Geoscience Reference
In-Depth Information
assessing the overall effect of ocean acidii cation on
marine organisms and ecosystems. The recent surge
of experimental data has made possible the use of
meta-analytical approaches (Hendriks
et al.
2010 ;
Kroeker
et al.
2010; Liu
et al.
2010 ). Despite their
strengths, such approaches have several problems
( Borenstein
et al.
2009), the most serious of which is
the inevitable use of a biased dataset. The 'i le
drawer problem' relates to the fact that studies
reporting relatively high treatment effects can be
easier to publish whereas those which are inconclu-
sive never get published. This bias may not be
extremely serious today as editors may i nd it excit-
ing to publish data that are inconclusive as they go
against earlier conclusions and could raise contro-
versy. It is nevertheless critical that all datasets,
inconclusive or not, are published in databases or in
data journals (e.g.
Earth System Science Data
). The
other reason for the use of a biased dataset is poor
data reporting, which is a serious issue in ocean
acidii cation research. The three recent meta-analy-
ses mentioned above could not use all the data
available because many are unavailable or unusa-
ble. In their data compilation, Nisumaa
et al.
(2010)
identii ed 185 papers of interest but data from 85 of
them could not be compiled for three reasons: only
one parameter of the carbonate system was meas-
ured (49 papers), data could not be obtained from
the authors (30 papers), or the data were lost (16
papers). The publication of the
Guide for best prac-
tices on ocean acidii cation research and data reporting
(Riebesell
et al.
2010a) will hopefully lead to better
data reporting in future publications.
PFT, nor for adaptation. This simplistic representa-
tion contrasts with the emerging complexity of the
biological responses to ocean acidii cation and chal-
lenges the modellers. The evaluation of the impacts
of ocean acidii cation and climate change might
well require the development of a new generation
of ecosystem models. An alternative to the classical
PFT approach consists in the random assignation of
physiological traits to a large number of plankton
types. This approach makes use of the variability of
physiological data reported by experimental stud-
ies, rather than identifying
a priori
a limited number
of PFTs which are described by representative mean
parameter values. It allows for the emergence of
marine communities.
While this approach appears promising when it
comes to studying the environmental controls on
microbial community structure and ecosystems,
its capacity to project the reorganization of com-
munities and ecosystems in response to global
warming and ocean acidii cation remains to be
demonstrated.
Future ocean acidii cation may alter the i tness of
organisms, and their differential response is likely
to modify the community composition, for example
through altered competition between species
(Section 15.2.2.6). Optimality-based adaptive mod-
els allow one to investigate the synergistic effects of
temperature changes and ocean acidii cation on the
basic life functions of organisms (e.g. growth, repro-
duction, maintenance, but also calcii cation), as well
as intra- and interspecii c competition (Irie e
t al.
2010). Conceptual frameworks such as this are
widely used in evolutionary biology and it is antici-
pated that they will be used increasingly in the con-
text of the effects of ocean acidii cation and climate
change. However, extending similar conceptual
approaches to the scale of the global ocean remains
a challenge.
15.4.8 Model development
Present-day state-of-the-art global biogeochemical
models mostly rely on the representation of plank-
ton functional types (PFTs). Following this approach,
major biogeochemical functions are identii ed and
assigned to specii c groups, e.g. CaCO
3
production,
biogenic silica formation, nitrogen i xation. The
growth of each group is directly controlled by the
availability of external nutrients. If environmental
conditions (light, temperature, nutrients) cross a
critical threshold, the plankton functional type dis-
appears from the model world. This approach does
not allow for shifts between species within a given
15.4.9
Need for a coordinated international
effort
The ongoing and planned research projects at the
national or international scale cover the i eld of
ocean acidii cation well and have begun to generate
critical data. Those research initiatives are at present
uncoordinated, which has several consequences.
