The Boscastle Flood

Some facts and figures
by Philip Eden

The flash flood at Boscastle, north Cornwall, on Monday 16 August, was the consequence of a violent four-hour downpour over the small catchment of the Valency river. There was no obvious culprit: some newspapers suggested that Hurricane Bonny was responsible, others blamed global warming, but in truth there was no evident external influence on the meteorological events of that afternoon.

A rather warm, moist, and unstable southwesterly airflow delivered a PmR (polar-maritime returning) air-mass which had travelled far to the south in mid-Atlantic before curving back towards the British Isles. The air-mass was thus unstable and cumulonimbus clouds (and therefore showers) developed readily. As the southwesterly flow crossed Cornwall, surface frictional effects caused the flow to decrease and back over the land, and a convergence line developed between the moderate southwesterlies over the sea and the light south-southwesterlies over Cornwall itself. This convergence line lay roughly parallel to the north Cornish coast from St Ives to Newquay, but it crossed the coast between Newquay and Boscastle. The uplift feeding the con vective process was therefore accelerated over the western flank of Bodmin Moor (known hereabouts as Davidstow Moor) resulting in a prolonged in-situ generation of active cumulonimbus clouds. This is why the downpour lasted for four hours.

Contrary to media reports, serious flooding has occurred before at Boscastle: in late-October 1996 when ex-hurricane Lili delivered a combination of heavy rain and high tides that inundated the lower part of the village; in the Junes of both 1958 and 1957 when sudden floods developed in circumstances similar to last Monday; and in July 1847 when a much broader region was affected.

The short, steep valleys of north Cornwall and north Devon are particularly vulnerable to localised summer downpours. They collect water efficiently from the surrounding moors, channel it rapidly into the main stream, and take it all out to sea in a matter of three or four hours. Because of their almost instantaneous response to a sudden cloudburst, these valleys are known in the trade as flashy catchments and they produce true flash floods. Some have speculated that the clearing of natural vegetation from the valley slopes contributed to the abruptness and intensity of the disaster, but it is certain that a destructive flood would have happened anyway. Now we all know what a real flash flood looks like, perhaps our traffic reports and news bulletins will stop using the term to describe 18 inches of water under a railway bridge in Neasden.

In the rush to find a scapegoat - climate change being the favoured one - most commentators ignored the fact that we were long overdue one of these catastrophes. There was a serious flash flood in mid-Wales two years ago, but before that we have to go back to Helston in June 1993 and Halifax in May 1992. The 1950s and 1960s gave us several major West Country floods, including the Wadebridge and Camelford flood in June 1950, the Lynmouth disaster in August 1952, another Camelford flood in June 1957, Porlock in July 1959, Wadebridge again in June 1963, and large parts of Somerset and east Devon in July 1968.

The media also largely ignored serious flooding in neighbouring river catchments. There was much damage a few miles further north in Crackington Haven, and also a little further up the coast at Millook, while to the south Bossiney, Tintagel and Trebarwith Strand were badly hit. At least half the water from the storm entered the larger, east-flowing catchments of the Ottery and Inney rivers and the south-flowing Camel, and many villages along these rivers suffered considerable (although less catastrophic) flooding.

It took several days before the quantity of rain responsible for the Boscastle disaster was accurately ascertained. We now know that the focus of the storm lay over the moors three or four miles east of Boscastle itself; 200.4mm fell at Ottersham, practically all of it inside four hours.

Maximum falls in specified periods were:
15 mins 32mm
30 mins 52mm
45 mins 73mm
60 mins 88mm
90 mins 112mm
2 hrs 124mm
3 hrs 162mm
4 hrs 197mm

None of these approaches a UK record.