ature decrease, the main thermocline. This results in enhanced stability of the
water column, which inhibits mixing from above or below. It is only in some polar
regions, particularly around Antarctica and in the Greenland and Norwegian Seas
in the North Atlantic, where the absence of the thermocline allows direct, and
therefore rapid, mixing of surface with deeper waters (see also Section 6.7).
The large, natural, two-way flow of CO 2 across the sea surface makes it very
difficult to measure directly the rather small additional flux (about 2% of the gross
flux in either direction) resulting from human additions of CO 2 to the atmos-
phere. Best estimates from this approach (which rely on measurements of pCO 2
(see Box 3.1) across the sea surface) are about 2 GtC yr-^1. In these circumstances
resort is often made to mathematical modelling approaches. These models can
be of considerable complexity—Box 7.1 shows the principles on which they
operate. From modelling studies, the best estimate of the amount of anthro-
pogenic CO 2 being taken up by the oceans is 1.9±0.6 GtC yr-^1 , which is in rea-
sonable agreement with the estimates from direct measurements mentioned
above.
As an illustration of how marine biological processes may affect the ability of
the oceans to take up CO 2 we now briefly discuss the results of some recent field
experiments on the role of iron in controlling photosynthesis in the sea (see also
Section 6.6). For many years it had been speculated that in some major ocean
areas availability of iron was the limiting factor for phytoplankton growth.
However, it was only recently that several direct tests of this idea were carried
out by adding iron (in the form of ferrous sulphate) to a small (about 100 km^2 )
area of ocean and observing any resulting effects over periods of days. What was
found was that addition of only a very small amount of iron led to a dramatic
increase in plankton growth (see Fig. 6.26) with resultant drawdown of CO 2 from
the water (and hence potentially from the atmosphere). This is clear in Fig. 7.4
where pCO 2 is lower inside the iron-fertilized patch in relation to the values
outside the patch. This is an exciting result in helping to understand what con-
trols marine biological activity. However, its importance for long-term uptake of
CO 2 by the oceans is not yet established since it may be that much of the extra
carbon incorporated into new phytoplankton growth is rapidly recycled by res-
piration/decomposition in the near-surface waters. Carbon will only be removed
from the atmosphere/surface ocean system for any length of time if dead plank-
ton remains sink into the deep ocean. In order to answer this question it will be
necessary to carry out iron-fertilization experiments which last for longer and
cover a greater area. This will take a considerable effort at the international level
and is currently being planned through programmes such as IGBP (Section 7.1).
Fossil fuel burningIt is relatively easy to quantify the amount of CO 2 that results from the burning
of fossil fuel and other industrial activities, such as the manufacture of cement (as
part of this process calcium carbonate (CaCO 3 ) is heated to a high temperature
and decomposes, yielding CO 2 ). This source is easier to estimate than those dis-
cussed earlier because there is no natural component. All that is required is quan-
Global Change 247