relatively small, while in fact the flux corrections in some coupled models are comparatively large." Surely this implies that there may be some distortion of a model's greenhouse gas response. The fundamental question is: is the distortion sufficiently great to negate the value of the GCM projection? Clearly, many modelers believe that the GCMs are still useful despite the presence of some distortion due to the flux corrections. This position is much more credible than the current statement which seems to imply that there is virtually no distortion created by the flux corrections. UNITED STATES Page 6.11, lines 30 and 38: The terms "internal variability" and "internal fluctuations" should be defined for non-specialist readers. UNITED STATES Page 6.12, line 14: How "small" must initial state differences be to prevent divergence of climate state simulations? A quantitative indication is needed here. UNITED STATES Page 6.12, line 18ff: These two paragraphs are unclear and should be re-written. What does "The warming at the time of doubling ranges from 2.6 to 1.5K respectively" mean? With respect to what? This result is based on an unpublished personal communication. The sentence that begins on line 27 is confusing as written. UNITED STATES Page 6.13, line 38: "...directly related to the net heating at the surface and the ability of the ocean to absorb this heat." This should be more clearly stated as proportional to the time integral of the net heating at the surface. UNITED STATES Page 6.14, line 6ff: The results described here are again based on preliminary, unpublished results. They should either be deleted or clearly labeled as being preliminary and subject to change. UNITED STATES Page 6.15, line 15: Change this to "reminiscent of the temperature pattern associated with El Nino events." UNITED STATES Page 6.15, line 25: The labeling of results as "preliminary" and "use with caution" should appear at several other points in section 6.2 (some of which are noted above) as well as here. UNITED STATES Page 6.21, Section 6.4.1, and page 6.22, Section 6.4.2: Chapter 5 and references therein indicate that the simulation of ENSO variability requires high resolution in the tropical region of the ocean model. Without that the variation in thermocline depth and upwelling patterns cannot be faithfully simulated. The ENSO-like variability displayed by the coarse resolution coupled GCMs surveyed in the present chapter may not behave like the true ENSO phenomenon. These model GHG simulations should therefore be interpreted with caution (See references in 5.2.4.2). Cross reference to Chapter 5 is also necessary when discussing inter-monthly variability (e.g. blocking in 5.4.1). UNITED STATES Page 6.23, line 16: The term "precipitation variability...is enhanced" is ambiguous here. Does it mean that +/- the absolute value of precipitation variability increases, or that the percent variability of precipitation increases? In most cases, changes in percent variability are probably of more interest and relevance. UNITED STATES Page 6.24, line 37ff: This sentence is rather confusingly worded. What it seems to say is that the sea ice cannot melt significantly until it has started melting significantly! UNITED STATES Page 6.26, Section 6.5: A much more careful definition of what is meant by "extreme event" is needed at the beginning of this section. Strictly speaking, the expected frequency of an extreme event, as usually defined (e.g., the most improbable one percent of occurrences, events with an average recursion time of greater than 100 years, etc.), cannot increase. The section is really concerned with how the frequency of certain threshold events, which are currently quite infrequent, will change in a global warming scenario. This distinction should be clearly stated. UNITED STATES Page 6.26, lines 35 and 36: It is not clear what the difference is between "the appearance of largescale meteorological phenomena (e.g., El Nino)" and "an intensifying or weakening of...large scale systems...". Please clarify. UNITED STATES Page 6.29, Section 6.5.4: Simulation of mid-latitude storms depends on horizontal resolution. Therefore, the reader should be alerted to differences in resolution of the various simulations that are compared in this section. UNITED STATES Page 6.30, line 32: Should "46 degrees C" be 4-6 degrees C? UNITED STATES Page 6.31, line 5: "Small changes in the mean climate can produce relatively large changes in the frequency of extreme events." See comment above. UNITED STATES Page 6.31, line 8: "It was assumed that temperature variability remained constant." Absolute variability or percent variability? See discussion above. UNITED STATES Page 6.32, line 28ff: Unpublished sources are given for the information in this and the next two paragraphs. UNITED STATES Page 6.32, line 31: "...longer waiting time for rain..." is awkward; use "reduced rainfall frequency" instead. UNITED STATES Page 6.33, line 2: "...1xCO2 but twice that in the 3xCO2 run." The omitted "but" changes the meaning of the sentence completely. UNITED STATES Page 6.33, Section 6.5.7: The preceding section on extreme events contains some excellent material and is well documented. However, it is lacking a cohesive statement which would bring this issue into focus for policymakers. Such a statement was made available to the Synthesis Report Drafting Team by Dr. John Houghton : "References have also been made to the possible increase of the frequency of extreme events that some maintain have occurred during the last few decades. Convincing evidence is still lacking and it is statistically difficult to prove that an increase has taken place. There may, however, be several reasons for the escalation of the costs for severe weather events: Developed countries have become wealthier, more people live in coastal areas with costly infrastructure, personal goods and business process have generally become more vulnerable to water damage. Even if the frequency of severe weather events has increased in some parts of the world, this might be due to natural variations of the weather." Some form of the above statement should be included in section 6.5.7 (Summary and Page 6.33, Section 6.6: The sidebar "Box 3" of the Summary for Policymakers should be repeated here to provide a concise summary of the regional results. UNITED STATES Page 6.34, Section 6.6.1: The regional temperature and precipitation biases of the models cited are large enough to make the projections of future regional climate scenarios of doubtful value. This weakness should be strongly emphasized in this section. UNITED STATES Page 6.36, line 1: The Pleym et al reference is given as 1994, not 1995, in the references. Why is this the only result discussed when several others are referenced in passing on the preceding page? UNITED STATES Page 6.37, line 7: "...10%-40% of observations..." What does this mean, that only 10-40 percent of observations have biases, or something else? The present wording is not clear. UNITED STATES Page 6.37, Table 6.6.1: Tasmania is omitted from the temperature part of the table. Was this intentional or an oversight? UNITED STATES Page 6.39, line 26: Change this to "...precipitation biases in the range of approximately -45% to +20% of observed values..." UNITED STATES Page 6.42, Section 6.7.1.3: It should be noted here that not all coupled ocean-atmosphere models require the use of flux adjustments in order to simulate realistic climate scenarios (e.g., Schopf et al). UNITED STATES Fig. 6.2.2: The text should note that the Cubash et al (MPI) transient model (identified as MPI CO2 anom in the figure) is suspected of simulating temperatures of the high latitudes of the Northern Hemisphere as too cold (by as much as 6C?) for roughly the first 65 years. If this error were eliminated, the warming it predicts would likely have begun well before now, leading to a disparity between this model and reality. UNITED STATES Page 6.68, Section 6.A.2: The text should indicate the best estimate and the uncertainty range regarding the forcing from each gas listed. (This comment applies to Chapter 2 if the section is moved to Chapter 2.) UNITED STATES Page 6.68, line 35: The sentence implies that Legett et al. did not project their results based on regional data. The text should clarify whether this is true. Given that the Legett et al. paper was prepared as an IPCC document, it would seem that the authors could have obtained unpublished regional results. UNITED STATES Page 6.73, line 8: Change the post-2100 assumption to an assumption based on constant emissions or constant concentrations, at least through the 22nd century--or make an assumption based on the published literature (e.g. Nordhaus or Cline). Justification: The basis for assuming that emissions decline linearly rom the year 2100 to the year 2200 seems extremely optimistic and is not supported by the economic analyses in Working group III. There is sufficient carbon in the ground to sustain year-2100 emissions for more than a century. In the absence of detailed modeling, a simple assumption is necessary--but a better assumption would be constant emissions, since the stated premise of the analysis is no policy. A linear decline in emissions would a technical or policy change. UNITED STATES Page 6.73: Somewhere on this page, a discussion on the possibility of an abrupt climate change, and the reasons for omitting it from the model, should be included. UNITED STATES Page 6.73, line 28ff: The discussion of long-term changes in sea level should be moved to Chapter 7, and compared to other analyses of long-term simulations of sea level rise, such as those by Thomas (1985), Lingle (1985) (both in Polar Research Board 1985), Wigley and Raper, Titus and Narayanan (1995). The text should note that the 1-D model ignores ice sheet dynamics, which may be a particularly important oversight for very long-term simulations. UNITED STATES Attachment to letter to Congressman George Brown from Dr. Robert Corell, March 4, 1996 ATTACHMENT 3 E-mail dated July 12, 1995 from J. Mitchell to Michael E-mail dated July 17, 1995 from J. Mitchell to Michael E-mail dated November 17, 1995 from J. Mitchell to Michael MacCracken Letter to Congressman Brown from Dr. Robert Corell, March 4, 1996 1 J Mitchell (by way of shenson@usgcr,7/12/95 7:59 AM,IPCCchpt6 Date: Wed, 12 Jul 95 08:59:44 EDT X-Sender: shenson@earth.usgcrp.gov To: mmaccracken From: J Mitchell <jfbmitchell@meto.govt.uk> (by way of shenson@usgcrp.gov (Susan Henson)) Subject: IPCCchpt6 >From jfbmitchell@meto.govt.uk 12 July 1995 Dear Mike, it was good to see you in Boulder. I think you organized an excellent session. I did have some difficulty with the general comments. It was not clear from the specific comments what instances there were of disconnected sentences.... not always logically related to one another". Nor am I quite clear to what extent we can provide a discussion of model results and what they mean where we have done so we seem to have got our wrists slapped eg interpreting the degree to which flux correction per se distorts the response of models. I would be grateful for guidance on this before next week when I will go over the section in detail before the Ashville meeting. As regards unpublished results, we have included references to results I should also like to rebutt the innuendo about release of results. John J F B Mitchell Hadley Centre for Prediction and Research Meteorological Office Bracknell RG12 2SZ, UK. NOTE CHANGE OF AREA CODE Fax +44 1344 856912 |