Assessing Earth system feedbacks in deep mitigation scenarios with activity-driven simulation of carbon dioxide removal

Assessing Earth system feedbacks arising from carbon dioxide removal (CDR) requires developing and simulating pairs of scenarios - a mitigation scenario with deployment of CDR and a corresponding no-CDR baseline. Both scenarios respect a specific long-term constraint on a carbon emission budget (i.e. emission reductions are pursued at the same level of ambition), but the latter describes a world where no CDR is deployed, such that net carbon emissions are larger and a given temperature threshold is missed. While over the past years a rich literature on deep mitigation scenarios with CDR has been emerging, the need for such no-CDR baselines has never been articulated explicitly. In idealized Earth system model (ESM) simulations of CDR, a no-CDR baseline is easy to imagine and implement, since socio-economic constraints are typically not taken into account. However, the deployment of CDR in deep mitigation scenarios, created by integrated assessment models (IAMs), is embedded in a consistent socio-economic description of plausible futures, and disallowing CDR may change many aspects of such scenarios, for example, the energy-system and land-use. Particularly, when moving towards an “activity-driven” representation of CDR in ESMs, where the activity that leads to a drawdown of CO₂ is explicitly modelled (rather than prescribed by using removal fluxes from the IAM simulation), the creation of no-CDR baselines comes with challenges. Here, we conceptualize how carbon cycle and biophysical feedbacks of CDR deployment can be determined from scenario simulations and corresponding no-CDR baselines. We show that different options exist for the creation of no-CDR baselines, which offer different insights and have their specific advantages and limitations. We argue that for certain applications (e.g., the determination of regional biophysical feedbacks) the use of idealized no-CDR baselines is unavoidable to some extent, implying that we have to accept some degree of socio-economic inconsistency in no-CDR baselines.