Conservation agriculture reduces climate change impact of a popcorn and wheat crop rotation

impacts for the three systems intersected after 23–28 years, when they became larger for Cons_CC and Cons_full than for Conv. 4. Discussion 4.1 C inputs and sequestration According to field data for the Nataïs conservation agriculture system, C input from cover crops was 0.8 and 1.7 t/ha in 2019 (fava bean) and 2020 (sorghum followed by fava bean), respectively (Table 3). The 2020 value is close to the mean C input by cover crops (1.9 t/ha) estimated from the meta-analysis of Poeplau and Don [39]. Considering both C inputs by crop residues and cover crops as well as mineralisation of soil C, mean annual C sequestration for the cropping system was 0.6 and 1.2 t/ha for 2017–2019 and 2018–2020, respectively. According to the meta-analysis of Sun et al. [40], the difference in mean annual SOC sequestration of conservation vs. conventional agriculture was 0.35 t C/ha. Conservation agri- culture practices do not include organic fertilisation, so for comparison with our results, the contribution of compost to SOC sequestration (0.46 (2019) and 0.69 (2020) t C/ha) should be subtracted. This yields differences in annual C sequestration of conservation vs. conventional system of 0.37 (i.e. 0.83 − 0.46) and 0.78 (i.e. 1.47 − 0.69) t/ha for 2019 and 2020, respectively. When a single cover crop was present, our results thus agreed with those from the meta-analy- sis, and when two consecutive cover crops were grown, they were twice as high. In the Nataïs conservation agriculture system, compost contributed 34% (2019) and 35%

(2020) to the flow of C to the soil, and was thus a major contributor to the system’s climate change impact (Table 3). Based on composting cost and compost price, we allocated 7.2% of impacts of the composting process to the compost. As composting cost and compost price can vary, a sensitivity analysis was conducted. A 50% decrease or 50% increase in the allocation percentage had a small effect (-5% and + 5%, respectively) on the climate change impact of conservation popcorn (Fig 4). However, when 100% of the impact of the composting process was allocated to the compost, the climate change impact increased by 140%. The French low-C label [41] developed a method to quantify SOC sequestration and reduc- tions in GHG emissions. This method uses the AGRIBALYSE LCI of compost of green waste, which allocates 100% of the impact of the composting process to the compost. The climate change impact of this compost is 694 kg CO2-eq./t. In our study, compost of green waste con- tained 172 kg C/t, 41% of which (i.e. 70.5 kg, corresponding to 259 kg CO2-eq.) was estimated to be sequestered [42]. Consequently, using this method, applying 1 t of compost of green waste increases the climate change impact of a given crop by 435 kg CO2-eq. (i.e. 694–259), not counting emissions associated with applying compost. This example illustrates the impor- tance of allocation choices on potential climate change impacts, and the need to establish a meaningful consensus on this. 4.2 SOC dynamics and climate change impact

4.2.1 SOC dynamics. In their recent review of 106 mainly short-term (2–3 year) experi- ments on the impact of cover crops, Abdalla et al. [43] found mean annual C sequestration of 0.54 t/ha. For our Cons_CC scenario, the short-term (2-year) annual C sequestration was 0.57 t/ha, close to their value. 4.2.2 Climate change impact. No tillage, permanent organic soil cover and crop rotation are the three crop management principles at the heart of conservation agriculture. The envi- ronmental impacts of each of these principles separately have been assessed using LCA or car- bon footprint studies. However, we found no LCA or carbon footprint studies of conservation agriculture systems, implementing these three principles simultaneously. This study is thus a PLOS ONE Conservation agriculture reduces climate change impact of a crop rotation PLOS ONE | https://doi.org/10.1371/journal.pone.0285586 May 23, 2023 12 / 18