PARENT SESSION
Oral Session - Forests Chair(s): Gustafson, Eric ¹, ¹ Landscape Ecology Unit, Rhinelander, WI
Friday, April 2, 2004 3:00 PM - 5:20 PM Apollo Room 7

Historical range of variability in deadwood biomass: scaling up stand-level biomass models to the landscape. ^*NONAKA, ETSUKO ¹, SPIES, THOMAS A. ², WIMBERLY, MICHAEL C. ³ and OHMANN, JANET L. ², ¹ 321 Richardson Hall, Corvallis, OR, USA² 3200 SW Jefferson Way, Corvallis, OR, USA³ 4-228 Warnell School of Forest Resources, Athens, GA, USA

ABSTRACT- The historical range of variability (HRV) in landscape structure created by natural disturbance has been proposed as a guide for evaluating managed landscapes. Previous studies, however, focused on variability based only on landscape patterns of age classes of live trees and did not address variation in stand structure including dead wood. The objective of this study was to investigate the HRV in live and dead wood biomass in the Oregon Coast Range and to examine variability in disturbance history and forest stand structure. We used a stochastic fire simulation model to simulate landscapes for 1000 yrs prior to Euro-American settlement (circa 1850) and calculated biomass as a function of disturbance history. The HRV was quantified as area of different levels of live and dead wood, and the current condition was compared with the HRV. Under the HRV, the majority of the landscape contained 500-700 Mg/ha of live wood and 50-200 Mg/ha of dead wood. The current dead wood condition is outside HRV. Stands with very low (<50 Mg/ha) dead wood currently cover almost 60% of the region, while historically these stands occurred only over 2.5% of the region. The model suggested that dead wood biomass was highly variable because of variation in disturbance frequency and severity. Approaches to evaluating landscape conditions using HRV need to include both landscape and stand characteristics to better distinguish managed and unmanaged forest landscapes.

KEY WORDS: biomass, historical range of variability, deadwood, regional scale, stochastic simulation modeling