Premium
A model of rapid preferential hillslope runoff contributions to peak flow generation in a temperate rain forest watershed
Author(s) -
Beckers Jos,
Alila Younes
Publication year - 2004
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/2003wr002582
Subject(s) - surface runoff , hydrology (agriculture) , environmental science , storm , watershed , streamflow , ecohydrology , return period , infiltration (hvac) , precipitation , interflow , groundwater , geology , drainage basin , flood myth , meteorology , ecology , geography , ecosystem , geotechnical engineering , cartography , machine learning , computer science , biology , oceanography , archaeology
A model for the 10 km 2 Carnation Creek watershed on Vancouver Island, British Columbia, is used to assess preferential hillslope runoff contributions to peak flow generation. The model combines the matrix flow algorithm of the distributed hydrology soil vegetation model with a Green‐Ampt formulation for calculating matrix and by‐pass infiltration, preferential hillslope runoff initiation controlled by rainfall depth, and downslope subsurface flow rates prescribed based on at‐site tracer tests. Model evaluation using 1972–1990 hydrometeorological data reveals that this formulation is successful in simulating subannual and larger peak flows. Model results suggest that preferential flow contributions to streamflow generation become greater than matrix flow contributions for unit area discharge values in excess of 2.8 mm/hr, corresponding to a peak flow return period of 2–3 months. This transition from matrix flow dominated runoff to preferential flow dominated runoff is consistent with an observed upper limit of groundwater response to precipitation for return periods in excess of 2 months. A break in slope in peak flow frequency curves at a return period of about 20 months appears to correspond to a change in storm characteristics. Thus at least three physically distinct populations of peak flows may exist at Carnation Creek. The ability of the model to simulate peak flows and groundwater responses for small and large storms suggests that it may be useful for addressing runoff process considerations in the debate whether forest management effects for annual and larger peak flows are similar to those inferred from analyses dominated by subannual peak flows.