Applying RUSLE 2.0 on burned-forest lands: an appraisal.

Forest fires can greatly accelerate soil loss. They deprive the forest soil of protection from rainfall impact and runoff over large areas, and they change soil properties in ways that may increase local runoff. Soil loss from a burned forest typically decreases rapidly with time postfire. Depending on climate and postburn condition, 3 to 10 years commonly will suffice to restore protection to the soil and lower soil losses to prefire values. The semiarid southwestern United States in particular is characterized by slow regrowth of vegetation and monsoonal rainstorms. There the risk of high soil-loss rates may continue for several years (Figure 1; McNabb and Swanson, 1990).

The Universal Soil Loss Equation (USLE) is commonly employed by the Burned Area Emergency Rehabilitation (BAER) teams to assess the risk of postfire erosion. Being a time-instantaneous prediction technique, the USLE fails to describe the long-term effect, and stresses the high risk of soil loss immediately following a wildfire. RUSLE 2.0, on the other hand, includes a time-varying option that may model seasonal or pluri-year variations in soil loss. RUSLE 2.0 does not explicitly account for a burned-forest scenario, and the official RUSLE database does not include forest vegetation. Nonetheless, given the long-standing success of the USLE/RUSLE erosion-prediction technique on varied land conditions, it seems reasonable to apply RUSLE 2.0 to burned-forest lands.

The main purpose of this paper is to explore whether RUSLE 2.0 is structurally capable of dealing with soil loss from burned forestlands, without need for significant changes. First, an overview is given of the RUSLE 2.0 functionality and mode of operation, emphasizing aspects that relate to the erosion of burned-forest soils. Data on fire-induced changes to the soil and the biomass, and on how they evolve with time postfire, are compiled and used in building a preliminary vegetation database on which to test RUSLE. An equation is proposed describing restoration of the canopy cover, a needed input to the RUSLE database, and test computations are made. In concluding, certain RUSLE limitations with respect to burned-forest lands are identified, and suggestions are made for improvement.

RUSLE 2.0. RUSLE 2.0 was developed by the U.S. Department of Agriculture's (USDA) Agricultural Research Service (ARS). The program, user's guide, and tutorial can be freely downloaded from the RUSLE website supported by the USDA-Natural Resources Conservation Service (NRCS): ftp://fargo.nserl.purdue.edu/pub/RUSLE2/.

RUSLE factors. USLE and RUSLE have the same mathematical structure: A = R*K*L*S*C*P, in which A is an estimate of the average annual soil loss from a hillslope, and is computed from the product of R, the average annual erosivity factor; K, the soil-erodibility factor; L, the slope-length factor; S, the slope-steepness factor; C, the cover-management factor; and P, the support-practice factor.

The average annual erosivity factor, R, represents the erosive effects of raindrop impact and overland flow. Its computation is based on depth and intensity of discrete rainfall events summed and averaged over many years. R lends USLE/RUSLE its stochastic nature. Problems arise in lands where snowmelt and freeze-thaw processes are common, and in regions characterized by monsoonal storm regimes; in these areas R is computed differently from the original USLE (cf. Renard et al., 1997; Foster et al., 2003).

The soil-erodibility factor, K, represents the combined effect of susceptibility to detachment and transportability of the mineral particles. Mathematically, K is a coefficient relating rainfall erosivity and soil loss, and has been measured systematically on standard unit plots for many soil types (Wischmeier and Smith, 1978; Foster et al., 2003). For loamy soil types K values may be obtained from the soil-erodibility nomograph (Wischmeier and Smith, 1978; Renard et al., 1997). In addition, values for K for soil map units in the United States are contained in the NRCS state soils database (STATSGO) as the kffact parameter.