Incorporating the influence of soil structure and horizons into parameterizations of distributed surface water/groundwater models remains a challenge. Often, only a single soil unit is employed, and soil-hydraulic properties are assigned based on textural classification, without evaluating the potential impact of these simplifications. This study uses a distributed physics-based model to assess the influence of soil horizons and structure on effective parameterization. This paper tests the viability of two established and widely used hydrogeologic methods for simulating runoff and variably saturated flow through layered soils: (1) accounting for vertical heterogeneity by combining hydrostratigraphic units with contrasting hydraulic properties into homogeneous, anisotropic units and (2) use of established pedotransfer functions based on soil texture alone to estimate water retention and conductivity, without accounting for the influence of pedon structures and hysteresis. The viability of this latter method for capturing the seasonal transition from runoff-dominated to evapotranspiration-dominated regimes is also tested here. For cases tested here, event-based simulations using simplified vertical heterogeneity did not capture the state-dependent anisotropy and complex combinations of runoff generation mechanisms resulting from permeability contrasts in layered hillslopes with complex topography. Continuous simulations using pedotransfer functions that do not account for the influence of soil structure and hysteresis generally over-predicted runoff, leading to propagation of substantial water balance errors. Analysis suggests that identifying a dominant hydropedological unit provides the most acceptable simplification of subsurface layering and that modified pedotransfer functions with steeper soil-water retention curves might adequately capture the influence of soil structure and hysteresis on hydrologic response in headwater catchments.