Nitrogen dynamics and carbon sequestration in cropping systems including legumes

Abstract

Soil carbon (C) and nitrogen (N) dynamics sustain to a large extent the productivity of cropping land to provide food, feed and fibre to satisfy the increasing requirements of the global population. At the plot level C and N dynamics are tightly bound and decoupling results in potential damages for soil, water and air quality. On this basis, decreased soil C and N stocks together with increased nitrate leaching in waters and gaseous N emissions towards the atmosphere, are of major concern. To reduce these environmental impacts, innovative and sustainable farming systems are promoted, such as conservation agriculture and organic farming, which favour soil organic matter restoration practices. Within these systems inclusion of legumes in crop rotations plays a prime role in providing multiple benefits in line with agorecological principles, with N provision through biological N fixation (BNF) topping them all. At a larger scale, the implementation of diversified crop rotations with legumes could play an important role in mitigating and adapting agriculture to climate change, assuming that holistic assessments in the short and long term prove beneficial. The objectives of this thesis were to 1) quantify the N dynamics of six grain legumes across pedoclimatic conditions 2) assess the contributions of legume cover crops to soil organic carbon (SOC) stocks. For this purpose, studies at the experimental farms of Gladbacherhof (Germany), Oberfeld (Germany) and El Dorado do Sul (Brazil) were conducted. The six tested grain legumes at the German sites performed differently across locations, underlying genotypic adaptation strategies of nutrient acquisition and nutrient transformations. At Gladbacherhof the mean rate of biological N fixation was 54.5% much lower than that at Oberfeld where some varieties were almost completely reliant on BNF; nevertheless tested legumes accumulated 99.7 kg ha⁻¹ at the first site and 52.4 kg ha⁻¹ at the second since, due to sound differences in biomass production. Allocation of N to grains was substantial (~5% N concentration) which led to N balances (N inputs from BNF – N exports through grains) between -214.9 to 41.9 kg ha⁻¹, highlighting a fine line between provisions of protein-rich grains at the expense of N rich residues. Nitrate leaching under legumes during the cropping season was low probably because no external fertilizers were applied and losses resulted from the mineralization of SOM. Sowing of winter legumes did not decrease winter nitrate or nitrous oxide losses, but extending the cultivation period, did lead to significantly higher shares of BNF. Weekly measurements of nitrous oxide emissions at Gladbacherhof shed important lights on timing and quantifying legume field scale emissions. I found that mineralization of crop residues can increase N₂O release when coupled to warmer and more moist conditions, therefore sowing of subsequent crops should be done shortly after legume harvest to maximize N transfer and decrease mineralization losses. In the Brazilian study, soil samples were taken at seven increments until the one meter depth from maize-based cropping systems with increasing shares of leguminous cover crops. After 39 years of continuous management, SOC concentrations increased linearly (R²=0.85) with amounts of C inputs suggesting that saturation may have not yet occurred. Significant correlations were also found between amounts of C inputs and thermally labile or thermally recalcitrant SOC stocks. These suggested that diversification of cropping systems with legumes to be a viable means to enhance top soil (0-30 cm) stocks, whereas only systems at 0N fertilizer applications also enhanced subsoil (30-100 cm) stocks. A further fractionation, combining physical and chemical steps, was performed to study the quality and the origin of SOC stocks. Plant inputs (POM) and mineral associated SOC stocks were enhanced, but not aggregation of soil particles suggesting organo-mineral associations to be the driving pathway of SOC sequestration at the site. δ¹³C data helped us confirm that increases in stocks were due to legumes by exploiting the unique isotopic signatures of the C₃ cover crops, and although shares of contributions were high in the top layers, these were low in the subsoil. This study challenges the common belief that legumes necessarily improve plot scale C and N dynamics and points out to the complex interactions occurring between sustainable cropping practices and the compensations which can occur between C and N processes. Some of which leading to increased biological N fixation or soil C stocks, or decreased leaching and gaseous N emissions. Further work should focus on combining the obtained molecular-scale and plot- level results into a conceptual framework of short, and long-term SOM dynamics, with the exploitation of legumes´ biological N and C provision as key contributor for the design of integrated cropping systems and the sustainable management of agricultural soils.