Jökull - 01.12.2007, Page 41

Rates of carbon ingrowth and nutrient release from young Icelandic basalts
Because of the value of C and P in revealing bio-
genic activity, we examined C ingrown and P content
and forms in soils derived from Icelandic basalts of
different ages. The primary objective of this studywas
to use biogeochemical analyses of C and P to deter-
mine the interplay between biogenic activity and soil
formation in Iceland. We hypothesized that biologi-
cal activity and weathering of parent rock will incre-
ase with age of the soil and expected to find evidence
to support this hypothesis in both the C and P ana-
lyses. We further hypothesized that older soils will
contain more organic C, higher concentrations of oc-
cluded and organic P, and lower amounts of mineral P
than younger soils.
GEOLOGIC SETTING
Iceland is located on top of the Mid-Atlantic Ridge
at the divergent boundary between the North Ameri-
can and Eurasian plates, above a mantle plume which
greatly enhances the volcanic productivity (Sæmunds-
son, 1979). Due to its tectonic location, the eruptive
products of Iceland are mostly comprised of basalt
(Jakobsson, 1979). Soil from lavas from three dif-
ferent volcanic systems were examined in this study:
Hekla, Eldgjá and Laki. Hekla is an active stratovol-
cano whereas the Eldgjá and Laki are eruption fissu-
res. The 1783 Laki eruption produced the second lar-
gest amount of basalt in the last 2000 years, second
only to the 934 AD Eldgjá fissure eruption (Thordar-
son et al., 2001; Thordarson and Self, 2003).
METHODS AND MATERIALS
Four historically dated basalt flows were sampled in
southern Iceland: the 934 AD Eldgjá flow, the 1300
AD Hekla flow, the 1554 AD Hekla flow, and the
1783 AD Laki flow (Figure 1). There were three sam-
ple sites at the 934 AD flow, two sample sites at the
1300 AD flow, two sample sites at the 1554 AD flow,
and one sample site at the 1783 AD flow. Soil and
bedrock samples were taken at each sample site. In
areas where there was well developed soil, a trench
was dug and samples were taken along various depths
of the trench. All of these soils can be classified into
Andosols (Arnalds, 2004).
Total organic C content of each unsieved soil sam-
ple was analyzed. Two grams of each bulk sample was
placed in crucibles and dried overnight at 106"C. The
samples were removed from the oven andweighed the
following day to determine the weight of the samples
before ashing. After recording the weight of the sam-
ple, the samples were ashed in a muffle furnace for 2.5
hrs at 550"C, with the combusted material re-weighed
after ashing. Loss on ignition (LOI) was determined
by subtracting the weight after ashing from the we-
ight before ashing. The LOI can be used to determine
the amount of organic C presenting the soil. Upon
combustion, C, N, H, and O are released. Approxi-
mately 40% of the mass lost during ashing is organic
C. Therefore, LOI is multiplied by 0.4 to approximate
the amount of organic C present. The resulting con-
centration of organic C was multiplied by soil density
from each sampling horizon and the depth of the sam-
ple trench to obtain total C in the soil (Oke, 1987).
In order to validate the LOI results, the organic
C content of five samples, which were chosen to re-
present the spectrum of organic C concentrations fo-
und in the LOI results, was determined using a CHN
Elemental Analyzer (FlashEA1112, Thermo Fisher
Scientific, Waltham, MA, US). Prior to analysis, sam-
ples were treated with dilute HCL for 24 hours and
then with water for 48 hours before being oven dried
(Midwood and Boutton, 1998). We found that the or-
ganic C values calculated from LOI were higher than
those determined by the CHN analyzer. However, the
difference is consistent across samples and is there-
fore not expected to affect the interpretation of the or-
ganic C results.
The P content of each sieved sample was deter-
mined by a four part sequential extraction as descri-
bed by Filippelli and Delaney (1996). This method
is similar to those of Walker and Syers (1976) and
Tiessen and Moir (1993). The steps of this procedure
include: 1) adding a citrate-dithionite-bicarbonate re-
ducing agent and magnesium chloride to acquire the
occluded fraction; 2) dissolutions with sodium ace-
tate+acetic acid solution and following dissolution in
hydrochloric acid to recover the mineral fraction, and
3) dissolution with hydrochloric acid after ashing to
obtain the organic fraction (Filippelli and Delaney,
JÖKULL No. 57 39