Follow this link to skip to the main content NASA Jet Propulsion Laboratory California Institute of Technology JPL HOME EARTH SOLAR SYSTEM STARS & GALAXIES SCIENCE & TECHNOLOGY BRING THE UNIVERSE TO YOU JPL Email News RSS Podcast Video
JPL Banner
Ocean Surface Topography from Space
SCIENCE
Using Satellite Altimetry to Improve Operational El Niño Forecasts

Image from Cheney science plan



Authors:


Robert Cheney, Laury Miller, C.K. Tai, John Lillibridge, John Kuhn
NOAA Laboratory for Satellite Altimetry, USA

Ming Ji, David Behringer
NOAA National Centers for Environmental Prediction, USA

Femke Vossepoel
LODYC, France




Abstract:

In order to improve the accuracy of NOAA's operational El Niño forecasting
system, we have devoted the past several years to developing methods for
processing and assimilating near real time sea level observations from multiple
satellite altimeter missions into an ocean general circulation model used to
initialize forecast runs. The problem of designing an assimilation scheme is
complicated by the fact that, although sea level variability in the equatorial Pacific is
primarily controlled by temperature, salinity plays an important role, particularly
on interannual time scales in the western half of the ocean. We have developed
techniques using TOPEX/Poseidon (T/P) altimeter data and subsurface temperature
data, primarily from the Tropical Atmosphere Ocean (TAO) array, to estimate the
vertical structure of this salinity variability. Here we present a brief description of
these techniques and the results from some experiments aimed at assessing the
impact of salinity variability on equatorial sea level and, most importantly, on the
accuracy of El Niño forecasting.

Background

Ji et al. (2000) noted that during 1996 in the western equatorial Pacific there
were 5-10 cm errors in sea level in the National Centers for Environmental
Prediction (NCEP) operational ocean analysis. This conclusion was based on
comparisons with both tide gauge and TOPEX/Poseidon (T/P) observations.
Temperature in the model was known to be accurate owing to the abundance of in-
situ thermal data already being assimilated. The model sea level error must
therefore have been the result of salinity variability not accounted for in the analysis
due to the absence of in-situ salinity observations. To help overcome the fact that
direct salinity measurements will remain rare, at least until the Array for Real-time
Geostrophic Oceanography (ARGO) is deployed, considerable thought has been
given to the idea of inferring salinity variability from altimetric measurements of
sea level. The general idea is to use any discrepancy between sea level estimated by a
model analysis and sea level measured by the altimeter to infer a correction to the
vertical salinity profile in the analysis.

Since much of the salinity variability in the western Pacific occurs in the
surface barrier layer and sea surface salinity (SSS) could be readily measured by
ships-of-opportunity, Vossepoel et al. (1999) devised a method for correcting
climatological T-S correlations with SSS and altimetric sea level observation. Maes
and Behringer (2000) approached the problem somewhat differently. Instead of
working with fixed T-S relationships, they generated synthetic salinity profiles from
a set of statistically computed joint empirical orthogonal functions (EOFS) that tie
salinity fluctuations to variations in temperature and sea surface height.

Experiments

To evaluate the impact of the Maes and Behringer technique on the NCEP
ocean analysis used to initialize El Niño forecasts, several experiments were carried
out. In the first, designated as Model, only wind forcing was applied. In the second,
Model(T), only in-situ temperature observations were assimilated in addition to
wind forcing. Finally, in the third experiment, Model (T,S), in-situ temperature and
synthetic salinity profiles derived from temperature and TOPEX/Poseidon sea level
observations were assimilated in addition to wind forcing.
Figure 1 illustrates the effects of these different runs on the vertical salinity
field along the equator in August 1996, at a time when the operational analysis, then
similar to Model(T), indicated dynamic heights 5-10 cm higher than either tide
gauge or T/P observations (Ji et al., 2000). The Model and Model(T,S) salinities are
similar to the monthly Levitus climatology, whereas the Model(T) field is weaker
overall and has a different vertical structure in the western Pacific. These results
suggest that the combined assimilation of temperature and synthetic salinity is
successful in counteracting whatever negative impact the assimilation of
temperature alone has on salinity. Indeed, sea surface height differences across the
Pacific in the Model(T,S) run are lower by as much as 10 cm compared to the
Model(T) run and corresponding changes in the strength of the equatorial
undercurrent are as large as 25 cm/sec.

Although we can demonstrate that the assimilation of temperature and
synthetic salinity improves the accuracy of the ocean analysis, we have not yet
shown significant improvement in the accuracy of the forecast model using this
analysis scheme. We are continuing to refine the background salinity error
covariance and expanding the assimilation system to handle temperature, synthetic
salinity (derived in part from T/P data), and sea level from T/P data directly. In the
near future, this mix of observations, particularly salinity, will be greatly improved
by the arrival of ARGO profile data.

References

Ji, Ming, R.W. Reynolds, and D.W.Behringer, Use of TOPEX/Poseidon Sea Level
Data for Ocean Analyses and ENSO Prediction: Some Early Results, J. Clim., 13, 216-
231, 2000.

Vossepoel, F.C., R.W. Reynolds, and L. Miller, Use of Sea Level Observations to
Estimate Salinity Variability in the Tropical Pacific, J. Atmos. Oceanic Technol., 16,
1401-1415, 1999.

Maes, C., and D.Behringer, Using Satellite-derived Sea Level and Temperature
Profiles for Determining the Salinity Variability: A New Approach, J. Geophys. Res.,
105, C4, 8537-8547, 2000.


Link to USA.gov
Site Manager: Margaret Srinivasan
Webmaster: Kristy Kawasaki
JPL Clearance: CL01-1707