Seedling response to gaps: separating effects of light and nitrogen

Light and nitrogen (N) availabilities are usually higher in large gaps than in small forest gaps. This covariation obscures how these resources and their interactions affect tree seedling growth and physiological response. In a greenhouse experiment, we altered light and N independently but at levels approximately representative of small and large gaps to separate effects on growth, biomass allocation, leaf characteristics, photosynthesis, and resource-use efficiencies. Betula papyrifera, B. lenta, Pinus strobus, Acer rubrum, Quercus rubra, and Tsuga canadensis seedlings were transplanted and potted in screened field soil, held dormant over the winter, and grown for 3 months in a greenhouse. We used a combination of two light levels (15 and 60% of incident) and two levels of N availability (field soil and N addition equivalent to 150 kg/ha). Foliar N concentration and light-saturated photosynthesis per unit mass (Amax(mass)) increased with N addition and leaf mass per unit area (LMA) increased and foliar N decreased at higher light. Seedling biomass and height growth responded strongly to N addition in B. papyrifera and height growth responded to light in P. strobus. The ratio of root mass to total biomass (RMR) decreased with N application overall, and significantly in three species, but it increased with light only in B. lenta. The increase in LMA at 60% light resulted in generally lower ratio of leaf area to aboveground biomass (LAR), while variation in the ratio of leaf biomass to aboveground biomass (LBR) varied relatively little. Increased N and light availabilities had opposite effects on foliar N and Amax(mass),suggesting that differences in carbon gain between small and large gaps may be obscured in the field. Variation in RMR was more strongly driven by N availability than light, although more extreme shade might reduce it more than we observed. Within a species, photosynthetic N-use efficiency (PNUE) was not higher at low N availabilities, but among species, those with lower N tended to have higher PNUE. Similarly, among species there was a negative correlation between PNUE and delta(13)C, suggesting a trade-off constrains resource use, but only two of the six species considered individually had a significant negative correlation. Our results indicate that nitrogen is as important as light in seedling responses to large gaps, that the differing effects of these resources may affect the interpretation of results from uncontrolled comparisons, and that trade-off patterns of resource-use efficiency appear to be regulated at the species level rather than varying within species. (C) 2004 Elsevier B.V. All rights reserved.