When Science Goes Wrong (9 page)

Still, the BBC Weather Centre has always had the reputation of being a scientific organisation, not just a group of presenters parroting forecasts prepared at Met Office headquarters.

Given that Fish mentioned the entertainment issue, it may be appropriate to say a word about his physical appearance at the time of his famous broadcast. A staple of TV forecasting since 1974, Fish was a balding, moustachioed 43-year-old, who liked to wear thick, dark-rimmed glasses and wool sweaters under tartan jackets. He sported a collection of ties with fish motifs and (if the
Sunday Herald
is to be believed) his underwear was also personalised, this time with weather-chart symbols. Fish had a love-it-or-loathe-it kind of style; fittingly, he was once voted Best Dressed Man and Worst Dressed Man on television in the same year.

Of course, I wanted to know the story behind the woman who called in to ask whether there was a hurricane on the way. This is a question that Fish has been asked many times before. In 2004, when he was interviewed for an article on the BBC Weather Centre’s website, he answered it as follows: ‘Nobody called in... My remarks referred to Florida and were a link to a news story about devastation in the Caribbean that had just been broadcast. The phone call was a member of staff reassuring his mother just before she set off there on holiday.’

Fish gave me pretty much the same account, and he went on to bemoan how the often-shown video clip had been edited to make him seem as if he was talking about the weather in England. ‘If you had the complete clip there it would be painfully obvious it was nothing to do with the situation [in England],’ he said. ‘The rest of the broadcast went on to say, “Batten down the hatches, there’s some extremely stormy weather on the way.” Which to me is a very good forecast.’

This account is at least partially incorrect, according to a study of the television and radio forecasts before the storm that was published in 1988 as part of the report of the official investigation. Fish didn’t say ‘Batten down the hatches...’ on that broadcast at all. He uttered that remark, or something like it, in the course of a different forecast that he gave 30 minutes later. This was a forecast for the European satellite television Superchannel. His exact words were, ‘It’s a case of batten down the hatches, I think, for some parts of Europe; some very, very stormy weather on the way indeed.’ In other words, Fish was saying that exceptional winds would occur over continental Europe – something that all the models were agreed on – and not that they would affect England. Later in that broadcast, he specified France and the Low Countries as the areas at risk.

On the BBC broadcast, Fish did say that it would get very windy, but he gave less emphasis to the wind than to the prospect for rain, in line with the existing concern about flooding. The charts that accompanied the television forecasts that day indicated sustained wind speeds of up to 50 mph, but only for the English Channel and North Sea, not for land areas.

What about the woman who called in? Was she really the mother of one of his colleagues who was planning a trip to Florida? Not at all, according to the
Daily Mail
. In the aftermath of the storm, that newspaper posted a monetary reward for the name of the woman involved. The answer soon came in: it was a Mrs Anita Hart from Pinner in northwest London.

I tracked down Mrs Hart and spoke with her by telephone in 2006. ‘Oh no, no!’ she cried in mock despair when I told her why I was calling. ‘This has been haunting us for the last 20 years.’

She told me that she saw the
Mail
’s reward offer but didn’t respond to it because she valued her privacy. ‘But we were shopped by one of our son’s friends. To get the reward he called the newspaper and identified us.’

Mrs. Hart’s story started a few days before the storm. She and her husband were planning a trip to Wales with their caravan. Their son Gaon was then studying meteorology at Manchester University. ‘We were in the habit, if we wanted to go away for the weekend – we would phone him and ask what the weather was going to be like. On this particular occasion he said, “Don’t laugh, but I think there’s going to be a hurricane.” He had tapped into the French computers, because our computers were not up to it.’

This was on Monday. By Wednesday there were still no storm clouds on the horizon, so Anita called the BBC Weather Centre and asked whether there was indeed going to be a hurricane as she’d been warned. A man whom she took to be Michael Fish replied, ‘No, we don’t get hurricanes in England.’

When I told Mrs. Hart of Fish’s explanation for his remarks on the broadcast – namely that he was talking about Florida, and that the video clip had been edited to make it seem as if he was referring to England – she laughed again. ‘That’s absolute nonsense. That’s not true at all. Obviously his story changed.’

Although Fish’s version of events is difficult to square with other accounts, it has established itself as authentic in many quarters, including the BBC Weather Centre’s website, the Wikipedia article on the storm, and so on. Thus many sources portray Fish as the innocent victim of a hatchet job by the media, when in reality he himself handed them the hatchet.

Fish’s broadcast did not endear him to his superiors. In talking with me, Bill Giles described Fish’s comments as ‘stupid’ and ‘silly’. Ewen McCallum seconded the ‘stupid’ and added ‘dumb’ for good measure. The official report published in the following year described Fish’s comments as ‘particularly unfortunate’. It wasn’t that Fish was single-handedly responsible for the bungled forecast, but that by issuing such an unqualified denial of the danger, he made any kind of defence of the Met Office’s performance impossible. ‘The Fish thing was very important because that was the damning piece of evidence,’ said McCallum. ‘That was the “No further questions, Your Honour.”’

Soon after Fish delivered his broadcast, the Met Office’s Cyber 205 embarked on yet another round of number-crunching, using the weather data for noon. This time, the results of the fine-mesh and global models were in reasonable agreement, and they also agreed with the ‘compromise’ forecast that had been issued earlier. This may have caused the Bracknell forecasters who had settled on the earlier compromise to congratulate one another on their good judgment, but if so their satisfaction was short-lived.

At any rate, they issued a forecast that was similar to the earlier one, and consequently Bill Giles’s presentation on the Thursday evening news was similar to the one given by Michael Fish at lunchtime – minus the hurricane story, of course. Again, the emphasis was on the prospect for heavy rain, with some mention of strong winds in south-eastern coastal areas. Thus most Britons went to bed without an inkling of the disaster that was about to strike. If they took any precautions, it was to station their wellies by their front doors, not to batten down any hatches.

Around the time of Giles’s evening presentation, the centre of the oncoming depression lay in the western approaches to the English Channel. It still seemed as if the Met Office’s forecast – a track up the channel and across the south-easternmost portion of England – was a good possibility. Between 7pm and midnight, however, the depression began to veer onto a more northerly path, and the depth of the depression intensified. By midnight, the depression was nearing the coast of Devon, and the sea-level atmospheric pressure at its centre had decreased to about 953 millibars. This corresponded to the pressure that (on a normal day) would be encountered at an altitude of about 1,500ft. It was at least 12 millibars below any estimate offered by computer models during the run-up to the storm. (By comparison, the pressure in the eye of Hurricane Floyd – the infamous Floyd – dropped to 921 millibars.)

With the extreme deepening of the depression, the winds picked up to very high speeds. The highest winds were far above the land surface and thus were never actually measured. From the spacing of the isobars to the south of the storm track, however, it can be calculated that ‘geostrophic’ winds at altitude would have blown at about 300 knots (345mph). The actual high-altitude winds would have been blowing at somewhat less than geostrophic speeds, because they would have been slowed by centrifugal forces as they followed the curved path of the isobars. Still, they must have been blowing at speeds far above the threshold for hurricane status (65 knots, or 75mph) or even the threshold for the most violent category 5 hurricanes (135 knots, or 155mph). Of course, as Michael Fish stated, this did not mean that the storm was in fact a hurricane.

At the Earth’s surface, winds were much slower, but they were still exceptionally strong. In the English Channel to the south of the depression, and along the adjoining English coast, winds reached 50 knots around midnight, making it a ‘severe gale’ or ‘storm’ on the Beaufort scale of wind strengths. A gust of 95 knots (109mph) was recorded on the French coast at about the same time.

During the small hours of Friday morning (October 16) the centre of the depression began to track across southern England, well to the north of its forecasted path. The depression lay over Bristol around 3am, Birmingham around 4am, Nottingham around 5am and Hull around 6am. These cities experienced only gentle winds, or even complete calm, as the depression passed over them, but it was a different story to the south. Steady winds of 60 knots or more were widely reported, and gusts reached much higher speeds. The strongest gust recorded on the English mainland was 106 knots (121mph) at Gorleston on the coast of Norfolk. Even in London, a gust of 94 knots (108mph) was recorded at the British Telecom tower at 2.40am, shortly before London’s electricity supply failed and the anemometer stopped recording. This was the highest wind speed ever recorded in the capital. The highest wind speed measured anywhere during the storm was a 117 knot (134mph) gust recorded on the coast of Normandy shortly after midnight.

Besides the wind and rain, southeast England experienced remarkable changes in temperature as the warm front and the pursuing cold front passed overhead. Shortly before midnight at the Met Office’s station in South Farnborough, southwest of London, the temperature rose by 9°C (16°F) in 20 minutes, and fell by the same amount a few hours later. Rapid pressure changes also occurred, particularly as the tail end of the depression swept across England: rises in pressure by as much as 12 millibars in one hour were recorded toward the end of the night. Both the temperature and the pressure changes were unprecedented in terms of their rapidity.

The human impact of the storm was felt earliest in the Channel. A catamaran carrying six people was put into difficulties off the coast of Dorset: the Weymouth rescue lifeboat set out in mountainous seas and, with support from a Royal Navy destroyer, was able to save the crew. A coaster, or small cargo ship, was damaged off the Isle of Wight and required rescue by two lifeboats. Farther to the east, it was even worse. A bulk carrier capsized and sank outside Dover harbour, taking the lives of two crew members. A cross-Channel freight ferry, the
Hengist
, ran aground near Folkestone, and the crew had to be brought to shore by an antiquated rope-and-harness contraption known as a breeches buoy. (The ship was saved and now plies the calmer waters of the Aegean under a different name.) Two passenger ferries, with 400 passengers aboard, were unable to enter port and had to ride out the storm at sea. A lifeboat set out from Sheerness, on the north coast of Kent, to save a fishing boat that was running aground; it rescued the crew but then was itself driven aground by a violent gust and had to remain there for the rest of the night.

Onshore, 16 people died. The most frequent cause of death was treefalls, which killed nine people in England. One of these, a homeless man, was killed directly by a tree that fell in a London park. Four people – including two firemen returning from an emergency call – were killed by trees that fell on their vehicles, and three others died when their vehicles collided with fallen trees or while swerving to avoid one. A woman in Chatham, Kent, died when a tree crashed through the roof of her home.

Other deaths included three people who were killed by falling chimney stacks, a fisherman killed by a flying beach hut and a motorcyclist who was blown into a barrier by a violent gust. Besides the British victims, four people died in France.

The main physical damage was to trees: an estimated 15 million of them were destroyed over a span of about six hours. Two factors conspired to maximise the damage. First, most deciduous trees were still in leaf, thus giving the wind an easy purchase. Second, the ground was sodden from weeks of rain, making the trees much less securely rooted than usual. Not even those trees whose roots held fast were safe, however: many of them succumbed to trunk breakage, a mode of failure that greatly reduced their salvage value.

In falling, the trees wreaked havoc. Many fell on their owners’ homes. Here is an excerpt from an account written by a meteorologist; HD Lawes:

We gave up trying to sleep and I made a cup of tea... Up to a point, one feels that your own home is the safest place to be but I had never seen any storm like this and I was beginning to feel uneasy... Another particularly strong gust started up, there was a huge rumbling crash and a rushing noise. The house shook, the ceiling cracked and outside masonry was falling. We realised that the cedar tree had fallen and rushed upstairs to where baby Sam had been asleep. There was a gaping hole in the wall of his bedroom and the ceiling light was blowing in the wind. Luckily Sam was still there in his cot with only a few pieces of plaster and brick on him. He was crying lustily, which was reassuring, and Sue plucked him from the cot. We rushed downstairs and out of the front door, pausing only to rescue the cat, my pipe and tobacco. Outside we had to clamber through a morass of branches and foliage to where a neighbour was holding a torch and we gratefully accepted their hospitality.