The PMRW hillslope is a characteristically well drained hillslope, remaining unsaturated for most of the water year. During the January to June 2002 period, average hillslope soil moisture shows a strong seasonal decrease starting in mid-to-late May, decreasing from 80% (relative values, see also the soil moisture data description) during winter to 40% during early summer, and values as low as 20% were measured during late summer (Tromp-van Meerveld and McDonnell, 2006a; Figure 3), consistent with the general seasonal wetness conditions of the watershed (Peters et al., 2003). Subsurface flow observed at the excavated trench face is ephemeral and occurs only in response to large rainstorms. Extensive analysis of the hillslope response to 147 rainstorms recorded over a >2 year period during 1996-1998 (Tromp-van Meerveld and McDonnell, 2006b) identified a precipitation threshold of 55 mm, above which all components of subsurface stormflow (total flow, matrix and macropore flow) increased by almost 2 orders of magnitude. Of the 147 rainstorms recorded during this period, only 6% produced significant (>1 mm total flow) subsurface flow at the trench face. Of this 6%, Tromp-van Meerveld and McDonnell (2006b) estimated that “...only a few rainstorms per year are large enough (larger than the threshold) to produce significant subsurface stormflow on the hillslopes...” resulting in delivery of hillslope subsurface stormflow to the downslope riparian area and stream channel. Their analyses also showed the importance of macropore flow from the five individual macropores in delivering subsurface stormflow. Macropore flow accounted for 42% of total trenchflow for 147 rainstorms and total storm macropore flow showed a strong linear relationship with total storm subsurface flow (r2=0.96). Detailed measurements from multiple studies at the PMRW hillslope have provided insight into the internal flowpaths of infiltrated water. For the PMRW hillslope, the spatial variation in trench outflow initially observed by Freer et al. (1997, 2002) and further verified by Tromp-van Meerveld and McDonnell (2006b) was much better correlated to bedrock topographic indices than ground-surface indices. These observations suggest flowpaths within the PMRW hillslope are controlled by subsurface bedrock topography. Further investigation of the spatial development of transient groundwater at the soil-bedrock interface confirmed the hypothesis of Freer et al. (2002) that saturation develops and expands from bedrock hollows and not from a saturated wedge at the base of the hillslope (Tromp-van Meerveld and McDonnell, 2006c). Further, the fill-and-spill mechanism in which lateral flow fills bedrock depressions and then spills downslope, can account for the spatial variability of flow observed at the trench face and the threshold response. From detailed observations of transient groundwater response and lateral extent, Tromp-van Meerveld and McDonnell (2006c) estimate hillslope-scale travel velocities of saturated flow along the soil-bedrock interface of 4 and 8 m hr-1 for two different rainstorms. These high lateral velocities suggest that functional lateral permeability, particularly at depth, may be a result of macropore flow or significant anisotropy. Loss of water to the underlying bedrock is estimated to be considerable at the PMRW hillslope. From sprinkler experiments performed on the lower hillslope, Tromp-van-Meerveld et al. (2007) estimate losses of greater than 20% to bedrock infiltration. However, the approximate two to three orders of magnitude difference in estimated soil and bedrock hydraulic conductivities (see also the site physical description) still promotes the development of transient saturation at the soil-bedrock interface.