1 Arctic terrestrial ecosystems, which are important components of the global carbon (C) cycle, are expected to undergo considerable future climate change. However, predicting arctic C budgets is complex because the landscape is highly diverse and plant biomass very variable.
2 We investigated the diversity of vegetation at 14 sites along an arctic transect in northern Alaska, including sites on wet coastal tundra, tussock tundra, heath and shrub tundra. We applied these data to a model of the soil–plant–atmosphere (SPA) continuum, to investigate how variations in vegetation structure might affect landscape patterns of gross primary productivity (GPP).
3 Although the dominant species varied, there was a highly significant relationship for vascular plants between the leaf area index (LAI) and total foliar nitrogen (N; g m–2 ground area) at each site.
4 A modelled response surface of GPP to changes in both N and LAI indicated that for tundra plants there is an almost constant ratio between the sensitivity of productivity to the two variables.
5 The conservative nature of canopy LAI–N ratios simplifies the task of generating regional predictions of C assimilation by vascular plants (for which LAI and N are key requirements) because measurements of LAI can be used to infer N.
6 Spatial and temporal patterns of LAI and N must be determined if the photosynthetic component of regional C budgets is to be modelled accurately for diverse arctic ecosystems. Patterns of solar irradiance, together with air temperature and soil moisture constraints, are also influential but less important. The contribution of bryophytes must be accounted for as well.