Editing IPCC:AR6/WGIII/Chapter-3 (section)

==== 3.8.2.1 Model Solvability ====

<div id="h3-30-siblings" class="h3-siblings"></div>

In order to be generated, scenarios must be coherent with the constraints and assumptions embedded in the models (i.e., deployment potential of given technologies, physical and geological limits) and in the scenario design (i.e., carbon budget). Sometimes, models cannot solve specific scenarios. This provides a first, coarse indication of feasibility concerns. Specific vetting criteria can be imposed, such as carbon-price values above which scenarios should not be reported, as in [[#Clarke--2009|Clarke et al. (2009)]] . However, model solvability raises issues of aggregation in model ensembles. Since model solving is not a random process, but a function of the characteristics of the models, analysing only reported outcomes leads to statistical biases ( [[#Tavoni--2010|Tavoni and Tol 2010]] ).

Although model-feasibility differs distinctly from feasibility in the real world, it can indicate the relative challenges of low-carbon scenarios – primarily when performed in a model ensemble of sufficient size. [[#Riahi--2015|Riahi et al. (2015)]] interpreted infeasibility across a large number of models as an indication of increased risk that the transformation may not be attainable due to technical or economic concerns. All models involved in a model comparison of 1.5°C targets ( [[#Rogelj--2018|Rogelj et al. 2018]] b) (Table S1) were able to solve under favourable underlying socio-economic assumptions (SSP1), but none for the more challenging SSP3. This interpretation of feasibility was used to highlight the importance of socio-economic drivers for attaining climate stabilisation. [[#Gambhir--2017|Gambhir et al. (2017)]] constrained the models to historically observed rates of change and found that it would no longer allow to solve for 2°C, highlighting the need for rapid technological change.

<div id="3.8.2.2" class="h3-container"></div>

<span id="scenario-feasibility"></span>