Spatial variation and temporal stability of soil water in a snow-dominated, mountain catchment

Soil is a critical intermediary of water flux between precipitation and stream flow. Characterization of soil water content (theta, m3 m-3) may be especially difficult in mountainous, snow-dominated catchments due to highly variable water inputs, topography, soils and vegetation. However, individual sites exhibit similar seasonal dynamics, suggesting that it may be possible to describe spatial variability in terms of temporally stable relationships. Working in a 0·36 km2 headwater catchment, we: (i) described and the spatial variability of theta over a 2 year period, (ii) characterized that variability in terms of temporal stability analysis, and (iii) related changes in temporally stable soil water patterns to stream flow generation. Soil water data were collected for 2 years at representative sites and quantified in terms of theta and water storage to a depth of 75 cm (S75, cm). Both S75 and theta were normally distributed in space on all measurement dates. Spatial variability was high relative to other studies, reflecting catchment heterogeneity. However, the ranking of S75 values displayed temporal stability for all site locations, seasonally and annually. This stability was attributed to soil texture. Further temporal analysis indicated that estimates of catchment mean and standard deviation of S75 may be characterized with relatively few measurements. Finally, we used temporal linear regression to define catchment soil water conditions related to stream-flow generation. Static, high S75 conditions in late winter and early spring indicate that stream-flow response is highly sensitive to inputs, whereas static, low S75 conditions in late summer and early fall indicate minimum stream-flow sensitivity to water inputs. The fall transition was marked by uniform Sd across the catchment. The late spring transition was marked by nonuniform S75 decreases, with the highest S75 sites decreasing most. Threshold S75 values identifying catchment sensitivity to water input were identified.