 |
| Titel |
The Faroe Islands geological stripping using weak assumptions |
| VerfasserIn |
Gabriel Strykowski, Morten Sparre Andersen |
| Konferenz |
EGU General Assembly 2010
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 12 (2010) |
| Datensatznummer |
250042830
|
|
|
|
|
|
| Zusammenfassung |
| In a complex geological-/geophysical setup with a general source distribution, the result of
the source modeling from the gravity data depends on how the available independent
information is being used. The model construction is implicitly hierarchic in a sense that,
ultimately, the final inversion for the unknown sources (us) is done from the residual
gravity signal Îgus. In order to make a successful inversion (so that the unknown
source model is a good approximation to the existing true source distribution) the
precondition must be that Îgus is modeled correctly from the measured gravity signal
Îg and using the independent information about the subsurface. Implicitly, the
gravity attraction Îgks from the known sources (ks) must be modeled correctly so
that:
Îgus = Îg - Îgks
yields as good as possible Îgus; a geological stripping. In short, the hierarchy in model
construction implies that any error in Îgks generates an error in Îgus. Consequently, a lot of
effort should be put into the correct modeling of Îgks.
One pitfall of the geological stripping is that the wrong use of either the mathematical
assumptions or the independent information only in very severe cases can be contradicted by
Îgus. However, the fact that Îgus does not contradict Îgks is not synonymous to that Îgks
(and thereby Îgus) is correctly modeled.
In our quest to do the “objective geological stripping” we advocate a cautious method in
modeling Îgks for a large marine area with complex geology around The Faroe Islands. One
such technique is to use the independent knowledge of bathymetry to “strip off” the
gravitational effect of the sea water without any assumption about the mass density contrast
to the sea bottom (that could bias Îgks). Another technique is to use in combination both the
gravity anomalies and the horizontal gradients, i.e. a transformation of Îg and
Îgks. A consequence of Green’s third identity of potential theorem is that a unique
solution cannot be obtained by simply transforming the external field. However,
we can safely assume that the two types of signals (the gravity anomalies and the
horizontal gradients) are generated by the same source distribution. Although a
unique model of the subsurface cannot be obtained, we can utilize that the weighting
between the contributions from the shallow/known sources and the deep/unknown
sources is different in these two types of the gravity data. Knowing independently,
e.g. from seismograms, the approximate depth to the “unknown sources” bears a
possibility to get a good model of Îgus even whenÎgksandÎgus are correlated. |
|
|
|
|
|
|