2. The Low Energy Demand (LED) scenario did not quantify costs "We have not explored [costs] in any detail, with the exceptions of the costs of supply-side ... However, this is a one-sided story without analogous quantifications of the demand-side ..."
4. IPCC AR5: Including more exogenous energy efficiency cuts mitigation costs in half (left bars), but "An important caveat to these results is that the costs of achieving these reductions were not considered nor were the policy or technology drivers that led to them"
5. Basically, studies that include more on the demand side have to force it into the model, it does not come naturally through the cost-optimal principles driving IAMs.
This is the background to the @wim_carton quote in the first tweet.
6. "While they don’t provide a cost analysis for these scenarios, one can assume that they would be significantly more costly – in the way IAMs assess this – than “standard” mitigation approaches." @wim_carton
7. "As Tavoni and Socolow note, this “should make the reader cautious about carrying modeling results into the real world”." @wim_carton
8. Most IAMers openly acknowledge that their scenarios are weak on the demand side, and this is a pretty critical weakness when it comes to interpreting those results in a policy context. IAMs tell us only half the story, the supply-side story.
9. Ok, that is enough tweeting on @wim_carton's paper.
Although the COVID-19 pandemic will cause a dip in 2020 emissions, this will not bring the world closer to the Paris Agreement goal of limiting global warming this century to well below 2°C & pursuing 1.5°C.
A 𝘁𝗵𝗿𝗲𝗮𝗱 on how mitigation works, why we probably need some level of carbon capture & storage (CCS) & carbon dioxide removal (CDR) - just not as much as in scenarios.
2. We start with a baseline or reference scenario, that assumes no or limited mitigation. If we want to stay "well below 2°C" we need to get rid of the dark grey & be net-zero!
We can argue about the baseline, but for the purposes here, it doesn't matter nature.com/articles/d4158…
3. The heavy lifting is done by conventional mitigation: behavioural change, energy efficiency, fuel switching (fossils to non-fossils), changed transport, dematerialisation, etc, etc...
But, scenarios suggest this is not enough to get rid of all greenhouse gases.
2/ "As the IPCC points out, aggregate mitigation costs in IAMs generally increase when action is delayed. ... The longer mitigation is delayed, ... the more investments and/or devaluations it will therefore take to eventually bring emissions down to net zero/net negative."
3/ "The cost of mitigation is therefore not a function of continued fossil fuel use per se, but of the steepness of the mitigation curve, that is, of how quickly fossil fuel consumption needs to fall in order to reach the specified temperature target."
1/ "the availability of BECCS proved critical to the cost-efficiency, & indeed the theoretical possibility, of these deep mitigation scenarios, leading to systemic inclusion of BECCS in RCP2.6 scenarios" says @katedooley0, Christoff, @KA_Nicholas
2/ "The incorporation of NETs in IPCC scenarios is one clear illustration of how, as @EstherTurnhout put it, “dominant political discourses compel scientists to create assessments that work within these discourses”..." writes @wim_carton
2/ Does it make sense to include current policies or NDCs across all SSPs? Doesn't the existence of current policies or NDCs begin to preclude some SSPs?
One could use SSP2 (current socioeconomic trends) with scenarios performed with varying SPAs.
3/ This is really a challenge of the SSP/SPA/RCP framework. The three axes are essentially assumed to be independent, this makes theoretical sense but not really practice sense. I understand why that decision was made, but does it make the framework too unrealistic?
2/ The Shared Socioeconomic Pathways (SSPs) are used as input to Integrated Assessment Models (IAMs) & combined with Shared Policy Assumptions (SPAs) to get different forcing levels (RCPs) in 2100. Earth System Models use pathways generated from the IAM SSP/RCP combinations.
3/ Not all combinations are possible. That is, IAMs often cannot solve for some SSP/RCP combination:
* Only SSP5 can get to RCP8.5 in IAMs
* Many IAMs cannot get to RCP1.9 or 2.6 with SSP3, 4, 5