Geophysical log characterization of the Flamborough Chalk Formation (Late Cretaceous, Santonian – Early Campanian), East Yorkshire, UK: implications for understanding the onshore and offshore occurrence of the late Campanian Rowe Chalk Formation

Although reasonable efforts have been made to obtain all necessary permissions from third parties to include their copyrighted content within this article, their full citation and copyright line may not be present in this Accepted Manuscript version. Before using any content from this article, please refer to the Version of Record once published for full citation and copyright details, as permissions may be required. Accepted Manuscript

Further north, lithological and stratigraphical changes mark the transition into the distinctly different Chalk of the Northern Province (sensu Mortimore 1983), covering Lincolnshire, Yorkshire and parts of north Norfolk (Wood & Smith 1978;Whitham 1991Whitham , 1993. For the Cenomanian to Coniacian parts of this succession, spanning the Ferriby, Welton and Burnham Chalk formations, the stratigraphical and marker-bed correlation of borehole geophysical logs is based on the Killingholme Borehole DG1 [TA 14981 19081] described by Barker et al. (1984). However, the succeeding Flamborough Chalk Formation, with an outcropping thickness approaching 200 m, dwarfing the typical thickness of the coeval Newhaven Chalk in the Southern Province (c. 60 -70 m; Mortimore et al. 2001, figs 3.68-3.69) and almost equalling the combined thickness of the Ferriby, Welton and Burnham Chalk formations, has never been geophysically characterized. Mortimore (2014) hinted at the potential influence of geological structure on changes in thickness, lithology and hardness of the Northern Province Chalk, and this seems the most plausible explanation for thickening of the Flamborough Chalk. During the Mesozoic, the Flamborough Fault Zone (Fig. 1) separated the more gently subsiding East Midlands Shelf to the south, from the more rapidly subsiding Cleveland Basin to the north, and later during the Late Cretaceous -Cenozoic, became the focus of both compression and extension related to uplift of the Cleveland Basin (Kirby and Swallow 1987;Starmer 1995;Mortimore et al. 2001). Evidence for several intra-Late Cretaceous tectonic episodes has been recognized across Southern England (Mortimore & Pomerol 1997;Mortimore 2018). One of these (Wernigerode The Flamborough Chalk is penetrated by deep hydrocarbons boreholes and shallower water boreholes, but almost none of these have both borehole core and geophysical logs that would allow an understanding of stratigraphical relationships. An exception is the c. 100 m deep Carnaby Borehole [TA 1505 6486] near Bridlington (Figs 1, 4), which combines lithological, biostratigraphical and geophysical log data (Woods 1997). Biostratigraphical data allowed provisional correlation of the succession in this borehole with the middle and higher parts of the Flamborough Chalk, but at that time there was no means of confirming this by geophysical log comparisons.
In the time that has elapsed since drilling of the Carnaby Borehole in the mid-1990s, no new data have emerged to allow geophysical log correlation of the Flamborough Chalk, but it has been proposed that stratigraphically younger chalk (Late Campanian, Rowe Chalk Formation; Fig. 1), with distinctly different physical properties, extends onshore beneath Glacial/Superficial Deposits in the Holderness area (Sumbler 1996;Hopson 2005;Mortimore 2014). With the growing desire to create 3D geological models of specific regions or intervals of strategically important stratigraphy (e.g. aquifers, resource-rich units, units with critically important engineering properties; Woods et al. 2016), it is now pressing to fully exploit geophysical log data in the Flamborough Chalk, and to understand the subsurface geometry of younger units. To answer this need, existing geophysical log data for deep hydrocarbons boreholes has been reinvestigated, and related to broad patterns of stratigraphical variation documented in published lithological logs of outcrops proving the Flamborough Chalk. The robustness of these interpretations has been tested by examining the lateral persistence of geophysical log features, consistency with regional thickness patterns, and consistency with raw stratigraphical data (including macrofossil biostratigraphy) for the Carnaby Borehole.  (Fig. 2). Here, borehole data support wider field evidence (Whitham 1991(Whitham , 1993Gaunt et al. 1992)  2) form well-defined markers for interpretation of the basal Welton and near-basal Burnham Chalk formations, respectively. The Black Band is an organic-rich unit formed during a protracted period of marine anoxia (Schlanger et al. 1987), and is typically manifest as a sharply defined, high gamma log peak. The North Ormsby Marl, locally 110 mm thick, is a volcanogenic clay unit (Wray & Wood 1998) with a high gamma log response emphasized by the very low gamma log values associated with super-and subjacent thick semi-tabular flints (Fig. 2). The appearance of these closely spaced flints is a feature of the basal Burnham Chalk, and their sharp low gamma log response is as valuable for geophysical log interpretation as the high response of the North Ormsby Marl that is closely associated with them. Figure 3 shows how these markers have been interpreted in the Barmston 1 Borehole beneath Holderness. The gamma logs of the Ella Crossroads Borehole and Barmston 1 Borehole show the same relative pattern of response, allowing confident interpretation of formational boundaries, but the amplitudes of the responses vary. This variation probably reflects differences in borehole logging environments, variable logging protocols (for example, logging speed), and that log acquisition in the Barmston 1 Borehole was probably optimized for units other than the Chalk.

Hydrocarbons borehole correlation in the Chalk beneath Holderness
Whilst the Burnham Chalk contains abundant flint and common marl seams, the Flamborough Chalk is virtually devoid of flint but with very common thin marl seams, ranging from a few millimetres to contrasts ought to be matched by corresponding changes in gamma and sonic/resistivity logs, ideally at distances above the base of the Burnham Chalk that are compatible with composite stratigraphical data documented in regional overviews, such as those of Whitham (1991Whitham ( , 1993 and Sumbler (1996). These contrasts are revealed on the gamma and sonic logs of the Barmston 1 Borehole, and the gamma log of the Hornsea 1 Borehole (Fig. 4).

The Barmston 1 Borehole shows an increase in average gamma log values and decrease in average
sonic log values at about 295 m depth. As well as the average shift in values, both logs show changes in their detailed character; neither this detailed change nor the general shift in log trend correspond with changes in borehole construction or logging procedure, and are interpreted as primary lithostratigraphical responses. The gamma log shows a much higher frequency of sharp gamma peaks above 306 m compared to the succession below. Correspondingly, the sonic log appears more highly serrated above 306 m; below there is less high frequency variability, but there are larger, broader shifts in the sonic log profile compared to the rather flat profile above 306 m. The change in the gamma log profile is consistent with the presence of Flamborough Chalk with its typical high frequency of marl seams. As well as marls, the Flamborough Chalk is also characterized by a profusion of stylolitic surfaces with thin coatings of marl that in combination probably help to produce the high serration of both the gamma and sonic logs. Large-scale shifts in sonic velocity can usually be related to changes in bed induration, but in the case of the Burnham/Flamborough Chalk boundary, it is probably the absence of flint in the Flamborough Chalk that causes the pronounced drop in average sonic velocity. The distance of this geophysical change above the interpreted base of the Burnham Chalk is c. 125 m. This compares favourably with a composite Burnham Chalk thickness of c. 135 m documented by Whitham (1991; includes an inferred exposure gap of 10 m), and a range of 85 -140 m recorded by Sumbler (1996).
As well as identifying the base of the Flamborough Chalk, the hydrocarbons boreholes along the Yorkshire coast also appear to show features that can be related to subdivisions of the Flamborough Chalk (South Landing, Danes Dyke and Sewerby members) described by Whitham (1993)

Correlation with the Carnaby Borehole
The Carnaby Borehole [TA 1505 6486], c. 3.5 km south-west of Bridlington, was tentatively inferred on lithological and sparse macrofossil evidence to equate with the higher part of the Sewerby Member of Whitham (1993) (Woods 1997). The lower part of the cored borehole contains locally abundant specimens of the oyster Pseudoperna boucheroni (Woods 1912 Whitham (1993, figs 8, 9), and these are largely absent above 80 m depth in the Carnaby succession. In the outcropping Flambrough Chalk described Whitham (1993), the change in inoceramid faunas occurs a short distance above the Flamborough Sponge Beds, and the record of Scaphites occurs in an interval of close-spaced marls (Sewerby Hall Marl, Sewerby Steps Marl, Nafferton Marls), matching a similar marl association in the higher part of the Carnaby Borehole. The data for the Carnaby borehole includes a focused resistivity log, which responds similarly to a sonic log in intervals of hard/soft chalk or marl-rich/marl-poor chalk (= elevated velocity/resistivity for hard and/or marl-poor chalk; reduced velocity or resistivity for softer and/or marl-rich chalk). These geophysical data have not hitherto been used to independently explore the correlation of the Carnaby Borehole, and in the context of the current work they provide a useful test of the assumptions used to interpret the geophysical logs of the Barmston and Hornsea borehole successions.
The sonic log response in the higher part of the Barmston 1 Borehole shows a series of well-marked embayments of alternating higher and lower sonic velocity. Part of this interval is correlated with the resistivity log of the Carnaby Borehole using three of these distinctive inflections (Fig. 4). This  fig. 4.54), geophysical log correlations of, for example, the Welton Chalk between the Hull area and the margin of the Yorkshire Wolds (Fig. 2), a distance of about 40 km, reveal no significant thickness variation. Consistency between the geophysically determined thickness with Whitham's (1991) composite measured thickness for the Welton Chalk is within a couple of metres. This is not only indicative of the remarkably consistent stratigraphy of the Chalk of the Northern Province, but testament to the meticulous work of Whitham (1991Whitham ( , 1993 in assembling an accurate composite Chalk succession from a disparate group of exposures.

Youngest Flamborough Chalk and Rowe Chalk Formation
Comparison of lithological and geophysical data for the upper part of the Carnaby Borehole suggests that the distinctive low resistivity embayments in the log profile are caused by concentrations of marl seams. The upward continuation of these geophysical features on the sonic log at Barmston might similarly be explained. The lowest of these features on the Barmston 1 log is around 130 m depth, which using the inferred geophysical interpretation for Barmston is c. 165 m above the base of the Flamborough Chalk, approximately coincident with the horizon of the Sewerby Hall / Sewerby Steps Marl on Whitham's (1993) composite thickness log. This geophysical log feature occurs at 40 m depth in the Carnaby succession, and its inferred interpretation agrees with the suggested marl seam correlation independently derived from lithological and biostratigrahical data (see above).
The highest 40 m of Flamborough Chalk Formation recorded by Whitham (1993) contains groups of closely spaced marl seams; features that are broadly consistent with the cyclicity of the sonic log for the Barmston 1 succession. The intervals of low sonic velocity are matched by peaks in the gamma log, suggesting that most if not all of the sonic log cyclicity is caused by clusters of marl seams or intervals of more marly chalk. Using Whitham's (1993) composite thickness data suggests that the Nafferton Grange Upper Marls, just below the top of the outcropping Flamborough Chalk Near Hornsea, Whitham (1993) reported the presence of Late Campanian, B. mucronata Zone Chalk in boreholes and in isolated rafts (of presumed glacial origin) within Quaternary deposits. Loose, flint-preserved fossils belonging to the Late Campanian or Maastrichtian were also reported from beach deposits at Barmston (Whitham 1993). Later, Sumbler (1996) suggested that flinty chalk of Late Campanian age might extend under Holderness as an arcuate subcrop beneath Quaternary deposits, and designated it Rowe Chalk following the use of this term for known offshore occurrences of flint-bearing chalk of comparable age (Lott & Knox 1994 (12 -41 m depth) (Fig. 5). Based on these data, illustrations of the onshore development of the Rowe Chalk beneath Holderness have been reproduced by other workers (Mortimore et al. 2001;Hopson 2005;Mortimore 2014).
Since the Rowe Chalk is essentially defined by a combination of its age and regular flint-bearing character, it is important to establish the extent of undoubted onshore occurrences. Whitham (1993) does not mention the presence of flint in the rafts at Holderness, only the record of Campanian fossils in plastic-textured, reworked chalk. The Campanian/Maastrichtian flint-preserved fossils from the beach at Barmston were considered by Whitham (1993) to have come from subsea outcrops or to have been derived from currently unexposed strata. Sumbler (1996) suggested that geophysical logs might be used to identify the Rowe Chalk, with flint-rich intervals represented by low gamma responses and peaks in sonic velocity. At face value, this is the signature seen in the higher part of the Barmston 1 Borehole, but the likely overlap with the Carnaby succession suggests that the distinct cyclicity in the sonic log is more likely a response to marl content rather than flint horizons (see above). The low-resolution lithological logs of both Atwick boreholes (Fig. 5) suggest that flint is localized and sporadic in these successions, with intervals of marly chalk exceeding 20 m in thickness between flint bearing horizons. The Atwick 10 Borehole [TA 18790 51650] shows a uniform annotation of chert/flint on the composite log, developed throughout the entire Chalk succession. This flint distribution seems highly unlikely in the context of the known, highly variable flint occurrence in outcrop successions (Whitham 1991(Whitham , 1993. These flint distributions may in part be schematic representations, or represent caved material logged in drilling mud. Apart from deep hydrocarbons boreholes, a number of shallower boreholes intersect the upper part of the Chalk in the Holderness area, and most lack any record of flint (Fig. 5).
Considering all the borehole and 'outcrop' data together for the Holderness area, it seems unlikely that regularly flinty chalk, analogous to the Rowe Chalk, is developed. The evidence of the geophysical log correlations between the Barmston and Carnably boreholes is for the presence of chalk that is largely comparable with the Flamborough Chalk Formation at outcrop, but significantly younger, with regularly developed marl seams and, perhaps on the evidence of the Atwick 1 Borehole, sporadic flint-bearing intervals. The presence of flints in the Flamborough Chalk is not without precedent. Berridge & Pattison (1994)

In a description and definition of Chalk stratigraphical units developed in the Southern and Central
North Sea, Lott & Knox (1994) illustrated a number of gamma and sonic logs through the Chalk succession. Offshore, the lithological and chronostratigraphical equivalent of the Flamborough Chalk is the Jukes Formation, described as moderately hard, variably argillaceous and with few cherts, of Santonian to Early Campanian age, and varying in thickness between 138 and 414 m. Geophysically, it is possible to match features of the sonic logs of offshore borehole successions with the sonic log of the Barmston 1 borehole (Fig. 7). The log correlations suggest that the highest Flamborough Chalk seen in the Barmston 1 Borehole (grey highlight on Fig. 7) can be traced into offshore successions; although the trends in the logs are similar, with a distinct sonic spike at the base of each of the highlighted intervals ('2a' on Fig. 7), the cyclicity is not as pronounced as the corresponding interval in the Barmston succession. Above the Jukes Formation, Lott & Knox (1994)  below the appearance of flint in strength can be proved, then the basis for this relationship will be strengthened. The correlation shown for offshore borehole 49/05-1 (Fig. 7) also supports the idea that the more than 400 m assigned to the Jukes Formation by Lott & Knox (1994) probably includes strata that are younger than any of the exposed onshore succession, rather than representing expansion of the onshore equivalent Flamborough Chalk.  (1993). Above this, the gamma and sonic log responses in the Chalk at Holderness can be matched with gamma and resistivity log responses in the Flamborough Chalk of the cored Carnaby Borehole.

Conclusions
Here, biostratigraphical data for the presence of key marl seams (Daneswood, Sewerby Steps, Sewerby Hall and Nafferton Grange Marls) support tentative thickness-based interpretations of geophysical log responses at Holderness.
The youngest Chalk in the succession below Holderness is geophysically characterized by a regular cyclicity of high and low sonic velocity. The gamma log indicates that low sonic velocity intervals correspond with one or more peaks on the gamma log, suggesting that the low sonic velocity intervals are marl-rich. Beneath Holderness, this Chalk has previously been interpreted as flint-rich The youngest Chalk seen beneath Holderness appears to be capable of correlation into offshore successions, but with a less marked cyclicity in its geophysical log signature that may indicate lateral transition into lithologies analogous to the Westermost Rough Member of Mortimore (2014).  Sumbler (1996). Holderness is the broad lowlying coastal and adjacent inland region between the Humber estuary in the south and the Yorkshire Wolds in the north and west.