Maize carbon dynamics are driven by soil erosion state and plant phenology rather than nitrogen fertilization form.

Abstract Carbon (C) stored in soils represents the largest C pool of terrestrial ecosystems and consequently plays a crucial role in the global C cycle. So far, it is widely unclear to what extent different land uses and land use change influence soil C storage. The hummocky ground moraine landscape of northeastern Germany is characterized by distinct small-scale soil heterogeneity on the one hand, and intensive energy crop cultivation on the other. Both factors are assumed to significantly influence gaseous C exchange; as such, they likely drive soil organic carbon (SOC) stock dynamics in terrestrial agricultural ecosystems. To date, it is not clear to what extent N fertilization forms, which are associated with energy crop cultivation (e.g., application of biogas fermentation residues) and soil type relative to soil erosion state, influence soil C dynamics, nor is it clear whether one of these factors is more important than the other. To investigate the influence of soil erosion state and N fertilization form on soil C dynamics, we present dynamic and seasonal net ecosystem carbon balances (NECB) as a proxy for changes in soil organic carbon stocks. Measurements were conducted for maize ( Zea mays L.) at five sites in the “CarboZALF-D” experimental field during the 2011 growing season. Measurement sites represent different soil erosion states (non-eroded Albic Luvisols, extremely eroded Calcaric Regosols and depositional Endogleyic Colluvic Regosols) and N fertilization forms (100% mineral fertilizer, 50% mineral and 50% organic fertilizer, and 100% organic fertilizer). Fertilization treatments were established on the Albic Luvisol. Net ecosystem CO 2 exchange (NEE) and ecosystem respiration (R eco ) were measured every four weeks using a dynamic flow-through non-steady-state closed manual chamber system. Gap filling was performed based on empirical temperature and PAR dependency functions and was used to derive daily NEE values. In parallel, daily above-ground biomass production (NPP shoot ) was estimated using a logistic growth equation, fitted on periodic biomass samples. Finally, C dynamics were calculated as the balance of daily NEE and NPP shoot based on the initial C input due to organic fertilization. Resulting NECB varied from pronounced soil C losses at the Endogleyic Colluvic Regosol (592 g C m −2 ) to soil C gains at the Calcaric Regosol (−124 g C m −2 ). Minor to modest C losses were observed for the Albic Luvisol. Compared to N fertilization forms, soil erosion states generally had a stronger impact on derived NECB. However, interannual variations in plant phenology and interactions between soil erosion states and fertilization forms might result in different NECB values over multiple years. Hence, long-term measurements of different fertilization treatments on characteristic soil landscape elements are needed.