HYDROLOGICAL PROCESSES Hydrol. Process. 23, 3631–3638 (2009) Published online 16 September 2009 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/hyp.7440 Impact of road-generated storm runoff on a small catchment response Daniel W. Woldie, 1 * Roy C. Sidle 2 and Takashi Gomi 3 1 Department of Geophysics, Graduate School of Science, Kyoto University, Kyoto, Japan 2 Department of Geology, Environmental Science Program, Appalachian State University, Boone, NC, USA 3 Tokyo Universityof Agriculture and Technology International Environmental and Agricultural Science 3-5-8 Saiwai-cho, Fuchu Tokyo, Japan Abstract: The impact of road-generated runoff on the hydrological response of a zero-order basin was monitored for a sequence of 24 storm events. The study was conducted in a zero-order basin (C1; 0Ð5ha) with an unpaved mountain road; an adjacent unroaded zero-order basin (C2; 0Ð2 ha) with similar topography and lithology was used to evaluate the hydrological behaviour of the affected zero-order basin prior to construction of the road. The impact of the road at the zero-order basin scale was highly dependent on the antecedent soil-moisture conditions, total storm precipitation, and to some extent rainfall intensity. At the beginning of the monitoring period, during dry antecedent conditions, road runoff contributed 50% of the total runoff and 70% of the peak flow from the affected catchment (C1). The response from the unroaded catchment was almost insignificant during dry antecedent conditions. As soil moisture increased, the road exerted less influence on the total runoff from the roaded catchment. For very wet conditions, the influence of road-generated runoff on total outflow from the roaded catchment diminished to only 5Ð4%. Both catchments, roaded and unroaded, produced equivalent amount of outflow during very wet antecedent conditions on a unit area basis. The lag time between the rainfall and runoff peaks observed in the unroaded catchment during the monitoring period ranged from 0 to 4 h depending on the amount of precipitation and antecedent conditions, owing mainly to much slower subsurface flow pathways in the unroaded zero-order basin. In contrast, the lag time in the roaded zero-order basin was virtually nil during all storms. Copyright  2009 John Wiley & Sons, Ltd. KEY WORDS zero-order basin; forest road; road-runoff; peak flow; antecedent moisture Received 7 November 2008; Accepted 17 July 2009 INTRODUCTION Most soil-mantled hillslopes in steep-to-moderately steep terrain of humid-to-semiarid landscapes are finely divided into small unchannelled valleys, which, because they con- centrate on runoff and sediment transport, are the dom- inant sites of saturation overland flow, periodic gullying and debris flow generation (Dietrich et al., 1987; Mont- gomery and Dietrich, 1988). Such unchannelled valleys or geomorphic hollows are typically located in the middle to upper portions of the hillslopes in steep terrain (Sidle, 1984; Tsuboyama et al., 2000). These hollows, alterna- tively called zero-order basins, are often overlooked but important components in catchment hydrology. Runoff from these basins appears to initiate after a threshold of groundwater has accumulated and this threshold appears to be directly related to soil depth (Sidle et al., 2000). The groundwater table that develops above the trough of zero-order basins does not always extend from the base to upslope but may appear simultaneously near the head hollow due to convergent flow (Sidle, 1984; Tsuboyama et al., 2000). Once they become saturated, zero-order * Correspondence to: Daniel W. Woldie, Department of Geophysics, Graduate School of Science, Kyoto University, DPRI, Gokasho 611-0011, Uji, Kyoto, Japan. E-mail: daniel@slope.dpri.kyoto-u.ac.jp This article was published online on 16 September 2009. Two errors were subsequently identified. This notice is included in the online and print versions to indicate that both have been corrected 3 November 2009. basins are potential source areas of runoff (Tsukamoto and Ohta, 1988) and solute transport (Luxmoore et al., 1990). Zero-order basins reflect the evolution of hills- lope processes (Anderson and Burt, 1978; Sidle, 1984; Tsukamoto and Ohta, 1988; Montgomery et al., 1997; Burges et al., 1998). Other studies have also shown that zero-order basins are areas of frequent discharge of weathered materials (Tsukamoto and Minematsu, 1987) and sites of recurrent landsliding and sometimes initia- tion of debris flows depending on failure volumes and channel linkages (Montgomery and Dietrich, 1988; Sidle and Ochiai, 2006). In addition they also act as local traps for a portion of the colluvium transported down hillslopes and tend to accumulate this colluvium for extended peri- ods of time (Reneau and Dietrich, 1991) before releasing it as a landslide or debris flow. Streamflow originates from various sources. Conven- tionally, groundwater held within the bedrock is the main source of baseflow while surface flow and shal- low subsurface flow are the main sources of storm runoff in streams. Both surface and shallow subsurface sources encompass several processes. These processes differ in degrees and combinations from one basin to another, depending on climate, bedrock, relief, soils, veg- etation and land use (Jones, 1997). Subsurface flow path- ways play a dominant role during stormflow generation from soil-mantled landscapes in humid areas (Louder- milk, 1934; Hursh, 1936; Hewlett and Hibbert, 1967; Copyright  2009 John Wiley & Sons, Ltd.