Geoscience Reference
In-Depth Information
for the past several decades, but whether the
decreases are due to ocean acidii cation, some other
environmental factor (e.g. warming), or a combina-
tion of factors remains unclear.
The issue of attribution is a well-known problem
in studies of the effects of climate change.
Rosenzweig
et al.
(2008) compiled a huge database
and used innovative statistical approaches to dem-
onstrate that changes in physical and biological sys-
tems are pervasive and could probably be attributed
to climate change caused by increasing concentra-
tions of greenhouse gases. However, this approach
is limited by the short time span of many datasets
(Zwiers and Hegerl 2008), especially from marine
systems, and does not provide the contribution
from the major variables such as CO
2
, temperature,
or nutrients.
maintain a set pH in a small volume (Kirkwood
and Sano 2009 ).
Each of these approaches has strengths and weak-
nesses. Natural high-CO
2
environments capture the
full scope of ecosystem interactions over long time-
scales (at least for the benthos), thereby providing
crucial information on the effects on ocean acidii ca-
tion on trophic and competitive interactions and the
potential for adaptation. The high spatial and tem-
poral variability in
p
CO
2
and pH, however, makes it
difi cult to determine a reliable dose-response rela-
tionship, complicating the use of this information in
projecting the response to future high-CO
2
scenar-
ios. The interpretation is also complicated by the
uncontrolled advection and recruitment of organ-
isms from unperturbed adjacent areas. The latter
problem also applies to FOCE-type experimental
systems. Both approaches are restricted to benthic
communities with non- or slowly migrating organ-
isms. In contrast, mesocosm enclosure experiments,
which range in size between one and a few hundred
cubic metres and can be used for both benthic and
pelagic communities, have the advantage of well-
controlled carbonate chemistry and the absence of
exchange with unperturbed surrounding waters.
Their limitation lies in the fact that mesocosm exper-
iments generally exclude migratory organisms and
their duration is limited as a result of undesirable
effects due to the growth of organisms on the walls,
which limit their scope with respect to interaction at
higher trophic levels and adaptational responses.
With none of the available approaches providing
information on the full range of ecosystem responses,
progress in our understanding of long-term, ecosys-
tem-level impacts of ocean acidii cation requires the
integration of: (1) community-level experimental
studies, (2) i eld observations in high-CO
2
environ-
ments, and (3) ecosystem modelling with (4) single-
species laboratory experiments addressing the
mechanisms underlying the observed sensitivities
and (5) long-term high-CO
2
exposure experiments
examining the potential for adaptation.
15.4.6
Limited work at the community level
Progress in our understanding of the possible
impacts of ocean acidii cation on marine life is
partly limited by the scarcity of information on
responses at the community and ecosystem levels.
Results obtained in single-species experiments are
not easily extrapolated to natural systems due to
the lack of information on the inl uence of com-
petitive and trophic interactions on single-species
responses. To close this gap two approaches are
particularly promising: (1) community-level stud-
ies in natural high-CO
2
environments and (2) CO
2
perturbation experiments at the community and
ecosystem level. Both approaches have provided
important information on the effects of ocean acid-
ii cation on natural marine communities. The best
known example for a natural high-CO
2
environ-
ment is a CO
2
venting site in the Gulf of Naples,
Italy, where a community shift was observed along
a
p
CO
2
gradient, with calcifying organisms succes-
sively disappearing from the community towards
the CO
2
venting site (Hall-Spencer
et al.
2008 ).
Community-level perturbation experiments have
been conducted on both pelagic and benthic com-
munities using mesocosm enclosures (Langdon
et al.
2010 ; Riebesell
et al.
2010b ; Widdicombe
et al.
2010). The so-called free ocean CO
2
enrichment
(FOCE) system uses injection of acid or of CO
2
-
enriched air with controlled-loop pH feedback to
15.4.7 Difi culties in performing meta-analysis
Meta-analysis, which statistically combines the
results of several studies that address a shared
research hypothesis, is a method of choice for
