The role of valley deposits in modifying runoff delivery through catchments

The management of near-stream areas to reduce sediment and associated nutrient delivery to stream networks involves the management of runoff delivery. The efficiency with which runoff is delivered from hillslopes into the stream network, and subsequently through the stream network is a reflection of a catchment's hydrologic connectivity. The role of valley fill deposits in modifying catchment runoff delivery is examined in this study to assess their significance as hydrologic buffers. Preliminary modeling of controls on near-stream hydrologic buffering suggest that buffering response is sensitive to convergence of the contributing area and to the antecedent moisture conditions of the buffer zone itself. Consistent with partial area models of surface runoff generation, the results imply that near-stream areas will be source areas of mnoff if they accumulate and store runoff. However, near-stream areas below planar and divergent slopes have greater buffering potential because specific catchment areas are lower. The identification of convergent areas as significant source areas of runoff, and hence pollutants, helped guide the selection of a buffer for hydrologic monitoring. An alluvial fan was selected because of its dispersive morphology and position at the mouth of a discontinuous stream. The 0.44 ha study fan-valley flat dispersal area subtends a 26 ha catchment. Hydrologic buffering by the fan was highly variable during the 20 month monitoring period, ranging from complete buffering during some of the smaller runoff events to about a 25% reduction in runoff inputs in one of the larger events. Response variability reflects antecedent moisture conditions on the fan. Under 'wet' conditions, runoff delivery from the catchment is rapid and represents a relatively large proportion of total rainfall. Under dry conditions the entire catchment is a significant store for rainfall, and catchment runoff yields are lower. Runoff losses on the fan are significant because of the greater available storage capacity. Results suggest that the potential for near-stream buffering is good where rain events are sufficiently spaced to permit catchment and near-stream drying. The buffering potential of alluvial deposits in the area is found to reflect their contributing area to dispersal area ratios. A small CA:DA suggests that buffering by the deposit will be significant across a wide range of rainfall events and catchment conditions. Small ratios are associated with interlink areas and low order tributary catchments feeding into high order streams (i.e. where valley floors are wide). As a proportion of total drainage area, the significance of fan buffering along a 5th order stream is found to be relatively minor (16%) compared to runoff contributions from sub-catchments drained by continuous streams (75%). Since runoff from continuous tributaries represents net yields following hydrologic buffering at the sub-catchment scale, however, it is argued that catchment hydrologic connectivity reflects the spatial extent and distribution of alluvial deposits throughout the catchment. A general pattern of increasing valley fill width with increasing stream order is found, which is consistent with the general pattern described in the literature. On average, hillslope-stream hydrologic connectivity decreases with increasing stream order, although there can be significant intra-order variability. The mean CA:DA for 1st order catchments is 70, indicating generally poor buffering potential. Mean CA:DAs for the higher order catchments are significantly lower. Once in the stream network, fans and floodouts below 1st and 2nd order streams are potentially significant buffers. Approximately 40% of 1st order streams discontinue. The area they drain constitutes 65% of the total 1st order stream drainage area and >40% of total drainage area. Fans and floodouts below 66% of 2nd order streams are potential buffers to >40% of total drainage area also. h