Geoscience Reference
In-Depth Information
reported a promotion of phytoplankton growth and
increased DMS production in response to increased
temperatures and irradiance. Clearly, the response
of the DMS system to changing global temperature
is likely to be variable and difi cult to predict or
generalize, without taking into account the effects
of ocean acidii cation which have not been included
in any model study to date. Changes to wind veloc-
ity are also likely to have an impact. Again, in the
modelling study performed by Bopp
et al.
( 2003 ) a
19% increase in the DMS l ux was observed from
30°S to 50°S, a quarter of which could be attributed
to an increase in wind speed. In brief, three other
climatic changes may inl uence DMS production.
First, a reduction in Arctic sea ice cover due to
increasing global temperatures. This would expose
more seawater to sunlight, and result in an increase
in phytoplankton productivity and DMS produc-
tion (Gabric
et al.
2005). Second, changes to atmos-
pheric dust input to the oceans due to changing
wind and land-use patterns. The iron fertilization
effect would enhance productivity, and increase
DMS production (Jickells
et al.
2005 ). Third, changes
to atmospheric convection patterns as a result of cli-
matic shifts may affect the transport of DMS to the
troposphere and subsequent CCN production
( Shaw
et al.
1998 ).
Thus, it is clear that the response of the DMS-
CCN-climate system to global environmental
changes is challenging to quantify, and despite more
than 20 years of research since the CLAW hypothe-
sis was i rst described, the sign of the feedback
mechanisms that may be involved is still uncertain.
At this stage, the extent of the impact and any
accompanying climate feedbacks associated with
changes in marine emissions of DMS as a result of
ocean acidii cation are uncertain.
(2007) reported that time-integrated concentrations
of chloroiodomethane (CH
2
ClI) increased under
high
p
CO
2
, with a 46% increase under two-times
present-day
p
CO
2
(~700 μatm) and a 131% increase
under three-times present-day
p
CO
2
(~1050 μatm).
Concentrations ranged from around 5 to 40 fmol l
-1
,
and maximum concentrations occurred 6 to 10 days
after the peak of chlorophyll
a
. The authors suggest
that differences in viral attack and phytoplankton
lysis may have produced the observed difference in
CH
2
ClI concentrations between present-day and
higher-
p
CO
2
treatments.
Following this, Hopkins
et al
. ( 2010 ) determined
the concentrations of methyl iodide (CH
3
I), ethyl
iodide (C
2
H
5
I), diiodomethane (CH
2
I
2
), and chloro-
iodomethane (CH
2
ClI) during an experiment at the
Bergen mesocosm facility in 2006 (Fig. 11.4). These
gases showed similar temporal trends, and maxi-
mum concentrations were observed over a 4-day
period immediately after the peak of the bloom. The
temporal development of these gases suggests a
close association with biological activity. However,
they do not appear to be directly related to phyto-
plankton growth, as maximum gas concentrations
generally occurred after the maxima in chlorophyll
a
. Most importantly, the data strongly suggest that
higher
p
CO
2
(lower pH) leads to a reduction in iodo-
carbon concentrations during the phytoplankton
blooms. Over the course of the experiment, large,
and in some cases signii cant, reductions in concen-
tration of all iodocarbons were observed (CH
3
I,
-44%; C
2
H
5
I, -35%; CH
2
I
2
, -27%; CH
2
ClI, -24%), and
these differences were even more pronounced dur-
ing the post-bloom phase of the experiment (CH
3
I,
-67%; C
2
H
5
I, -73%; CH
2
I
2
, -93%; CH
2
ClI, -59%).
In this experiment, CH
3
I and C
2
H
5
I concen trations
within the mesocosm were representative of realis-
tic open-ocean and coastal seawater (Abrahamsson
et al.
2004 ; Chuck
et al.
2005 ; Archer
et al.
2007 ). C
2
H
5
I
is considered to be a minor iodocarbon in seawater
( Archer
et al.
2007), and similarly it made up less
than 1% of the total iodocarbon pool in the meso-
cosm. CH
2
I
2
and CH
2
ClI concentrations were some-
what elevated compared with most (but not all)
oceanic measurements, and they dominated the
iodocarbon pool, in common with a number of
other i eld studies (Klick and Abrahamsson 1992;
Abrahamsson
et al.
2004 ; Archer
et al.
2007 ).
11.3 Impacts of ocean acidii cation on
organohalogen production and
atmospheric chemistry
11.3.1
Summary of experimental evidence
The response of iodocarbons to ocean acidii cation
has been assessed in two mesocosm experiments
( Wingenter
et al
. 2007 ; Hopkins
et al.
2010 ). During
the PeECE III experiment in 2005, Wingenter
et al
.
