Fallen, broken tree in a park. Credit: RenegadeStudio

UK's Great Storm of 1987 - 30 years on

15 October 1987 11:00 UTC–16 October 03:00 UTC

Fallen, broken tree in a park. Credit: RenegadeStudio
Fallen, broken tree in a park. Credit: RenegadeStudio

In mid-October 1987 a powerful storm ravaged many parts of the UK.

Last Updated

05 April 2022

Published on

15 October 2017

By Vesa Nietosvaara (EUMETSAT) and Sancha Lancaster (Pactum)

The powerful storm which hit parts of the UK overnight 15/16 October 1987, became infamous in the country. Winds gusted up to 100mph, causing major destruction, particularly across Southern England. Eighteen people were killed and at least 15 million trees were blown down.

 Meteosat-2 Water Vapour, 15 Oct 23:00 UTC
Figure 1: Meteosat-2 Water Vapour, 15 October 23:00 UTC

The high winds were caused by a depression which formed in the Bay of Biscay on the afternoon of 15 October. Within hours, as it moved towards the south-eastern coast of England the depression rapidly deepened to 964mb. This was what is know as rapid or explosive cyclogenesis, when surface pressure falls by more than about 24 millibars per day.

By the time the storm passed over South East England in the early hours of 16 October it had gusts exceeding 180km/h. The strongest gust of 100kts (185km/h) was recorded at Shoreham on the Sussex coast and gusts of more than 90kts (166km/h) were recorded at several other coastal locations.

Satellite imagery in 1987

In 1987, very few observations were received from satellites and knowledge on how to interpret the satellite data for forecasting was more limited.

Looking at imagery from then experts can now spot the meteorological signs of a potentially severe storm.

 Met-2, 15  Oct 11:00 UTC–16 Oct 03:00 UTC
Figure 2: Meteosat-2 MVIRI, Infrared, 15 October 11:00 UTC–16 October 03:00 UTC
 Met-2, 15  Oct 11:00 UTC–16 Oct 03:00 UTC
Figure 3: Meteosat-2 MVIRI, Water Vapour, 15 October 11:00 UTC–16 October 03:00 UTC

In the Meteosat-2 MVIRI Water Vapour (WV) and Infrared (IR) image sequences (Figures 2 and 3) we can see some signs typical for rapid cyclogenesis.

At 11:00 UTC on 15 October 1987 in both the Water Vapour and Infrared channels a well-developed cloud head can be seen (Figures 4 and 5).

 Met-2, 15  Oct 11:00 UTC
Figure 4: Meteosat-2 MVIRI, Infrared, 15 October 11:00 UTC
 Met-2, 15  Oct 11:00 UTC
Figure 5: Meteosat-2 MVIRI, Water Vapour, 15 October 11:00 UTC

In the WV channel image there is also a distinct black stripe, indicating dry air moving downwards in the upper troposphere. During the day this black stripe only becomes darker and darker, and on 15 October at 23:00 UTC this dark stripe already extends over parts of the UK (Figure 1).

More information about rapid cyclongenesis signatures is available in the Manual of Synoptic Satellite Meteorology. Rapid Cyclogenesis, Cloud Structure in Satellite Images.

30 years on

According to the Met Office this storm was 'a wake-up call ' for the organisation and helped change many things about the forecast and warning process.

In 1987, only a few satellite images were available, and these were not used for the standard forecasts, instead they relied on observations from ships, buoys and planes, of which few were operating in the Bay of Biscay where the great storm was brewing.

Ken Mylne, Head of Verification, Impacts and Post Processing said: "We now have much better observations to initialise our forecasts. In 1987 we did not use satellite observations at all in the computer models; now we ingest over 14 =Gb of satellite observations per day, as well as thousands of automatic observations from aircraft and other sources, giving us a much more detailed analysis of what is happening over the Atlantic and the globe."

Now they have access to a vast array of data from EUMETSAT's geostationary and polar orbiting satellites, and to third party data services from all global providers of meteorological satellite data through EUMETCast. This has had a tremendous impact on the forecast skill that Numerical Weather Prediction models have nowadays.

In 1987 the Met Office's computer had a processing capacity that allowed it to make about 4 million calculations a second — about one-fifth of the power of an average smartphone.

Now, the models calculate at a rate of 14,000 trillion a second, can be run several times a day, and, most importantly, be compared to 50 other models around the world, meaning forecasters can generate ensemble forecasting using the best and most up-to-date data, including a wealth of satellite data. Ensemble forecasts produce a spread of possible outcomes of any given weather situation.

As well as increased computing power, the way the storm warnings are communicated has changed in recent years. In 2015 the Met Office started to formally name storms, to help raise awareness of severe weather and ensure greater safety of the public.

The most recent storm was Storm Aileen on 12/13 September 2017. The animation of Metesat-10 Airmass RGB images shows Storm Aileen forming south of Iceland at the beginning of the sequence and moving west across the UK and Ireland and into the North Sea by 12:00 UTC on the 13th.

Connected content

30th Anniversary of the 1987 Storm (Met Office)
Great Storm of 1987 (Liam Dutton/YouTube)
Weather forecasting's post-1987 revolution (BBC News)
Michael Fish would not dismiss great storm of 1987 today, says Met Office (The Guardian)
Met Office supercomputer would not have missed The Great Storm of 1987 (The Telegraph)