The effect of three fertilisers that delay the bioavailability of nitrogen (N) in the soil was compared with ammonium nitrate and a zero N control in two irrigated turfs in NE Portugal. The fertilisers used were: Floranid permanent 16-7-15 (slow-release, IBDU/Isodur fertiliser); Basacote plus 9M 16-8-12 (controlled-release fertiliser, copolymer ethylene acrylic); Nitroteck 20-8-10 (stabilized fertiliser, dicyandiamide as nitrification inhibitor + coating with polyterpene) and Nitrolusal (ammonium nitrate, 20.5% N), applied all at a rate of 120 kg N ha-1. Nitrolusal was split into two fractions of 60 kg N ha-1. Phosphorus (P) and potassium (K) rates were balanced among treatments by using superphosphate (18% P2O5) and potassium chloride (60% K2O). The turf dry matter (DM) yield and N concentration in dry material were determined from several cuts of biomass throughout the growing season. Based on DM yield, N concentration in dry material and fertilisation costs, indices of N use efficiency and economic efficiency were estimated. Soil nitrate levels were monitored by using anion exchange membranes inserted directly into the soil. Basacote gave significantly lower DM yields than the other fertilised treatments. The apparent N recovery of Basacote was also the lowest. The results showed that Basacote released less N than that required for an adequate plant growth in the beginning of the growing season, hampered the flush of spring growth. Furthermore, the release period of this Basacote formulation, in the environmental conditions of these experiments, seemed to be longer than the length of the growing season. Nitroteck and Floranid yielded similar or even higher DM and apparent N recovery values than did Nitrolusal. The indices of economic efficiency ordered the fertilisers as Nitroteck > Nitrolusal > Floranid > Basacote or Nitrolusal > Nitroteck > Floranid > Basacote, if the costs of P and K fertilisers used to balance the P and K rates in the experimental design were, respectively, taken or not taken into account.