Abstract

Abstract. Wildfire is a dominant disturbance agent in forest ecosystems, shaping important biogeochemical processes including net carbon (C) balance. Long-term monitoring and chronosequence studies highlight a resilience of biogeochemical properties to large, stand-replacing, high-severity fire events. In contrast, the consequences of repeated fires or temporal variability in a fire regime (e.g., the characteristic timing or severity of fire) are largely unknown, yet theory suggests that such variability could strongly influence forest C trajectories (i.e., future states or directions) for millennia. Here we combine a 4500-year paleoecological record of fire activity with ecosystem modeling to investigate how fire-regime variability impacts soil C and net ecosystem carbon balance. We found that C trajectories in a paleo-informed scenario differed significantly from an equilibrium scenario (with a constant fire return interval), largely due to variability in the timing and severity of past fires. Paleo-informed scenarios contained multi-century periods of positive and negative net ecosystem C balance, with magnitudes significantly larger than observed under the equilibrium scenario. Further, this variability created legacies in soil C trajectories that lasted for millennia. Our results imply that fire-regime variability is a major driver of C trajectories in stand-replacing fire regimes. Predicting carbon balance in these systems, therefore, will depend strongly on the ability of ecosystem models to represent a realistic range of fire-regime variability over the past several centuries to millennia.

Highlights

  • Wildfire is a pervasive disturbance agent in forest ecosystems, strongly shaping ecosystem structure and function, including vegetation composition, nutrient cycling, and energy flow

  • In the context of wildfire, biogeochemical resilience is determined by pool sizes prior to a fire event, elemental losses and transformations that occur during and shortly after a fire event, and post-fire changes in elemental pools, which in turn are determined by the rate and composition of post-fire revegetation (McLauchlan et al, 2014; Schlesinger et al, 2015; Smithwick, 2011)

  • Through a series of paleo-informed and control modeling scenarios, we address two key questions about the biogeochemical impacts and legacies of wildfire activity: (1) how does centennial-to-millennial-scale variability in fire activity impact biogeochemical processes that regulate slowly varying carbon pool (soil C) and Net ecosystem carbon balance (NECB), and (2) for how long does the legacy wildfire activity impact current biogeochemical states? In addition to testing the general hypothesis that forest carbon storage will differ between equilibrium and paleo-informed simulations, we evaluate the impact of increasing or decreasing fire frequency, relative to that inferred from the paleo record

Read more

Summary

Introduction

Wildfire is a pervasive disturbance agent in forest ecosystems, strongly shaping ecosystem structure and function, including vegetation composition, nutrient cycling, and energy flow. Characterizing biogeochemical resilience emphasizes understanding pool sizes and changes to inputs or outputs of key elements (McLauchlan et al, 2014; Smithwick, 2011). In the context of wildfire, biogeochemical resilience is determined by pool sizes (e.g., carbon, nitrogen) prior to a fire event, elemental losses and transformations that occur during and shortly after a fire event (e.g., from volatilization and erosion), and post-fire changes in elemental pools, which in turn are determined by the rate and composition of post-fire revegetation (McLauchlan et al, 2014; Schlesinger et al, 2015; Smithwick, 2011). Increased disturbance frequency can deplete key growth-limiting nutrients (Yelenik et al, 2013), potentially influencing ecosystem trajec-

Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call