NWP Satellite Application Facility

The Numerical Weather Prediction Satellite Application Facility (NWP SAF) is a collaborative project that develops tools to exploit satellite data for NWP. Funded by EUMETSAT, the NWP SAF is one of eight SAFs, each of which concentrates on a different meteorological application.

The NWP SAF is led by the Met Office, with partners ECMWF, KNMI and Météo-France. The SAF was started in February 1999 and is planned to continue to at least 2012.

Objectives

• Improve the benefits derived by European National Meteorological Services from NWP through use of satellite data
• Enable effective exploitation of EUMETSAT's satellite programmes and related programmes of other agencies

Activities

• Assess user requirements and prioritise activities to meet them
• Develop and deliver satellite data processing modules
• Provide user support for those modules
• Monitor operational satellite data quality and publish results on the NWP SAF website

A popular overview of the NWP SAF appeared in EUMETSAT's newsletter (PDF, 225 kB).

Deliverables

The SAF's main deliverables are software packages and data monitoring reports.

The ATOVS and AVHRR Pre-processing Package (AAPP) produces files of quality-controlled brightness temperature or radiance data from the HIRS, AVHRR, AMSU, MHS  and IASI instruments on polar-orbiting satellites.

The SSMIS Pre-processing Package (SSMIS_PP) processes data from the Special Sensor Microwave Imager/Sounder (SSMIS) instrument on board the US DMSP series of satellites. It corrects for systematic effects in the SSMIS data stream, ready for direct input into NWP assimilation systems.

Three One-Dimensional Variational Analysis (1D-Var) retrieval packages

• Met Office 1D-Var
• ECMWF 1D-Var
• SSMIS 1D-Var

These retrieve physical parameters from satellite instrument data by using a first guess taken from an NWP computer model. The three packages have different characteristics and cover a range of instruments.
 
Scatterometer data processors (the SDP for SeaWinds and the AWDP for ASCAT) process radar backscatter data to produce quality-controlled wind vector field solutions, including full wind vector probability distributions and ambiguity removal.

RTTOV is a very fast radiative transfer model for infrared and microwave satellite sensors, and is designed to be incorporated into users' applications. Given an atmospheric profile, RTTOV will compute the top of atmosphere radiances in each of the channels of the sensor being simulated.

Atmospheric profile data sets selected from ECMWF model fields, with many applications, including statistical regressions, provision of first guesses for inversion algorithms and the validation of various models, particularly in the field of radiation.

AIRS/IASI Cloud Detection Software based on a pattern recognition algorithm developed at ECMWF, detects and distinguishes clouds and aerosols. It is sufficiently modular to ‘plug-in’ to most NWP systems.

IASI PCA-based Compression, an IASI pre-processing package which calculates a covariance matrix based on a set of training data, from which principal components (PCs) are derived. These PCs may then be used to noise-filter other spectra such that a reduced set of channels can be used for data assimilation.

Monitoring Reports on the NWP SAF web site for various operational satellite products and instruments, including the EUMETSAT and Regional ATOVS Retransmission Services (EARS and RARS), atmospheric motion vectors, scatterometers, SSMI, and the advanced infrared sounders AIRS and IASI.