The radiation use efficiency (RUE) model is one of the most used tools to generate large spatial and temporal scale net primary productivity (NPP) estimations by remote sensing. It involves two key issues to make accurate estimations of NPP: the estimation of the fraction of photosynthetically active radiation (PAR) intercepted by vegetation (fPAR) and the estimation of the plant RUE. The objectives of this work were to quantify the above-ground RUE under optimal water and nutrient conditions in two C3 and one C4 grass species and to analyse the effect of restrictions in these factors upon RUE by comparing both metabolic pathways. Grasses were cultivated from seeds and four treatments combining contrasting availabilities of water and nutrients were applied. RUE values were calculated from measurements of the incoming PAR, fPAR and productivity. In each of the species, plants with sufficient water and nutrients showed the highest RUE (2.61–3.52 g MJ−1), whereas those with deficiencies in both resources presented the lowest RUE (1.15–2.39 g MJ−1). Cynodon dactylon (C4) was the species with higher value of RUE and no significant differences were detected between treatments. However, no significant differences were detected between C. dactylon and D. glomerata under no stress treatment (N1W1) and between C. dactylon and L. perenne under water stress treatment (N1W0). RUE values of Dactylis glomerata (C3) diminished if only one of the two stress factors was presented, while Lolium perenne (C3) only when both stress factors were present. The decreases under stress treatments were between 35% and 60% compared with the no stress treatment. When regional NPP is estimated it is therefore important to take into account the decrease in the RUE, especially in areas under severe stress.