Abstract The concentration of carbon dioxide in the atmosphere is influenced by land use and management. Here, the carbon (CO 2–C) budgets of a managed forest (spruce), a grass site and a crop site (crop rotation) have been compared to examine the effects of different management practices on net ecosystem exchange (NEE) and net biome productivity (NBP). This approach enables a more comprehensive carbon budgeting as it takes into account carbon exports and imports for particular land uses. Based on NEE measured with the eddy covariance (EC) method, we quantified NBP by also considering the carbon exports due to thinning or harvest and the carbon imports due to organic fertilisation. The sites are part of the Tharandt cluster in eastern Germany, characterised as having only slight differences in weather and climate. As a result, the management effects on the C budgets can be distinguished from other drivers. For the forest site, continuous CO 2 flux measurements are available from 1997 to today, and the common study period of the grassland and cropland sites is 2005–2008 (one crop rotation). Only when considering the forest's annual NEE was there a variability in the net sink from −698 g C m −2 (1999) to −444 g C m −2 (2003), whereas the grassland and cropland sites exhibited small sinks between −177 g C m −2 (2004) and −62 g C m −2 (2005) and between −115 g C m −2 (2005) and −32 g C m −2 (2007 and 2008), respectively. The forest site is a carbon source with an NBP +221 g C m −2 in 2002 because ∼43 m 3 ha −1 solid wood was removed. The grassland alternated between carbon sources and sinks, with NBP ranging from +25 g C m −2 (2008) to −28 g C m −2 (2006) due to carbon export through several cuts per year. The cropland site was mainly influenced by the cultivated crop species and the application of organic fertiliser (manure), resulting in NBP values between +484 g C m −2 (2007) and −89 g C m −2 (2006). The different timing and length of the growing season for winter and spring crops result in different intra-annual patterns of NEE as well as lower annual net CO 2 sinks for the spring crops. Alternative accounting periods have been defined for all sites starting immediately after the harvest date at the crop site. Based on the full crop rotation period (around 4 years), the NBP showed a C sink for the spruce site and a C source for the crop rotation. The grass site was C neutral according to the calculated NBP.