Snowy London

Analysis of a 38-year-long data record of the polar jet

Winter 2010 and winter 2012

Photo credit: Robert Bye

Snowy London
Snowy London

Time-series of Atmospheric Motion Vectors (AMV) can be used to detect changes in location and intensity of the polar jet, which is related to inter-annual variations of the North Atlantic Oscillation (NAO) and the Arctic Oscillation (AO).

Last Updated

28 October 2021

Published on

21 October 2021

By Alessio Lattanzio, Marie Doutriaux Boucher, Roger Huckle, Oliver Sus, L. Medici, Mike Grant, Jaap Onderwaater, Régis Borde, Rob Roebeling and Joerg Schulz

Schematic Illustration of North Atlantic Oscillation (NOA)
Figure 1: Schematic Illustration of North Atlantic Oscillation (NOA) with an example of a positive NAO mode (upper panel) and a negative NAO mode (lower panel). The zonal flow is the west-east wind component, while the meridional flow is the south-north component (courtesy, Bradbury et al., 2002).

The NAO, in connection with the polar vortex, directly steers the location and intensity of the polar jet. As illustrated in Figure 1, a positive NAO directs the polar jet towards northern Europe, while a negative NAO directs it southward.

The jet is the outer edge of the polar vortex; it is a permanent feature of atmospheric circulation that is present at both poles. This use case aims to demonstrate the use of Meteosat-derived AMVs for detecting known changes in the NAO, and discusses the potential of using AMVs for monitoring climate change.

The NAO is a large-scale atmospheric pressure see-saw driving the weather and climate patterns in the northern hemisphere (Hurrell and Deser, 2010). The phase of the NAO, is expressed by the NAO index. This index is calculated from the difference in sea-level pressure between the Arctic (commonly over Iceland) and the subtropical Atlantic (commonly over the Azores) regions.

Even if it always expresses the same atmospheric pattern, the NAO index can be calculated using pressure values from different ground stations. In this case, we use the daily NAO index provided by the National Centers for Environmental Prediction (NCEP) Climate Prediction Center (Barnston and Livezey, 1987). The daily NAO index values, as derived by NCEP, are plotted for the period 2009-2013 in Figure 2.

In the same figure, two winters are highlighted:

  1. Winter 2009-2010, a colder and longer winter than normal in Western and Northern Europe (Cattiaux et al., 2010), was characterised by a strong negative NAO index.
  2. Winter 2011-2012, a particularly mild winter over Europe, was characterised by a strong positive NAO index.
 Daily NAO derived by NCEP
Figure 2: Daily NAO as derived by NCEP. A monthly running average is superimposed to show the variation at larger time scale (Credit: NCEP)

NAO variations have a direct impact on the polar jet or jet stream. Jet streams are geostrophic winds. i.e., their strength only depends on the pressure gradient and the Coriolis force. In the northern hemisphere, they flow from west to east, at speeds higher than 30m/s (108 km/h) and they are located in the upper level of the troposphere between 100 and 400 hPa (Kington and Ley, 1999).

First results

Doutriaux-Boucher et al., 2016 first attempted to show the link between the NAO and jet stream derived using meteosat observations comparing two single days: 1st December 2010 and 1st December 2011. For this analysis the Meteosat Second Generation (MSG) AMV Release 1 data have been exploited (EUMETSAT, 2015).

This connection has been analysed in more detail by Lattanzio et al., 2019 extending the comparison for a complete December January February (DJF) season. They first compared AMV patterns for a day during a season with a strong negative NAO phase (20 December 2009) and a day during a season with a strong positive NAO phase (20 December 2011). Figure 3 presents the AMV patterns during these two days, only considering AMVs with a Quality Indicator (QI) higher than 30. The QI provides an indication of the robustness of the retrieval (Holmlund, 1998). During the day with a negative NAO phase (20 Dec 2009), the median latitude of the detected AMVs is about 14 degrees lower and their corresponding median speed is about 5 m/s slower, compared to the day with a positive NAO phase (20 Dec 2011).

Jet streams retrieved on the same day (20 December) in 2009 during a strong negative NAO phase (upper panel) and 2011 during a strong positive NAO phase (lower panel).
Figure 3: Jet streams retrieved on the same day (20 December) in 2009 during a strong negative NAO phase (upper panel) and 2011 during a strong positive NAO phase (lower panel).

In their conference paper, Lattanzio et al., 2019 also analysed the seasonal anomaly of AMV latitude and AMV wind speed for the winter (DJF) with mostly negative NAO indices (2009-2010), and the winter with mostly positive NAO indices (2011-2012). The results of this comparison are shown in Figure 4, which presents the time-series of NAO indices, median AMV latitudes, and median AMV wind speeds during the months December, January, February, and March of these winters.

omparison of time-series of AMV wind speed and latitude during Winter 2009-2010 (left plot) and Winter 2011-2012 (right plot)
Figure 4: Comparison of time-series of AMV wind speed and latitude during winter 2009-2010 (left plot) and winter 2011-2012 (right plot). Dashed blue lines show the median wind speed and median latitude values during winter 2009-2010, dashed red lines show the median wind speed and median latitude values during winter 2011-2012.

It can be seen that the AMV latitudes during the winter with negative NAO indices (2009-2010) are about 15 degrees lower than during the winter with positive NAO indices (2011-2012). Similarly, a slight difference in AMV wind speeds is observed, with about 5 m/s slower wind speeds during periods with negative NAO indices.

The inclusion of Meteosat First Generation (MFG) in the latest release allowed extending the analysis back to 1981, covering a period of 38 years. The trend of both position and speed (see Figure 5) of the jet stream is in line with the prediction of several model and reanalysis (Irvine et al., 2016).

Time series for JS from 1981 to 2019 for mean latitude (top panel) and mean zonal speed (bottom panel).
Figure 5: Time series for JS from 1981 to 2019 for mean latitude (top panel) and mean zonal speed (bottom panel). The zonal speed is the west-east wind component. The analysis distinguishes between annual, boreal summer and boreal winter. The annual mean latitude decadal trend is slightly positive. The annual mean zonal speed shows a clear increase in the decadal trend.

This case shows that the patterns in AMV latitude and wind speed are consistent with the expected behaviour due to changes in NAO, and may serve as proxy for observing NOA changes related to shifts in the polar jet. The data record used for this use case was released in September 2021 (refer to Table 1 for details) with the following digital object index:


This study shows the importance that geostationary data such as the Meteosat play in analysing atmospheric climate pattern and in assessing their evolution during time. In particular, the analysis of the polar jet streams provides a key proxy data for the study of the Northern Atlantic Oscillation.

Data used

For this study, the following data has been used:

General Data record name Atmospheric Motion Vectors Release 2
Data record digital identifier DOI: 10.15770/EUM_SEC_CLM_0020 (0-degree)  
Data record short description Reprocessed level-2 geostationary atmospheric motion vector from Meteosat first and second generation
Record type Thematic Climate Data Record
Content Meteosat level-2 atmospheric motion vectors (TCDR)
Coverage Spatial
  • Meteosat disk
  • each pixel (IFOV) ground resolution of  2.5 km (3.0 km) for MVIRI (SEVIRI) at sub-satellite point.
Time period 01 January 1982 – 31 December 2017
Temporal frequency
  • MSG/SEVIRI hourly
  • MFG/MVIRI bihourly
Instrument Instruments names Meteosat Visible and Infrared Imager (MVIRI) Spinning Enhanced Visible and Infrared Imager  (SEVIRI)
Instruments descriptions
  • MVIRI is a passive imaging radiometer, the optical system consists of a scanning Ritchey-Chretien telescope with a primary aperture of 400 mm diameter (140 mm secondary aperture) and focal lengths of 3650 mm for VIS (resolution 2.5 km) and 535 mm for WV and IR ranges (resolution 5 km).
  • SEVIRI is a compact telescope and scan assembly, allowing the implementation of a large passive cooler which improves IR detector performances by lowering their operating temperature. The imaging SEVIRI radiometer is equipped with a patented three-mirror (3M) telescope of compact design (focal length of 5367 mm) which allows the insertion of a small black body for full-pupil calibration. It has 12 spectral channels, 11 provide measurements with a resolution of 3 km at the sub-satellite point. The twelfth, so-called HRV (High Resolution Visible) channel of SEVIRI (Spinning Enhanced Visible and Infrared Imager), provides measurements with a resolution of 1 km.
Instrument Data Input data
  • MVIRI  level-1.5 IR and WV channel images
  • SEVIRI level-1.5 IR10.8 (channel 9) and WV6.2 (channel 5) images
  • ERA-Interim forecast
  • Cloud top height at pixel resolution
Output data Geostationary atmospheric motion vector retrieved using the EUMETSAT algorithm
Format The products are provided in BUFR format and NetCDF 
Access EUMETSAT Data Centre The data set is available from EUMETSAT Data Centre ( EXT)
  • ftp push
  • online pull


Barnston, A. G. and Livezey, R. E.: Classification, Seasonality and Persistence of Low-Frequency Atmospheric Circulation Patterns, Mon. Wea. Rev., 115(6), 1083–1126, doi:10.1175/1520-0493(1987)115<1083:CSAPOL>2.0.CO;2, 1987.

Bradbury, J. A., Dingman, S. L., and Keim, B. D.: New England drought and relations with large scale atmospheric circulation patterns, J. Am. Water Resour. As., 38, 1287–1299,, 2002

Cattiaux, J., Vautard, R., Cassou, C., Yiou, P., Masson‐Delmotte, V. and Codron, F.: Winter 2010 in Europe: A cold extreme in a warming climate, Geophysical Research Letters, 37(20), doi:, 2010.

Doutriaux-Boucher, M., A. Lattanzio, O. Hautecoeur, R. Borde, and J. Schulz, Reprocessing of atmospheric motion vectors at EUMETSAT, 13th International Wind Workshop, Monterey, 2016,

EUMETSAT, 2015: Atmospheric Motion Vectors - MSG - 0 degree (CF-015 Release 1)

Hartmann, D.L., A.M.G. Klein Tank, M. Rusticucci, L.V. Alexander, S. Brönnimann, Y. Charabi, F.J. Dentener, E.J. Dlugokencky, D.R. Easterling, A. Kaplan, B.J. Soden, P.W. Thorne, M. Wild and P.M. Zhai, 2013: Observations: Atmosphere and Surface. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

Holmlund, K.: The Utilization of Statistical Properties of Satellite-Derived Atmospheric Motion Vectors to Derive Quality Indicators, Wea. Forecasting, 13(4), 1093–1104, doi:10.1175/1520-0434(1998)013<1093:TUOSPO>2.0.CO;2, 1998.

Hurrell, J. W. and Deser, C., 2010: North Atlantic climate variability: The role of the North Atlantic Oscillation, Journal of Marine Systems, 79(3), 231–244,

Irvine, E. A., Shine, K. P., and Stringer, M. A., 2016: What are the implications of climate change for trans-Atlantic aircraft routing and flight time?, Transportation Research Part D: Transport and Environment, 47, 44–53,

Lattanzio, A., Doutriaux-Boucher, M., Schulz, J. and Borde, R., 2019: Representation of Regular Climate Variations in the Eumetsat Atmospheric Motion Vector Climate Data Record. [online] Available from:

Related links

AMS conference proceedings 2019
Definition NOA UK Met Office
EUMETSAT satellite user conference 2021 Jet Stream presentation