[1] Hyporheic hydrodynamics are a control on stream ecosystems, yet we lack a thorough understanding of catchment controls on these flow paths, including valley constraint and hydraulic gradients in the valley bottom. We performed four whole-stream solute tracer injections under steady state flow conditions at the H. J. Andrews Experimental Forest (Oregon, United States) and collected electrical resistivity (ER) imaging to directly quantify the 2-D spatial extent of hyporheic exchange through seasonal base flow recession. ER images provide spatially distributed information that is unavailable for stream solute transport modeling studies from monitoring wells alone. The lateral and vertical extent of the hyporheic zone was quantified using both ER images and spatial moment analysis. Results oppose the common conceptual model of hyporheic ''compression'' by increased lateral hydraulic gradients toward the stream. We found that the extent of the hyporheic zone increased with decreasing vertical gradients away from the stream, in contrast to expectations from conceptual models. Increasing hyporheic extent was observed with both increasing and decreasing down-valley (i.e., parallel to the valley gradient) and cross-valley (i.e., from the hillslope to the stream, perpendicular to the valley gradient) hydraulic gradients. We conclude that neither cross-valley nor down-valley hydraulic gradients are sufficient predictors of hyporheic exchange flux nor flow path network extent. Increased knowledge of the controls on hyporheic exchange, the temporal dynamics of exchange flow paths, and their the spatial distribution is the first step toward predicting hyporheic exchange at the scale of individual flow paths. Future studies need to more carefully consider interactions between spatiotemporally dynamic hydraulic gradients and subsurface architecture as controls on hyporheic exchange.

[1] Patterns of riparian hydraulic gradients and stream-groundwater exchange in headwater catchments provide the hydrologic context for important ecological processes. Although the controls are relatively well understood, their dynamics during periods of hydrologic change is not. We investigate riparian hydraulic gradients over three different time scales in two steep, forested, headwater catchments in Oregon (WS01 and WS03) to determine the potential controls of reach-scale valley slope and cross-sectional valley geometry. Groundwater and stream stage data collected at high spatial and temporal resolutions over a period encompassing a 1.25 year storm and subsequent seasonal baseflow recession indicate that hydraulic gradients in both riparian aquifers exhibit strong persistence of down-valley dominance. Responses to rainfall do not support the simple conceptual models of increased riparian hydraulic gradient toward streams. Hydraulic gradient response in WS01 to both the seasonal baseflow recession and the storm suggested the potential for increased stream-groundwater exchange, but there was less evidence for this in WS03. Results from four constant-rate tracer injections in each stream showed a high baseline level of exchange overall, and both a slight seasonal increase (WS01) and slight decrease (WS03) in the riparian intrusion of tracer-labeled stream water as stream discharge receded. These results indicate that steep headwater valley floors host extensive stream water exchange and very little change in the water table gradients over 3 orders of magnitude of stream discharge.

[1] Measurements of transient storage in coupled surface-water and groundwater systems are widely made during base flow periods and rarely made during storm flow periods. We completed 24 sets of slug injections in three contiguous study reaches during a 1.25 year return interval storm event (discharge ranging from 21.5 to 434 L s À1 ) in a net gaining headwater stream within a steep, constrained valley. Repeated studies over a 9 day period characterize transient storage and channel water from prestorm conditions through storm discharge recession. Although the valley floor was always gaining from the hillslopes based on hydraulic gradients, we observed exchange of water from the stream to the valley floor throughout the study and flow conditions. Interpretations of transient storage and channel water balance are complicated by dynamic in-stream and near-stream processes. Metrics of transient storage and channel water balance were significantly different (95% confidence level) between the three study reaches and could be identified independently of stream discharge via analysis of normalized breakthrough curves. These differences suggest that the morphology of each study reach was the primary control on solute tracer transport. Unlike discharge, metrics of transient storage and channel water balance did not return to the prestorm values. We conclude that discharge alone is a poor predictor of tracer transport in stream networks during storm events. Finally, we propose a perceptual model for our study site that links hydrologic dynamics in 3-D along the hillslope-riparian-hyporheicstream continuum, including down-valley subsurface transport.

No abstract

No abstract