YORK AS A TIDAL PORT
by Colin Briden
Almost a generation of York archaeologists has now been puzzled by the discovery that Roman waterside structures in the bottoms of their trenches lie below the Mean Summer Level of the River Ouse. Interesting explanations for this phenomenon have appeared in print such as: the river has moved; the river was wider and shallower; things have changed since those days. In fact the immediate explanation is more prosaic. In 1757 the construction of Naburn weir fixed Mean Summer Level at 4.44m OD; this was raised in 1835 to 4.90m OD and again, in 1876, to 5.0m OD; at which level it now remains. The construction of the weir, prompted by a chronic lack of water in the upper Ouse, has had the effect of isolating the City from the tidal regime of the lower sections of the river. This paper is an attempt to reconstruct that regime as it existed in York until the C18 and, more importantly, to establish a set of tidal constants which can be applied to the analysis of waterfront deposits.
Tidal factors were of the utmost importance to the commercial life of the City.
Such is the strength of the tides in the lower Ouse that in the days before steam, at anyone point on the river, at anyone time, a boatman had but two options: to go with the flow, so to speak, or to wait for the turn of the tide; and, as any York-based voyager who has sat out a nine-and-a-half hour ebb at Selby will know, waiting can be a tedious business.
2. The Nature of the Tides in the Humber and Ouse
Any consideration of the upper reaches of a tidal river must begin with its end the estuary. At the end of the last ice advance the drainage pattern of East Yorkshire, much altered, began to reassert itself Initially, with vast quantities of water still locked up in continental ice sheets, the sea-level remained low, leading to considerable down cutting by rivers draining into the North Sea. This, of course, included the Humber, the bed of which at this period probably lay up to 25m below its present level. The subsequent eustatic rise in sea-levelled to the deposit and accumulation of clays, silts, sands and gravels in these river valleys, burying forests, which had formed alongside the rivers, to form bands of peat. Fresh water peat dating to about 6,900 BP is known at about - Om OD beneath Hull; a similar deposit at - gm OD below Immingham iR dated to C 6,700 BP. Peat and submerged forest producing Neolithic finds near the modern low tide muk along the Lincolnshire coast show that sea-levels were still relatively low prior to about 4,000 BP but that by 3,000 BP the sea had reached a level maintained, 'with small variations, until modern times.
Since sea-level determines the long profile and tidal regime of a river it may also be assumed that in the very broadest terms the tidal Ouse has been behaving in mud the same way for the last three thousand years.
The tides are the outcome of the gravitational pull on the oceans exerted by the moon and the sun: a combination of factors, of which the diurnal rotation of the earth is the most important, produces high water every 12.4 hours at a point on the coast. Twice in each lunar month the combined effect of moon and sun working together raises tides which are both higher and lower than average. These tides Mean High Water Springs, and Mean Low Water Springs (MHWS and MLWS) occur about 2 days after Full Moon and New Moon. For any inland port short of water - and as will be seen this is a description which almost certainly applied to York - these spring tides are vitally important. They allow heavily laden vessels to reach the port, unload, and then return, empty and less deep in the water, on a succeeding neap tide. The skipper of any vessel which failed to make port on the top of a spring tide would find himself without sufficient water to proceed on every subsequent high water for almost two weeks. The tides which surge in and out of the Ouse and Humber are generated not in the North Sea itself - it is too small- but in the Atlantic.
These Atlantic tides pass around the North of Scotland and down into the North Sea: as each passes the Humber mouth, a pulse of water is sent into the estuary. The form of the tide at the Humber mouth is that of a wave: a wave with a length of about 400 km, but a height of only 6m or so. As this tidal wave enters the river the ebb current is first checked, then reversed, until high water has passed; then the ebb begins. Thus every spot on the tidal river sees high water in sequence: if HW Spurn is at, say, midnight, then HW Hull will be at 0100, HW Selby at 0320, and HW Naburn at 0500. Since Naburn is about 120 km from the sea it follows that HW moves up the river at an average speed of about 25 km per hour. Like all waves, however, the speed of this tidal wave is reduced by the shoaling of the water: thus at Spurn the duration of the rise of the tide - the time between LW and HW - is 6 hrs and 20 mins, but at Goole it is only 3 hrs and 30 mins 4Ild at Naburn a mere 2 hrs.
The wave-like nature of the tides in the river is well illustrated by the fact that HW Hull occurs at the same moment as LW Selby, and that HW Naburn comes about an hour before LW in the North Sea.
This demonstrates how the rise of the tide becomes shorter as it travels up-river and also, therefore" how the ebb lasts for correspondingly longer. Most importantly, extrapolation of the times of HW gives us the beginnings of our table of tidal constants for the port of York:
HW York Duration of Rise = HW Hull + 4 hrs 45 mins 1 hr 45 mins
The next question to be considered is: what was the level of MHWS in York before the construction of Naburn weir?
3. THE TIDAL REGIME AT YORK
The speed with which the tidal wave carries High Water up the river should not be confused with the velocity of the currents themselves. Nevertheless these currents are fast: in the lower Humber the incoming flood tide may reach 6 knots, while at Selby bridge a spring flood tide boils in at around 8 knots - nearly 15 km/hour. With the incoming flood comes tens of thousands of tons of sediment, much of which falls out of suspension during the gentler ebb phase of the tidal cycle. Thus the bed of the river is built up during the summer, when water-levels are low, and scoured out during the winter and early spring when freshwater levels are at their highest. The superimposition of this annual cycle of deposition on the monthly tidal cycle makes any attempt to reconstruct water depth in the tidal river at Ouse Bridge a rather difficult thing to do. Nevertheless it is possible to arrive at some conclusions as to level AOD of the river itsdf,at MHWS. The following table shows river heights at MHWS for a number of places on the tidal Humber and Ouse:-
The consistency in these figures encourages extrapolation to York, five miles upriver from Naburn; which in itself is seven miles up-river from Cawood, thus:-
York MHWS 3.4m AOD
The former tidal range - the difference in height between MHWS and MLWS - is harder to assess. Naburn is currently around 2m, although the presence of the weir may have distorted this figure upward by ponding back the tide. At Cawood the tidal range is 3m; taken together these measurements suggest that a tidal range at York of around I.5m may not be wide of the mark. The table of tidal constants for York is, therefore:-
HW York Tidal range at Springs Duration -of Rise MHWS= HW Hull
4 hrs 45 mins I.5m
1 hr 45 mins 3.4m OD
Spring tides themselves vary rhythmically, for astronomical reasons, and more sporadically for other reasons, discussed in the following section. On many occasions MHWS will have been significantly higher than this. The gentler neap tides, falling midway in the month between spring tides, have lower current velocities and a smaller tidal range; thus low water is higher and high water is lower than on springs. The tidal range at neaps at Naburn Lock is currently around 1.0m, occasionally less:
under certain circumstances -increased fresh water in the river, for example - high water neaps would hardly have been noticed at York.
4. FACTOR INFLUENCING THE HEIGHT OF HIGH WATER AT YORK
All the figures quoted for high water so far have been mean figures; we must now examine some of the factors which may in practice have influenced tide heights in the past. Some of these are so long term as to appear permanent, at least over a single lifetime; others are purely ephemeral.
It will be clear by now that the most important single factor governing the height of the tide in York was the height of the North Sea at high water. There is some evidence that sea levels in the Humber estuary and North Sea have fluctuated since 3000 BP and this would clearly have a profound effect: if, in the Roman period, sea level was appreciably lower then tides may have failed to reach the City more than twice a month.
(ii) Channel Width
Evidence from excavation in the City suggests that the Ouse in the Roman period was considerably wider than it is today: this observation is occasionally coupled with the idea that the river was therefore 'shallower'. However, this statement has little meaning in the context of a tidal river, since not low water but high water was the factor governing shipping movement; and high water depends less on channel width than on sea level. Between 1851 and 1961 the width of the Humber between Brough and Hessle varied by as much as 30% either side of average, but with no marked effect on the tidal regime.
Encroachment into the river, however, would have the important effect of increasing current velocity and, therefore, of deepening the bed. This may have been the purpose of Roman encroachment into the Ouse, recorded from at least two sites in the City. It is also likely that the drainage and embankment of such vast wetland areas as Walling Fen, north of Trent Falls, from the C13 on also increased the velocity and duration of the incoming flood tide by channeling the river and reducing the sponge-like effect of the Humber head marshes. Mean high water levels, however, were probably only slightly affected.
(iii) Ephemeral Factors
On a day-to-day basis High Water levels in the City would be quite noticeably altered by such factors as strong winds in the Humber and heavy rain in the upland catchment areas of the Ouse. A south-westerly gale can delay high water in the river by as much as half an hour; an onshore gale will increase the height of the tide considerably. Conversely a high level of freshwater in the river would have masked any tidal effects and, by eliminating the flood tide, would have made the City approachable by river, albeit against a strong downstream flow.
5. EVIDENCE FROM EARLY SURVEYS AND FROM CONTROLLED ARCHAEOLOGICAL EXCAVATION
The tidal constants we have derived for the river at Ouse Bridge can be checked against two other kinds of information: soundings taken in the river before 1757, and recorded levels at which Roman riverside structures have been found during controlled excavation close to the river, both in York and elsewhere.
A. EARLY HYDROGRAPHIC SURVEY DATA
In 1688 the sluice at the mouth of Dutch River, below Goole, collapsed. This allowed the incoming flood to move freely up the River Don as it has ever since and also, claimed the Yorkshire antiquary, de la Pryme, deprived York of two feet of water on spring tides. Since the sluice had been built at Vermuyden's own expense no-one could be found to repair it and it was with a sense of real grievance that in 1699 the Corporation invited Thomas Surbey to 'view' the river between York and Hull Roads. Surbey began on 5th May and despite the temporary loss of his boat at Howdendyke was back in York to report on 22nd May. At low water, in Fulford, he found 8 inches of water and he noted a tidal rise of no more than three feet in that stretch of the river. He noted also that ships drawing seven feet could reach Shenton's Pit (about five miles below York) but no further. A second set of soundings was produced by J Perry in 1727: these confirm Surbey's findings. Both sets, metricised, can be tabulated as follows:-
TABLE 3 TIDAL RANGE AT MHWS Surbey Perry 1690 1727
Perry also remarks that at neaps the rise of the tide at Fulford was '2" - or not at all'. If, by the late C 17, the tides really did run two feet lower than before - and there seems no reason to doubt it since it was a source of universal complaint - then we may be justified in adding this figure to that given by Perry to derive a tidal range at earlier periods at MHWS in the City of about 1.40m. This is close to our projected figure of 1.50m. The difference is more than adequately accounted for by the complicating factors in the tidal regime described above.
B. LEVELS OBTAINED FROM THE CONTROLLED EXCAVATION OF ROMAN WATERSIDE STRUCTURES
a) The Ouse and Humber
(ii) however, that the level of the tides postulated by Wacher is a little lower than that of modern MHWS and so it is not likely that mean levels were above 3.4m OD in York even during this late Roman transgression.
58-9 Skeldergate, York This riverside site within the colonia was excavated by YAT in 1973-5. A machine-cut trench located a riverside road: unfortunately levels are missing from the published section but the road seems to have been above 5.00 OD with the top of the camber at around 5.80m. Thus the road was at least 1.5m above projected MHWS.
Rougier Street, York A site within the Roman colonia, excavated by YA'f, revealed a natural stream at about 5.00m OD and the surface of a Roman road at 6.5m OD.
.North Street, York Roman deposits at this riverside site included timber and later stone structures, enveloped in river silts, which have been interpreted as the base ofrevetments or wharves:-
Timber slot and piles:
Late Roman Stone Structure:
1.95m OD - 2.39m OD 2.48m OD - 2.75m OD 2.99m OD - 3.15m OD
b) The Foss
Excavations here in 1950 revealed the 'old course' of the Foss at 3.35m -4.58m OD. The base for a crane was found at 5.26m OD: this was associated with a possible wharf and a double row of wooden piles in the bed of the river adjoining.
It will be seen that in general these levels accord well with our postulated level for MHWS at York of 3AOm OD. The rather higher level of (presumed) Roman riverside structures on the Foss is interesting and hints at the existence of a tidal barrage or flashlock at its junction with the Ouse.
What aspect then, would the river in York present to an interested observer on Ouse Bridge at almost any date before the construction of Naburn Lock in 1757?
Let us take as our example a tide 2 days after a new moon in September. This is an equinoctial spring tide, one of the lowest and highest tides of the year; after a dry summer the bed of the river has accumulated a substantial thickness of sediment by deposition out of the sluggish current. High water in Hull that day is at 0800 hrs; as we have seen, it is simultaneously low water in Selby. Between the North Sea and Hull the tide is ebbing; between Hull and Selby the water is rising. Between Selby and York the last of the previous tide is ebbing. At 1100 hrs it is low water at York possibly very low indeed, only inches deep. Almost imperceptibly the ebb slows and stops and the river begins to rise. With little freshwater to prevent it, the river begins to flow backwards, rising all the time. At half-tide, around 1130, the first heavily laden ships begin to appear in the river from Selby and the Humber, taking advantage not only of the favourable current but also of the fact that should they happen to touch on a shoal - the infamous clay hutts - the rising tide coming up behind will float them off. Equally heavily-laden ships are moving off down-river, trading the disadvantage of punching a foul tide against the need for deep water in the upper reaches of the river. At 1245 it is high water York, and the river has risen around 1.5m to 3.40m OD, in 1 3/4 hours. Almost immediately the tide turns, picking up speed to the point where sailing or rowing vessels cannot overcome the current. Any unfortunates who have missed the tide and run aground down river now have little option but to transfer some of their load to lighten ship and hope to make it on the next, lower, tide.
This was the pattern of life on the river in York for centuries; as it still is today in Selby, a tidal port. The water-borne commerce of both towns depended before the age of steam not merely on the existence of the river but on the powerful tides that moved up and down the river, carrying all before them. The pattern of these tides seems, in three thousand years, to have changed only in detail, adjusted here and there by changing sea levels, the raising of levees and embankments, and the draining of wetlands. If the Ferriby boats ever came to York, this is how they did it.
©Colin Briden 1990
Reproduced with permission
|Last updated May 2012|