Perspectives for best practices in boron-based CO₂ reconstruction

Since atmospheric carbon dioxide (CO₂) is a key driver of global temperatures over geological time, determining past CO₂ and fluxes among Earth's carbon reservoirs is critical in resolving drivers and feedbacks within Earth's climate system. Such CO₂ reconstructions are challenging prior to ice core records of the last 800 kyrs, and yet it is beyond this time where the most promising analogs for our warmer, higher CO₂ future lie. Amongst proxies for past CO₂ reconstruction, boron isotopes (δ¹¹B, primarily measured in foraminiferal carbonate) have become one of the most well-established, having demonstrated their ability to replicate atmospheric CO₂ during periods where ice core estimates are available for comparison. Although analytical reproducibility across laboratories has greatly improved in recent years, the approaches taken to derive CO₂ estimates from these boron isotope measurements remain variable. For example, there are significant inconsistencies in how seawater temperature, boron isotope, and elemental composition are estimated, in how a second carbonate system parameter is used to constrain the carbonate system, in how ecophysiological imprints on proxy signals are accounted for, and in how uncertainties are quantified and propagated. The need to better harmonize practices in the community motivated a series of virtual workshops and a PAGES-supported meeting, and ultimately this AGU Special Collection “Advances and Best Practices in Boron-based Paleo-CO₂ Reconstruction.” In this synthesis article, we briefly summarize the history of approaches to address these issues, current limitations and strengths of the proxy, and highlight some important points to consider when applying the boron isotope proxy.