Denitrification is well known being the most important nitrate-consuming process in water-logged peat soils, whereby the intermediate compound nitrous oxide (N(2)O) and the end product dinitrogen (N(2)) are ultimately released. The present study was aimed at evaluating the release of these gases (due to denitrification) from a nutrient-poor transition bog ecosystem under drained and three differently rewetted conditions at the field scale using a (15)N-tracer approach ([(15)N]nitrate application, 30 kg N ha(-1)) and a common closed-chamber technique. The drained site is characterized by a constant water table (WT) of -30 cm (here referred to as D30), while rewetted sites represent a constant WT of -15 cm, a constant WT of 0 cm (i.e. waterlogged), and an initial WT of 0 cm (which decreased slightly during the experiment), respectively, (here referred to as R15, R0, and R0(d), respectively). The highest N(2)O emissions were observed at D30 (291 µg N(2)O-N m(-2) h(-1)) as well as at R0d (665 µg N(2)O-N m(-2) h(-1)). At the rewetted peat sites with a constant WT (i.e. R15 and R0), considerably lower N2O emissions were observed (maximal 37 µg N(2)O-N m(-2) h(-1)). Concerning N(2) only at the initially water-logged peat site R0d considerable release rates (up to 3110 µg N(2)-N m(-2) h(-1)) were observed, while under drained conditions (D30) no N(2) emission and under rewetted conditions with a constant WT (R15 and R0) significantly lower N(2) release rates (maximal 668 µg N(2)-N m(-2) h(-1)) could be detected. In addition, it has been found that natural WT fluctuations at rewetted peat sites, in particular a rapid drop down of the WT, can induce high emission rates for both N(2)O and N(2).