Present climate and biological change - Climate Change: Biological and Human Aspects - page 233

cycling even faster than thought. (See also Cuntz, 2011, for an overview of this work
and also other earlier evidence for increased global primary production discussed in
section 6.1.3.)
That we need to treat climate computer models with caution was highlighted in
2007 by Stefan Rahmstorf in Germany and Anny Cazenave in France together with
colleagues in Britain and the USA. They looked at (then) recently observed climate
trends for carbon dioxide concentration, global mean air temperature and global sea
level and compared them to previous model projections as summarised in the IPCC
2001 assessment report (IPCC, 2001b). They reported that data for the period since
1990 to 2005 raised concerns that the climate system, in particular sea level, may be
responding more quickly to climate change than our current generation of models
indicates. (This research just missed getting included in the IPCC's 2007 assessment
but will no doubt help inform its 2013 report.)
In 2010 the Royal Society (Britain's academy promoting scientific excellence) also
urged caution when considering climate models. In its Climate Change: A Summary
of the Science report, it said
The ability of the current generation of models to simulate some aspects of regional
climate change is limited, judging from the spread of results from different models;
there is little confidence in specific projections of future regional climate change,
except at continental scales.
As noted, one of the biggest problems with global computer models is that to date
(2012) they do not successfully reflect climate change at high latitudes, both when
modelling the present Earth and in the past with a different continental distribution.
Their failure to capture the mid-Cretaceous warm Arctic climate (see Chapter 3)
is a case in point, although we know from palaeoclimate proxies (fossil remains,
biomolecular climate proxies and oxygen isotope analysis) that the Arctic was warm
at that time. It was even warmer later, during the Initial Eocene Thermal Maximum
(section 3.3.9), and global models also have difficulty in capturing this. So, even
being aware of polar amplification from actual measurements, the implication from
palaeodata is that high latitudes will warm far more than models currently suggest.
Indeed, Antarctica is currently the fastest-warming place on Earth, having warmed
some 3 ◦ C in the 50 years up to 1990 (which compares to a global average warming
of about 0.6 ◦ C for the entire 20th century). Whereas model failures may not so
markedly affect the IPCC's climate-scenario forecasts for tropical and temperate
zones, implications for the IPCC's projections of sea-level rise are considerable, as
the major ice caps (Antarctica and Greenland) capable of contributing to sea-level
rise are at high latitudes. (We will return to this in the next section as well as next
chapter.)
Part (although not all) of this high-latitude, polar amplification problem - especially
with regard to the northern hemisphere, which has a greater land area just outside the
Arctic Circle compared to the Antarctic Circle - is the effect of biology on albedo.
Again, it must be emphasised that computer climate models have been, and are,
getting ever more sophisticated (and including more biological dimensions). Up to
the end of the 20th century models did include surface cover in a simplistic sense and
surface-cover albedos so that land without snow (and sea without ice) absorbed more,