108 

Andrew Hudak, USFS Rocky Mountain Research Station, ahudak@fs.fed.us (Presenter)
Terrie Jain, USDA Forest Service, tjain@fs.fed.us
Jonathan Sandquist, USFS-RMRS, jsandquist@fs.fed.us
Tess Pinkney, USFS-RMRS, tpinkney@fs.fed.us
Robert Liebermann, USFS-RMRS, rjliebermann@fs.fed.us
Michael Battaglia, USDA Forest Service -RMRS, mbattaglia@fs.fed.us
Paula Fornwalt, USFS-RMRS, pfornwalt@fs.fed.us
Russ Parsons, USFS-RMRS, rparsons@fs.fed.us
Russ Graham, USFS-RMRS, rtgraham@fs.fed.us

 

Research has demonstrated the tremendous utility of LiDAR data for mapping forest structure attributes, and many managers have implemented LiDAR-derived products for improved forest management. The sensitivity of LiDAR to canopy structure variation makes LiDAR-derived metrics useful predictor variables for mapping canopy fuel attributes (e.g., canopy bulk density) but much less so for mapping surface fuel attributes (e.g., coarse woody debris). However, imputation-modeling methods allow surface fuel (or other plot-level measurements) to be mapped as ancillary variables by virture of their association with the forest structure and canopy fuel attributes that comprise the response variables in the model. We applied imputation modeling to predict plot-level basal area, canopy bulk density, and coarse woody debris from LiDAR-derived height, intensity, density and topographic metrics and to map these attributes in 0.04 hectare (400 m2) units across Priest River and Deception Creek Experimental Forests in Northern Idaho. We hypothesized that resilient landscapes are heterogeneous within mixed fire regimes. We quantified heterogeneity with the LiDAR at these two Experimental Forests, where we have developed and implemented a series of silvicultural concepts that are designed to create heterogeneity from the site to the landscape. In this study we used LiDAR to quantify variation in the response variables within treatment units and untreated stands and contrasted the effect of different silvicultural and fuel treatments on stand variability. We will present the coefficient of variation in plot-level basal area, canopy fuel, and surface fuel, mapped at the 400 m2 scale and aggregated to the stand level.