Microbial Nitrogen Cycling Becomes Conservative and Resilient to Long-Term Warming in High-Latitude Carbon-Limited Soils

Abstract

High-latitude soils are warming rapidly due to climate change, raising concerns about long-term impacts on nitrogen (N) and carbon (C) cycling. Here, we investigate how decadal soil warming affects microbial N transformations in subarctic grasslands using natural geothermal gradients with soil temperatures ranging from ambient to +12.3°C. Seasonal measurements of N-pools and gross N transformation rates-including the production and uptake of amino acids, ammonium, and nitrate-were used to characterize microbial responses across warming intensities and time. Warming enhanced microbial turnover of amino acids by accelerating both gross amino acid production and uptake, while net depolymerization remained unchanged. In contrast, ammonium production remained stable, but its microbial uptake increased significantly with temperature. These decoupled responses suggest a microbial shift toward preferential use of organic N sources under warming, likely driven by reduced soil C availability. This strategy provides a dual source of C and N, enabling microbes to sustain high metabolic activity while limiting additional N losses. Supporting this, total soil N stocks declined early in the warming period-by 0.11 tons of nitrogen per hectare per degree Celsius over 5 years-but remained stable thereafter, indicating a transition toward more conservative microbial N cycling. Together, these findings reveal that long-term warming restructures microbial N use strategies, favoring tight organic N recycling and mineral N conservation. These physiological adjustments may buffer N losses under future warming and should be integrated into models predicting high-latitude ecosystem responses to climate change.

Keywords: carbon and nitrogen losses; climate change; high‐latitude ecosystems; plant–soil interactions; soil microorganisms; soil warming.

FIGURE 1

Effect of soil warming intensity and seasonal variations on (a) total dissolved nitrogen, (b) microbial nitrogen, (c) total free amino acids, and (d) ammonium concentrations in soil. p values indicate the effect of warming intensity, season, and their interaction according to linear models on previously transformed variables. Lines represent significant (p < 0.05) effects of warming intensity for each season. Shadowed areas around lines represent the 95% confidence intervals of the regressions. Box plots indicate significant (p < 0.05) effects of season only. Different letters indicate significant differences among seasons according to the post hoc Sidak test for multiple comparisons.

FIGURE 2

Effect of soil warming intensity and seasonal variations on (a) mass‐specific gross amino acid production, (b) mass‐specific gross amino acid uptake, (c) mass‐specific net protein depolymerization, and (d) amino acid turnover time. p values indicate the effect of warming intensity, season, and their interaction according to linear models on previously transformed variables. Lines indicate significant (p < 0.05) effects of warming intensity for each season, while the grey trendline indicates an overall significant (p < 0.05) effect of warming. Shadowed areas around lines represent the 95% confidence intervals of the regressions. Box plots indicate significant (p < 0.05) effects of season only. Different letters indicate significant differences among seasons according to the post hoc Sidak test for multiple comparisons. Data for the snowmelt season is unavailable due to a methodological error that occurred during sample processing.

FIGURE 3

Effect of soil warming intensity and seasonal variations on (a) mass‐specific gross ammonium production, (b) mass‐specific gross ammonium uptake, (c) mass‐specific net N mineralization, and (d) ammonium turnover time. p values indicate the effect of warming intensity, season, and their interaction according to linear models on previously transformed variables. Lines show significant (p < 0.05) effects of warming intensity for each season, and the grey trendline represents an overall significant (p < 0.05) effect of warming. Shadowed areas around lines represent the 95% confidence intervals of the regressions. Box plots indicate significant (p < 0.05) effects of season only. Different letters indicate significant differences among seasons according to the post hoc Sidak test for multiple comparisons.

FIGURE 4

Correlation between (a) mass‐specific gross ammonium production and soil ammonium concentration and (b) mass‐specific gross ammonium uptake and soil ammonium concentration. p values indicate the significance of Spearman's correlation. Grey lines indicate linear correlations, and shaded areas represent 95% confidence intervals for the fitted trend lines.

FIGURE 5

Effect of soil warming intensity on total N stocks measured after 5 years (2013), 10 years (2018) and more than 50 years of soil warming. Lines indicate significant (p < 0.05) effects of warming for each sampling year. Shadowed areas around lines represent the 95% confidence intervals of the regressions.