Training and Outreach


Once of the main functions of the NWP SAF is to provide educational training courses in the assimilation of satellite data and encourage development through the organisation of science workshops. This is primarily done through the annual EUMETSAT/ECMWF NWP-SAF Satellite data assimilation course as well as other, shorter workshops. Details are given below.

EUMETSAT/ECMWF NWP-SAF Satellite data assimilation

2-6 March 2020

This course is sponsored by EUMETSAT and comprises 18 hours of lectures plus 5 hours of practical sessions. It is aimed primarily at graduate students and other research scientists (e.g. from National Meteorological Services) who are relatively new to the subject of satellite data assimilation for NWP. However, it also caters for more experienced scientists who wish to advance their knowledge in specific areas. Lectures focus on aspects that are important for the successful exploitation of satellite observations in NWP from fundamental theoretical concepts through to detailed practical implementations in modern state-of-the-art data assimilation systems.

Topics include:

  • Global Observing System overview
  • What do satellites measure?
  • Data assimilation algorithms and key elements
  • The infrared spectrum, measurement, modelling and information content
  • Introduction to GPS radio occultation
  • Wind information from satellites  (PMW, SCAT, AMV and PT)
  • Observation errors
  • Microwave observations in clear-sky conditions
  • Microwave observations in all-sky conditions
  • The detection and assimilation of clouds in infrared radiances
  • Estimation and correction of systematic errors

More details of the course are available on the course’s ECMWF homepage.

Previous courses

Resources from recent courses are publicly available (available via the external links below).

Upcoming Workshops

NWP SAF/JCSDA Workshop on assimilating satellite cloud and precipitation observations for NWP

3-6 February 2020, ECMWF, Reading, UK.

Satellite observations sensitive to cloud and precipitation are key to improving global, regional and convective scale weather forecasts. All-sky assimilation of microwave radiances has been proven through operational assimilation. Further, all-sky infrared and visible radiances are in development, particularly to make use of rapidly updated geostationary observations. Cloud and precipitation radar, cloud lidar and lightning data should all start to contribute to NWP in the next decade. However, the assimilation of cloud and precipitation pushes data assimilation beyond its current design assumptions. Models often have state dependent biases in cloudy areas and do not represent details of the microphysical and sub-grid variability that are needed to drive the observation operators. These, based on scattering radiative transfer, require major simplification to make them fast enough for operational use. The random part of the observation error is dominated by the lack of predictability of cloud and precipitation processes. It is hard to get the most from increasing satellite instrument resolutions when the predictable scales of cloud are far coarser. Further, these predictability or representation errors are correlated in space and time and likely with the background errors, but to diagnose these errors is extremely difficult. The problem is not just correlations, because cloud and precipitation also give rise to nonlinear, non-Gaussian bounded processes and errors. To progress our ability to use cloud and precipitation data we may need to ‘learn’ better physical models and microphysical assumptions, improve fast scattering radiative transfer, and allow data assimilation to handle errors that are non-Gaussian, heterogeneous and heavily correlated.

Further information about this workshop

NWP SAF Workshop on the treatment of random and systematic errors in satellite data assimilation for NWP

2-5 November 2020, ECMWF, Reading, UK.

Dealing with random and systematic errors in observations and models is at the heart of satellite data assimilation for Numerical Weather Prediction (NWP). This invitation-only workshop will capture recent progress and the state of the art in this area, identify key issues and limitations, and discuss avenues for further developments. The workshop will cover the following main themes:

  • Estimating uncertainties
  • Correction of observational and model biases in data assimilation
  • Representing observation errors in data assimilation

Futher information about this workshop

Past Workshops

EUMETSAT NWP SAF mesoscale wind data assimilation workshop (2018)

A EUMETSAT NWP SAF mesoscale wind data assimilation workshop was held on 18 September 2018 in Tallinn, Estonia. 2018 sees an unprecedented increase in the number of wind-measuring satellites:

  • EUMETSAT’s MetOp-C will bring the 3rd ASCAT into orbit, where the three ASCATs will provide, for the first time, close to 100% wind coverage of the seas around 9:30 local time (LT), both in the morning and evening;
  • ESA’s Aeolus mission will, also for the first time, provide wind profiles from space;
    The Chinese-French Ocean SATellite, CFOSAT, includes a rotating fan-beam wind scatterometer, SCAT, and combines it with a wave-measuring instrument, SWIM;
  • India will launch OceanSat-3 with on-board its well-proven wind scatterometer, providing ocean wind coverage at 12:00 and 00:00 local time;
  • China launched HY-2B, providing ocean wind coverage at 6:00 and 18:00 LT.

Satellite observations have been fundamental in improving weather forecast skill over the past two decades on all terms, while few high-quality wind observations were present in the Global Observing System. To depict and initialize the flow on scales smaller than 500 km, wind observations are essential, however, and adding satellite wind observations for predicting dynamical weather has proven to be beneficial. But how are we going to exploit all these wind observations?

The mesoscale wind data assimilation workshop addressed this question and raised expectations for the beneficial and practical use of these novel wind observing systems. It addressed the spatial scales observed and the errors in scatterometer winds. For satellite ocean winds it addressed the calibration wind reference at high and extreme winds, ambiguity removal, and a method to obtain weather-dependent error covariances, both for buoy, satellite and NWP data. For NWP models it provided examples of the spatial scales deterministically modelled, but also the spatial scales which are only realistically modelled and not well initialized. This sets the scene to discuss spatial representation errors, which often dominate wind measurement observation errors, be it from radiosondes, aircraft winds or wind scatterometers. The pros and cons of thinning, superobbing and “supermodding” have been discussed. The workshop also addressed well-known biases of weather models, but which violate the BLUE data assimilation principle: Best Linear Unbiased Estimate. While air-mass dependent bias correction schemes are fundamental to the current beneficial use of satellite observations, they are not in place for satellite winds. A way forward has been suggested and discussed with the attendees.

The presentations and accompanying documents are now available:

Presentation held at the workshop:
Summary of the workshop, presented at the EUMETSAT satellite conference:
Wind Bias Correction Guide, v1.3 (NWPSAF-KN-UD-007):
High resolution data assimilation guide, v1.2 (NWPSAF-KN-UD-008)

Assimilation of hyperspectral geostationary satellite observations (2017)

A NWP SAF workshop on ‘Assimilation of Hyperspectral Geostationary Satellite Observations‘ was held at ECMWF, Reading, UK, on 22-25 May 2017. The workshop was intended to assess the state of readiness ahead of the launch of the METEOSAT Third-Generation geostationary satellite and focused on the latest progress in instrument calibration and spectral characterisation, advances in radiative transfer modelling and the various assimilation strategies being considered to exploit this unique and exciting source of observations.

Efficient representation of hyperspectral infrared satellite observations (2013)

An NWP SAF workshop on the ‘Efficient representation of hyperspectral infrared satellite observations‘ was held at ECMWF, Reading, UK, on 5-7 November 2013. The workshop focused on optimizing dissemination practices to allow data compression with minimal information loss, together with the development of novel data assimilation techniques that can efficiently convey this information to the analysis.

Scatterometer Data Assimilation Workshop (2011)

In April 2011, the NWPSAF ran a Scatterometer Data Assimilation Workshop at EUMETSAT. The goal of the workshop was to provide NWP data assimilation colleagues insight in the main relevant characteristics of the scatterometer data, provide guidance for data assimilation and collect NWP user requirements. Slides, posters and training material have been made available here.

Assimilation of IASI in NWP (2009)

An NWP SAF workshop on the Assimilation of IASI in NWP was held at ECMWF, Reading, UK, on 6-8 May, 2009. The workshop considered the progress that had been made in the exploitation of these unique hyper-spectral observations. The methods that have been adopted to assimilate clear and cloud affected radiances were covered, in addition to novel techniques based on the use of principal components and level-2 retrievals. Reflecting the recent expansion of NWP systems to cover environmental monitoring applications, the workshop also dealt with the assimilation of atmospheric constituent information from IASI.

Bias estimation and correction in data assimilation (2005)

The NWP SAF Workshop on ‘Bias estimation and correction in data assimilation‘ was held at ECMWF, Reading, UK on 8-11 November 2005. The workshop focused on the careful handling of biases in data assimilation in order to make ffective use of current and future satellite observations.

Other outreach activities

Representatives from the NWP SAF also participate in international scientific conferences such as ITSC, IWW, etc. and other externally organised workshops to gather information about user requirements for NWP SAF software packages.

Recent ITSC links

The International TOVS Study Conference (ITSC) is held every 18-24 months to provide a forum to facilitate the exchange of information on methods for extracting information from these data on atmospheric temperature and moisture fields and on the impact of these data in numerical weather prediction and in climate studies. Below is a table of resources from recent conferences:

ITSC XXI (December 2017, Darmstadt, Germany) Report to follow Program to follow
ITSC XX (October 2015, Lake Geneva, Wisconsin, USA) Report Program
ITSC XIX (March 2014, Jeju Island, South Korea) Report Program
ITSC XVIII (March 2012, Toulouse, France) Report Program

Recent IWW links

The International Winds Working Group (IWWG) was established in 1991 and became a formal working group of the Coordination Group for Meteorological Satellites (CGMS) in 1994. Workshops (IWW) are held approximately every 2 years. The aim is to promote exchange of scientific and operational information between the producers of satellite derived wind products, the research community and the user community

IWW14 (April 2018 Jeju City, South Korea) Proceedings to follow Program
IWW13 (June 2016, Monterey, California, USA) Proceedings Program
IWW12 (June 2014, Copenhagen, Denmark) Proceedings Program
IWW11 (February 2012, Auckland, New Zealand) Proceedings Program

EUMETSAT Training and Conferences

EUMETSAT Conference 2017
Conferences and Events

If you would like to add a link to your satellite-related event or training please contact the helpdesk.