Fifty years of contrasted residue management of an agricultural crop: Impacts on the soil carbon budget and on soil heterotrophic respiration

► Crop residue export, manure input or residue restitution were applied for 50 years. ► Long-term (SOC budgets) and short-term (soil respiration) approaches were combined. ► SOC content stabilized after 20 years in the residue export and manure treatments. ► Residue restitution did not affect SOC content. ► Below-ground and weed inputs contributed largely to SOC budgets. Crop management exerts a strong influence on the soil carbon (C) balance. This study investigated a long-term experiment initiated in 1959 at a site in the Hesbaye region of Belgium and focused on three contrasted treatments: residue export (RE), farmyard manure (FYM) addition and residue restitution (RR) after harvest. The objectives were to quantify the components of the C budget of croplands from a 50-year perspective and to identify the impact of the treatments on this budget and soil C sequestration, given the relatively low levels of residue application. The soil C budget was calculated for each treatment on the basis of total soil organic C (SOC) content measurements and C input data collected since the experiment had begun and drawn from the literature. To evaluate the robustness of this approach, the budget-based output estimates were compared with annual heterotrophic respiration (HR) averages extrapolated from seasonal field HR measurements carried out at the same experimental site in 2010. The soil C budget-based output estimates accorded well with field-based HR measurements and with most HR estimates in the literature, suggesting that, despite the many uncertainties affecting the soil C budget, these results were robust. The three treatments investigated in this study had different impacts on SOC stocks, mainly during the first 20 years of the experiment. RE and FYM caused significant SOC decreases (on average, −7±5gCm−2year−1 over the 50 years) and increases (10±5gCm−2year−1), respectively, whereas RR had no significant impact on the SOC stocks. The study also showed (i) the very large part (about two-thirds of the total input) that represented the below-ground input, weeds and other left-over residues in the C budget, (ii) the important role probably played by residue quality in C sequestration and (iii) the large proportion of C lost annually from the soil (which represents 93–98, 100 and 102–107% of the amounts of fresh residue brought to the soil each year in the FYM, RR and RE treatments, respectively).