SOIL NITROGEN TRANSFORMATIONS IN BEECH AND MAPLE STANDS ALONG A NITROGEN DEPOSITION GRADIENT

A gradient of atmospheric nitrogen deposition exists across the northeastern United States due to the concentration of urban and industrial sources of nitrogen oxides in the southern and western parts of the region. We examined possible effects of N deposition on N cycling in forests by measuring potential net mineralization and nitrification of soils under single-species plots of sugar maple and American beech along this gradient. The total atmospheric deposition of nitrogen was estimated to range from 11.1 kg N·ha−1·yr−1 at a site in southern New York to 4.2 kg N·ha−1·yr−1 at a site in western Maine. Although potential net mineralization and nitrification rates were extremely variable, highly significant positive correlations were observed between N deposition and mineralization and nitrification rates in organic horizons in maple plots, but not in beech plots. The correlation between deposition and N cycling variables was weaker in mineral horizons than in organic horizons for maple plots, and no significant correlations between these variables were found for beech mineral horizons. Many beech soils showed no net nitrification even under the higher deposition conditions. Percentage nitrogen (%N) of the organic horizons increased with increasing deposition in sugar maple, but not in beech plots. In organic horizons of both species, mineralization and nitrification increased with increasing %N, although the slopes of the increases were steeper for maple than for beech. Nitrogen deposition, mean annual temperature, and mean annual precipitation were intercorrelated across the sites of this study, but the data indicate that the observed patterns of N mineralization in maple plots resulted from the N deposition gradient rather than the climate gradient. These results suggest that the two species respond differently to N accumulation from atmospheric deposition. Species differences in the responses of forests to N deposition should be considered in both the prediction of forest response and the management of forest composition which could affect that response.