Select Page


What Do Data Snapshots Miss? The Case for Time-Series Measurements

By Kara R. Scheu, Ph.D., Consultant
Samuel McWilliams, E.I.T., Consultant
Craig A. Jones, Ph.D., Managing Principal
Linda J. Baker, R.G., L.H.G., Principal

Poster presented at Battelle Sediments Conference, Jan 9-12, 2023, Austin, Texas


The physical dynamics of a contaminated sediment site govern the dispersal and transport of contaminants, the stability of sediments over time, and the efficacy of remedial design.  Physical forcing often varies on a wide range of spatial and temporal scales.  For instance, Portland Harbor, a Superfund site on the Willamette River in Portland, Oregon, is influenced by tidal dynamics (hours), precipitation events (days), seasonal snowmelt (months), and climate variability (years to decades).  Despite this wide range of time scales, the physical forcing of a system is often evaluated using data snapshots in time, such as sediment samples, water samples, bathymetric surveys, and transect measurements.  These snapshot data provide valuable time-specific information and are sometimes extrapolated to characterize temporal dynamics within contaminated sediment sites. While this type of evaluation can provide a useful baseline for understanding broad changes within a system, comparing snapshot data may miss some of the critical dynamic variability within a site due to shorter time scale variability.  This study focuses on the dynamics of Portland Harbor and highlights the benefits of physical time-series measurements when evaluating transport dynamics within a contaminated sediment site.


The Willamette River, particularly downstream of the Multnomah Channel diversion, is a complicated stretch of river that is governed by competing forcing conditions of Willamette River flow, Columbia River water level, and ocean tidal forcing.  In the lower portion of the Willamette River (downstream of the Multnomah Channel diversion), the flows can stagnate and even reverse during periods when the rest of the Willamette River is flowing downstream.  These dynamics vary over annual, seasonal, and tidal time scales.  To evaluate the complex dynamics over a broad range of time scales, a diagnostic hydrodynamic and sediment transport model was developed using Delft 3D, a widely adopted modeling platform used by both industry and academia. The hydrodynamic model relied on time series measurements of water level and water column currents for model calibration and validation.  These measurements were collected over a period of 3 months at two different locations.  The model was first calibrated using the water level measurements and produced an excellent fit with available data.  However, when the model was also calibrated using the measured water column current measurements, the resulting predicted flow down Multnomah Channel, in particular, was significantly altered.  In fact, without the water column current time-series measurements, an otherwise “well-calibrated” model using measured water level misses periods of flow reversal in the downstream portion of the Willamette River when compared with measured water column currents.

Results/Lessons Learned

Time-series measurements provide valuable resolution of short time scale dynamics that can be missed when looking at data snapshots alone.  While the region downstream of the Multnomah Channel is highly depositional, the results of this case study indicate that the Multnomah Channel can, at times, divert more than 100 percent of the flow within the Willamette River, leading to low velocity reversal of surface water currents downstream of the Multnomah Channel diversion.  Our study shows the critical importance of time-series measurements to accurately characterize complex surface water flow dynamics over a wide range of temporal scales, and it highlights the need for both spatial and temporal data when developing critical lines of evidence at a complex contaminated sediment site.

Download Poster

Related Services :

Related Content :