Runoff from a semiarid Ponderosa pine hillslope in New Mexico

Abstract

Abstract. The mechanisms by which runoff is generated in semiarid forests have been little studied. Over the past 4 years we have been investigating runoff processes in semiarid regions by continuously monitoring runoff, both surface and lateral subsurface, from ari 870-m • ponderosa pine hillslope in northern New Mexico. We have found that runoff accounts for between 3 and 11% of the annual water budget. We have also found that lateral subsurface flow is a major mechanism of runoff generation, especially following periods of above-average fall and winter precipitation. In one winter, lateral subsurface flow was equivalent to about 20% of the snowpack (about 50 mm). When antecedent soil moisture was high, lateral subsurface flow was extremely responsive to snowmelt and rainfall events and was much more dynamic than would be suggested by the low (laborat6ry determined) hydraulic conductivity of the soil. The rapidity with which lateral subsurface flow follows these events suggests that macropore flow is occurring. In the case of surface runoff, the major generation mechanisms are intense summer thunderstorms, prolonged frontal storms, and snowmelt over frozen soils. Surface runoff at our site took the form of infiltration-excess overland flow; this type of surface runoff has not been found to dominate at other ponderosa pine sites studied. These detailed and continuous investigations are increasing our understanding of runoff processes in semiarid forests and are thereby laying the groundwork for improved predictions, not only of runoff, but also of the concomitant transport of sediment and contaminants within and from these zones.

Patterns of water movement on a logged Gray Luvisolic hillslope during the snowmelt period

“…The approximately diurnal cycling of liquid moisture content in the Ae horizon for instance suggests that some of the meltwater refroze after infiltrating. The sites that have reported overland flow during the snowmelt [e.g., Wilcox et al (1997) or Nyberg and Fahey (1988)] lacked LFH horizons, and had organic matter contents in surface horizons ranging from 6 to 9% (Knight et al 1985). Frozen infiltration capacities in these montane forest settings may not be comparable to that in upland Boreal soils where LFH horizons are the norm.…”

“…Horton overland flow did occur in forest sites in relatively fine-textured soil in New Mexico (Wilcox et al 1997). In this setting, snowmelt runoff occurred in 3 of 4 yr, both as surface and subsurface flow, and increased with antecedent moisture content as the model of Granger et al (1984) predicted for fine-textured prairie soils.…”

“…On the western Canadian prairies, seasonally frozen soils often limit infiltration (Granger et al 1984) and produce snowmelt runoff as infiltration excess overland flow. The snowmelt period is the most important period for annual runoff in prairie (Hayashi et al 1998), montane (Wilcox et al 1997), and arctic regions (Young and Woo 2000). Kane and Stein (1983) showed that soil frost reduced infiltration capacity of silty textured Alaskan forest soils.…”

“…On the Boreal Shield Ecozone of eastern North America, for instance, Horton overland flow occurred on rock outcrops whereas subsurface flow occurred in areas of shallow soil (Allan and Roulet 1994). Hillslope flow generation in forest settings often includes interflow in locales of shallow soil cover (Renzetti et al 1992) or deep (Gaskin et al 1989;Mulholland et al 1990;Wilson et al 1990;Wilcox et al 1997). The key requirement is a horizon of reduced hydraulic conductivity (Whipkey and Kirkby 1978;Ahuja and Ross 1983).…”

Representation of landscape variability and lateral redistribution processes for large-scale hydrological modelling in semi-arid areas

“…While the outline so far focused on surface runoff, lateral subsurface flow processes may also be relevant although they are usually not considered in semi-arid environments (for an overview and a critique see Beven, 2002). Lateral subsurface flow in the semi-arid is generated for specific conditions, for instance in the presence of soil pipes or other macropores (Torri et al, 1994;Sandström, 1996), during the development of a perched water table in wet periods (Wilcox et al, 1997;Van de Giesen et al, 2000;Chamran et al, 2002) or during saturation of alluvial zones next to the main channel (Ceballos and Schnabel, 1998). …”