Seasonal variation in the temperature sensitivity of soil nitrogen transformations in New England forests

Background/Question/Methods

Increasing temperature may alter nitrogen (N) transformations and N availability in soils, which may have large effects on Net Primary Production in temperate forests, as these forests are generally N limited. However, the temperature sensitivity of soil N transformations has received little study relative to related processes such as soil carbon mineralization. We expected that the temperature sensitivity of soil N transformations would depend on (1) the recalcitrance of soil organic matter (SOM), which may affect substrate quantity and quality, and (2) seasonality, which may be related to substrate quantity, microbial activity, and thermal acclimation of microbial activity. We utilized laboratory incubations of net N mineralization and proteolysis at six temperatures from soils sampled across a seasonal time-course from five species assemblages in Harvard Forest (central MA) and Pisgah State Forest (Southern NH). The species assemblages were chosen to capture variation in litter and SOM recalcitrance; in order of increasing litter and soil C:N values, the species utilized were sugar maple (Acer saccharum), white ash (Fraxinus americana), a mix of red maple and red oak (Acer rubrum and Quercus rubra), American beech (Fagus grandifolia), and Eastern hemlock (Tsuga canadensis).

Results/Conclusions

Net N mineralization (N_min) increased exponentially with incubation temperature across all species and sampling dates, and showed clear effects of species composition (i.e. soil C:N) and seasonality. In comparison, rates of proteolysis were weakly temperature sensitive and more variable. N_min was negatively correlated with soil C:N (i.e. N_min decreased with increasing "recalcitrance") and this pattern was generally maintained at each temperature and sampling point. At a given incubation temperature, N_min declined for all species during the Fall and increased in the Spring. Arrhenius-plot analysis indicated that variations in the temperature sensitivity were largely driven by species and seasonal changes to the intercept parameter; the slope parameter (i.e. the activation energy (E_o) or Q₁₀) was less variable across species and seasons, and a common slope parameter fit the data almost as well as independently fitting slopes for each species (single slope model r² = 0.73, species-specific slope r² = 0.76). A simple mathematical model demonstrated that variation among species could largely be explained by temporal variation in substrate availability. Thus, it may be possible to incorporate variation in the temperature sensitivity of N_min into ecosystem models as a relatively simple function of soil C:N and organic N availability.