Ecological Modelling 180 (2004) 89–102 Fuel load reductions and fire risk in central hardwood forests of the United States: a spatial simulation study Bo Z. Shang a,* , Hong S. He a , Thomas R. Crow b , Stephen R. Shifley c a School of Natural Resources, University of Missouri-Columbia, 203 Anheuser-Busch Natural Resources Building, Columbia, MO 65211-7280, USA b USDA Forest Service, North Central Research Station, 1831 Highway 169 E. Grand Rapids, MN 55744, USA c USDA Forest Service, North Central Research Station, University of Missouri-Columbia, 202 Anheuser-Busch Natural Resources Building, Columbia, MO 65211-7260, USA Received 7 July 2003; received in revised form 5 January 2004; accepted 5 January 2004 Abstract With wildfire suppression, accumulation of fine fuels and coarse fuels increases fire risk in Central Hardwood Forests. Assessment of fire risk provides guides for fuel treatments (prescribed fires, or coarse woody debris reduction) to avoid severe wildfires. A spatially explicit landscape model, LANDIS, was used to simulate fine and coarse fuels and their dynamics in the Missouri Ozarks as influenced by forest succession, species establishment, tree mortality, wildfire disturbance and fuel treatments. Three scenarios were selected and simulated for 200 years: (I) wildfire suppression only; (II) prescribed fire with wildfire suppression; and (III) prescribed fire and coarse fuel reduction with wildfire suppression. LANDIS evaluated the potential fire risk for each forest stand by rating the potential fire intensity and fire probability. About 5% of the stands with the highest potential fire risk were selected each decade for fuel treatment under scenarios II and III. Simulation results showed that fuels and potential fire risk gradually built up to high levels in scenario I. Prescribed fires in scenario II reduced potential wildfire risk at the beginning of the 200-year simulation, but became less effective in limiting wildfires in later years. Additional coarse fuel reduction coupled with prescribed fire in scenario III effectively controlled the coarse fuel loading and potential fire risk. The simulated mean wildfire return interval increased from 325 years in scenario I, to 496 years in scenario II, to 637 years in scenario III. © 2004 Elsevier B.V. All rights reserved. Keywords: Fuel; Fire risk; Fuel treatment; Prescribed fire; Landscape model; LANDIS * Corresponding author. Tel.: +1 573 884 9410; fax: +1 573 882 1992. E-mail address: shangz@missouri.edu (B.Z. Shang). 1. Introduction Since 1940, forests in Missouri and throughout the Central Hardwood Region have experienced ac- tive wildfire suppression (Westin, 1992; Guyette et al., 2002), significantly extending the mean wildfire return 0304-3800/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.ecolmodel.2004.01.020