A reinterpretation of the Neogene emersion of central Belgium based on the sedimentary environment of the Diest Formation and the origin of the drainage pattern

The 20 to 25 m thick sand deposits crowning the Flemish Hills from Cassel to Flobecq are geologically mapped as Diest Formation (Upper Miocene), but they are now argued to belong to a different formation, for which the new name of Flemish Hills Formation is proposed. They are interpreted as a vertical succession of inner shelf, lower shoreface to upper shoreface deposits of an exposed, wave-dominated coast at the south shore of an open shelf sea. They constitute a normal regressive progradation fed by a mixture of shelf and continent-derived clastics. At least near Flobecq, the top may contain a second normal regressive deposit indicative of a southwards coastal shift. The top sands at Oudenberg and Kesterberg, also traditionally mapped as Diest, are reinterpreted as outliers of the Middle-Eocene Lede Sands, based on a sedimentological revision of the immediately underlying Gentbrugge Formation. The Hageland Diest Sands are thought to be the sedimentary fill of submarine current scour troughs, created at the opposite side of a semi-enclosed tidal embayment, tributary to the Roer Valley Graben, that was rapidly filling from the NW during high base level under a high sediment supply. The depositional environments of the Flemish Hills Sands and the Hageland Diest Sands are incompatible. An analysis of the central Belgium river pattern, including the smallest dry upper thalwegs, suggests that it originated on a Miocene, long-lived, slowly tilting surface draining to the Roer Valley Graben in the NE. The Flemish Hills were no morphological barrier and hence were buried at that time. Additional arguments based on basin architecture and the possible link of the coarse continental clastics to the Pyrenean tectonic phase, lead to the hypothesis that the Flemish Hills Sands may be outliers of the lower cycles of the Upper-Eocene Bassevelde Member of the Zelzate Formation. If confirmed, many elements of the geomorphological genesis of central Belgium need revision