Russell Creek Annual Report 2006 / 07

2 FOREWORD The Russell Creek Research Project is a joint multi-year project between the University of British Columbia and the B.C. Ministry of Forests, with an objective of increasing our understanding of hydrological processes by focusing on subsurface flow, spatial variation of climate data and snow accumulation and ablation dynamics. This process based experimental setup will provide data to justify future physically based modeling of the hydrological response and the subsequent effects on stream flow at different scales, ranging from the hillslope to watershed. The project consists of the three major parts subsurface/preferential flow, spatial variation of climate data and snow dynamics. The spatial variation and dynamics sub-projects have been started in mid 2005. In order to achieve the set objectives the year 2005 and the first part of 2006 have been used for climate station and instrumentation design and setup as well as for testing the instruments. This report consists of 3 parts. The first part is describing the spatial variability analysis of climate data in the watershed. Parts 2a and 2b investigate two aspects of subsurface/preferential flow. Part 3 is dedicated to ongoing project on snow accumulation and ablation dynamics in rain-on-snow dominated watersheds. 3 Part I Spatial variability of climate data Page 4 Part IIa Scaling of subsurface flow velocities through a hillslope with preferential flow pathways Page 44 Part IIb Revealing the secrets of a lateral preferential flow network with a dye staining and excavation. Page 72 Part III Snow accumulation and ablation dynamics in watersheds dominated by rain-on-snow processes: Innovative Measurement Techniques. Spatial variability of climate data 5 INTRODUCTION Spatial variability of climate data Watershed modeling aims at realistic and reliable description and prediction of runoff from catchments. The complexity, i.e. the spatial variability and the non-linear character, of the mechanisms involved in rainfall-runoff modeling cause uncertainties in the modeling process. These uncertainties are widely acknowledged and can be attributed to one of the following groups: (1) model structural uncertainty, (2) model specification imperfect understanding of rainfall-runoff processes (Wagener et al., 2004). It can be reduced by a careful choice of the model that is the most suitable for the existing conditions. Model specification uncertainty on the other hand is caused by the absence of a single best parameter set (Sorooshian et al. Several approaches are used to minimize and quantify the uncertainty including multi-objective and multi-criteria calibration (e.g. dynamic choice of parameter sets (Wagener …