Herbert Wood c.1935

To begin with the Humber.

There are certain facts about the River Humber, the estuary for the combined waters of the Ouse and Trent, of which most of us are more or less aware: that it drains one-fifth of England, from Birmingham in the south-west to Richmondshire in the north; that it is very shallow; that its tides run fast; and that it is abnormally muddy. All of these features, to some extent, have always affected not merely the technical aspects of navigation on the lower Ouse – by which I mean the river from Trent Falls to York – but also that less tangible factor: the mentality of those that travel on it. The faltering of the spirits so graphically described by Herbert Wood, an early member of the Humber Yawl Club, who of course was out there trying to enjoy himself, has been experienced countless times by those intent on the more important matters of fishing, warfare, trade, or actual survival. But what is it that has made our local river system such a difficult one?

This is not an easy question to answer: assuming, as it does, that life on the river has remained virtually unchanged. This is clearly not the case. Minor adjustments in sea level, major programmes of land reclamation from the early mediaeval period onward, and unrecorded climatic change have all contributed to a modern river regime which must differ in certain important respects from that known and exploited two or three thousand years ago. Nevertheless there are three factors which I believe to be so overwhelmingly powerful in their influence that they are worth investigation: the rapidity of the tides; the vast burden of silt carried at all times in the lower Ouse and Humber; and the often severe seasonal flooding.

Tides

Casual visitors to the seaside naturally notice that the tide goes up and down; visitors to a tidal estuary cannot fail but notice that it also moves sideways. In the Humber and Ouse it does this with useful regularity but also with extraordinary violence.

The pulse of water that brings high tide to the ports of the north-east coast is generated in the Atlantic; whence it travels around the north coast of Scotland, and down the east coast of Britain, before meeting a similar pulse which has moved up the English channel. The ebb tide returns by the same path; at least so far as coastal waters are concerned. This means that the actual time of high water on the north-east coast becomes a little later the further south one travels. High water at Aberdeen at (say) 0620 is followed (simplifying matters a little) by high water at Whitby at about 0910, at Bridlington at about 0930, and at Spurn at about 1020. The tidal wave has travelled 260 miles in 4 hours, at an average speed of about 65 miles an hour. This should not be confused with the speed of the tidal streams; except around headlands, where spring rates of 3 knots are not uncommon, they are slight on this coast. In Bridlington Bay even spring rates rarely approach 1 knot.

The tide then sweeps into the estuary, taking the shortest possible route and consequently gouging a deep channel – the deepest water for miles around – only a stone’s throw from Spurn Head. Things begin to hot up. High water Hull that day will be at around 1100, followed by high water at Trent Falls (the junction of the Ouse and Trent with the Humber) at 1130, and at Selby at a little after 1300. High water Naburn, the present head of the tideway, will be at 1500, while by extrapolation high water York, before the construction of Naburn weir in 1757, would have been at around 1545; high water Poppleton (an unlikely concept, but that sleepy suburb was declared in 1698 to be the tidal limit) would have been a few minutes later. But while, due to the retarding effect of a narrowing estuary and its shallow bed, the average speed of the tidal wave between Spurn Head and Trent Falls has dropped to about 25mph, and between Trent Falls and Naburn to about 12mph, the length of the flood has also decreased. At Hull the incoming tide runs for nearly six hours; at Selby, for only two and a half. The shorter time taken by the flood means that 1,850,000,000 litres of water has to run with increasing velocity; resulting in some truly astonishing spectacles. In former times, before the construction of training walls at Trent Falls after 1920, one of these was the Aegre, or bore; which remains a distinctive feature in the Trent and one which can still toss barges and coasters about as far up-river as Gainsborough. On the Ouse at Selby the flood tide, at springs, runs up at speeds of around 8 knots; similar rates are achieved at Trent Falls. The whirlpools and counter-currents generated as these bigger tides pour around difficult areas such as Howden Dyke, Asselby, and the sharp bend immediately above Selby, still occasionally lead to what is locally known as pilot-assisted collision; and they probably always have. At various points on the lower river, just to add to the confusion, the ebb and flood currents run in different channels, with a shoal between them, and the height of the bed in each of these channels also varies seasonally. A large-scale if relatively static example of this arrangement is to be seen at Howden Dyke island.

After a very short period of slack water the gentler ebb begins. Like high water, low water occurs at different times on the river with the rather remarkable outcome that it is low water at Selby at precisely the same time that it is high water at Hull. Since the length of the flood tide decreases with distance from the sea it follows that the length of the ebb must increase if the tidal cycle, of 12.4 hours, is to be maintained. This leads to an ebb at Selby of about ten hours; and also to a very considerable difficulty with the bed of the river, as will be seen.

The sea state generated by these fast-flowing tides in the lower Ouse and Humber can be remarkably bad – worse, in a sense, than anything experienced outside the estuary; at any rate under similar weather conditions. Wood’s experience off Hull is not exceptional:

John Taylor 1662

Taylor was being rowed from Trent Falls to Hull on the ebb tide (quite a big one – his boat achieved seven knots over the ground) and against a brisk easterly, as he is careful to point out. These are precisely the conditions which generate the chilling hollow seas, of remarkably short wavelength, for which the Humber is famous. These seas are killers; and must have given the local builders of open boats much pause for thought. Certain craft are clearly unsuitable for the Humber under any but the most settled conditions. The Ferriby boats, for example, were found beside the river; but there must be countless more like them beneath it.

Albert Strange c.1900

The lower Ouse also has its fair share of wind-over-tide problems – particularly at Trent Falls and at Sandhall Reach (just above Goole) where a westerly gale can make life difficult on the flood and on the ebb respectively. But most seamen in the days before steam would probably have breathed a sigh of relief once they reached the relative calm of even the lower reaches of the Ouse and Trent.

On waters such as these the skippers of boats powered by sail and oar had but two options: they could either go when they felt like it and arrive when they got there, or they could plan a passage to use the tides. Before charts and tide tables became available this demanded considerable local knowledge. For example, it is at the present time impossible to leave any of the upper Humber havens, including Brough, a former Roman naval base, much before local HW –2 hours. A passage from Brough to York would therefore begin either with a whole ebb spent at anchor outside the haven, followed by a departure on the first of the flood, or with a later departure and the need to sit out an even longer ebb somewhere higher up the river. In times of emergency this must have led to logistical nightmares – it would not be impossible, for example, for an entire Roman fleet to be stuck on the mud at Brough while ships crewed by those entertaining different ambitions for the province cruised past at high speed on their way to York. Once past, it would be impossible to catch them. Later, at the upper end of the tideway, shipping would, at some point, meet the following ebb: since no unpowered vessel can outrun the flood. This would inevitably mean a long wait at anchor; it may also explain the otherwise rather odd and ultimately inconvenient decision of Harald Hardrada to leave his ships at Riccall when assaulting York, immediately before the battle of Stamford Bridge, in 1066; he simply had no alternative.

Travel downstream from York was not necessarily any easier. Shoals immediately below the city had to be passed, and passed early. By the time of high water at York the tide was already ebbing fast at Naburn and depending on draft this might therefore require a departure from the city during the flood. A ship travelling at a speed of six knots over the ground would meet the flood tide somewhere in the vicinity of Howden Dyke (where the Bishop of Durham owned a strategically-placed staithe); here it would be necessary to moor up for three hours or so to let the flood run up before the journey could be resumed. The next leg would end with the onset of another flood tide in the neighbourhood of Hull, and a further period at anchor of five or six hours depending on the size of the tide. Only on the third ebb would it be possible to leave the Humber. Strong ground tackle and a deck of cards were clearly prerequisites for this kind of voyaging.

Silt

It has been calculated that at any one time, during the winter, there is a total suspended load in the Humber of over three million tonnes of sediment. Since, as described above, the flood tide dominates, this means that there is a net input of sediment, derived from the crumbling Holderness coast, into the river system; which as a result is steadily decreasing in volume. One estimate suggests that 63,400 tonnes of sediment are deposited in the Humber each year – sufficient to have created, over the last 6000 years, a deposit which varies from 2m to 20m thick. Where does all this material end up? And how is it possible to remove it to improve navigation?

In the sixteenth century the first question became easier to answer as the keels of deeper and heavier ships repeatedly ground their way onto the river bed. By the middle of the century complaints about the Ouse– and particularly about the stretches immediately below York – were coming thick and fast from those in the shipping business. In 1544 the Corporation noted that its 35-ton sailing barges – clearly sea-going vessels, since one of them had been pressed to Boulogne – were unable to navigate reliably from York to Hull due to lack of water. Generous contracts to clean the river immediately below York – usually by dredging, or ploughing, the shoals – came to nothing as the primitive tools of the river-improvers failed against the tough clays of the river bed. The ears of James I himself were assailed, on Ouse Bridge in 1617, by the voice of the River: a costumed figure informed the monarch that his money was urgently required to return the city to her mediaeval glory in the matter of river trade. Silting of the river, of course, was not the only problem: it was compounded by the urgent need of the merchants of York to compete more effectively in larger vessels. But by the late C17 it had been reluctantly accepted that unless something were done the laden ships and barges of the time would only be able to reach the City twice a month – on spring tides.

The problem was a fundamental one as it was caused by the nature of the tidal regime. The fast-moving flood tide, laden with silt from the estuary, is capable of carrying material for a considerable distance up-river. The much longer and gentler ebb, on the other hand, lacks the power under normal conditions to remove it. On each tide the bed level of the river therefore rises a little. The current itself assists the process. The fastest-moving water is at the surface of the river while the speed of the current reduces as the bed is approached. As particles begin to descend they find their downward progress is made easier and deposition increases. This process is naturally at its greatest in the slacker sections of the river, such as the insides of bends, where considerable and dangerous nesses form – examples abound, but perhaps the greatest is at Faxfleet – although the pattern of ebb and flood channels can result in mid-river shoals which have a tendency to become islands. Examples can be seen at Asselby (now a wooded promontory following further changes in the 1930’s) and at Whitton, in the upper Humber.

Improvement of the upper Humber, in particular, has never been viewed as a serious proposition. The underlying causes of the frequent channel changes are not understood and little pattern is discernible; the most that can be done at the present day is to provide regular navigational information concerning a ship channel which at low water off Whitton Ness is no more than 0.5m deep. But the upper Humber has a tidal range of about 6.7m at springs; and with proper attention to times of passage very large ships indeed can make their way up to Goole and the Trent ports with only occasional moments of embarrassment. The lack of human interference, and of subsequent industrial development, has helped the upper part of the estuary to retain an extraordinary, remote, beauty.

The Ouse is a different matter entirely. Below York the silt piles up on the bed of the river throughout the summer. Since before 1757 the rise of even a spring tide seems only to have been of the order of a metre or so –the surveyor Perry, in 1727, found only two feet six inches at HW springs at Ouse Bridge – the city was effectively cut off from all but lighter-borne trade. The evidence for this is conclusive since it is based on a fine survey of the river, carried out at the request of the Corporation, made by Thomas Surbey in 1699. The Corporation was aggrieved, to say the least, by what they alleged to be the loss of up to two feet of tide after the collapse of the sluice at the mouth of Dutch River in 1688. Surbey – an obscure figure, but evidently a very tough and determined one – spent a week from 5^th May on the river between York and Hull and his journal, preserved in the City archives, records his journey and his work in great detail. His instructions were to find a way of restoring the tide to York and it seems that the preferred method was to make a series of huge artificial cuts across several of the large bends below Selby. By patient levelling Surbey was able to demonstrate that the additional flow of tide gained by this expensive solution would give York no more than 4" more water on spring tides; nowhere near the seven feet or so that the city fathers hoped for. His soundings also show that at low water at Fulford the river was only eight inches deep, and the rise of tide about three feet at springs and a miserable two inches at neaps. Further shoals near Cawood and Selby were also noted although with a much greater rise of tide in the lower sections of the river navigation was less of a problem. Surbey’s advice was that the Corporation should provide a pound lock at Naburn; and sixty years later, with the benefit of several more very expensive opinions all telling them the same thing, they finally acted upon it.

Flooding

In recent years minds have once more been bent to the question of river improvement. Not, sadly, to ease navigation – the technology is with us, but the ships are not – but to reduce the impact of large-scale flooding. For, as Baron Duckham remarked, the Ouse does not so much drain the Vale of York as water it.

The question of ancient flooding is a vexed one, not to be lightly entered into here. Wacher found convincing evidence for an increase in the mean level of high water at Brough in the late Roman period, and postulated, reasonably enough, that silting as a result of increased sea levels may have led to the abandonment of the Roman port in the middle of the fourth century. But with care boats drawing up to 2m can still use the upper Humber havens; leading one to suspect that military considerations may have played a larger part. Any discussion of late Roman or later flooding in York must take account of the fact that the river was tidal, and that a flood might therefore consist more of a series of pulses than of a lengthy period of high water. Today, there comes a point in any major flood when the river below Naburn weir makes a level with the river above it. At that moment the Ouse in York is once more in direct communication with the North Sea; it becomes a tidal river. The engineers of the Environment Agency believe that the influence of the tides in the lower river makes itself felt as a periodic peak in flood levels in York; and there is no reason to suppose that things were any different two thousand years ago. From the perspective of a river pilot such catastrophic flooding is nothing but beneficial since it removes, in days, shoals that have grown up over months. Minor flooding can also help deep vessels to move on the river; it seems certain that much of the lead brought from Boroughbridge, for example, must have been carried in this way. If sufficient horsepower is available then heavier ships than usual can also be hauled up-river.

The seasonal nature of the flooding is reflected in seasonal change in river bed levels; reference to two floods, one in 1892, and the other in November 2000, will best illustrate this. In 1892, over the weekend of October 15^th-16^th, the river rose to 16 feet above Mean Summer Level at the Lendal Bridge gauge. A report on this flood was submitted by Mr A Creer, the city engineer, to the Ouse Navigation Committee of York Corporation and was subsequently published in the York Herald. Creer examined earlier flood levels and calculated that in 1625 the river rose to 17 feet 10 inches above MSL; in 1636 to 17 feet one and a half inches; in 1763 to 16 feet 3 inches; and in 1831 to 15 feet 7 inches. Even allowing for a certain inaccuracy in those figures that pre-date the establishment of Mean Summer Level at Naburn weir these results encourage the view that large-scale flooding is not a recent phenomenon. In 1905 Creer went on to produce a longitudinal section (Fig. 1) which places the bed levels of the river, between Naburn and Selby, in the spring following the 1892 flood, against those of August 1905: the difference is startling. Assuming, reasonably enough, that bed levels in August 1892 were comparable to those in August 1905, the survey shows that the October flood stripped the river bed of from six to ten feet of accumulated sediment over a distance of thirteen miles.

In June 2001 a similar survey was carried out by British Waterways following the flood of November 2000; when, on the 4^th of the month, the river level peaked at 5.35m above MSL. The same dramatic results were recorded. Generally between two and three metres of material were removed but in some areas the bed was lowered by up to four metres. This is an enormous quantity of material – but all of it, in archaeological terms, redeposited, and almost none of it more than a few years old (Fig. 2).

The archaeology of the river Ouse is still little studied. Work to the banks, for example, either by engineers or by the river itself, clearly has the potential to reveal significant finds. Immediately below Cawood bridge, on the right bank of the river, the wreck of a wooden ship, visible at low water springs, is being steadily disinterred as the channel moves inexorably into the outside of a bend. All that can be said of this vessel is that it probably predates the construction of Cawood bridge itself, when surveys place the right bank further to the east; and the bridge was only built in 1871, so the ship may not be ancient. However there may be others like it – this region has after all produced some of the oldest boat finds in Europe. Dredging, should it ever take place – which is unlikely, since it has never successfully competed with summer aggradation of the river bed – does not appear to be as potentially damaging as it may seem. It would only be sensible to dredge the river in the late summer, and as noted above few deposits in the upper couple of metres of the river bed would be more than a year or two old. A winter flood is nature’s own dredger; and a mighty effective one too.

The anticipated effects of global warming have already led to the raising and strengthening of flood defences along the Humber and lower Ouse. At mean high water springs the Humber at Brough reaches 4.2m AOD, while large areas of farmland to the north of the river are at 2m-4m AOD. Much of Hull itself lies at around 3m AOD. Records at Immingham, a standard tidal port, indicate that the extreme water level there rose by about 7mm a year between 1910 and 1972. If the trend continues then mean sea level in the Humber will stand at about 4.99m AOD by the year 2100; and we shall all have rather more to worry about than the occasional winter flood.

©Colin Briden 1990

Tide predictions 1066 Tidal York The tidal river Ouse