A Rationale for Old Water Discharge Through Macropores in a Steep, Humid Catchment

Abstract

Simultaneous observations of rapid preferential flow through macropores and isotopically "old" water displacement remain unresolved in the Maimai (M8) catchment. Continuous, three-dimensional soil moisture energy conditions were monitored in two discrete catchment positions for a series of storm events in 1987. TenSiometric response was related to the soil water characteristic curve, hillslope throughflow, and total catchment runoff. For events yielding <<2 mm hr -1 peak runoff, near-stream valley bottom groundwater systems discharged water volumes sufficient to account for storm period streamflow. This process was assisted by regular low (<-40 cm H20) matric potential conditions and rapid filling of available soil water storage. For events yielding >2 mm hr-1 peak storm flow, hillslope hollow drainage into steeply sloping first-order channels dominated old water production and most of the catchment storm flow. Highly transient macropore-driven processes of crack infiltration (bypass flow), slope water table development, and lateral pipe flow enabled large volumes of stored water to be delivered to the first-order channel bank at the appropriate time to satisfy catchment storm flow volumes and water isotopic and chemical composition. INTRODUCTIONProgress in understanding processes of storm runoff generation in humid headwater catchments is hampered by discrepancies often found between physical, chemical, and isotopic approaches. Furthermore, the results from chemical and natural isotope separations of stream water into old (pre event) and new (event) water sources often appear to contradict results from hydrometric studies of pathways of water movement on hillslopes.Recent work in a highly responsive catchment at Maimai (M8) on the west coast of New Zealand has yielded conflicting results regarding the importance of new versus old water in through-flow transmission. Hydrometric studies employing dye tracing and subsurface flow measurement [Mosley, 1979[Mosley, , 1982 concluded that rapid flow of new water through macropores was capable of accounting for storm period streamflow. In more recent natural stable isotope and chemical tracing studies, Pearce et al. [1986] and SMash et al. [1986] explicitly refuted this earlier interpretation by indicating that old water dominated throughflow in all storm events monitored. On the basis of previous computer simulations [SMash and Farvolden, 1979] and physical principles [Gillham, 1984], they suggested that a rapid matrix flow displacement mechanism, occurred through (1) saturated wedges on the lower slopes and (2) groundwater ridges in the valley bottoms. This process involved capillary fringe response [Ragan, 1968], which increased local hydraulic gradients and promoted increased gravity drainage of old water to the stream channel.For this study area it now appears that there may be no need for rapid macropore flow transport of new water to explain the subsurface transmission of water downslope, because stored water is discharged into the stream channel. However, the c...

Source areas and mixing of runoff components at the hillslope scale—a multi-technical approach

“…The fact that only minor contributions of direct runoff could be detected demonstrates an extensive mixing of event water with water stored in the soil and drift cover prior to the event (cf. McDonnell, 1990). The runoff generation processes seem to be accelerated by increasing precipitation volumes and intensities; only after exceeding a certain threshold was a significant rise in the discharge volumes detectable at both springs.…”

“…Thus, a three-component separation was calculated using silica and deuterium as tracers. The three components are: (a) direct runoff, with low silica concentrations and the deuterium composition of rainwater (temporal variable, incremental mean according to McDonnell, 1990); (b) shallow groundwater; and (c) deep groundwater. The deuterium and silica concentrations for the deep and shallow groundwater (Fig.…”

Effects of bedrock groundwater on spatial and temporal variations in soil mantle groundwater in a steep granitic headwater catchment

“…Especially since the 1960s, increasing attention has been paid to subsurface flow as a dominant process for runoff generation in humid, soilmantled, and vegetated catchments. Detailed observations have been carried out at various study sites, including the Maimai catchment, New Zealand [Mosley, 1979[Mosley, , 1982Sklash et al, 1986;McDonnell, 1990], LI1 catchment at Llyn Brianne, Wales [Soulsby, 1992], Plastic Lake basin 1-08, Canada [Peters et al, 1995], and Tatsunokuchi-yama Experimental Forest, Japan [Tani, 1997]. The water flow processes in headwater catchments can generally be described as follows: in humid, soil-mantled, and vegetated catchments, most of the rainwater infiltrates into and flows downward through the permeable soil layers, eventually reaching a less permeable layer and forming a transient perched groundwater table.…”

What happens after the catchment caught the storm? Hydrological processes at the small, semi-arid Weatherley catchment, South-Africa

“…(ii) Soil physical and hydrometrical studies, using tensiometers and piezometers, were executed at the plot scale (e.g. McDonnell, 1990). (iii) Sprinkling experiments at hillslopes, often in combination with tracer tests, were conducted (e.g.…”