The effects of soil piping on contributing areas and erosion patterns

Natural piping doubles the dynamic contributing area on the upper Maesnant stream in mid-Wales, mainly through linking points well beyond the riparian zones of seepage to the stream. Both discharge and sediment transport rates in the major pipes are closely related to the size of shallow surface microtopographic hollows in which they lie, and which themselves are largely created by piping erosion. However, pipe dischrges are frequently generated by contributing areas larger than these surface depressions and some pipes run counter to the surface topography. The redistribution and acceleration of hillslope drainage processes by piping has implications for theories of hillslope development, especially through plan-form modifications, and also for channel discharge and erosion.     

Contributions to storm quickflow in a small headwater catchment—the role of natural pipes and soil macropores

Analysis of hydrographs from a 4·3 hectare stream head catchment indicates that storm runoff is generated from dynamic source areas. The volume and timing of contributions from different parts of the catchment show, when compared with the extent of surface saturation, that pipeflow generated from areas not saturated at the soil surface is a significant component of the quickflow hydrograph. A simple model of pipeflow generation and contribution is discussed in the light of field results.     

Modelling pipeflow contributions to stream runoff

Runoff from natural soil pipes has been shown to be a significant contributor to stream discharge in parts of upland Wales. Attempts have been made to model pipeflow contributions using both theoretical and empirical approaches, but little progress has yet been made towards producing generally applicable models of complete pipeflow systems. The paper identifies some of the problems of devising general models with reference to data from the Maesnant Experimental Catchment. The data suggest that in-pipe travel times are more important here than inferred elsewhere, and that neither simple hydraulic models nor kinematic wave theory can adequately explain the patterns of response. In particular, stormflow tends to begin first at the network outfall and even peak discharge often occurs at the outfalls before the headward zones.