Equatorial Waves in High-Resolution Satellite Observations and Reanalysis Products

Equatorial waves drive global-scale circulation changes that are relevant to climate. A spectrum of equatorial waves drives the quasibiennial oscillation in tropical lower stratospheric zonal winds. Equatorial waves are also partly responsible for driving the upwelling motion across the tropical tropopause layer, which is the input to the global stratospheric transport circulation known as the Brewer-Dobson Circulation. Stratospheric water vapor, an important climate variable, is transported from the troposphere through the tropical tropopause layer. Other trace gases important to climate and stratospheric chemistry are also transported to the stratosphere through this layer. Fundamental to describing the role that equatorial waves play in these processes is the spectrum of Eliassen-Palm flux and the divergence of this flux. Reanalysis products have unique difficulties in representing these equatorial waves. Most assimilated measurements provide only temperature, yet geostrophic wind balance does not apply near the equator. Wind data come from only a relatively few radiosonde sites. Near the tropopause and lower stratosphere, equatorial waves are observed with very short vertical wavelengths, yet much of the temperature information comes from near-nadir sounders that do not have sufficient vertical resolution. In fact the vertical wavelengths of the observed waves are a challenge to resolve even at the vertical resolution of modern reanalysis systems. Temperature measurements from the HIgh Resolution Dynamics Limb Sounder (HIRDLS) instrument on the Aura satellite have exceptionally high vertical resolution and high spatial sampling. The vertical resolution, at approximately 1 km, is comparable to the resolution for temperatures derived from GPS radio occultation data, while the number of samples in the tropics is more than a factor of three higher for HIRDLS than the current GPS coverage. HIRDLS data span only a three year period 2005-2008, and they are not assimilated in reanalyses. We exploit the high resolution and spatial coverage of HIRDLS to evaluate the representation of tropical waves in reanalysis products including ERA-Interim and MERRA. We have developed a method for computing the Eliassen-Palm (EP) flux spectrum for equatorial waves directly from HIRDLS measurements without auxiliary wind data. HIRDLS EP flux spectra and their seasonal variation will be compared with the reanalyses. Our results test the accuracy of reanalysis data for studies of upper tropospheric and stratospheric equatorial waves.