97th ESA Annual Meeting (August 5 -- 10, 2012)

COS 104-8 - Assessing disturbance severities: Wind and cumulative disturbance data from Eastern Deciduous forests

Wednesday, August 8, 2012: 4:00 PM
D136, Oregon Convention Center
Luke J. Snyder, Department of Plant Biology, University of Georgia, Athens, GA and Chris J. Peterson, Dept. of Plant Biology, University of Georgia, Athens, GA
Background/Question/Methods

Disturbance severity is a focal concern of researchers and forest managers who can use natural severities as guidelines to prescribe management actions. Wind is a dominant source of disturbance within many forested systems, and within this type of disturbance, extent and severity are highly variable, from single tree injury to nearly complete blowdown of 1000’s of hectares. Salvage harvesting timber is a common forest management reaction following wind disturbance, but may increase and change patterns of disturbance severity generated by the wind. Here we use a three dimensional conceptual framework of disturbance severity generated by Roberts1 to compare wind damaged stands both before and after salvage logging. We provide the first direct measurements to apply to this framework and the first measurements of cumulative severity in the context of Roberts’ model. Roberts’ framework of disturbance severity is a composite of three components: % forest canopy removed (canopy openness), % understory vegetation removed, and % soil disrupted or removed (mineral soil exposed)1.  We examined the ability of these metrics to differentiate three disturbance situations: 1. Four patches of 3-12 ha within Daniel Boone National Forest, KY, each with both straight-line wind damage and salvage logging. 2. A single patch which also experienced two disturbance events in short sequence (4 ha stand within Clinch Mountain Wildlife Management Area, VA- tornado and salvage logging). 3. Three large heterogeneous patches in Chattahoochee National Forest, GA generated by the same large (F3) tornado. 

Results/Conclusions

Straight-line wind disturbed patches within a landscape (KY) clustered tightly and differed from the more severe tornado damaged sites (VA & GA). Salvage logging greatly increased severity for all three metrics. Our data also elucidates the high variability of severity within disturbed patches, which has implications for the emulation of natural disturbances in forest management. Roberts1 model effectively separates wind disturbances but needs modifications to include the addition and survival of bent or uprooted trees to the forest floor, which may be important in many forest systems. 

1Roberts, M.R. 2004. Response of the herbaceous layer to natural disturbances in North American forests. Canadian Journal of Botany 82, 1273-1283.