St Osyth Gravel Pit And Holland-on-Sea Cliff

[TM 120 174] and [TM 211 166]

D.R. Bridgland

Highlights

Sediments at these two sites record the events immediately prior to and during the glaciation of the Thames valley, leading up to the diversion of the river. At both sites, gravels of the pre-diversion (Kesgrave Group) are overlain by sediments rich in outwash from the ice sheet that blocked the course of the river in central Essex and the Vale of St Albans. At St Osyth the sediments were laid down immediately upstream from the Medway confluence, whereas the gravels at Holland-on-Sea are of Thames-Medway type. The upper gravel at Holland-on-Sea demonstrates that the Medway was unaffected by the glaciation that blocked the Thames.

Introduction

The St Osyth and Holland-on-Sea GCR sites are situated near the southern edge of the Tendring Plateau, in the vicinity of Clacton-on-Sea (Figure 5.2). The St Osyth pit lies 6 km west of the coast, whereas the Holland cliffs are near the northern end of a 6 km length of erstwhile coastal exposure running north-eastwards from the West Cliff at Clacton, itself part of another GCR site (see Part 2 of this chapter). The St Osyth and Holland sections are both in the lowest of the four Low-level Kesgrave Subgroup formations recognized on the Tendring Plateau (Bridgland, 1988a; (Figure 5.2); see above, Introduction to Part 1). St Osyth lies upstream from the contemporary confluence between the Thames and the Medway, whereas Holland lies within the confluence area (Bridgland et al., 1988, 1990; (Figure 5.4)E).

The sites at St Osyth and Holland-on-Sea provide complementary evidence enabling the reconstruction of events in the lower part of the Thames basin during the Anglian glaciation, when the diversion of the river took place. At both sites deposits typical of the Kesgrave Group are overlain by later gravels, believed to have been laid down at the time of the Lowestoft glaciation (Bridgland, 1980, 1983a, 1988a; Bridgland et al., 1988). Considered together, the sites are of considerable stratigraphical significance, since they provide a basis for correlating the terrace sequence in southern East Anglia with the succession in the Middle Thames and the Vale of St Albans, which can also be related to the Anglian glaciation (Table 1.1) and (Figure 1.3).

Description

There have been relatively few descriptions of the Pleistocene deposits in this area, with the notable exception of the Clacton interglacial sediments. Wood (1866b) attributed gravels underlying the coastal district, between St Osyth and Clacton, to his 'East Essex Gravel', equating them with deposits south of the Blackwater estuary (see Part 2 of this chapter). The only available geology map (Old Series, Sheet 48) shows the Tendring Plateau largely covered by 'Glacial Gravel' and 'Glacial Loam', but with the patches of 'Post Glacial' drift fringing the valleys of the Colne and Stour and in the extreme south-east of the area, between St Osyth and the coast. The 'loam' comprises post-Anglian loess mixed with stones from the underlying gravels (Eden, 1980). Misinterpretation of this material has led to the unfounded suggestion that till occurs on the Tendring peninsula (Geological Survey, 1:625,000 Quaternary sheet).

The St Osyth site falls within an outcrop mapped as 'Glacial Gravel', separated from deposits to the east, which were mapped as 'Post Glacial', by the valley of a stream flowing into the St Osyth Creek. The 'Post Glacial' deposits extend to the coast, where they are synonymous with the 'Holland Gravel' of Warren (1923a; 1955). However, it is apparent from clast-lithological studies that the gravels on either side of the above-mentioned stream were formerly continuous (Bridgland, 1983a; Bridgland et al., 1988), despite the fact that they were classified differently by the Geological Survey. Indeed, Oakley and Leakey (1937, fig. 10) classified the St Osyth deposits, in common with all the gravels east of a line from Brightlingsea to Great Oakley, as fluviatile.

Warren (1923a, 1924b, 1933) had already interpreted the gravels of the Tendring Plateau as fluvial deposits and attributed them to the Thames, although at that time he was not aware of their considerable antiquity. The discovery of the Cromerian channel-fill at Little Oakley (see above, Little Oakley) led Warren (1940, 1955, 1957) to realize that these gravels were the products of Thames drainage prior to the diversion of the river into its modern valley. This anticipated the inclusion of these deposits by Rose et al. (1976) in the newly-defined Kesgrave Sands and Gravels, which they attributed to the pre-diversion Thames.

There are, in fact, significant differences between the Kesgrave Group deposits at St Osyth and Holland-on-Sea, but these represent downstream compositional changes within a single gravel formation. These early Thames terrace deposits are locally overlain by later gravels of a different type, which prove to be of considerable stratigraphical significance.

St Osyth Gravel Pit

Various commercial workings have exploited the spread of gravel to the north-west of St Osyth over the past few decades. The full thickness of the aggradational sequence is only preserved in a small part of the area, the upper horizons having been widely denuded as a result of later dissection. The GCR site occupies approximately the highest point on the outcrop and appears to preserve the most complete sequence. The lower and major part of the sediments here comprise up to 10 m of typical coarse, predominantly matrix-supported gravels of Kesgrave type. Palaeocurrent data from cross-bedded sandy intercalations indicate flow to the south-east, in keeping with the interpretation of the sediments as products of the early Thames, which flowed from the Colchester area towards Clacton (Bridgland, 1980, 1983a). This, the Lower St Osyth Gravel, is overlain by up to 3 m of sand, into which is channelled 1–2 m of gravel of a quite different character, the Upper St Osyth Gravel (Figure 5.14). The latter comprises fine gravel material scattered in a matrix of coarse sand. It contains a higher proportion of flint than the lower gravel, but a much smaller proportion of rounded pebbles reworked from the Palaeogene. Furthermore, the Upper St Osyth Gravel contains fewer exotic rock-types than Kesgrave Group gravels (including the Lower St Osyth Gravel) and those present show greater affinities to Anglian glacial gravels than to the underlying Kesgrave Thames deposits. In particular, the exotic suite in the Upper St Osyth Gravel includes Rhaxella chert, a rock derived from the Oxfordian of north Yorkshire that, in the London Basin and southern East Anglia (outside the Crag Basin), is present only in Anglian glacial deposits or, reworked, in post-Anglian sediments (Bridgland, 1983a, 1986b). It is uncertain at present whether the intermediate sand has closer affinities with the upper or lower gravel. It is hoped that an analysis of heavy minerals, presently being undertaken, will answer this particular question.

Holland-on-Sea Cliff

The Pleistocene sequence exposed in the Holland-on-Sea cliffs is closely comparable to that at St Osyth. Again there is a dominant lower gravel, c. 5 m thick, of Kesgrave type, although with a marked increase in southern clasts (Lower Greensand chert) in comparison with the Lower St Osyth Gravel. Such material constitutes 3–11% of the Lower Holland Gravel, an increase from a maximum of only 2% at St Osyth. A fine-grained, sandy upper gravel is again recognized, separated from the underlying coarser gravel by sands. However, this separation is less well-marked, with alternations of sand and gravel, of which the lower horizons have clast compositions transitional between the Lower and Upper Holland Gravels (Bridgland et al., 1988; (Table 5.2), sample 2B), possibly reflecting reworking of the former. The Upper Holland Gravel has been sampled over a wide area inland, as well as at the type section. At some localities it contains large proportions of southern lithologies, one analysis yielding nearly 32% of such material (Bridgland, 1983a; see below). The deposit is otherwise similar in composition and sedimentary characteristics to the Upper St Osyth Gravel.

Interpretation

The principal scientific interest at the St Osyth and Holland-on-Sea GCR sites is stratigraphical. Each is the type site of two lithostratigraphical units: St Osyth pit is the type locality for the Lower St Osyth Gravel and the Upper St Osyth Gravel, whereas Holland Cliffs provide the type section for the Lower Holland Gravel and the Upper Holland Gravel. The interrelations between these deposits and the palaeogeographical interpretation that has been determined from their clast composition provide an illustration, in north-east Essex, of the glacial interruption of Thames drainage that occurred, further upstream, during the Anglian Stage. This provides an important means of correlation with the sequence in the Vale of St Albans (see (Figure 1.3)).

The Lower St Osyth and Lower Holland Gravels together represent the lowest (and therefore the youngest) formation within the Kesgrave Group (Bridgland, 1983a, 1988a; (Figure 5.2)). They are separately named because they differ from one another in clast-lithological composition; the Lower St Osyth Gravel is a typical Kesgrave Group Thames gravel, but the Lower Holland Gravel contains a higher proportion of southern material. This change is considered to reflect the contemporary confluence of the Kesgrave Thames with the extended River Medway ((Figure 5.4)E; see above, Introduction to Part 1). The upstream limit of the contemporary Thames-Medway confluence area has been recognized, for each formation, from changes in clast-lithological content. This is best illustrated by differences in the ratio of southern material to quartz and quartzites (Bridgland et al., 1988, 1990; see above, Little Oakley). Kesgrave Thames gravels upstream from the Medway confluence are characterized by values for this ratio below 0.10, whereas the ratio increases progressively eastwards from the upstream edge of the confluence area (Bridgland et al., 1988; (Table 5.2)). The Lower St Osyth/Lower Holland Gravel has been sampled at various points between Fingringhoe [TM 042 202] and Holland-on-Sea, enabling a more detailed appraisal of these compositional changes than has been possible so far in any of the higher-level formations (Bridgland et al., 1990). This has shown the change from Lower St Osyth to Lower Holland Gravel to occur within the eastern extension of the St Osyth pit (TM 120 170; (Figure 5.14)), less than 0.5 km to the east of the GCR site (Bridgland et al., 1988).

The fact that the Lower St Osyth/Lower Holland Gravel is the lowest of the pre-diversion (Low-level Kesgrave) Thames formations suggests that its deposition closely preceded the diversion of the river, particularly since temperate sediments of 'Cromerian Complex' age are recognized within higher formations in the area. This led Bridgland (1983a, 1988a) to ascribe this formation to the Anglian Stage and propose a correlation with the Winter Hill Gravel of the Middle Thames. It may be significant that the 'Tendring Association', the soil unit believed to contain relict elements of the Valley Farm Soil (see above, Wivenhoe), has not been mapped on the St Osyth/Holland Formation (Kemp, 1985a), nor were the Valley Farm or Barham Soils recognized in exposures recorded at St Osyth by Rose et al. (1976). This may indicate that no temperate-climate interval separated the aggradation of the Lower St Osyth/Lower Holland Gravel and the Anglian glaciation, as is implied by the ascription of the former to the Anglian Stage. However, J. Rose (pers. comm.) has pointed out that palaeoargillic soils are developed at the land surface on both the Lower and Upper St Osyth and Holland Gravels and that these are difficult to distinguish from the Valley Farm Soil. They are, however, developed on dissection slopes and are independent of the original St Osyth/Holland Formation terrace surface, so they might be expected to be of relatively recent origin. They may be comparable with palaeoargillic and rubified soils developed on the Anglian till of central Essex (Rose et al., 1978; Sturdy et al, 1978).

There is strong stratigraphical support for the correlation of the St Osyth and Holland Gravel with the Winter Hill/Westmill Gravel of the Middle Thames from the interpretation of the two upper units at St Osyth and Holland-on-Sea. The Upper St Osyth Gravel, recognized at the type site and at Fingringhoe, is a fine sandy gravel with closer compositional affinities to Anglian glacial gravel than to Kesgrave Group Thames deposits. In particular, it is characterized by a low Palaeogene (rounded) flint content, a relatively large non-quartzose exotic component and the occurrence of Rhaxella chert, all features that typify Anglian outwash gravels but would be unusual in the Kesgrave Group (Rhaxella chert is found consistently in Kesgrave Group deposits only within the Crag Basin, where the lithology has been reworked from the Crag). The deposit lacks the nondurable, calcareous component of ice-proximal gravels associated with the Lowestoft Till; calcareous clasts would probably not have survived fluvial transport from the ice front, which lay c. 20 km west of St Osyth (Figure 5.4)F. The Upper St Osyth Member is therefore interpreted as a distal outwash gravel.

East of St Osyth, in the former Thames-Medway confluence area, the Lower Holland Gravel is overlain by another fine-grained gravel, similar to that at St Osyth in that it is relatively poor in Palaeogene flint and quartzose exotics. However, this Upper Holland Gravel contains large amounts of southern material (Bridgland, 1983a; Bridgland et al., 1988). In fact, the deposit contains as much Lower Greensand chert as the early Medway gravels to the south of the Blackwater estuary ((Table 5.3) and (Table 5.5); see Part 2 of this chapter). The favoured interpretation of the Upper Holland Gravel is that it represents the confluence of the Medway, not with the Thames, but with the outwash stream, occupying that river's former course, that deposited the Upper St Osyth Gravel. A further observation can be made from the clast composition of these various deposits: whereas Thames-derived sediment considerably dominates the various Thames-Medway Kesgrave Group gravels, such as the Lower Holland Gravel, Medway-derived material completely dominates the Upper Holland Gravel (Bridgland, 1983a, 1988a; (Table 5.2)). The Upper Holland Gravel has been sampled over a wide area between the St Osyth pits and the type section. Its southern component generally increases in size downstream, although the highest figure, 31.4%, was encountered in a sample from Burrs Road [TM 193 173], 2.5 km to the north-west of the Holland-on-Sea GCR site (Table 5.2). The Upper Holland Gravel appears to be much more widely distributed than the Upper St Osyth Gravel, the occurrence of which is extremely localized. This again suggests that the outwash stream responsible for the Upper St Osyth Gravel was much smaller than the contemporary River Medway, into which it flowed to the east of St Osyth.

The Upper St Osyth Gravel, even as far upstream as Fingringhoe, also contains a significant southern component (Table 5.2). This is unexpected in what is believed to be the product of an outwash stream that issued from the Anglian ice sheet. Anglian Stage glacial deposits in Essex generally contain such material in small quantities, reworked from earlier sediments such as the Kesgrave Sands and Gravels, in which Greensand chert has been traced as far north as Norfolk (Hey, 1980). However, other features of the clast composition of the Upper St Osyth Gravel (the paucity of reworked Palaeogene flint pebbles, for instance) imply that reworking of material from the Kesgrave Group has been insufficient to account for the high southern count. This is an obvious fact, since the amount of southern material is higher in the Upper St Osyth Gravel than in any Kesgrave Group formation upstream from the Medway confluence. The provenance of this extra southern material was probably the area to the south of Colchester, which, in the Middle Pleistocene, would probably have been covered by high-level left-bank terraces of the early Medway. Work in south-eastern Essex and north Kent has indicated that the Medway is a river of considerable antiquity and that throughout its early course from the Medway Towns to the Blackwater estuary it was progressively migrating eastwards (Bridgland, 1980, 1983a, 1988a; Bridgland and Harding, 1985). It is therefore likely that an extensive terrace system existed to the west of the Anglian course of the Medway, which is depicted in (Figure 5.4) (E and F). It is possible that a small river system drained northwards from this area into the old Thames valley. Its contribution would have been insignificant whilst the Thames continued to supply huge quantities of gravel to the area, but, as with the Medway in the case of the Upper Holland Gravel, this contribution made a significant difference to the gravel load of the Upper St Osyth Gravel outwash stream.

(Table 5.5) Clast-lithological composition of gravels described in Chapter 5, Parts 2 and 3.

* Not separately recorded

⁽¹⁾ For comparison only — non-durables excluded

⁽²⁾ Feather edge

⁽³⁾ From the lower gravel at the GCR site

⁽⁴⁾ Subsample

See also notes to (Table 4.2), page 181

Detailed analysis of the clast composition of these various gravels in north-eastern Essex therefore shows that the Thames was replaced, in Upper St Osyth/Upper Holland Gravel times, by an outwash stream. The explanation of this remarkable change lies in events during the Anglian glaciation of the northern London Basin. Gibbard (1977, 1979) showed that, during this glaciation, ice blocked the early Thames course through the Vale of St Albans, leading to the diversion of the river into its modern valley through London (see Chapter 3). The Upper St Osyth and Upper Holland gravels have been interpreted as deposits laid down during this period when the Thames valley was blocked and the river was not reaching the Tendring Plateau. The Medway, however, was unaffected and continued to flow to Clacton and beyond without hindrance from the glaciation, receiving the Upper St Osyth Gravel outwash stream as a west-bank tributary (Figure 5.4)F.

This interpretation of the sequences at St Osyth and Holland-on-Sea is strengthened by the identification of the next (and final) aggradation in the terrace succession of the Tendring Plateau as the product of the post-diversion Thames-Medway ((Figure 5.5)B; see Part 2 of this chapter). The Upper St Osyth/Upper Holland Gravel can be correlated with deposits in other parts of the Thames drainage system that were affected by the (Anglian) glacial diversion of the river. It has been considered (Bridgland et al., 1988) to be a time-equivalent of the (lacustrine) Moor Mill Laminated Clay and the (deltaic) Winter Hill Upper Gravel, both laid down in a proglacial lake that formed when the Thames was blocked, immediately prior to its diversion (Gibbard, 1977; see Chapter 3, Moor Mill). It is probable that the time interval during which the Moor Mill lake existed was relatively brief, perhaps only a few centuries (Bridgland et al., 1988). The reinterpretation of events during the glaciation of the Vale of St Albans, by Cheshire (1981, 1986a), suggests that the Thames was diverted following the formation of the Watton Road Lake, near Hertford, rather than at Moor Mill (see Chapter 3, Part 2). This would indicate precise correlation between the Watton Road lacustrine sediments and the Upper St Osyth/ Upper Holland Gravel, making the latter slightly older than is indicated by the correlation with the Moor Mill lake beds. Whichever of these two correlations is correct, the Upper St Osyth/ Upper Holland Gravel is clearly one of the most closely datable gravel units in the Thames basin.

Summary

At St Osyth, an important section reveals the lowest Kesgrave Thames formation, the Lower St Osyth Gravel, overlain by fine-grained sandy gravel (Upper St Osyth Gravel), interpreted as distal outwash laid down while the Thames was blocked further upstream. The palaeogeographical interpretation of this sequence cannot be made without reference to the evidence further downstream in the area of the contemporary Thames-Medway confluence. The sequence there is revealed in equally important cliff sections at Holland-on-Sea, where a comparable sequence to that at St Osyth is exposed, with equivalent Lower Holland Gravel overlain by Upper Holland Gravel. The composition of the former is typical of the Kesgrave Sands and Gravels, but the latter is dominated by Medway-derived material. The only way that gravel so closely resembling the Medway deposits further south could have been deposited at Holland, which was clearly within the contemporary Thames-Medway valley, is for there to have been no contribution from the Thames. The only time when the Thames and Medway have not joined, either before or after the diversion of the former, was during the brief period when the Thames was blocked by the Lowestoft Till ice sheet (immediately prior to its diversion). The Upper Holland Gravel, and its upstream equivalent the Upper St Osyth Gravel, are therefore correlated with this glacial event. These deposits thus provide a key stratigraphical marker within the terrace sequence of north-east Essex, one that assists correlation within the Thames system as a whole and with the Pleistocene sequence in other areas.

Conclusions

Using only the evidence of the rock-types present in the gravels at St Osyth and Holland-on-Sea, it is possible to demonstrate the rapid and catastrophic changes that affected the Thames as a result of the most extensive Pleistocene glaciation, during the Anglian Stage (around 450,000 years ago). Comparison of the gravels at these two sites reveals that the north-eastward-flowing Thames abruptly ceased to reach this area, because it was blocked by ice upstream, in what is now Hertfordshire and central Essex. The lower gravel at St Osyth is a typical pre-diversion Thames deposit. The Holland-on-Sea section is downstream of the contemporary confluence with the Medway, so that the Lower Gravel there is a Thames-Medway deposit, although much dominated by Thames material. The upper (later) gravels at both sites are significantly different. In particular, they contain material carried by meltwater streams from the Anglian ice sheet. The River Medway, lying beyond the direct influence of the ice, continued to flow northwards. This is demonstrated by the composition of the upper gravel at Holland-on-Sea, which is very much dominated by Medway material, in marked contrast to the lower gravel. The implication of this is that the Medway was very much larger than the meltwater river that replaced the Thames at that time. After it was diverted into its modern course, the Thames joined with the Medway in the Southend area, approximately at the location where the estuaries of the two rivers join today.

References