
the literary sources (Figures 3a and 4a). The solid
flow-lines indicate that the historical source provides
precise information on the location of the flow front
by mentioning place names and/or estimate distances
from known locations, hence we are more certain of
these emplacements. Whereas the dashed lines indi-
cate greater uncertainty because only vague directions
of the lava-flow direction are given (e.g. "19 Septem-
ber 1766: .. and lava-flowed both to the west and
east and in the direction of Krakatindur" Thórarins-
son, 1976).
Fieldwork and geochemical analyses
During fieldwork August 7–10, 2018 we sampled
rock specimens and studied flow morphologies of the
lava-flows 1766–68, 1845–46 and 1947–48. Four to
five specimens were sampled for each of the three
lava-flows representing a range of emplacement ages,
ideally from oldest to youngest. The samples were
analysed at the mineral laboratory Bureau Veritas
Commodities Canada Ltd. by XRF and ICP-MS meth-
ods to reveal whole rock and trace elements. Analy-
ses of certified reference materials show the relative
deviation from the reference values are less than 1%
for the major elements and less than 7% for the trace
elements discussed in this project. Repeated analy-
ses of samples demonstrate reproducibility within 1%
for the major- and minor elements and within 7% for
most trace elements.
Viscosity estimates
Several models predict viscosity of silicate melts as a
function of temperature and composition (e.g. Shaw,
1972; Bottinga and Weill, 1972; Giordano et al.,
2008; Hack and Thompson, 2011). In this study vis-
cosity estimates have been calculated after the model
by Giordano et al. (2008). This is a multicomponent
chemical model spanning most of the compositional
range found in natural-existing volcanic rocks. It is
based on experimental measurements of viscosity at
different temperatures of known melt compositions at
atmospheric pressure (105 Pa) (Giordano et al., 2008).
They use the Vogel-Fulcher-Tammann equation (Vo-
gel, 1921; Fulcher, 1925) to estimate the temperature
and composition dependent viscosity (η):
log η = A+
B
T (K) − C
(1)
2. mynd. – Mynd af hraunþykktarmælingum með dæmum um hæðarsnið B), C), D) og E) ásamt upplýsingum
um það hvort mælingarnar uppfylli kröfur sem gerðar voru. Hraunþykktarmælingarnar eru merktar með "ac-
cepted" (samþykkt) eða "not accepted" (ósamþykkt) neðst til vinstri á hæðarsniðunum. Hæð við botn hraunsins
er merkt með "level 0" (stig 0). Gert er ráð fyrir að þykkt hraunsins sé náð þar sem bratti hraunsins er orðinn
lítill. Þetta er táknað með " level 1" (stig 1). Mismunur á hæð stigs 1 og stigs 0 er þykkt hraunsins fyrir það
tiltekna snið. Meðalþykkt er áætluð út frá viðurkenndum hæðarsniðum og er notuð í rúmmálsútreikningum. F)
Kort af hraunbreiðum frá 1845–46, skipt niður í níu svæði. Nærmyndin sýnir þykktarmælingar (svartar línur).
Svæði 1: Gígasvæði og Hekluhryggurinn (halli svæðis er um það bil ≤55◦). Svæði 2: flókin formgerð breytist
í hraunrás (≤40◦). Svæði 3: hraunfossar (≤40◦). Svæði 4: hraunflæði í hraunrásum (≤15◦). Svæði 5 og 6:
mikil hraunbelging og undanrennsli kviku, greinilegast á svæði 5 (≤15◦). Svæði 7 og 9: undanrennsli (≤10◦).
Svæði 8: hraunflæði í hraunrásum, hugsanlega hraunbelging og undanrennsli (≤10◦). Bakgrunnur og útlínur
eins og á 1. mynd.
JÖKULL No. 70, 2020 41