Geomorphic controls on hyporheic exchange flow in mountain streams

Abstract

Simulations of stream-subsurface water exchange (hyporheic exchange) using a three-dimensional steady state groundwater flow model and a particle tracking model in unconstrained and constrained reaches of small (2nd-order) and intermediate (5th-order) mountain streams were conducted to estimate the effects of geomorphic features on the extent, volume and residence time of hyporheic exchange flow. Study sites were located in the Lookout Creek drainage in the western Cascades of Oregon. Stream water and water table elevations and saturated hydraulic conductivity were collected from four field sites. Steps were the dominant geomorphic features creating vertical complexity in 2nd-order streams and dominating hyporheic exchange flow. The removal of steps from the simulation models of 2nd-order stream reduced the total volume of hyporheic exchange by 54 %. Furthermore, hyporheic exchange flows resulting from steps had relatively short residence times. Horizontal complexity, resulting from secondary channels and channel splits, had strong effects on hyporheic exchange, as did vertically extensive features, such as riffles, in 5th-order streams. Removal of secondary channels, channel splits and riffles from the simulation models reduced hyporheic exchange flow by 25 %, 30 % and 40 %, respectively. Secondary channels contributed to hyporheic exchange with relatively long residence time, where as channel splits and riffles contributed to hyporheic exchange with short residence time. Thus, multiple features strongly contributed to the creation of hyporheic exchange flow and drove various types of the exchange flows. Also, the simulation results and stream survey showed that the interactions between multiple features, for example secondary channels and riffles, enhanced hyporheic exchange. Because horizontally extensive features were important in driving hyporheic exchange flow in the studied unconstrained reaches, the width of the valley floor strongly controlled hyporheic exchange flow in 5th-order streams. Geomorphic controls on hyporheic exchange flow differed between the two stream sizes sampled. Second-order streams had a single key geomorphic feature (i.e. steps) influencing the hyporheic exchange flow, whereas multiple geomorphic features (i.e. riffles, secondary channels and channel splits) were important in 5 b-order streams. As a result, residence time distribution of hyporheic exchange flow was narrow in 2nd- order stream and wide in 5th-order stream.