Ecologists have long been intrigued by the ways co-occurring species divide limiting resources. Such resource partitioning, or niche differentiation, may promote species diversity by reducing competition. Although resource partitioning is an important determinant of species diversity and composition in animal communities, its importance in structuring plant communities has been difficult to resolve. This is due mainly to difficulties in studying how plants compete for below-ground resources. Here we provide evidence from a 15N-tracer field experiment showing that plant species in a nitrogen-limited, arctic tundra community were differentiated in timing, depth and chemical form of nitrogen uptake, and that species dominance was strongly correlated with uptake of the most available soil nitrogen forms. That is, the most productive species used the most abundant nitrogen forms, and less productive species used less abundant forms. To our knowledge, this is the first documentation that the composition of a plant community is related to partitioning of differentially available forms of a single limiting resource.

Abstract. Nitrogen (N) and phosphorus (P) are tightly cycled in most terrestrial ecosystems, with plant uptake more than 10 times higher than the rate of supply from deposition and weathering. This near-total dependence on recycled nutrients and the stoichiometric constraints on resource use by plants and microbes mean that the two cycles have to be synchronized such that the ratio of N:P in plant uptake, litterfall, and net mineralization are nearly the same. Disturbance can disrupt this synchronization if there is a disproportionate loss of one nutrient relative to the other. We model the resynchronization of N and P cycles following harvest of a northern hardwood forest. In our simulations, nutrient loss in the harvest is small relative to postharvest losses. The low N:P ratio of harvest residue results in a preferential release of P and retention of N. The P release is in excess of plant requirements and P is lost from the active ecosystem cycle through secondary mineral formation and leaching early in succession. Because external P inputs are small, the resynchronization of the N and P cycles later in succession is achieved by a commensurate loss of N. Through succession, the ecosystem undergoes alternating periods of N limitation, then P limitation, and eventually co-limitation as the two cycles resynchronize. However, our simulations indicate that the overall rate and extent of recovery is limited by P unless a mechanism exists either to prevent the P loss early in succession (e.g., P sequestration not stoichiometrically constrained by N) or to increase the P supply to the ecosystem later in succession (e.g., biologically enhanced weathering). Our model provides a heuristic perspective from which to assess the resynchronization among tightly cycled nutrients and the effect of that resynchronization on recovery of ecosystems from disturbance.

Summary We are developing a process‐based modelling approach to investigate how carbon (C) storage of tundra across the entire Arctic will respond to projected climate change. To implement the approach, the processes that are least understood, and thus have the most uncertainty, need to be identified and studied. In this paper, we identified a key uncertainty by comparing the responses of C storage in tussock tundra at one site between the simulations of two models – one a global‐scale ecosystem model (Terrestrial Ecosystem Model, TEM) and one a plot‐scale ecosystem model (General Ecosystem Model, GEM). The simulations spanned the historical period (1921–94) and the projected period (1995–2100). In the historical period, the model simulations of net primary production (NPP) differed in their sensitivity to variability in climate. However, the long‐term changes in C storage were similar in both simulations, because the dynamics of heterotrophic respiration (RH) were similar in both models. In contrast, the responses of C storage in the two model simulations diverged during the projected period. In the GEM simulation for this period, increases in RH tracked increases in NPP, whereas in the TEM simulation increases in RH lagged increases in NPP. We were able to make the long‐term C dynamics of the two simulations agree by parameterizing TEM to the fast soil C pools of GEM. We concluded that the differences between the long‐term C dynamics of the two simulations lay in modelling the role of the recalcitrant soil C. These differences, which reflect an incomplete understanding of soil processes, lead to quite different projections of the response of pan‐Arctic C storage to global change. For example, the reference parameterization of TEM resulted in an estimate of cumulative C storage of 2032 g C m−2 for moist tundra north of 50°N, which was substantially higher than the 463 g C m−2 estimated for a parameterization of fast soil C dynamics. This uncertainty in the depiction of the role of recalcitrant soil C in long‐term ecosystem C dynamics resulted from our incomplete understanding of controls over C and N transformations in Arctic soils. Mechanistic studies of these issues are needed to improve our ability to model the response of Arctic ecosystems to global change.

Abstract. We calibrated the Multiple Element Limitation (MEL) model to Alaskan arctic tundra to simulate recovery of thermal erosion features (TEFs) caused by permafrost thaw and mass wasting. TEFs could significantly alter regional carbon (C) and nutrient budgets because permafrost soils contain large stocks of soil organic matter (SOM) and TEFs are expected to become more frequent as the climate warms. We simulated recovery following TEF stabilization and did not address initial, short-term losses of C and nutrients during TEF formation. To capture the variability among and within TEFs, we modeled a range of post-stabilization conditions by varying the initial size of SOM stocks and nutrient supply rates. Simulations indicate that nitrogen (N) losses after the TEF stabilizes are small, but phosphorus (P) losses continue. Vegetation biomass recovered 90% of its undisturbed C, N, and P stocks in 100 years using nutrients mineralized from SOM. Because of low litter inputs but continued decomposition, younger SOM continued to be lost for 10 years after the TEF began to recover, but recovered to about 84% of its undisturbed amount in 100 years. The older recalcitrant SOM in mineral soil continued to be lost throughout the 100-year simulation. Simulations suggest that biomass recovery depended on the amount of SOM remaining after disturbance. Recovery was initially limited by the photosynthetic capacity of vegetation but became co-limited by N and P once a plant canopy developed. Biomass and SOM recovery was enhanced by increasing nutrient supplies, but the magnitude, source, and controls on these supplies are poorly understood. Faster mineralization of nutrients from SOM (e.g., by warming) enhanced vegetation recovery but delayed recovery of SOM. Taken together, these results suggest that although vegetation and surface SOM on TEFs recovered quickly (25 and 100 years, respectively), the recovery of deep, mineral soil SOM took centuries and represented a major ecosystem C loss.

Fire frequency has dramatically increased in the tundra of northern Alaska, USA, which has major implications for the carbon budget of the region and the functioning of these ecosystems, which support important wildlife species. We investigated the postfire succession of plant and soil carbon (C), nitrogen (N), and phosphorus (P) fluxes and stocks along a burn severity gradient in the 2007 Anaktuvuk River fire scar in northern Alaska. Modeling results indicated that the early regrowth of postfire tundra vegetation was limited primarily by its canopy photosynthetic potential, rather than nutrient availability, because of the initially low leaf area and relatively high inorganic N and P concentrations in soil. Our simulations indicated that the postfire recovery of tundra vegetation was sustained predominantly by the uptake of residual inorganic N (i.e., in the remaining ash), and the redistribution of N and P from soil organic matter to vegetation. Although residual nutrients in ash were higher in the severe burn than the moderate burn, the moderate burn recovered faster because of the higher remaining biomass and consequent photosynthetic potential. Residual nutrients in ash allowed both burn sites to recover and exceed the unburned site in both aboveground biomass and production five years after the fire. The investigation of interactions among postfire C, N, and P cycles has contributed to a mechanistic understanding of the response of tundra ecosystems to fire disturbance. Our study provided insight on how the trajectory of recovery of tundra from wildfire is regulated during early succession.