Modelling long-term peatland dynamics. I. Concepts, review, and proposed design

This article reviews the existing conceptual and simulation models for peatlands and develops a model framework for simulating peatland dynamics in boreal regions. Clymoʹs peat bog growth model has been used as the conceptual foundation in published simulation models, but in these models the structural and functional layers of litter, acrotelm and catotelm are not dynamically integrated. Organic-matter decomposition processes in these layers operate at different rates over various time scales. The processes in more active litter and acrotelm layers largely pre-determine the net peat accumulation in the catotelm. On the basis of evaluating litter and peat-core data, we propose a model design that connects different layers through changing water-table depth, driven by effective moisture. The model framework includes single and double negative exponential decay functions that together produce net peat accumulation or degradation and hence carbon balance. The model is designed to be generic in structure, but can be validated using estimated parameters and observed peat accumulation data for continental western Canada. The objectives of this model exercise are (1) to understand the interactions of different biological and environmental factors in boreal peatlands, and (2) to realistically simulate peat accumulation and decay over the last several millennia using proxy paleoclimate data as drivers and peat profiles for validation.