Separating surface storage from hyporheic retention in natural streams using wavelet decomposition of acoustic Doppler current profiles

Abstract

The ability to separate surface storage (e.g., due to in‐channel features such as eddies and pools) from retention due to hyporheic exchange is important in many solute transport considerations; however, current stream tracer approaches do not allow such separation. We examined transient storage processes in a fourth‐order Michigan stream using tracer studies, numerical flow and transport models, and hydrodynamic data obtained from acoustic Doppler current profiler (ADCP) surveys. Since the high‐resolution, three‐dimensional velocity fields obtained from an ADCP relate to in‐channel processes, we used wavelet decomposition to separate the flow into regions of slow and fast moving zones and to estimate the relative sizes of the main channel (A) and the storage zones (AS). Transport modeling based on the tracer data provided estimates of storage zone sizes that included contributions from both surface storage and hyporheic exchange. By coupling the estimates from tracer data with those obtained from an ADCP we were able to assess the relative importance of surface storage in different stream reaches. Estimated (AS/A) values in three test reaches ranged from 0.12 to 0.22, and transient storage residence times varied from approximately 4 min in a run reach dominated by surface storage (reach A) to about 13 min in a reach with some potential for hyporheic exchange (reach C). In reach A the (AS/A) values estimated from tracer and ADCP data were in good agreement, indicating that in‐channel processes were the main mechanism responsible for storage in this reach. Reach C estimates, however, showed that surface storage (AS/A = 0.05) accounted for only a fraction of the transient storage estimated using tracer data (AS/A = 0.12), which indicated that hyporheic exchange contributed to transient storage in the reach. The wavelet decomposition approach based on the ADCP data provides a framework to better constrain transient storage models and to eliminate unrealistic parameters.