Wetter, but not wet enough—Limited greenhouse gas mitigation effects of subsurface irrigation and blocked ditches in an intensively cultivated grassland on fen peat

• We tested effects of water management and grassland renewal on a fen grassland. • Subsurface irrigation (SI) raised water levels, but ditch blocking did not. • SI slightly reduced CO 2 , but has increased emissions of N 2 O and CH 4 . • Grassland renewal has increased N 2 O emissions for almost two years. • Results do not support SI as a measure to reduce GHG from high-intensity grassland. High-intensity grassland farming on peatlands is a profitable land use option in Western and Central Europe. This highly productive land use requires extensive drainage measures and regular grassland renewal. The drainage practice in particular substantially increases peat mineralisation, resulting in high emissions of the greenhouse gases (GHG) carbon dioxide (CO 2 ) and nitrous oxide (N 2 O). Against this, a controlled raising of the water level (WL) by subsurface irrigation (SI) or ditch blocking (DB) has been proposed as a compromise between reducing the GHG emissions and maintaining grassland use on peatlands. We tested this assumption by measuring the full set of GHGs over four years for three water management systems (SI, DB, ditch drainage as control) in combination with three grassland renewal treatments (direct sowing, shallow ploughing, original sward as control) on an intensively used fen grassland in Northwest Germany. The mean annual WL was successfully raised by SI to −0.25 m below the soil surface, while the DB unit remained at a similar level (−0.37 m) as the control (−0.38 m). However, CO 2 emissions were only marginally reduced by SI due to high variability between sites and years. Partially higher CO 2 emissions may have been caused by a higher temperature sensitivity of the heterotrophic respiration at intermediate WLs. Partially lower CO 2 emissions may reflect increased carbon uptake by root growth ( Juncus effuses ) rather than reduced peat mineralisation. The GHG mitigation potential of the SI system remained negligible in this study, as the small CO 2 reduction was offset by increased CH 4 and N 2 O emissions. The average emissions of the DB system were similar to those of the control unit. Both renewal treatments increased N 2 O emissions for approximately two years. Overall, our study results do not support the use of SI as a GHG mitigation measure for intensively used fen grasslands.