In the face of rising levels of atmospheric carbon dioxide and the progressive degradation of cultivated soils, cover crops have emerged as a crucial strategy to enhance soil organic carbon sequestration. Cover cropping involves the cultivation of non-economic crops between the main crops, replacing bare fallow during winter. While it is widely acknowledged that cover crops increase soil organic carbon through greater carbon inputs, there is less clarity regarding various species’ total carbon inputs, including rhizodeposition, and their contributions to different soil organic carbon pools with varying residence times (i.e. relative stabilization) in the soil. Hence, the overarching goal of this PhD project was to evaluate cover crop and soil factors relevant to soil organic carbon input and stabilization in a Danish agricultural context.

Study I aimed to quantify the total carbon (C) input, including rhizodeposition, from different cover crop pure stands and mixtures at varying soil N availability. This field trial involved five cover crop treatments and two soil residual N levels, with carbon deposition to soils measured using 14C multipulse labeling of cover crops. Among the three pure stands (winter rye, oilseed radish, and hairy vetch), winter rye exhibited the highest carbon inputs at both soil residual N levels, primarily due to significantly higher root C production. Two legume-based mixtures, combining vetch with either rye or rye and radish, were also evaluated, albeit only at the low soil residual N level. The mixtures displayed the highest total carbon input among all the treatments, with high shoot C similar to vetch and substantial root- and rhizodeposition C similar to rye. As a percentage of total plant C input, the carbon lost via phyllo- and rhizodeposition (%ClvPR) accounted for 7-14%, with no significant differences between treatments. There was also a trend of higher %ClvPR with lower soil N availability, which could be attributed to N-induced root or microbial responses or simply to greater C allocation to shoots when N is more available. However, root C could only explain half of the variation in rhizodeposition under field conditions, suggesting that other factors are important in governing rhizodeposition.

Study II sought to understand how cover crop-derived belowground carbon inputs, including rhizodeposition, translate into distinct soil organic carbon pools, specifically into particulate organic matter (POM) and mineral-associated organic matter (MAOM). This study, conducted in a climate chamber combining a living plant and incubation trial, traced 14C-labeled inputs to various plant and soil C pools. The results revealed significantly higher root C input from the fast-growing winter rye and oilseed radish in the topsoil (0-25 cm), while winter rye and radish showed the highest and lowest root C in the subsoil (25-50 cm), respectively. Root morphology seemed to play a significant role in determining carbon lost via rhizodeposition (%ClvR), with branched root systems exhibiting three times greater %ClvR than thick tap root systems. Non-legumes displayed significantly higher %ClvR in the subsoil, hinting at an N-induced effect. After one year of incubation, SOC pools were fractionated based on size, and the results indicated the highest MAOM formation efficiency for vetch and rye, emphasizing the role of rhizodeposition in MAOM formation. However, despite vetch demonstrating the highest MAOM formation efficiency, it could not compensate for the substantially lower belowground inputs, emphasizing the importance of cover crop productivity in increasing SOC, including MAOM.

Study III explored the association between cover crop species, plant organ and residue quality (chemical composition) and the formation of various soil organic carbon pools, including free POM, occluded POM, MAOM, and C deposited outside the detritusphere. These fractions were mainly separated based on density before and after aggregate disruption. This microcosm incubation study tracked the fate of 14C-labeled shoot and root residues from rye and vetch in medium-fertility and low-fertility soils. The findings highlighted a strong link between higher residue quality (lower C:N, higher content of solubles, lower lignin content) and increased formation efficiency of MAOM, oPOM, and distant deposited C in medium-fertility soil. Plant organ (shoot, root) had a more significant influence on residue-C fate than species identity (rye, vetch), probably due to the higher lignin content in roots. While soil fertility did not alter cover crop-derived respiration patterns, medium-fertility soil showed higher oPOM formation efficiency, possibly due to a greater aggregating capacity.

Collectively, these studies demonstrate that traits conducive to maximizing slower-cycling SOC fractions are associated with high biomass production, C deposited as rhizodeposition, and high-quality residues, and probably in that order. However, a sole focus on SOC sequestration may compromise other critical ecosystem services provided by cover crops, such as prevention of N leaching and nonCO2 greenhouse gas emissions, and N transfer to subsequent crops. Legume-based mixtures may offer a solution in this regard, as mixtures were shown to perform similar or outperform a weighted average of the pure stands on the investigated parameters, and thus mitigating trade-offs among ecosystem functions when selecting cover crops.