Stability, Bistability, and Critical Thresholds in Fire-prone Forested Landscapes: How Frequency and Intensity of Disturbance Interact and Influence Forest Cover

Abstract:

Many aspects of disturbance processes can have large impacts on the composition of plant communities, and associated changes in land cover type in turn have biogeochemical feedbacks to climate. In particular, changes to disturbance regimes can potentially change the number and stability of equilibrial states, and plant community states can differ dramatically in their carbon (C) dynamics, energy balance, and hydrology.

Using the Klamath region of northern California as a model system, we present a theoretical analysis of how changes to climate and associated fire dynamics can disrupt high-carbon, long-lived conifer forests and replace them with shrub-chaparral communities that have much lower biomass and are more pyrogenic.

Specifically, we develop a tractable model of plant community dynamics, structured by size class, life-history traits, lottery-type competition, and species-specific responses to disturbance. We assess the stability of different states in terms of disturbance frequency and intensity, and quantitatively partition long-term low-density population growth rates into mechanisms that influence critical transitions from stable to bistable behavior.

Our findings show how different aspects of disturbance act and interact to control competitive outcomes and stable states, hence ecosystem-atmosphere C exchange. Forests tend to dominate in low frequency and intensity regimes, while shrubs dominate at high fire frequency and intensity. In other regimes, the system is bistable, and the fate of the system depends both on initial conditions and random chance. Importantly, the system can cross a critical threshold where hysteresis prevents easy return to the prior forested state. We conclude that changes in disturbance-recovery dynamics driven by projected climate change can shift this system away from forest dominated in the direction of shrub-dominated landscape. This will result in a large net C release from the landscape, and alter biophysical ecosystem-climate interactions.