A data assimilation method was used to estimate the variability of three ecologically significant features of the Columbia River estuary and plume: the size of the plume, the orientation of the plume, and the length of the salinity intrusion in the estuary. Our data assimilation method was based on a reduced-dimension Kalman filter that enables fast data assimilation of nonlinear dynamics in the estuary and plume. The assimilated data included measurements of salinity, temperature, and water levels at 13 stations in the estuary and at five moorings in the plume. Our experimental results showed that data assimilation played a significant role in controlling the magnitude and timing of dynamic events in the Columbia River estuary and plume, such as events of extreme salinity intrusion and events of regime transitions in the plume. Data assimilation also changed the response of the salinity intrusion length to variations in the Columbia River discharge, hence imposing a new dynamic on the simulated estuary. The validation of the assimilated solution with independent data showed that these corrections were likely realistic, because the assimilated model was closer to the true ocean than the original, non-assimilated model.
- Columbia River
- Data assimilation
- Reduced-dimension Kalman filter
ASJC Scopus subject areas
- Aquatic Science