Fire is an important natural disturbance, shaping vegetation structure and constraining ecosystem C storage.  However, in many regions fire suppression and other land management choices are leading to fire regimes that are novel to the local ecosystems.  An important challenge in ecology is to understand how ecosystems vary in their resilience to these novel fire regimes.  We used chronosequences, repeated sampling, and tree rings to measure decadal changes in soil and plant C storage following stand-replacing fires in ponderosa pine-bunchgrass ecosystems in Arizona and Aleppo pine-shrub ecosystems of Spain.  Both of these woodlands have experienced a recent increase in the number and size of stand-replacing fires that cause dramatic decreases in ecosystem C storage that lasts at least 3 decades. 

Results/Conclusions

In ponderosa pine forests, a single large fire can decrease C storage; unburned forests contain ~60 MgC/ha in live trees and ~15 MgC/ha in soil O horizons.  Three decades after a single stand-replacing fire, live tree and soil O horizon C storage are less than 5 MgC/ha.  The lack of resilience in C storage is related to poor ponderosa pine regeneration and patchy colonization by grasses.  In contrast, C storage in Spanish woodlands is resilient to a single stand replacing fire, but not to repeated burning.  In this case, unburned forests have ~60 MgC/ha in plants and ~10 MgC/ha in soil O horizons.  Within 3 decades of a single fire, O horizon and shrub C are near pre-fire levels and Aleppo pine C is ~ half of pre-fire levels.  However, if stand-replacing fire return intervals become less than 15 years, Aleppo pine regeneration fails, and plant C storage declines dramatically.  These data illustrate how two very different ecosystems lack resilience to fire regime change, shifting from high C storage woodlands to low C storage grasslands or shrublands.  In both cases, land management, the physical environment (lack of water), and plant life history strategies interact to determine C storage resilience.