date: Wed, 24 Nov 2004 11:07:17 +0000
from: Sarah Raper <sraper@awi-bremerhaven.de>
subject: my zero draft for CH10
to: Tom Wigley <wigley@cgd.ucar.edu>

Dear Tom,

At the end of this email please find my zero draft for ch10. I would 
much appreciate it if you would look through it. I have reviewed 
literature since the TAR and how that reflects on what we did in the 
TAR, and outlined what might be done in the AR4.
It's only zero draft and likely most of it won't stay but it does set 
the tone for where we are going so its quite important for that.
It is already about 3/4 the length of what I am allowed!

cheers,

Sarah

PS will have some space for forcing in Collin's section.

10.5.2 Range of responses from different scenarios.

The TAR projections with a SCM presented a range of warming over the 
21st Century for all the SRES scenarios. The construction of the TAR 
Figure 9.x was pragmatic. It used a simple model tuned to AOGCMs that 
had a climate sensitivity within the long-standing range of 1.5 - 4.5 
advocated by the IPCC. Models with CS outside that range were discussed 
in the text and allowed the statement that the presented range was not 
the extreme range indicated by AOGCMs. The figure was based on a single 
anthropogenic forcing estimate for 1750 to 2000, which is well within 
the range of values recmommended by TAR ch 6, and is also consistent 
with that deduced from model simulations and the observed temperature 
record (TAR ch 12.). To be consistent with TAR Ch 3. climate feedbacks 
on the carbon cycle were included. The resulting range of global mean 
temperature change from 1990 to 2100 given by the full set of SRES 
scenarios is 1.4 to 5.8 degC.

Since the TAR several studies have examined the TAR projections and 
attempted  probabilistic assessments. Allen et al, 2001 show that the 
forcing and SCM tunings used in the TAR give projections that are in 
agreement with the observationally constrained probabilistic forecast 
(Allen et al. 2000), reported in TAR ch x, stating that under the IS92a 
scenario anthropogenic warming is likely to lie in the range 0.1 deg to 
0.2 degC over the next few decades.

As noted by Schneider (2001), Jones (2000) and Moss and Schneider 
(2000), giving only a range of warming results is potentially 
misleading unless some guidance is given as to what the range means in 
probabilistic terms. Wigley and Raper (2001) interpret the warming 
range in probabilistic terms, accounting for uncertainties in 
emissions, the climate sensitivitiy, the carbon cycle, ocean mixing, 
and aerosol forcing. They give a 90% probability interval for 1990 to 
2100 warming of 1.7 deg to 4.9 degC. As pointed out by Wigley and Raper 
(2001), such results are only as realistic as the assumptions upon 
which they are based. Key assumptions in this study were; that each 
SRES scenario was equally likely, that 1.5 to 4.5 corresponds to the 
90% confidence interval for the CS, that carbon cycle feedback 
uncertainties can be characterised by the full uncertainty range of 
abundance in 2100 of 490 to 1260 ppm given in the TAR. The aerosol pdf 
was based on the uncertainty estimates given in the TAR together with 
constraints based on fitting the SCM to observed global- and 
hemispheric-mean temperatures.

Several studies have used observational constraints to determine the 
range of likely future climates under specific emissions scenarios.  
Because CS is only weakly constrained by the observations Knutti et al 
(2002) admit higher warming for the specific scenarios studied compared 
to the TAR SCM projections. However, when they also constrain CS to be 
in the range 1.5 to 4.5 deg C they get results consistent with the 
those of the TAR. Stott and Kettleborough 2002 bypass the need to 
specify the CS and scale scenarios on the assumption that a model that 
over- or under-estimates the response by a ceratin fraction now will 
continue to do so by a similar fraction in the future. They give 
probabilistic results for specific emissions scenarios which admit 
higher warming than given in the TAR.

Stott and Kettleborough 2002 mention that the reduction of SO2 
emissions in the SRES scenarios in the latter half of the century 
increases the uncertainty range consistent with past observations. 
However, Wigley and Raper 2001 with their method, report that the 21st 
C decline in SO2 emmissions leads to a reduction in the effect of the 
very large present-day uncertainty range in aerosol forcing on future 
projections.

Webster et al. (2003) use the probabilistic emissions projections of 
Webster et al.(2002) which consider present uncertainty in SO2 
emissions, and allow the possibility of continuing increases in SO2 
emissions over the 21st C, as well as the declining emissions 
consistent with SRES. Their main results give a CS pdf not unlike that 
used by Wigley and Raper (2001) but for aerosol forcing their pdf gives 
substantially smaller forcings compared to both Wigley and Raper (2001) 
and Knutti et al. (2002). This is likely to be a compensatory effect 
because they did not explicitly consider natural forcings. Since their 
climate model pdfs were constrained by observations and are mutually 
dependant the effect of the lower present day aerosol forcing on the 
projections is not easy to determine but there is no doubt that their 
projections tend to be lower where they admit higher SO2 emissions.

Only the first of these studies (Wigley and Raper, 2001) considers the 
effect of carbon cycle feedbacks. None of the studies consider abrupt 
changes which are examined in another section.

The aim of this section is to bring together information on emissions 
scenarios from WGIII, on forcings from Chapter 2, on the carbon cycle 
from Chapter 7, on attribution from chapter 8, and on model assessment 
in Chapter 9, together with the AOGCM responses examined in this 
chapter and to make projections with simplified models that are 
consistent with that information.

There will be a figure 10.x1 comparable to TAR Figure 9.14, so that the 
new projections can be compared to the old. Observational constraints 
will be considered at least for the near term using information from 
Chapter 8 (attribution).

As well as the SRES scenarios there may be new information from WGIII, 
with the possiblility of probabilistic and longer-term scenarios. It is 
likely that Chapter 2 will provide probabilistic forcings. In TAR 
Figure 9.14 a climate feedback on the carbon-cycle was included. We 
will seek the latest information on the strength of this feedback and 
its uncertainties from Chapter 7. These uncertainties will be 
concatenated where possible in probabilistic form to produce a new 
presentation of results in Figure 10.x2 (SCM) and Figure 10.x3 (EMIC, 
possible contribution from Knutti). It may be necessary to have an 
additional figure showing longer-terms scenarios (Figure 10.x4).

The SCM will be tuned to emulate the AOGCMs using the PDMCI AR4 AOGCM 
modelling exercise data. For Figure 10.x1 the climate sensitivities 
will be the AOGCM effective climate sensitivities at the time of CO2 
doubling and the ocean heat uptakes in the SCM will match those of the 
AOGCMs in the 1% CO2 increase experiment. Figures 10.x2 and 10.x3 may 
present results based on pdfs of the climate sensitivity and other 
inputs.

Probabilistic temperature projections do not give true probabilities of 
occurrance but are conditional on the assumptions made in their 
construction. To convey this it may be wise to present more than one 
set of probabilistic projections, using for example different pdfs for 
the climate sensitivity (cf Box on pdfs of climate sensitivity).

