Do oversimplified durability metrics undervalue biochar carbon dioxide removal?

Soil amendment of biochar-the solid product of biomass pyrolysis-is one of few engineered strategies capable of delivering carbon dioxide removal (CDR) today. Quantifying CDR for biochar projects hinges critically on the durability of biochar materials once amended in soil. However, consensus on the definition of durability is still evolving, and as a result, standards developing organizations have generated a variety of different methodologies to assess the removal value of biochar projects. These methodologies primarily rely on single-parameter regression models to link the molar H/C ratio-an easily measurable bulk chemical metric-to the modeled durability of biochar materials. Specific deployment variables are not commonly considered. Thus, although H/C-based methodologies simplify project development and CDR assessment, questions remain as to how well they predict real project outcomes. Via a re-analysis of existing biochar incubation data and several case studies, we show that durability standards based on bulk compositional metrics are biased towards particular feedstocks and may not account for key environmental drivers. Without provisions for these factors, we find that existing assessment models appear to discount the removal value of biochar projects significantly. However, our conclusions rely on predictive models with important weaknesses and unknown uncertainty-pointing to a need to develop a use-aligned database. Limitations notwithstanding, our findings ultimately suggest the biochar 'durability problem' may be an artifact of the desire to simplistically define it. To reliably credit CDR, we propose a series of recommendations, including the creation of representative distributions for current feedstocks and environmental gradients to better align experimental data with real-world practices. Further, we suggest an approach to integrate in-field measurement protocols with existing strategies to evaluate CDR value, with potential to co-generate data to guide deployment, maximize agronomic co-benefits, and improve confidence in project integrity.