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

First, we test whether using the GWP100 yields the same CDR rates as explicit modelling, the latter based on the concept of effective radiative forcing (ERF). Second, we compare the use of two alternative metrics (GWP20 and GWP�) to offset methane. Finally, we explore the interna- lisation of the framework’s costs, such as with a tax on methane and nitrous oxide emissions. Unlike previous studies [22, 23, 28, 58], we do not focus on using the tax to reduce emissions through highly inelastic demand-side effects and changes in production systems, but rather on using the tax as a revenue-raising tool to fund CDR offsetting. As part of this, we assess the change in price, over current retail price benchmark for a United States and global scenario, of three exemplary agricultural products (beef, milk, and rice) in the presence of such offsetting. Methods The RCP2.6 leads to an expected temperature increase by 2100 in the range of 1.5–2˚C [59] and is hence an emission pathway broadly compatible with the Paris Agreement’s target. We investigate the amount of CDR required to achieve radiative forcing and consequently warm- ing rates as in the RCP2.6 for alternative scenarios of agricultural emissions. Our reference sce- nario, the “SSP1-2.6”, achieves radiative forcing as in the RCP2.6 assuming agricultural emissions and CDR rates as in the SSP1-2.6 (“Sustainability–Taking the Green Road” reaching radiative forcing by 2.6Wm-2) [7, 60, 61]. In this scenario, methane emissions from the agricul-

tural sector decrease by 56% by 2100 from their observed values in 2015, while agricultural nitrous oxide increases by 3.5%. The reduction in methane is slightly higher than the average reduction in agricultural and land use methane emissions of the IPCC scenarios compatible with the 1.5˚C goal and limited overshoot (-48±20%). On the opposite, the 1.5˚C scenarios contain on average larger reductions in nitrous oxide than the SSP1-2.6 (-23±40%) [19]. We additionally explored two alternative pathways of agricultural non-CO2 emissions: a “con- stant” scenario, where agricultural methane and nitrous oxide emissions remain stable at their 2015 levels, and a “worst-case” scenario, where they increase as in the baseline scenario of the Shared Socioeconomic Pathway with the least mitigation in the agricultural sector (the SSP3- 7.0) [7]. We chose the SSP3-7.0 scenario to explore the upper boundary of offsetting require- ments within the agricultural sector, since it is the SSP scenario with the highest challenges in both mitigation and adaptation in the land use sector [7]. The details of the different scenario constructions are summarized in Table 1. Fig 1 depicts the methane and nitrous oxide emissions pathways associated with these sce- narios. We use agricultural methane and nitrous oxide, as well as CDR rates, from the harmo- nized CMIP6 emissions data set [62, 63] from year 1765 to 2100 for the SSP1-2.6 (for the “SSP1-2.6” scenario) and from the SSP3-7.0 (for the “worst-case” scenario), as detailed in

Table 1. We hold net emissions of climate-forcing species other than methane, nitrous oxide, and CO2, which is also affected by CDR, at levels envisioned by the RCP2.6 scenario. Table 1. Details of alternative scenarios used in this study. “SSP1- 2.6” “Constant” “Worst-case” Agricultural methane SSP1-2.6 Agricultural emissions constant at 2015 observations as long as they are higher than in SSP1-26 scenario Agricultural emissions as in SSP3-7.0 Agricultural nitrous oxide SSP1-2.6 Agricultural emissions constant at 2015 observations as long as they are higher than in SSP1-26 scenario Agricultural emissions as in SSP3-7.0 Other species (including non- agricultural methane and nitrous oxide) RCP2.6 RCP2.6 RCP2.6 https://doi.org/10.1371/journal.pone.0247887.t001 PLOS ONE Offsetting future agricultural emissions with CO2 removal PLOS ONE | https://doi.org/10.1371/journal.pone.0247887 March 17, 2021 4 / 19