Species composition interacts with fertilizer to control long-term change in tundra productivity

Fifteen years of N and P fertilizer addition to an Alaskan moist tundra inc reased aboveground biomass and primary production by 2.5 times. Species com position of the fertilized vegetation also changed dramatically, from a mix of graminoid, evergreen, deciduous, and moss species to strong dominance b y a single, deciduous shrub species, Betula nana. Analysis of these simulta neous changes allows insights into the interactions between changes in reso urce availability and changes in species composition in regulating vegetati on biomass, production, and element use. By the 15th year (1995), both new leaf production and total leaf mass were lower in fertilized than in control plots, although leaf area in fertilized plots was twice that of controls. This occurred because Betula produced th inner leaves than other species, with a high specific leaf area (SLA, leaf area per unit leaf mass). Woody stem mass also increased dramatically in fe rtilized plots, with secondary growth accounting for over half of abovegrou nd net primary production, NPP. The large increase in wood production was m ade possible, in part, by the low cost of production of Betula's thin leave s, allowing greater allocation to secondary growth. Wood also had lower N c oncentrations than leaves, allowing large accumulations of wood at low N co st. Overall, aboveground N concentration in Betula did not change in fertil ized relative to control plots, because its low-N wood mass increased more than its high-N leaf mass (with high SLA). Because Betula was so strongly d ominant an the fertilized plots and was better able to dilute its greater N supply with new growth, community production and biomass in fertilized plo ts were higher, and N concentration was lower, than would have been the cas e if species composition had not changed. Aboveground biomass and leaf area of individual species and functional type s were predicted accurately by regression against the number of hits per po int-frame pin across the full range of data, including both treatments. Cha nges in overall canopy structure and leaf display due to fertilization were thus due mainly to changes in species composition, with no detectable effe ct of treatment on size/structure relationships within species or functiona l types.