**4. Soil Loss Characterization**

Characterization of soil loss is very important for environment and natural resources. In erosion control planning, soil loss estimates for a particular site are determined using a prediction model and compared with a T-value for that site [31]. The Universal Soil Loss Equation (USLE) is an example of a model used extensively to predict erosion from crop‐ lands and rangelands. More recently, the Agricultural Research Service, Forest Service, and the Bureau of Land Management have joined in a cooperative effort, the Water Erosion Prediction Project (WEPP). WEPP has been implemented to develop an improved model based on modern technology for estimating soil erosion by water. WEPP technology is based on fundamental hydrologic and soil erosion processes and is designed to replace the wide‐ ly used USLE [8].

Until recently, prediction of soil loss rates on National Forest lands involved using the USLE [8, 22]. Soil losses were evaluated in the context of potential soil losses, natural soil losses, current soil losses and tolerable soil losses. Potential losses were those that would occur af‐ ter complete removal of the vegetation and litter. Natural losses were associated with the potential natural vegetation community. Current losses were those occurring with current management. Tolerable loss was assumed to be the rate that can occur while sustaining in‐ herent site productivity [8].

The Universal Soil Loss Equation (USLE) is a widely used method for calculating annual soil losses, based on rainfall, runoff, slope, runoff length, soil type and landuse parameters. The equation originally developed on small agricultural plots, but has been adopted for evaluat‐ ing erosion from large watersheds under a wide range of land uses. [41]

$$A = R \times K \times L \times S \times C \times P \tag{1}$$

where *A* represents the soil loss, commonly expressed in tonnes ha-1 year-1. *R* refers to the rainfall erosivity factor, calculated by the summation of the erosion index EI30 over the peri‐ od of evaluation. EI30 is a compound function of the kinetic energy of a storm and its 30 min maximum intensity. The latter factor is defined as the greatest average rainfall intensity experiences in any 30-min period during a storm. *K* is the soil erodibility factor reflecting the susceptibility of a soil type to erosion. It is expressed as the average soil loss per unit of the R factor. *L* is an index of slope length, *S* is a slope gradient index, *C* is an index for the pro‐ tective coverage of canopy and organic material in direct contact with the ground. It is meas‐ ured as the ratio of soil loss from land cropped under specific conditions to the corresponding loss from tilled land under clean-tilled continuous fallow conditions. Finally, the protective factor P represents the soil conservation operations or other measures that control the erosion, such as contour farming, terraces, and strip cropping. It is expressed as the ratio of soil loss with a specific support practice to the corresponding loss with up-anddown slope culture [41].

Soil loss rates have been generally estimated in agricultural areas up to now. Various USLE and GIS combinations have been used to estimate soil loss in forest land [25]. But in this kind of studies, soil loss was determined by quantitatively. For example; in study realiz‐ ed in Taiwan estimating watershed erosion using GIS coupled with the USLE in agricul‐ tural areas. Furthermore a WinGrid system was developed to calculate slope length factor (L) in USLE [4].

Samar [30] developed three soil loss prediction models (WEPP, EPIC, ANSWERS) and used them for simulating soil loss and testing their capability in predicting soil losses for three tillage systems (rigde-till, chise-plow, and no-till). In other study (leave a space after point), USLE and GIS combination were used to predict long-term soil erosion and sediment trans‐ portation from hillslopes to stream networks under different climate conditions and forest management scenarios. Soil erosion was predicted by the USLE watershed level. The GIS utilities are employed to calculate total mass of sediment moving from each cell to nearest stream network [35]. Mısır et al. [25] developed a soil loss model applicable for forest man‐ agement scenarios for forested areas in northern Turkey.

Forest values including soil protection function need to be determined quantitatively in multi-objective forest management planning. Relationships between soil loss and stand structure on a particular must be determined before incorporation of soil protection values into multi-objective forest management plans.
