Violent thunderstorms in the Thames Valley and south Midlands in early June 1910

Authors


A very thundery period affected England and Wales between 5 and 10 June 1910, the storms being especially prolonged and violent across the Thames Valley and the south Midlands on the 7th and 9th. Although over a century ago, it still seems worthwhile studying this event in some detail especially with the recent availability of reanalysis data. The localized extreme precipitation can be compared to events in the past decade such as at Boscastle (Burt, 2005) and Ottery St Mary (Grahame et al., 2009). All times quoted in this article are in utc and the counties are as in their post-1974 ceremonial designations.

Thunderstorms of 7 June 1910: many hours of intense electrical activity, evening deluge

A large anticyclone covering northern Scotland and southern Scandinavia in early June receded northeast and declined while a shallow area of low pressure over Biscay drifted slowly into southern Britain (Daily Weather Report, 1910). Despite a good deal of cloud, temperatures rose to 26–28°C in places between the 7th and 9th. Surface charts on 7 June show a light to moderate northeasterly airflow across the Midlands and the 500mbar reanalysis charts (Compo et al., 2011) indicate a large cut-off vortex northwest of Iberia with southwest England within its circulation. Increasingly humid air edged northwards into southern England: at Oxford (Oxford, Radcliffe Observatory, 1911) the 0900h dew point was 11°C on the 6th, 15°C on the 7th and 16°C on the 8th.

Unusually prolonged thunderstorm activity, with widespread lightning damage, occurred in a broad swathe from Surrey to Worcestershire on the 7th. The most severe downpours and local flooding occurred in the evening, affecting especially the north Cotswolds. The northeasterly surface wind may well have been undercutting the advancing warm and humid air, as happened in a more recent instance of a prolonged thunderstorm event in the area (Pike, 1994). Indeed, as noted by Mill (1911), the most severe storm downpours occurred almost simultaneously throughout this swathe between 1930 and 2330h, a feature indicative of a convergence zone.

The Henley Standard reported that the rumbling noise of the thunder up the Thames Valley prevailed all day. In Oxfordshire, Churchill School, near Chipping Norton, recorded a daily rainfall total of 108mm. At Stow-on-the-Wold (Quar Wood) most of the 90mm of rain recorded fell in two hours in the evening and in nearby valleys houses were flooded to the tops of doors on ground floors. The Evenlode Vale was reported to be in flood for miles. At Kingham (Fowler, 1913) the great thunderstorm began at 1930h with incessant thunder and lightning from 2000–2330h. The brass vane on the manor house was struck and crumpled. At the railway level crossing between Kingham and Churchill the crossing keeper was opening the gates for a train at the very moment that a tall elm tree was struck by lightning 20m away: this was one of several trees in the area which were struck and had bark ripped off. At Swerford, also near Chipping Norton, (British Rainfall 1910) the observer reported seven separate thunderstorms (three directly overhead) during this extraordinary day, culminating in a prolonged and alarming storm from 2045 to 2330h; thunder was heard during 12 hours of the day. Further south, at Wantage, three periods of violent overhead electrical activity occurred: around noon, 2000–2100h, and 2200–2300h when a large barn was struck and set alight.

There were indeed many damaging lightning strikes in Oxfordshire. Deddington parish church was struck, with damage to the pinnacles, and houses were struck in Kingham, Lyneham, Churchill, Deddington, Hook Norton, South Newington, Ducklington (Witney) and Boars Hill (Oxford). At Chipping Norton two hayricks were struck and set on fire. Farm livestock were killed by lightning at Great Rollright, Hook Norton and Salford. In neighbouring north Gloucestershire lightning dislodged the chimney of a house at Oddington and struck and fired a group of farm buildings at Moreton-in-Marsh.

The storms and extreme rain and hail on 9 June

Pressure was uniform on the morning of the 9th (Figure 1) and the area of warm, humid, stagnant air was conducive to the development of convergence zones. The Monthly Weather Report indicated a very shallow low centre tracking eastnortheast into the eastern English Channel, whilst the 500mbar reanalysis chart for 1200h indicates the cut-off cold vortex southwest of the British Isles within the axis of a deep upper trough just to the west of these islands. This corresponds closely to the scenario of the ‘modified Spanish Plume’ described by Lewis and Gray (2010).

Figure 1.

Surface weather chart for 0700h on 9 June 1910, from the Daily Weather Report (UK Met Office).

Storms of exceptional intensity occurred in the Reading and Oxford areas. At Caversham torrential rain commenced at 1215h, changing to intense hail between 1225 and 1300h. At Peppard 74mm fell from 1200 to 1330h and at Assendon 48mm between 1315 and 1415h. At Wheatley the deluge commenced at 1242h, while at nearby Waterstock the storm began at 1300h with a hurricane (sic) followed by the hailstorm which lasted from 1315 to 1415h; rain continued until 1530h though most of it fell before 1500h. At Pyrton Hill, pioneering meteorologist WH Dines wrote thunder was first heard soon after noon and was practically continuous from 1230 to 1600h, very severe from 1400 to 1500h. Nearby, three trees were struck by lightning. The storms appeared to move slowly from south to north. Rain (35mm) fell from 1352 to 1620h with 20mm coming between 1400 and 1430h (Mill, 1911). At the Radcliffe Observatory, Oxford, heavy thunderstorms were reported from 1200 to 1600h.

Two independent rain gauges at Wheatley, Oxfordshire, recorded over 125mm of rain on the 9th: School House with 132mm and Holton Cottage with 127mm, the former including a carefully measured 110mm in one hour. However there are caveats to both readings since the former was from a non-standard gauge and the latter from a gauge evidently in an unorthodox exposure (Mill, 1911). The nearby standard gauge at Waterstock recorded 100mm but it had become completely choked by hail (indeed it overflowed) and it is estimated that the true fall was close to that of the Wheatley recordings. Another station at Woodperry registered 81mm. The consistency between the Wheatley readings and the support derived from neighbouring rainfall stations (NB rainfall footprint in Mill, 1911) do indicate that the peak recorded figures at Wheatley were very close to the truth (see also Jackson, 1979). Moreover, with reference to the Hampstead storm of 14 August 1975, Tyssen-Gee (1976) indicated that about 30% of hard hailstones of 20mm diameter were likely to bounce out of a standard rain gauge. While the most destructive hail was at Waterstock, marble-size hailstones were also reported to have stripped foliage in Wheatley. An even more extreme rainfall reading of 140mm, of which 125mm was measured between 1230 and 1330h, was reported from Kidmore End, just north of Reading. However, this rain gauge was also non-standard and the reading has less support from adjacent stations (e.g. Peppard and Caversham noted above). Nevertheless the accounts of runoff reported in British Rainfall indicate a very significant storm event.

The flood and damage at Wheatley is described in the Henley Chronicle of 17 June 1910. Flowing currents were 45–90cm deep in most streets of the village, the water carrying earth and debris washed down from the surrounding small hills as well as many household goods, and also live ducks and geese! The main street became a raging torrent within 30 minutes with many houses flooded to depths of 45–90cm. One woman was rescued from drowning. In Crown Street, where water was 180cm deep, tenants in one dwelling escaped via a bedroom window. Water even damaged a Dutch clock hanging near the ceiling of the house! A wall was washed away at the west end of the village, a 180cm gulley was scoured in the meadow turf and a hayrick weighing nearly two tonnes was shifted several yards by the water. The village Post Office was inundated, cutting communications. The Oxford Journal Illustrated (1910) published photographs showing the peak flood-tide marks in Crown Street (with a resident pointing to one at 180cm), the displaced hayrick and damage to roads which had been under 30–60cm of debris. In Waterstock, a deluge of hail and water raced through the village and the houses; a large elm tree was uprooted by the force of water. Motor cars were stranded in floods at Great Milton.

The various observations of hail all indicate stones of 20–35mm diameter: at Caversham (Reading), marble to large walnut size (c. 20–35mm diameter); at Assendon (Henley), half an ounce (corresponding to about 32mm diameter) and at Waterstock, walnuts (20–30mm across). The hail wrecked glasshouses and frames at all three localities and at Waterstock hailstones were piled in broad mounds 90–120cm deep near obstructions with a level depth of nearly 10cm on open ground. The glass dome on Waterstock House was shattered by hailstones and plants were reduced to bare stalks with the complete devastation of vegetation and glass in the Rectory garden, as described in a letter in British Rainfall 1910 which was also reproduced in The Times.

Numerous incidents of lightning damage were reported, some involving serious injuries. In Caversham a female teacher was struck in her home at the time the local Post Office was hit and she suffered scorched hands and loss of movement. In Wheatley another woman was struck and lost the use of her arm for several days. In Farthinghoe, north Oxfordshire, a man was struck in a horse-driven cart: he was thrown out of it as the startled horse fled. The famous windmill at Brill, Buckinghamshire, was struck with one of the sails torn off and broken into pieces.

Comparisons of 9 June 1910 storms with the severe storms of 17 May 1997

More recently, the Thames Valley was the focus of prolonged thunderstorm activity featuring destructive hail on 17 May 1997 (Webb, 1998). The surface, and 500mbar (Figure 2), charts for 1200h on the 17th are both similar to those for the 1910 Oxfordshire storms, with a deep upper trough just west of Britain and Ireland tilted northwest to southeast and a very slack surface area of low pressure across central and southern England and Wales. In the south, early low cloud on 17 May 1997 broke up and temperatures rose to 23–27°C: a convergence zone became evident between light east-to-northeast winds across the Thames Valley and rather more southerly winds nearer the south coast.

Figure 2.

500mbar analysis for 1200h on 17 May 1997, data provided courtesy of ECMWF-ERA Interim Analysis Geopotential 500 (http://data.ecmwf.int/data/, accessed 23 October 2010).

Explosive cumulonimbus development occurred from 1300 to 1400h, initially over the north Wessex downlands but later from just west of London to east Somerset, storms then moving northnorthwest. Widespread hail, with stones 10–20mm in diameter, fell between Whitchurch, north Hampshire, and Wheatley, Oxfordshire. Hailstones of 40mm diameter were repor-ted at Thame, Oxfordshire, and 50–65mm diameter in villages just north of the town on the Oxfordshire/Buckinghamshire border (e.g. Long Crendon and Chearsley). There was extensive damage to greenhouses, windows, car bodywork and even roofing tiles. At Woburn, Bedfordshire, 86mm of rain fell in about two hours with hail 25–38mm across, while tornadoes occurred at Towersey, near Thame, and at Wootton, near Bedford.

Concluding remarks

Two remarkable thunderstorm episodes affected Berkshire, Oxfordshire and adjoining counties on 7 and 9 June 1910. Excep-tionally long periods of electrical activity were observed on the 7th while severely-damaging hailstorms were a feature on the 9th, with intensities of at least H4 on the TORRO international scale (TORRO, 2008). Daily rainfalls of 100mm or more occurred in Oxfordshire twice in three days. This threshold has only been reached in this county on three standard (09h-09h) rainfall-days since then: 10 July 1968, 26 May 1993 and 20 July 2007 (on one other occasion, 23 June 1960, over 100mm was recorded within five hours (at Duns Tew) but this was split between two standard rainfall-days (Pike and Webb, 2004)). Collective evidence from the Wheatley area indicates that at least 125mm fell on the 9th, of which over 100mm came within one hour during the peak of the thunderstorm. The latter is comparable to the Boscastle storm of 16 August 2004 (Burt, 2005) when the peak hourly rainfall extrapolated from measured falls was 95–100mm (estimates from radar indicated an epicentre maximum of approximately 125mm in an hour). The violence of the rain and hail storms on 9 June 1910 can be attributed to destabilisation of a very humid air mass by surface heating, local convergence and general ascent ahead of a major upper trough.

Acknowledgements

Thanks are due to Steve Jebson, Met Office Library, for copies of Daily Weather Report charts and the Monthly Weather Report for June 1910. The 500mbar charts the author researched for 1910 came from the Twentieth Century Reanalysis Project data set, which is supported financially by the US Department of Energy, the Office of Science Innovative and Novel Computational Impact on Theory and Experiment (DOE INCITE) program, the Office of Biological and Environmental Research (BER) and the National Oceanic and Atmospheric Administration Climate Program Office. It can be accessed at http://www.esrl.noaa.gov/psd/

Ancillary