Detecting small-scale spatial heterogeneity and temporal dynamics of soil organic carbon (SOC) stocks: a comparison between automatic chamber-derived C budgets and repeated soil inventories

Carbon (C) sequestration in soils plays a key role in the global C cycle. It is therefore crucial to adequately monitor dynamics in soil organic carbon (deltaSOC) stocks when aiming to reveal underlying processes and potential drivers. However, small-scale spatial (10–30m) and temporal changes in SOC stocks, particularly pronounced in arable lands, are hard to assess. The main reasons for this are limitations of the well-established methods. On the one hand, repeated soil inventories, often used in long-term field trials, reveal spatial patterns and trends in deltaSOC but require a longer observation period and a sufficient number of repetitions. On the other hand, eddy covariance measurements of C fluxes towards a complete C budget of the soil–plant– atmosphere system may help to obtain temporal deltaSOC patterns but lack small-scale spatial resolution. To overcome these limitations, this study presents a reliable method to detect both short-term temporal dynamics as well as small-scale spatial differences of deltaSOC using measurements of the net ecosystem carbon balance (NECB) as a proxy. To estimate the NECB, a combination of automatic chamber (AC) measurements of CO2 exchange and empirically modelled aboveground biomass development (NPPshoot) were used.To verify the method, results were compared with deltaSOC observed by soil resampling. Soil resampling and AC measurements were performed from 2010 to 2014 at a colluvial depression located in the hummocky ground moraine landscape of northeastern Germany. The measurement site is characterized by a variable groundwater level (GWL) and pronounced small-scale spatial heterogeneity regarding SOC and nitrogen (Nt) stocks. Reported data sets contain: (1) weather data (PAR, air temperature, precipitation, groundwater level (GWL)); (2) calculated CO2-fluxes; (3) gap-filled and modelled daily sums of Reco, GPP, NEE, C content in aboveground biomass (NPPshoot) and NECB; (4) data regarding chamber position specific LAI-measurements, parcel biomass sampling campaigns and harvests at chamber position. Tendencies and magnitude of deltaSOC values derived by AC measurements and repeated soil inventories corresponded well. The period of maximum plant growth was identified as being most important for the development of spatial differences in annual deltaSOC. Hence, we were able to confirm that AC-based C budgets are able to reveal small-scale spatial differences and short-term temporal dynamics of deltaSOC.