Implications of carbon saturation model structures for simulated nitrogen mineralization dynamics

Abstract

Carbon (C) saturation theory suggests that soils have a limited capacity to stabilize organic C and that this capacity may be regulated by intrinsic soil properties such as clay concentration and mineralogy. While C saturation theory has advanced our ability to predict soil C stabilization, few biogeochemical ecosystem models have incorporated C saturation mechanisms. In biogeochemical models, C and nitrogen (N) cycling are tightly coupled, with C decomposition and respiration driving N mineralization. Thus, changing model structures from non-saturation to C saturation dynamics can change simulated N dynamics. In this study, we used C saturation models from the literature and of our own design to compare how different methods of modeling C saturation affected simulated N mineralization dynamics. Specifically, we tested (i) how modeling C saturation by regulating either the transfer efficiency ([straight epsilon], g C retained g-1 C respired) or transfer rate (k) of C to stabilized pools affected N mineralization dynamics, (ii) how inclusion of an explicit microbial pool through which C and N must pass affected N mineralization dynamics, and (iii) whether using [straight epsilon] to implement C saturation in a model results in soil texture controls on N mineralization that are similar to those currently included in widely used non-saturating C and N models. Models were parameterized so that they rendered the same C balance. We found that when C saturation is modeled using [straight epsilon], the critical C : N ratio for N mineralization from decomposing plant residues (rcr) increases as C saturation of a soil increases. When C saturation is modeled using k, however, rcr is not affected by the C saturation of a soil. Inclusion of an explicit microbial pool in the model structure was necessary to capture short-term N immobilization-mineralization turnover dynamics during decomposition of low N residues. Finally, modeling C saturation by regulating [straight epsilon] led to similar soil texture controls on N mineralization as a widely used non-saturating model, suggesting that C saturation may be a fundamental mechanism that can explain N mineralization patterns across soil texture gradients. These findings indicate that a coupled C and N model that includes saturation can (1) represent short-term N mineralization by including a microbial pool and (2) express the effects of texture on N turnover as an emergent property.

[A & I plus PDF only]