Silkeborg gravity high revisited: Horizontal extension of the source and its uniqueness

Silkeborg Gravity High is a dominant positive gravity anomaly in Denmark. It is associated with an igneous intrusion within the crust. A deep refraction seismic profile locates the top of the intrusion in depths between 11 km and 25 Inn. The present contribution should be read together with two other papers by the author (Strykowski, 1998; Strykowski, 1999) dealing with the modelling problems of the same area.

Strykowski (1998) discusses an advanced method of geological stripping. The focus is on coupling various types of piecewise information (depth to the top/base of geological bodies/layers obtained from depth converted seismograms and interpolated to a horizontal grid, sui-face gravity data, and mass density information from boreholes).

The objective is to model the surface gravity response of known sediments to a depth level of 10 km. Illustrated by the practical example (modelling of the source of Silkeborg Gravity High) Strykowski (1999) discusses methodological aspects of extracting information about the geometry of the source body (in 3D) from (geologically stripped) surface gravity data and from a cross-secting deep seismic profile.

The average mass density contrast between the source body (the intrusion) and the surroundings is estimated. The used geometrical information from the seismogram is weak (only the depth interval). A remarkable result of this investigation is that the along profile cross section of the obtained (3D-)structure agrees with the geometrical information of the refraction seismic profile.

The present paper is an attempt to extend this result to the rest of the sedimentary basin. Of particular interest is another positive gravity anomaly (another intrusion?) located to the north-west of the studied anomaly. A "final model" obtained here estimates the depth to the source body to 14 km.

Nevertheless, the focus of the present paper is not on finding a particular "best model" of the subsurface, but on ambiguity considerations. Especially, on how the different assumptions alter the obtained model? The interesting aspect is whether the used assumptions are supported by the available information.