Vegetable production systems are often characterized by excessive nitrogen (N) fertilization and the incorporation of large amounts of post-harvest crop residues. This makes them particularly prone to ammonia (NH3) and nitrous oxide (N2O) emissions. Yet, urgently needed management strategies that can reduce these harmful emissions are missing, because underlying processes are not fully understood. The present study therefore focuses on the effects of residue placement on NH3 and N2O emissions. For this, cauliflower leaf residues (286 kg N/ha) were either applied as surface mulch (mulch) or mixed with the topsoil (mix) and in situ NH3 and N2O emissions were investigated. The experiment took place on a sandy soil in Northeastern Germany during summer 2012. Residue application created a high peak in N2O emissions during the first 2 weeks, irrespective of residue placement. There was no significant difference in the emission sums over the experimental period (65 days) between the mix (5·8 ± 0·68 kg N2O-N/ha) and the mulch (9·7 ± 1·53 kg N2O-N/ha) treatment. This was also the case for NH3 emissions, which exhibited a lower initial peak followed by a prolonged decline. Measured emission sums were 4·1 ± 0·33 (mix) and 5·1 ± 0·73 (mulch) kg NH3-N/ha. It was concluded that substantial NH3 and N2O emissions can occur after high input of available organic carbon and N even in a coarse-textured soil with low water-holding capacity. Other than expected, surface-application does not enhance NH3 emissions at the expense of N2O emissions compared with residue mixing into the soil, at least under the conditions of the present study.