Beyond “the limits to peat bog growth”: cross‐scale feedback in peatland development

The postglacial development of peatland systems has had a strong influence on the global carbon cycle. Peatland effects on carbon cycling involve changes in both large‐scale structure and community composition. The rate of C sequestration declines as a peat deposit grows, and methane emission increases as wet microhabitats expand in area. We examine the theoretical basis, underlying assumptions, and empirical evidence for two highly influential models of peatland form and development: Clymo's bog growth model (BGM) and Ingram's groundwater mound hypothesis (GMH). Our detailed theoretical analysis reveals previously unrecognized consequences of the models and, for the BGM, unrealistically stringent conditions for application. Our review of model assumptions and empirical evidence identifies the principal limitations of these models: they ignore spatial heterogeneity in peatland structure and function, fast processes occurring near the peatland surface, and interactions between peat accumulation and hydrological conditions. As a result, these models are unable to predict the effects of bog growth on hydrological conditions or peat‐forming processes. We introduce five conceptual models of peatland development that differ in how bog size and shape change over time, and we examine the consequences of each model for hydrological conditions. This exercise shows how bog height growth and lateral expansion change the boundary conditions constraining peatland dynamics, and so determine the direction of ecosystem development. We link peatland properties to the four general properties of “complex adaptive systems” (CAS): spatial heterogeneity, localized flows, self‐organizing structure and nonlinearity. We also present a framework for modeling peatlands as CAS. In this framework, the system is disaggregated, both vertically and horizontally, into a set of components that interact locally through flows of energy and resources. Both internal dynamics and external forcing drive changes in hydrological conditions and microhabitat pattern, and these autogenic and allogenic changes in peatland structure affect hydrological processes, which, in turn, constrain peatland development and carbon cycling. We conclude by outlining four areas in which further empirical research is urgently needed.