Service of the atmospheric contribution to the geopotential


Contents:

Abstract

This page provides links to files relevant to the service of the atmospheric contribution to the geopotential provided by the Astrogeo Center. The service provides time series of Stokes coefficients of the atmosphere contribution to the geopotential into a series of spherical harmonics up to degree/order 72/72 and 20/20 and updates them every night.

Although initial tests showed good agreement with the series computed by T. Johnson, and including them in the model reportedly reduced residuals of analysis of GRACE satellite data, you should be aware that this is still experimental series.

Introduction

Variations of the atmosphere pressure causes changes in the Earth's gravity field. These changes should be taken in account for reduction of high precision observations.

Service of the atmospheric contribution to geopotential

On 2005.01.24 the service of atmospheric contribution to geopotential was established.

The series of the Stokes coefficients of the harmonic expansion of the contribution of the atmosphere to the geopotential to degree/order 72/72 and 20/20 are computed. The time resolution of the series is 6 hours. The series start on 1976.01.01_00:00. NCEP Reanalysis pressure field with spatical resolution 2.5ox2.5o and the 0.25ox0.25o land-sea mask are used for computations.

Files for downloading

These series are updated daily. The update process starts at 02:00 local time. 
[an error occurred while processing this directive]

File Format Size Comment
agra72_2m.tar.bz2 AGRA
Data are archived with tar and bzip2. 		7 Mb This compressed tar-file contains a directory of ascii files with Stokes coefficients of degree/order 72 from 00:00 of the first day of the previos month through 2009.01.16-18:00:00
agra72_ty.tar.bz2 AGRA
Data are archived with tar and bzip2. 		7 Mb This compressed tar-file contains a directory of ascii files with Stokes coefficients of degree/order 72 from 01-JAN 00:00 this year through 2009.01.16-18:00:00
agra72_py.tar.bz2 AGRA
Data are archived with tar and bzip2. 		1090 Mb This compressed tar-file contains a directory of ascii files of Stokes coefficients of degree/order 72 from 1976.01.01 through 31-DEC 18:00 of the previous year.
agra72 AGRA
9730 Mb This is the link to the directory with uncompressed ascii files of Stokes coefficients of degree/order 72 from 1976.01.01_00:00 through 2009.01.16-18:00:00
    
agra20_2m.tar.bz2 AGRA
Data are archived with tar and bzip2. 		0.5 Mb This compressed tar-file contains a directory of ascii files with Stokes coefficients of degree/order 20 from 00:00 of the first day of the previos month through 2009.01.16-18:00:00
agra20_ty.tar.bz2 AGRA
Data are archived with tar and bzip2. 		0.5 Mb This compressed tar-file contains a directory of ascii files with Stokes coefficients of degree/order 20 from 01-JAN 00:00 this year through 2009.01.16-18:00:00
agra20_py.tar.bz2 AGRA
Data are archived with tar and bzip2. 		93 Mb This compressed tar-file contains a directory of ascii files of Stokes coefficients of degree/order 20 from 1976.01.01 through 31-DEC 18:00 of the previous year.
agra20 AGRA
840 Mb This is the link to the directory with uncompressed ascii files of Stokes coefficients of degree/order 20 from 1976.01.01_00:00 through 2009.01.16-18:00:00

The time series have zero mean and no signal at S1 and S2 frequencies. The total atmospheric contribution to geopotential is the sum of the time series and the model of harmonic site geopotential variations caused by atmospheric tides at the S1 and S2 frequencies.

Other related files

  1. Source code of the programs for computing the Stokes coefficients of the harmonic expansion of the contribution of the atmosphere to the geopotential.

  2. Gridded values of the mean surface pressure in NetCDF format.

  3. Gridded values of the S1 and S2 atmospheric tides in NetCDF format which are present in raw NCEP Reanalysis pressure field. This model of atmospheric tide has been subtracted from the raw NCEP Reanalysis surface pressure field before computing Stokes coefficients.

  4. Gridded values of the S1 and S2 atmospheric tides in NetCDF format according to the model of Ponte and Ray (2002) model.

  5. Gridded values of the 2.5ox2.5o land-sea mask in NetCDF format.

  6. Gridded values of the 0.25ox0.25o land-sea mask in NetCDF format.

  7. Load Love numbers. The first column is the load Love h, the second is the load Love number h, the third column is the load number k.

  8. Brief log file of the service. Time tag in local time.

  9. Verbose raw log file of the service. Time tag in local time.

Smart downloading

Time series are updated every day automatically. It is assumed that regular users also use automatic routines for retrieval the series. Some tips: the date of the latest epoch of the time series is stored in http://astrogeo.org/agra_res/agra_dates.txt. Examining the date stored in this file, the client process can determine whether the time series have been updated. Package wget is suitable for data retrieval. An example of the client program which is scheduled by cron every hour can be found here.

Latency

The Stokes coefficients are computed within 24 hours after availability of the surface atmospheric pressure field of the NCEP Reanalysis model. NCEP Reanalysis model lags about 3 days. At 90% cases the Stokes coefficients time series has lag 3.5 days. At 10% cases the lag is in the range of 4-15 days. The plot of latency for last 2 months is available here.

Method

The Stokes coefficient are obtained by expansion of the input pressure field into series of spherical harmonics. The Stokes coefficients are of this expansion are multiplied by a factor which depends on the load Love number of the certain degree. We used the load Love numbers derived by P. Gegeot on the basis of the preliminary Earth model (PREM) with of the surface pressure field with 2.5ox2.5o resolution with 6 hour time resolution provided by NCEP/NCAR Reanalysis numerical whether model after subtraction the mean surface pressure and the model of diurnal and semi-diurnal variations of surface pressure computed from the NCEP reanalysis surface pressure field over the period of 1980-2002. The oceanic response was modeled according to the inverted barometer hypothesis (the variant with conservation of the total ocean water mass). The FES99 land-sea mask with resolution of 0.25ox0.25o was used. The loading Love numbers are expressed in the reference frame of the total Earth, i.e. the solid Earth and the atmosphere. Therefore degree 0 and degree 1 Stokes coefficients are zero. The Stokes coefficients using so called "4pi normalization over the unit sphere". This means that the integral of the square of each spherical harmonics is 1/(4pi).

Citation instruction

If you use this time series in your analysis, you are welcome to acknowledge in your publication that you used "the Stokes coefficients of the harmonic expansion of the contribution of the atmosphere to the geopotential provided by the Astrogeo Center which are available on the Web at http://astrogeo.org/agra (Petrov & Boy, 2005)".

References

  1. J.-P. Boy, L. Petrov, Memo on computing Stokes coefficients of the expansion of the atmosphere contribution to the geopotential into a series of spherical harmonics, Internal memo, unpublished, 2005. Available here.

  2. Kalnay E., M. Kanamitsu, R. Kistler, W. Collins, D. Deaven, L. Gandin, M. Iredell, S. Saha, G. White, J. Woollen, Y. Zhu, A. Leetma, R. Reynolds, M. Chelliah, W. Ebisuzaki, W.Higgins, J. Janowiak, K. C. Mo, C. Ropelewski, J. Wang, R. Jenne and D. Joseph, The NCEP/NCAR 40-Year Reanalysis Project, Bullet. Amer. Meteorol. Soc., vol. 77, pp. 437--471, 1996.

  3. Lefevre, F., F.H. Lyard, C. Le Provost and E.J.O. Schrama, FES99: a global tide finite element solution assimilating tide gauge and altimetric information, J. Atmos. Oceanic Technol., vol. 19, pp.1345--1356, 2002.

  4. Wunsch, C. and D. Stammer, Atmospheric loading and the "inverted barometer" effect, Rev. Geophys., vol. 35, pp. 117--135, 1997.

Back to loading page.

This page was prepared by Leonid Petrov ()
Last update: 2012.10.11_22:54:52