Offsetting unabated agricultural emissions with CO2 removal to achieve ambitious climate targets

revised formula for GWP� by Cain et al. (2019) [77] may offer a better estimation of the CDR requirements. We conclude that only an explicit computation of CDR rates derived from the total ERF budget leads to the complete neutralization of the effects of additional agricultural methane. GWP20, GWP100 and GWP� provide estimates of variable quality when benchmarked against explicit modeling. Compared to the RCP2.6 scenario, the ERF-based approach increases the cumulative CDR use by 28% in the constant and 102% in the constant agricultural emission scenarios by 2100. The GWP100 based increases in cumulative CDR are approximately half as large (16% and 63% in the constant and worst-case scenarios, respectively), and thus fail to fully offset the additional climate forcing. The ERF-based offsetting explored in this study comes, however, with challenges. Firstly, its operationalization within international climate conventions and agreements, such as the UNFCCC and the Paris Agreement, would require a change of current practices relying on the simple heuristic of GWP100-based conversion. Secondly, the initial CDR rates that the approach yields might be too high to be feasible, especially if agricultural emissions follow a worst-case pathway. While currently the maximal estimated capacity for CDR ranges between 0.75–1.5 GtC/yr [79], the ERF approach requires CDR rates over 0.75 GtC/yr already in 2021 and over 1.5 GtC/yr already in 2025 (for the “worst-case” scenario). Such CDR rates are con- tained in the SSP1-2.6 scenario only starting from 2038–2046, leaving CDR technologies an

important time window to develop and scale up. Since the current rates of carbon removal are substantially lower than this maximal estimated capacity, the upscale in CDR in the first years since the introduction of the policy would need to be massive, potentially leading to delays in the early years. On the longer-run, however, the growth rates of CDR converge to those con- tained in the SSP1-2.6 scenario and, despite additional CDR, higher emission scenarios do not significantly overshoot the maximal CDR capacity in 2100 (5–20 GtC/yr). Internalizing CDR costs in the form of a tax on agricultural emissions would have two effects: disincentivize the production and consumption of commodities with a negative effect on the climate, hence decreasing agricultural emissions and the amount of CDR needed for their offset, and create an opportunity to finance early deployment of CDR technologies. We examined the latter of these two effects. Average taxes on agricultural non-CO2 emissions would range between $59-119/tCO2eq, which is in the range of the carbon prices, reported in the IPCC SR15 database [19], needed to limit maximal warming below 2˚C. The scheme, which however does not consider the price elasticity of agricultural products, could maximally finance only a small share of the total CDR in 2100 but it would greatly contribute to the financing of CDR’s early deployment. Ensuring early finance-flows to CDR would create an environment of investing security that would lead to more efforts in R&D, technology

advances, and learning effects reflecting in sinking technology costs [39]. Yet, there is a large gap between the modelled assumptions of a globally harmonized price on agricultural emis- sions and the reality of agricultural climate policies. To date, no single country currently imposes carbon prices on agricultural emissions [15]. Public opinion research, as well as the unresolved decades-long debates in New Zealand, the only country discussing their introduc- tion, suggest that stringent agricultural climate policies will encounter large socio-political resistance [15, 29]. Even if the opposition to agricultural emissions prices were overcome and CDR techniques were to grow fast in the next few years, offsetting additional agricultural emissions comes with costs and threats to society. The increased CDR rates have an average direct financial cost cor- responding to 0.1–0.2% of the global GDP (as projected in the SSP1-2.6) in 2020, ramping up to 0.2–0.5% by 2050 [19]. This cost is likely to be higher if we considered indirect effects on employment and innovation. Internalizing the cost of the offsetting strategy leads to higher PLOS ONE Offsetting future agricultural emissions with CO2 removal PLOS ONE | https://doi.org/10.1371/journal.pone.0247887 March 17, 2021 12 / 19