Jökull - 01.01.2012, Page 33

Deciphering eruption history and magmatic processes from tephra in Iceland
Figure 5. a) Relationship between SiO2 content of
tephra during a few historical Hekla eruptions and the
repose period before each eruption. The two semi-
parallel lines before the 1510 and 1593 eruptions have
slopes reflecting the observed early glass composition
for these eruptions and a steeper slope based on compa-
rable lines for younger eruptions. Redrawn from Tho-
rarinsson (1967). b) Magnitude of the explosive open-
ing phase of Hekla eruptions in historical time, using
the equation: magnitude = log10(erupted mass, kg)−7
(Pyle 2000). Volume is based on Thorarinsson (1954,
1967), Thorarinsson and Sigvaldason (1972), Grönvold
et al. (1983), Larsen et al. (1992, 1999) and Lacasse
et al. (2004). Density values used are 550 kg/m3 for
H-1104 and H-1158 and 700 kg/m3 for the remain-
ing eruptions (Thorarinsson, 1967). The tephra fall
from the H-1947 and H-1510 eruptions (those with the
longest repose time shown as squares in the figure) was
towards south and their mass may be underestimated.
Hence, these two eruptions are eliminated when calcu-
lating the best fit curve. Extrapolation of this curve sug-
gests a repose period before the large H-1104 eruption
of at least 250 years.
– a) Samband upphafsstyrks SiO2 og lengdar undanfarandi goshlés í nokkrum sögulegum Heklugosum. Heilu
og brotnu línurnar eru byggðar á efnagreinigum en punktalínurnar sýna SiO2 innihaldið í beinu hlutfalli við
lengd hlés. Endurteiknað eftir S. Thorarinsson (1967). b) Samband stærðar sögulegra Heklugosa (reiknað
með aðferð Pyle, 2000) og lengdar undangengis goshlés. Rúmmál samkvæmt S. Thorarinsson (1954, 1967),
Thorarinsson og Sigvaldason (1972), Grönvold o.fl. (1983), Larsen o.fl. (1992, 1999) og Lacasse o.fl. (2004).
Rúmþyngd sem notuð var við útreikninga er 550 kg/m3 fyrir H-1104 og H-1158 og 700 kg/m3 fyrir önnur gos
(Thorarinsson, 1967). Magn gjósku frá 1947 og 1510 (ferhyrningar) sem dreifðist til suðurs er hugsanlega
vanmetið og er því ekki haft með í útreikningum á bestu kúrfu. Út frá bestu kúrfunni má meta lengd goshléss
fyrir gosið 1104 en það mat gefur a.m.k. 250 ár.
al., 2012). From 14 to 19 April the tephra contained
three glass types of basaltic, intermediate, and silicic
compositions recording rapid magma mingling with-
out homogenisation, involving evolved FeTi-basalt
and alkaline rhyolite with composition identical to
that produced by the 1821–1823 AD Eyjafjallajökull
summit eruption. The time-dependent change in the
magma composition is best explained by a binary
mixing process with changing mixing end-member,
silicic and basaltic in compositions, during the course
of the eruption. Tephra erupted in early May (Figure
6) showed that a new injection of deep-derived basalt
had occurred that caused increased activity and illus-
trated the prominent role of the deep basaltic intrusion
during this eruption. Decreasing mafic end-member
proportions with time in the erupted mixture strongly
suggests that the basaltic injections remobilized half-
solidified residual silicic magma beneath Eyjafjalla-
jökull and that the 2010 eruption was shut off by de-
clining injection of basaltic magma rather than emp-
JÖKULL No. 62, 2012 31