Dear Ron, Thanks for your questions. I am going to post some answers below point-by-point for clarity's sake. I hope these answers help clarify and please feel free to contact me directly ([email protected]) if they don't.
Kate 2. I looked carefully only at the South Asia region - in part because it > was large and stood out (apparently [in figs 1 and 2] always (all years?) > wanting about .19 rather than .2 for preferred aerosol optical depth). > I read that Fig1 says that South Asia comprises 25% (by population?) of the > 7-member winning coalition when population is the criterion but only 10% > (of the GDP?) of the excluded 16 members, should the decision be made on a > GDP basis (when the coalition is only 6 members). > KLR: The numbers in Figure 1 show projected shares of world population and GDP in the 2070s (the last decade we simulated the game in). You can read more about the data sources and methodology we used to calculate these projected shares in the supplementary materials. 3. Can you explain the numbers +3% and -6% for South Asia in Fig 2? > How do I get to 100% from these numbers? > KLR: The numbers for each region in Figure 2 show the share of that regions damages from anthropogenic climate change in the absence of geoengineering that are gained or lost when geoengineering is implemented through an exclusive coalition rather than an open global one. This is why the winning coalition members (in blue) always have positive values and the excluded regions generally (though not always) have negative values. 100% is equal to the damages South Asia would suffer if there were no geoengineering. These values are shown in percentages rather than absolutes so you can get a sense of the incentive each region has, from its own internal perspective, rather than relative to the smaller or larger economies of the other regions. Just to be clear the numbers on these axes are obtained with the following formula -- for region N, Change in damage reductions due to exclusive coalition_N = (Damages reduced in exclusive coalition game_N - Damages reduced in open membership coalition game_N) / Damages from climate change absent geoengineering_N > 4. Figure 3 has three parts; not clear to me is part b). The above > numbers for South China look close to, but not exactly the same as the > differences of the red-dashed and blue-solid curves from the "Grand > Coalition" curve. That is - is the "Grand Coalition" curve of Figure 3 > also the origin on the ordinate of Figure 2? > > KLR: These percentages, again, are relative to climate damages absent geoengineering. So this panel shows the share of damages that can be negated (using the climate damage function we chose) using the amount of geoengineering implemented by the grand coalition (in black) or either of the exclusive coalitions. When displaying this result for the exclusive coalitions, we split the results into damages reductions achieved by the winning coalition members versus the excluded parties. So you can see that the winning coalition members have higher damage reductions than the excluded regions. > 5. The supplementary material gives more data for all 22 regions. For > South Asia, Figure S1 (open) shows four numbers close to 91%. Is the > complement (9%) similar to the -6% number above? > KLR: This figure only shows results for open membership coalition games (with and without transfers via Nash bargaining). The point of this figure was just to show how much damage reductions change with transfers allowed in the open coalitions game (which was generally not much) since the open coalitions game is used as a point of comparison as explained above. > In Figure S-2, the values of about 6% (in a coalition) and 14% (out) > have same meaning as in Figure 1 of main paper? (But surprisingly, in > opposite order of magntude) > KLR:The values in this figure have nothing to do with whether a given region is in or out of a coalition. They are simply the power share of that region (an input to the coalition game, not an output). These are values that came from the data sources cited (and partitioned into the Giorgi climate regions according to the method in section S1) > Figure S-3 is much like Main Fig 2 and Figure S-4 like Figure 3a - but > just different decision criteria? > I have looked up "IFP" but do not yet know who would favor its use over > the other three and why? > KLR: It the same as Fig 2, but using a different power criterion. Power share influences the game in two ways : 1) a viable coalition must have >50% of world power, and 2) intracoalitional transfers are a function of intracoalitional power share. The IFP index in just another potential power metric. Its advantage is that it takes into account GDP, population, military power and other considerations. In this way it is a better power metric than any of the others, because true power in the international relations arena is not a function of any one characteristic of a player. It was a reviewer that suggested we try the game with this power metric as well, which we did because it was not too hard and interesting to see how the results differed. We chose not to show it in the main text because it is not as intuitive a metric as GDP or population. > 6. This last question is what really interests me. I don't see too > much difference from being in or out of the "power" coalition, but if I > were representing "South Asia", I clearly would rather have the decisions > being based on population and not GDP. Could your team now rank order the > four criteria you studied on any ethical basis? > > KLR: For the purposes of the paper, I don't want to rank these on an ethical basis. We tried to choose metrics that could plausibly be a proxy for the power of the players in the game, recognizing that all players are not created equal in international negotiations. This paper is intended to be a scientific exercise (economics is a science), not a normative commentary. Where values enter into our discussion, eg, at the end of the paper, we try to be explicit that any conclusions are conditional on those value judgements. Personally (not as a scientist, but as a person), I agree that using population is more equitable than GDP, but I don't think that it is a more realistic metric than GDP. To the extent that such highly idealized models as ours can represent reality, we tried to do that. That's why we calibrated our model with real socioeconomic and climate model data. > 7. I also have a gut feeling that the "Grand Coalition" looks "better" > in some average sense than any of the four decision criteria you studied. > . Might you agree? Should population be the criteria, in your six decadal > decisions, South Asia would be in and out three times each. Maybe > logical, but the rationale for this result must be pretty hard to figure > out My gut reaction is based only on increasing likely inter-country > disagreements - and the fact that the differences were not large (as > measured in the "desired" optical depth or years to achieve a specific > depth. > KLR: Yes, we draw this qualified conclusion at the end of our paper for all the reason listed in the Discussion section. I think there are both positive and normative reasons you might draw this conclusion -- in our paper we try to focus on the positive ones, e.g., the obvious benefits of good international relations with regions you might otherwise exclude due to trade & security considerations, etc.; direct costs; fickle outcomes. > 8. My reason for digging this deeply was to see if I could transfer > your very novel approach over to the other side of geoengiineering: CDR. > So far, I have failed to see any answer clearly. I guess (no proof) that > the decision should be like the one you recommend for mitigation - all > regions should always do as much as they can. Would that be your "guess" > also? > KLR: I think this approach would not be particularly illustrative for CDR because, in the important ways for this game, CDR more closely resembles mitigation than SRM. It is slow and expensive and everyone would like to do as much of it as possible, so interests are aligned -- there are strong incentives to be inclusive. > 9. It seems there were no costs at any time for any region. True? > It seems that some cost should be involved somewhere, but I missed it. > KLR: True -- Direct costs for stratospheric SRM are estimated to be very low (see McClellen et al 2012) compared to climate damages, so we excluded them from our model due to its many other complex (and in our judgement essential) components, and a desire keep things as simple as possible. Of course, direct costs may turn out to be significantly higher than currently estimated, in which case they would absolutely need to be incorporated in order to correctly illustrate strategic incentives. > > 10. I think this gaming effort was noble, but I have not been > convinced that we will such a decision process develop - and I don't think > you are recommending it necessarily either .But I think a lot of helpful > dialog could result if people could see more detail on the geographic > regions of most interest to them - and how they would benefit from lobbying > for one decision criterion or another. Are any regions left behind every > year for each criterion? > Might the full set of your output data be available anywhere (now > or later)? > KLR: Again, we tried our best not to make any normative judgements in this paper about what is right or wrong. We are trying to illustrate strategic economic incentives for regions to behave one way or another -- just one thing any player would take into account in the real world. This model is a huge simplification of the reality. There are many important factors that are not included in the model, for the sake of some dynamic transparency -- the more complicated the model, the more difficulty in teasing out the mechanisms driving the results. All of the climate model output that fed the game model are available through the climateprediction.net results site. There are several papers that have been published on the regional results of the se particular climate simulations: Ricke KL, Morgan MG and Allen MR. 2010. Regional climate response to solar radiation management Nature Geosci. 3 537–41 Moreno-Cruz J, Ricke K and Keith D 2011. A simple model to account for regional in equalities in the effectiveness of solar radiation management. Clim. Change 110 649–68 Ricke KL, Rowlands DJ, Ingram WJ, Keith DW, Morgan MG (2012) Effectiveness of stratospheric solar-radiation management as a function of climate sensitivity. Nature Climate Change 2:92-96 There are also many additional papers that have been written purely about the physical effects of SRM, including at the regional level. I would caution against relying on the regional results from any one model. I generally interpret the regional results from these climate models as an illustration of the kinds of heterogeneities that could occur, not the ones that will occur, if we implement solar geoengineering. -- You received this message because you are subscribed to the Google Groups "geoengineering" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/geoengineering?hl=en. For more options, visit https://groups.google.com/groups/opt_out.
