Change in a woodland/degraded woodland system with oaks as the keystone species over 22 years of variable fire management: Is it predictable and unsurprising?

Background/Question/Methods

Oak dominated ecosystems are a unique but imperiled ecosystem type across five continents because of fragmentation, lack of proper management and a changing climate. For example, and for the focus of our research, oak forests and woodlands extend over a large range of the central and eastern U.S. Fire is a primary management tool in many oak woodlands to maintain a relatively open canopy, which allows light into the understory and maintains a “healthy” ecosystem. While prescribed fire effects can be managed with respect to extent, intensity and season, climate change adds a dimension that may or may not be easily understood, predictable or coupled with management.We have examined processes at the individual, population, community, and ecosystem level in three different oak woodlands that vary in fire history and overstory cover in a 2,200 ha area of oak woodland in southwest MO since 1999. “Degraded” woodland areas have not been burned in 80 years and overstory cover is between 85-100%. Restoration with fire in some degraded woodland areas was initiated in 1999 and cover is currently between 70-85%. Woodland areas that were historically considered “savannah” have a cover between 35-50%, with periodic burns for the past 62 years.

Results/Conclusions

We have recorded 26 spring ephemeral species with the highest diversity in open woodlands; only 3 species in open woodlands are shared with the other two woodlands. 22 years of prescribed fire in degraded woodland has not “restored” the spring ephemeral guild. Yearly differences in diversity are significantly different but cannot be predicted by precipitation or temperature at any time scale. Total overstory production and decomposition rates are greater in degraded than burned degraded than open woodlands. Temperature across any time scale accounts for less than 5% of variation between years. Precipitation and phenology (phenocam collected) can predict yearly open woodland production (30-40% of variation), but not production in the other two woodlands (< 8%). Sassafras sapling density in burned degraded woodlands has increased from 0 (1999) to 50 (2014) to 560 (2021) per 0.1 ha; there are no sassafras in the other two woodlands. In comparison, oak sapling density is declining in all three woodlands since 1999. The reason for sassafras sapling dominance and effects on ecosystem processes now must become a focus of our research. For example, in 2021 leaf moisture content was 25% higher and specific leaf area was 52% higher in sassafras than oak leaves, respectively.