Modelling Nutrient Limitation During Litter Decomposition
Abstract
As soil carbon (C) cycling models are improved to account for microbial processes and their effect on soil C storage, it becomes increasingly important to quantify when microbial decomposers are nutrient limited and what are the consequences of nutrient limitation on C fluxes. In fact, microbial decomposers often face stoichiometric imbalances, especially when degrading nutrient-poor litter. These imbalances may cause a range of responses, from increased respiratory losses (flexible C-use efficiency), to decreased decomposition rates of C-rich compounds, adjustments of cellular composition, or increased retention of limiting elements. These four resource use modes affect litter C and nitrogen (N) dynamics differently, because they either lead to higher removal of C from the system, eventually reducing C storage in soil, or promote C retention by slowing down decomposition and nutrient mineralization. This raises the questionwhich nutrient limitation modes are most dominant during decomposition? In this contribution, we use a dynamical model of litter decomposition where these four resource use modes are implemented as alternatives. The model is then calibrated with approximately 80 published litter decomposition datasets with contrasting qualities and from different climate zones. The four calibrated model variants perform well when calibrated on individual datasets, whereas the performance after a global calibration is lower. However, with both local and global calibrations, they capture most of the variability in the dynamics of litter C, N, and lignin, independently of the selected resource use mode. We can thus conclude that different modes lead to comparable decomposition trajectories, making the identification of the most dominant mode difficult when using standard litter chemistry data. In conclusion, our results show that all resource use modes can be important, suggesting that more detailed microbial physiological data are needed to assess which N limitation responses are most likely. These results also point to potential equifinality issues with current models including nutrient limitation, because it might not be possible to select optimal model structures using standard decomposition data.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021
- Bibcode:
- 2021AGUFM.B43B..08M