Mixing models have long been used in catchment studies to partition streamflow runoff into individual source components and characterize hydrologic flow paths (Pinder and Jones 1969; Sklash and Farvolden 1979; Bazemore et al. 1994; Buttle 1994). These models assume that catchment runoff is a mixture of unique runoff sources whose contributions can be determined using tracers that behave conservatively. The key benefit of using these models is that they describe the integrated catchment response as opposed to point-specific information provided by, for example, groundwater wells. These models are especially valuable for understanding watershed behavior when used in conjunction with hydrometric data (Bonell 1998; Buttle 2005). The use of mixing models in catchment hydrology has evolved over time with a focus on different types of runoff components, changes in computational methods, and a greater recognition of the limitations and benefits of these models.
