As we count down to the 4th International Conference on Carbon Dioxide Removal in Milano, we are hosting a series of discussions on the research that will be shaping our sessions this June! 
This study introduces CDRMIP2, a global modelling initiative designed to improve how carbon dioxide removal 
methods are represented in Earth System Models 
. By comparing approaches such as afforestation 
, BECCS 
, ocean alkalinity enhancement 
, and macroalgae farming 
across harmonised simulations, the project aims to better understand carbon removal efficiency 
, Earth system feedbacks 
, and potential side effects 
while improving transparency and consistency across modelling efforts 
.
Full Abstract: The next phase of the carbon dioxide removal model inter-comparison project (CDRMIP) - Experimental design, proof of concept and CDR modelling repository
Authors: Nadine Mengis JörgSchwinger Anna Harper Raffaele Bernadello EsteladeAlmeidaMonteiro Makcim De Sisto Justus Perschon Giang Tran
We introduce the second phase of the Carbon Dioxide Removal Model Intercomparison Project (CDRMIP2), a coordinated international effort to advance understanding of Earth system responses to large-scale carbon dioxide removal (CDR) in Earth System Models (ESMs). Building on the previous phase of CDRMIP and complementing ongoing model intercomparison initiatives, this phase focuses on improving the representation of activity-driven CDR methods and systematically assessing their associated climate, biogeochemical, and biophysical responses.
We present the experimental design of CDRMIP2, including its key research questions, proposed simulation protocols, and core diagnostic outputs. The experimental framework encompasses a broad range of terrestrial- and marine-based CDR methods, including afforestation and reforestation, bioenergy with carbon capture and storage, ocean alkalinity enhancement, and macro-algae farming. Implementations span a wide range of complexity, from idealised experiments aimed at process understanding to more realistic, scenario-constrained applications. Participating models aim for fully coupled land–ocean–atmosphere ESMs, enabling interactive carbon cycle and climate feedbacks and allowing assessment of both local and global responses.
We will be able to present first results from proof-of-concept simulations conducted within this framework. These results illustrate sensitivities in carbon removal efficiency, climate responses, and associated side effects across CDR approaches and modelling configurations, providing early insights into the role of implementation choices and Earth system feedbacks.
To enhance transparency, comparability, and synthesis across modelling efforts, we introduce a comprehensive, community-driven repository documenting how different CDR methods are implemented across models and levels of complexity. The repository is designed to facilitate systematic model intercomparison, enable reproducibility, and highlight differences in implementation choices that influence simulated outcomes. It is also intended as a practical resource to support modelling centres in developing and improving their representations of CDR, thereby lowering barriers to participation and broadening engagement in CDRMIP. Initial analyses based on the repository identify key gaps between current modelling practices and potential future implementation pathways, as well as poorly constrained processes that limit robust assessment of CDR efficacy and side effects.
Through global collaboration, shared infrastructure, and harmonised protocols, this next phase of CDRMIP aims to support more robust, transparent, and policy-relevant assessments of CDR efficacy, side effects, and Earth system feedbacks in ESMs.
How can we confidently assess the effectiveness and risks of carbon dioxide removal if different Earth System Models represent these approaches in different ways? 