FY 1988

Model-data comparisons for the 1982–83 El Niño: The XBT tracks

Harrison, D.E., W.S. Kessler, and B.S. Giese

NOAA Tech. Memo. ERL PMEL-79, NTIS: PB88-221668, 61 pp (1988)

Five different analyses of 1982-83 monthly average surface wind stress fields have been used to force an ocean general circulation model of the tropical Pacific, in a series of El Niño hindcast experiments like that reported by Philander and Seigel (1985). The results of these hindcasts are compared here with the variability of upper ocean dynamic height, sea surface temperature and subsurface temperature as observed from XBT data obtained along the three main ship of opportunity tracks. During 1982-83 there were prominent departures from climatology in the surface wind stress field. However, there are so few observations of surface wind available that efforts to produce fields of wind stress must fill in considerable gaps in data coverage, with the result that the various tropical surface analyses tend to differ considerably; the differences between analyses, along the ship tracks, are described. Generally the operational analyses produced by the meteorological centers (National Meteorological Center (NMC), European Center for Medium-range Weather Forecasting (ECMWF), and Fleet Numerical Ocean Central (FNOC)) had weaker surface stress and weaker spatial gradients than the special research products (Florida State University (FSU) and Sadler (SADLER)). The ocean underwent several major changes during this period according to the XBT data; these changes are summarized. The model hindcasts are examined to determine the extent to which the observed major ocean changes were reproduced. Within the equatorial waveguide, dynamic height changes are hindcast with quantitative skill using each wind stress field; the best hindcasts differ from the observations by only a few dyn cm more than the estimated uncertainty in the observations. The large waveguide dynamic height hindcast skill found here indicates that the major elements of the 1982-83 El Niño are contained in the 1982-83 surface wind stress field, rather than in any particular aspect of the state of the ocean during late 1981. Sea surface temperature changes are generally hindcast with some qualitative skill; the correlation between hindcast and observed SST is usually significantly positive, but the RMS difference between any hindcast and the observations is generally greater than the RMS signal in the observations. Subsurface temperature variability is hindcast with differing levels of skill, depending upon stress field, region and depth. The vertical temperature gradients and mixed layer temperatures, as well as the depth of the thermocline, undergo substantial changes (especially in the eastern Pacific); primitive equation physics appear necessary to model these observations. Outside the waveguide, hindcast skill is generally much reduced; although qualitatively correct behavior is often hindcast, amplitudes can be seriously in error. The most striking inconsistency found involves the NMC hindcast in the region of the North Equatorial Counter Current. The special research products generally give more accurate hindcasts of dynamic height, but the operational fields often produce better SST hindcasts. A clear deficiency of the operational fields is the character of their wind stress curl fields, compared either with climatology or the special research analyses; because Ekman pumping is a major factor in thermocline adjustment outside the waveguide, wind stress curl must be correctly represented if adequate hindcasts are to be obtained. Although improved parameterization of upper ocean mixing, and better knowledge of the surface heat flux appear needed to improve SST hindcast skill, these results estabish that a most serious need for improved hindcast performance is better knowledge of the surface wind stress field.