Search Results

You are looking at 1 - 2 of 2 items for :

  • Author or Editor: David W. Reynolds x
  • Journal of Atmospheric and Oceanic Technology x
  • Refine by Access: All Content x
Clear All Modify Search
Christophe Maes, David Behringer, Richard W. Reynolds, and Ming Ji


Empirical orthogonal functions of the combined variability of temperature and salinity have been used as basis functions for the indirect reconstruction of salinity from observations of temperature alone. The method employs a weighted least squares procedure that minimizes the misfit between the reconstructed temperature and the observed temperature, but also constrains the variability of the reconstructed salinity to remain within specified bounds.

The method has been tested by fitting to temperature profiles from the Tropical Atmosphere Ocean array along 165°E in the western equatorial Pacific Ocean (8°N–8°S) for the 1986–97 period. Comparisons of the reconstructed salinity field with sea surface salinity and conductivity–temperature–depth data and of the reconstructed dynamic height with TOPEX/Poseidon observations of sea level demonstrate the reliability of the method. The reconstructed data successfully capture the upper-ocean variability at annual to ENSO timescales. The impact of neglecting salinity variability on the dynamic height anomaly in the western tropical Pacific Ocean is addressed.

Full access
David Halpern, Ming Ji, Ants Leetmaa, and Richard W. Reynolds


Equatorial Pacific current and temperature fields were simulated with and without assimilation of subsurface temperature measurements for April 1992–March 1995 and compared with moored buoy and research vessel current measurements. Data assimilation intensified the mean east–west slope of the thermocline along the equator in the eastern Pacific, shifted eastward the longitude of the mean Equatorial Undercurrent (EUC) maximum speed 800 km to 125°W, and produced a 25% stronger mean EUC core speed in the eastern Pacific. In the eastern Pacific the mean EUC core speed simulated with data assimilation was slightly more representative of observations compared to that computed without data assimilated; in the western Pacific the data assimilation had no impact on mean EUC simulations.

Data assimilation intensified the north–south slope of the thermocline south of the equator in the western Pacific to produce a thicker and more intense westward-flowing South Equatorial Current (SEC) in the western Pacific. In the western Pacific the mean SEC transport per unit width simulated with data assimilation was more representative of observations compared to that computed without data assimilation. However, large differences remained between the observed SEC transport per unit width and that simulated with data assimilation. In the eastern Pacific, the data assimilation had no impact on mean SEC simulations.

The temporal variability of monthly mean EUC core speeds and SEC transports per unit width were increased significantly by data assimilation. It also increased the representativeness of monthly mean SEC transports per unit width to the observations. However, the data representativeness of monthly mean EUC core speeds was decreased. Results could be explained by the coupling between zonal gradient of temperature and EUC and between meridional gradient of temperature and SEC. Longitudinal variations along the Pacific equator of the impact of data assimilation on the EUC and SEC precludes the choice of a single site to evaluate the effectiveness of data assimilation schemes.

Full access