ºù«Ӱҵ

We aim to

• quantify behaviour of participating models in response to key drivers of weathering and carbon capture 
• elucidate major sources of model uncertainties 
• develop best practices for benchmarking and validation
• facilitate transparency via open-source, well-documented code when papers published

Participants will be provided with the following input data:

• Precipitation 
• Air and soil temperature
• Atmospheric pCO₂
• Feedstock data
  • Treatment timings and quantities
  • Treatment mineralogies with standard kinetic and thermodynamic data
  • Secondary phases to be included/reported
• Soil data
  • Texture data, cation exchange capacity, organic carbon, 
  • Soil thickness to be modelled

Participating models will deliver the following time-series outputs:

• Carbon dioxide removal (CDR), i.e. bicarbonate in effluent (analogous to CO₂ consumption measured using river chemistry data)
• CDR potential, i.e. release of major cations from applied minerals during weathering 
• Individual mineral mass losses (percent)
• Selected secondary phases (e.g. calcite)
• Soil and effluent solution chemistry: pH and major ion concentrations

Each part of the project will answer the following research questions:

• How similar are the results from the participating models?
• Can we explain any major differences between models?

Where observational data are available, we will also quantify:

• Root-mean-square error for each model and each available type of observation
• Multivariate root-mean-square error (R has a package that calculates this as the mean of all the individual RMSEs scaled by their standard deviations)