Geoscience Reference
In-Depth Information
are then run with a series of different assumed greenhouse gas climate forcings
(scenarios) out into the future. Although one obtains a series of projections based
on different emissions scenarios, as noted in
Section 9.1
, many modeling centers
also provide ensembles - a set of simulations that start from slightly different ini-
tial conditions which enable comparisons between the forced component of a cli-
mate change with natural climate variability. For example, in the case on the NCAR
CESM, the CMIP5 archive contains five historical (1850-2005) ensemble members
started on January 1 from years 863, 893, 983, 937, and 1031 of a long-term prein-
dustrial (pre- 1850) spin-up integration.
Simulations for the IPCC ARC4 contained in the CMIP3 archive made use of a
suite of emissions scenario based on models of economic and population growth
and technological development. The A1 group of scenarios describes a future world
of rapid economic growth, a global population that peaks in the middle of the cen-
tury and then declines, and a rapid introduction of new and more efficient technolo-
gies. A1 is in turn subdivided into fossil intensive (A1FI), non-fossil energy sources
(A1T), or a balance across all energy sources (A1B). Other scenarios include A2,
B1, and B2. B1, for example, describes a world like A1 in which global population
peaks in mid-century and then declines, but in this case with a rapid change in eco-
nomic structures toward a service and information economy, with the introduction of
clean and resource-efficient technologies. Simulations for the IPCC AR5 contained
in the CMIP5 archive by comparison made use of Representative Concentration
Pathways (RCPs), which are based on a possible range of radiative forcing in
W m
−2
in the year 2100 (RPC 2.6, RPC 4.5, RPC 6.0, and RPC 8.5). The range cov-
ered by the RCPs is wider than that used in earlier IPCC Assessment Reports. In the
extreme RPC 8.5 scenario, carbon dioxide levels by the year 2100 reach 1,300 ppm
(today's value is a little more than 400 ppm). RPC 2.6 is as modest as RPC 8.5 is
extreme, with greenhouse-gas emissions dropping to zero by about 2070, and then
continuing to fall, with emissions becoming negative (carbon is removed from the
atmosphere).
Figure 9.12
shows projected (2006-2100) and hindcast (1900-2005) anomalies
in annual mean surface air temperature for the Arctic (taken here as the region pole-
ward of 65
o
N) from the CMIP5 models, along with the Arctic time series from the
ERA-Interim reanalysis for the 1979-2011 period (see
Section 9.7
). Anomalies are
expressed with respect to the 1986-2005 period. For the hindcasts and projections,
the solid line represents the mean from all models and their ensemble members,
while the shading indicates the one standard deviation range between the all models
and their ensemble members. The top (bottom) of the bars on the right side of the
plot represent the average of the +1 (−1) standard deviation values from all models
and their ensemble members for the last decade of the twenty-first century. The
salient points of the figure are that: (1) for the next couple of decades, projected
temperature changes are not strongly dependent on the RPC (there is much overlap
between all of the projections); (2) for a given scenario, there is a large range in
projected warming by the end of the twenty-first century, but how warm the Arctic
becomes depends strongly on human behavior.
