European croplands under climate change: Carbon input changes required to increase projected soil organic carbon stocks.

Increasing soil organic carbon (SOC) stocks in agricultural systems is a pivotal strategy for promoting soil health and mitigating climate change. Global initiatives have set ambitious targets, aspiring to achieve an annual SOC stock increase of 4 ‰. In the European Union, the recently approved Nature Restoration Law aims to increase SOC stock trends in the top 30 cm of cropland mineral soils. However, current monitoring and reporting practices in some countries rely on simplistic SOC models with default parameters, which may not provide reliable predictions. In this paper, we study the feasibility of a 4 ‰ target in European croplands (i.e., an aspirational target proposed by The international "4 per 1000" Initiative), through estimations of required C input changes. To ensure robust predictions, we propose a novel calibration approach that links model parameters to pedo-climatic variables via statistical relationships from 16 long-term experiments. The effectiveness of the method is evaluated for three SOC models across 4281 sites from the European LUCAS soil survey. Our findings demonstrate that the statistical calibration of the multi-model ensemble improves the accuracy of 2015 and 2018 SOC stock predictions, compared to default parameterization. This improvement was however mainly due to the substantial enhancement of one of the models. According to the weighted multi-model mean, median C input changes to reach a 4 ‰ target for Northern, Central, and Southern Europe stand at 1.85, 1.20, and 0.13 Mg C ha yr under RCP 2.6, and 2.21, 1.26, and -0.10 Mg C ha yr under RCP 6.0, respectively. To achieve the aspirational 4 ‰ target, estimated C input change requirements exceed the predicted changes in net primary productivity under RCP 2.6 and RCP 6.0. This emphasizes the importance of strategic land-use and land-management interventions to enhance SOC stocks.