Litterfall and nutrient return in five tree species in a common garden experiment

Canopy litterfall is a significant pathway for return of nutrients and carbon (C) to the soil in forest ecosystems. Litterfall was studied in five even-aged stands of Norway spruce, Sitka spruce, Douglas-fir, European beech and common oak at three different locations in Denmark; two sandy sites, Ulborg and Lindet in Jutland, and one loamy site, Frederiksborg on Zealand. Litterfall was collected during three years from 1994 to 1996 in all five species and during six years from 1994 to 1999 in Norway spruce, Sitka spruce and European beech. The average total litterfall was in the range of 3200–3700 kg ha−1 yr−1 and did not differ significantly among tree species. There were no significant differences in total litterfall among sites during the short period, but during the longer period the richer site Frederiksborg had significantly higher total and foliar litterfall amounts compared to the more nutrient-poor sites Lindet and Ulborg. There were close relationships between foliar and total litterfall suggesting that foliar litterfall can be reliably estimated from total litterfall. Beech and oak bud scale litter was significantly related to foliar litterfall. The amount of branch and twig litter was significantly higher in oak than in other tree species. The average foliar litterfall was well related to the annual volume increment. The relationship differed markedly from previously reported relationships based on global litterfall data suggesting that such relationships are better evaluated at the regional level. Nutrient concentrations and fluxes in foliar litterfall were not significantly different among the five tree species. However, there was a significant effect of site on most nutrient concentrations of the three litterfall fractions, and foliar fluxes of P, Ca and Mn were all significantly highest at Frederiksborg and lowest at Ulborg. The similarity in litterfall inputs to the forest floor under these five tree species suggested that previous reports of large variability in forest floor accumulation should primarily be attributed to differences in litter decomposition.