Changes in live plant biomass, primary production and species composition along a riverside toposequence in Arctic Alaska, USA

In the arctic landscape, vegetation composition and structure are strongly affected by topographic position and associated variation in microclimate. Along a single riverside toposequence in northern Alaska, six distinct plant communities were studied including a riparian shrub community, a wet sedge tundra, a footslope Equisetum community, a hillslope shrub/lupine community, a hilltop birch-heath community, and a moist tussock tundra. Total live plant biomass varied threefold along the toposequence (450-1400 g m(-2)) while live vascular plant biomass (including belowground stems and rhizomes but not roots) varied sevenfold (160-970 g m(-2)). Aboveground vascular plant production varied less than fourfold (80-265 g m(-2)). Although all six communities showed some signs of nutrient Limitation, measures of soil nutrient availability were highly variable among communities. Contrary to expectations, the most productive community along the toposequence was the hillslope shrub/lupine community, where a late-lying snowbank delayed the start of the growing season by 2 wk. The second most productive community was the hilltop birch-heath, which was exposed to winter winds and where snow melted earliest; most of the production in this community occurred in relatively protected depressions where there were dense accumulations of plant mass. A conclusion is that soil fertility, soil thaw, and protection from wind are more important than length of the snow-free season in regulating productivity along the toposequence. Also contrary to expectations, overall production:live biomass ratios of the six communities varied little despite large differences in growth form composition among communities and in biomass turnover among growth forms. High-biomass, highly productive communities had the lowest production:live biomass ratios, and thus the lowest biomass turnover. Because production and live biomass were linearly correlated over the entire range sampled, it may be possible to use live biomass and/or leaf area as a reasonably accurate predictor of productivity at the landscape or regional level in the Arctic.