Cold climate ecosystems are particularly vulnerable to warming, with ongoing climate change posing high risks to their essential functions. The differential warming responses of ecosystem components such as plants and soil microbes and the resulting shift in their interactions can strongly influence ecosystem functions such as carbon (C) storage, potentially leading to non-linear ecosystem responses. As one aspect of these warming-induced shifts in aboveground-belowground interactions, we hypothesized that the competitive ability of plants and soil microbes to assimilate limiting nutrients such as nitrogen (N) and phosphorus (P) play a key role in determining the warming responses of ecosystem carbon dioxide (CO₂) fluxes. Given that soil microbial biomass generally has lower C:N and C:P ratios than plant biomass, a microbial dominance in soil nutrient uptake would likely lead to increased respiration of excess C from soils, thereby increasing CO₂ fluxes and contributing to C loss from the ecosystem under warming.
To test this hypothesis, we collected intact columns of tundra ecosystems, including plants and soils from the heath tundra in Abisko, Sweden. These columns were subjected to five treatments in climate chambers: an ambient temperature control and four levels of warming with temperature increases up to +9 ℃, over a period of 20 weeks. We measured ecosystem CO₂ fluxes alongside various variables likely influencing these fluxes and the NP uptake by plants and soil microbes, such as soil nutrient concentrations, plant leaf and root functional traits, soil enzymatic activities, and dark respiration of CO₂. Additionally, we measured the NP concentrations of plant and soil microbial biomass, to determine whether their relative nutrient concentrations shifted with warming.
We found that several variables, such as soil dissolved organic carbon concentration, beta-glucosidase activity, plant leaf:root N concentration ratios, and ecosystem CO₂ fluxes showed quadratic non-linear correlations with warming. Other variables, however, showed linear warming responses. While the nutrient concentration ratios of plant and soil microbial biomass were not directly correlated with warming, they were correlated with ecosystem CO₂ fluxes. These results indicate that the warming responses of tundra ecosystems can indeed be non-linear, but the relative nutrient assimilation ability of plants and soil microbes are likely not directly related to these warming responses. Instead, nutrient concentration ratios of plants and soil microbes may serve as a useful control variable for explaining part of the context-dependency in ecosystem CO₂ flux responses to warming.