30 years of N fertilisation in a forest ecosystem - The fate of added N and effects on N fluxes

We investigated the fate of added N and its effect on N fluxes in a long-term nitrogen fertilisation experiment. Ammonium nitrate was added annually (30 years) at mean rates of 0 (NO), 35 (N1), 73 (N2) and 108 (N3) kg N ha(-1) yr(-1) to a spruce forest in Sweden, which initially showed signs of N deficiency. Net N mineralisation and N leaching were measured in situ together with soil N pools. We used the PnET-CN model to model the maximum sustainable net N mineralisation rate. The short-term fate of added N was studied by addition of (NH4Cl)-N-15. In N1 and N2 most of the added N (80-120%) was retained in the system, compared to 45% in N3. A major fraction was retained in the organic horizons (58-79%). The internal N fluxes had increased considerably as a result of the N additions, Net N mineralisation in N1 had increased by a factor 10 and litterfall N flux by a factor 4. The PnET-CN model could not mimic the fast changes in tree growth and N mineralisation, but the maximum N mineralisation rate seems realistic. The ratio of actual to maximum mineralisation rate indicates that the N1 treatment now is close to N saturation, and nitrate was occasionally found in soil solution from the B-horizon in N1. The N retained was probably to a great extent immobilised directly by mycorrhizal fungi, as indicated by the high amounts of N-15 found in the L and F layers and by the great fraction of N-15 found in amino sugars compared to amino acids.