Jökull - 01.01.2019, Page 97

Nikkola et al.
0 20 40 60 80 100 0 20 40 60 80 100
0 50 100 150
80
81
82
83
84
85
86
0 50 100 150
0 20 40 60 80 100
0 20 40 60 80 100
80
81
82
83
84
85
86
80
81
82
83
84
85
86
80
81
82
83
84
85
86
80
81
82
83
84
85
86
80
81
82
83
84
85
86
Ol81_1
t = 22.3 days
Fo-core = 82.9
Ol82_1
t = 9.2 days
Fo-core = 82.45
Ol45_1
t = 36.7 days
Fo-core = 80.8
Ol76_1
t = 17.9 days
Fo-core = 84.4
Ol83_2
t = 33.4 days
Fo-core = 81.65
Ol89_1
t = 23.1 days
Fo-core = 82.9
Ol91_1
t = 10.3 days
Fo-core = 84.0
Fo Fo
Fo Fo
Fo
Fo
Distance ( m)μ Distance ( m)μ
Distance ( m)μ Distance ( m)μ
Distance ( m)μ Distance ( m)μ
Distance ( m)μ
a b
c d
f
g
80
81
82
83
84
85
86
Fo
0 20 40 60 80 100
e
’mixing plateau’
’core plateau’
Figure 6. a-g)Fe-Mg zonation in
the seven complex reverse zoned
olivine macrocrysts used for dif-
fusion modelling. The blue dots
show the measured forsterite (Fo)
content near the olivine rims,
the orange stippled line is the
assumed initial olivine compo-
sition, and the orange curve is
the calculated end result of dif-
fusive re-equilibration after time
(t). – a-g) Fe-Mg-beltun í
sjö ólivíndílum með flókna öfuga
beltun og notaðir voru við líkan-
reikninga efnasveimis. Bláu
punktarnir sýna mælt forsterít-
gildi (Fo) nærri rimum ólivíns.
Gula brotalínan á hverri mynd er
áætluð samsetning ólivíns í upp-
hafi líkanreiknings og heila gula
línan er reiknuð samsetning af
völdum efnasveimis eftir tíma t.
Kahl et al., 2015); however, the timescales are sen-
sitive to the selected model temperature. Consider-
ing the primitive nature of the Brattaskjól ankaramite,
we surmise that 1170◦C is fair, although conservative
(potentially low) estimate of the diffusion T (see Fig-
ure 5). We prefer a lower model temperature to avoid
underestimating the diffusion time scales, because, ir-
respective of the model conditions, the resulting diffu-
sion timescales are the maximum time scales of sub-
surface storage and ascent (after magmatic recharge)
as the crystals are hosted by a lava flow. Lava flow
fields insulate themselves effectively (e.g., Hon et al.,
1994; Self et al., 1998; Harris and Rowland, 2009)
and hence lava-hosted olivine crystals may remain at
high but gradually decreasing temperature for weeks
to months after eruption. We cannot separate diffu-
sion in a lava flow/lava lake and diffusion occurring
at depth during subsurface ascent; this is the caveat
of working with crystalline lava samples instead of
tephra.
96 JÖKULL No. 69, 2019