In oil palm plantations, addition of nitrogen (N) via legume cover crops and fertilisers is a common practice to achieve the yield potential of the crop. It is associated with effects on climate change through emissions of N2O (Choo et al., 2011). As oil palm is the most rapidly expanding tropical perennial crop, and is expected tokeep expanding in the next decades (Corley, 2009), this raises environmental concerns. We reviewed the available measurements for N2O and other N fluxes in oil palm plantations on mineral soils (Pardon et al., under review, a). We saw that direct N2O emissions were the most uncertain N flux, ranging from 0.01-7.3 kgN.ha-1.yr-1, with a tendency to be higher during the immature phase (Ishizuka et al., 2005 ; Banabas, 2007). However, only very few measurements were available on mineral soils, and data is still lacking to better understand the potential effects of spatial heterogeneity in plantations (soil properties, soil cover) and management practices (e.g. fertiliser application timing, splitting, placement). We compared 11 existing models and 27 sub-models to simulate oil palm N budget and losses, among which 7 sub-models were specific to N 2 O emissions (Pardon et al., under review, b). We saw that direct N2O emissions estimates were some of the most variable across models, ranging from 0.8-7 kgN.ha-1.yr-1 (Mosier et al., 1998; Bouwman et al., 2002b ; IPCC 2006, from Eggleston et al., 2006 ; Crutzen et al., 2008 ; Meier et al., 2012 ; APSIM from Huth et al., 2014 ; Shcherbak et al., 2014) . Mineral fertiliser was identified as the main contributor to the emissions, but plant residues and soil N mineralisation were also important. The models accounting for felled palms decomposition, em pty fruit bunches applications, and biological N fixation also estimated a peak of N2O emissions during the immature phase. Therefore direct emissions of N2O in oil palm plantations seemed not to be negligible in terms of environmental effect (2.98 to 2,17