The Balder and Ringhorne Tertiary oilfields of the Utsira High are a cluster of prominent Palaeocene mounds, whose presence has confounded geologists since they were first observed on 2D seismic data back in the 1960s and 1970s. Until recently, the Balder Field geologic model consisted of distal, deep‐water Heimdal, Hermod and Balder Formation sandy‐debrite and turbidite sands, compensationally stacked along the flanks of the Utsira palaeo‐high, with intervening layers of hemipelagic shale. Remobilization of the sands by large‐scale fluidization accentuated the high‐relief mounds and sand injections linked reservoirs that were originally isolated. Reprocessed seismic data show strong primary reflectors that cannot be reconciled with this model; continuous sand bodies are observed to cross‐cut normal, biostratigraphically constrained sequence‐stratigraphic reflectors. The implication is that many, if not most, of the Hermod and Balder sands are not depositional, but were emplaced by injection. Furthermore, most of the Palaeocene mounds are associated with major ‘tears’ in the normally smooth Top Cretaceous chalk seismic reflector, and overlying chalk ‘rafts’. The chalk rafts were formerly thought to be detached glide‐blocks, shed from distant scarps to the east, but the improved data show that they are always positioned above matching depressions in the main Chalk, implying a local origin. Some of the rafts are immense, exceeding half a kilometre in width and millions of tonnes in mass. We believe the association of the chalk features with the mounds is more than a coincidence and suggest that they are genetically related. A number of driving mechanisms have been proposed for the formation of the Palaeocene mounds of the Utsira High, including normal post‐depositional compaction, fluidization by earthquakes, overpressuring due to gas migration and gravity‐sliding back into the Viking Graben. Our observations are, however, more consistent with the mounds resulting from episodic supra‐lithostatic pressure escape from beneath the Chalk.

SynopsisThe geological structure and history of the Rockall Trough margin to the west of the Outer Hebrides (57°N-59°N) has been inferred from the analysis of bottom samples, seismic, sonar, magnetic, satellite altimeter and surface-ship gravity observations. Much of the continental shelf consists of a shallow platform of Precambrian Lewisian basement, covered by a thin (<300m) blanket of Cenozoic sediments. Apart from areas near the shelf break and immediately west of Lewis, the basement has not undergone the large-scale subsidence typical of many regions adjacent to the Rockall Trough. Beneath the outer shelf, the Lewisian dips steeply towards the deeper water and is covered by a westwardthickening succession (>3km) of Mesozoic and Cenozoic sediments.The eastern margin of the Rockall Trough appears to have been well established as a site of deposition by the end of the Triassic. Further subsidence occurred during the late Jurassic and continued into the Cretaceous when the axial province of the Rockall Trough off the Hebrides was invaded by basic volcanic material. Subsequently, the region was profoundly affected by early Cenozoic igneous activity. The long-term buoyancy of the Lewisian basement under the Outer Hebridean platform and early-formed structural discontinuities within it have influenced the overall pattern of sediment accumulation and the distribution of volcanic activity along the margin. During the later stages of deposition, sedimentation was closely controlled by the presence of early Cenozoic volcanics, by bottom currents and by glacial processes.

The Balder Field reservoir sandstone has been interpreted as remobilized from a Mesozoic parent bed. This paper seeks to address the questions raised about this unusual origin. Research and broadly analogous processes are reviewed leading to the proposal that the parent beds may have been fluidized by bedding-parallel retrogressive entrainment of Statfjord Formation sands by a connected larger source of overpressured fluids. These fluids are identified as most likely derived by lateral migration from the Viking Graben, initiated in response to early Eocene basin inversion related to North Atlantic rifting. The event probably involved breaching of the topseal in multiple places over a large section of the Utsira High. The geometry of large sills formed from small breach points and internal differentiation seen may show that the sills inflated by lateral accretion from a medial active flowing zone of turbulent, transitional or laminar flowing suspension. It is suggested that sequence-stratigraphic and structural context should be considered as additional criteria to discriminate between depositional and intruded sands.