The impact of ATOVS on NWP
The accuracy of NWP has improved very significantly in recent years. Satellite
data has been a major contributor to this improvement and of the various
satellite data available to NWP centres ATOVS is found, at present, to have
the largest impact. It is generally accepted that it is the microwave data
which supplies much of the impact of ATOVS. However the infrared components,
and for that matter other satellite data types still have an important role
to play in specific situations (e.g. tropical storms).
This figure shows the improvement in NWP since 1997 as a yearly average of RMS error of 250 hPa wind for the southern hemisphere verified against radiosondes at T+72. Most NWP centres who made effective early use of ATOVS have seen substantial reduction in forecast RMS error since 1998. The reductions in error are similar for all centres ranging from 30% to 40% with the largest falls being for those centres using radiances from ATOVS. There are similar falls for almost all meteorological variables for which verification is exchanged under CBS (commission for basic systems) rules. Combined with evidence from trials (e.g. English et al. 2000, QJ Royal Meteorol. Soc. 126 (569): 2911-2931) it is generally accepted that ATOVS data has been an important factor in the large improvements seen in recent years.
To find out more the ITSC proceedings are a vital source of information. Regular progress reports and reports on OSEs are provided by met services. For results of OSEs including satellite data OSEs the WMO OSE workshop proceedings are a useful source of information. There was also an important workshop at ECMWF in 1999 (A.P. McNally. ECMWF/EUMETSAT Workshop on Use of ATOVS Data for NWP Assimilation, 2-5 November 1999, Shinfield Park, Reading, 1999.)
A number of papers have been written on the impact of ATOVS on NWP. A selection are given below. Click on titles for abstracts and full reference details.
- Zhang H, Xue JS, Zhu GF, Zhuang SY, Wu XB, Zhang FY, 2004:
Application of direct assimilation of ATOVS microwave radiances to typhoon track prediction.
- Langland RH, Baker AL, 2004:
Estimation of observation impact using the NRL atmospheric variational data assimilation adjoint system.
- Ahn MH, Kim MJ, Chung CY, Suh AS, 2003:
Operational implementation of the ATOVS processing procedure in KMA and its validation.
- Fourrie N, Doerenbecher A, Bergot T, Joly A, 2002:
Adjoint sensitivity of the forecast to TOVS observations .
- Le Marshall JF, Seecamp R, Harris B, Tingwell C, Kelly G, 2001:
First results from the use of local ATOVS data in the Australian region
- Bouttier F, Kelly G, 2001:
Observing-system experiments in the ECMWF 4D-Var data assimilation system.
- English SJ, Renshaw RJ, Dibben PC, Smith AJ, Rayer PJ, Poulsen C, Saunders
FW, Eyre JR, 2000:
A comparison of the impact of TOVS and ATOVS satellite sounding data on the accuracy of numerical weather forecasts.
- McNally AP, 2000:
Estimates of short-range forecast-temperature error correlations and the implications
for radiance-data assimilation
- Chen MH, Rood RB, Joiner J, 1999:
Assimilating TOVS humidity into the GEOS-2 data assimilation system.
- Prasad VS, Ramesh KJ, Bohra AK, Bhatia RC, 1998:
Assessment of quality and impact of full resolution TOVS temperature
profile data on the operational global data assimilation forecast system of India.
- Eyre JR, 1997:
Variational assimilation of remotely-sensed observations of the atmosphere
- McNally AP, Vesperini M., 1996:
Variational analysis of humidity information from TOVS radiances
- Gadd AJ, Barwell BR, Cox SJ, Renshaw RJ, 1995:
Global processing of satellite sounding radiances in a numerical weather prediction system.
- Andersson E, Pailleux J, Thepaut JN, Eyre JR, McNally AP, Kelly GA,
Courtier P, 1994:
Use of cloud cleared radiances in 3-dimenstional 4 dimensional variational data assimilation.
- Eyre JR, Kelly GA, McNally AP, Andersson E, Persson A, 1993:
Assimilation of TOVS radiance information through one-dimenstional variational analysis.
- Thepaut JN, Moll P, 1990:
Variational inversion of simulated TOVS radiancs using the adjoint technique.