Two giant oil tanks (79.3 m diameter by 19.5 m high) were constructed at Fawley in 1967-68. The tanks were founded on a reinforced concrete slab supported by piles of varying lengths taken through alluvium and end-bearing in Valley Gravel which overlies Barton Clay (Eocene). Totals of 1580 and 1652 piles, respectively, were used for each tank. The foundations failed during water-loading tests with a maximum depression of 0.425 m in one of the reinforced concrete slabs. Post-mortem examination of the foundations revealed defects in a number of piles, and after further tests and investigations, Professor G. A. Leonards concluded that the failure was due to defective piles.
The success of the original design was critically dependent on the geology. Particular problems on which attention was focussed included the continuity and thickness of the gravels, their variation in density, the character of the interface between Valley Gravel and Barton Clay and the lithological variations within the alluvium. These problems are examined here in relation to the existing geological knowledge of the region. Of particular significance is the conclusion that, in part, the variation in density within the gravels is due to the existence of Flandrian gravel overlying Pleistocene terrace gravel. Isopachytes of the gravels have been drawn along with contours on the upper and lower surfaces. These are compared with the settlement contours of the failed foundation slab and it is concluded that there is no correlation between them.