 |
| Titel |
Stress field conditions for ring-fault initiation: A new perspective |
| VerfasserIn |
A. Geyer, J. Gottsmann, H. Kinvig, J. Martí |
| Konferenz |
EGU General Assembly 2009
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 11 (2009) |
| Datensatznummer |
250022244
|
|
|
|
|
|
| Zusammenfassung |
| During the last decades, calderas have been analyzed through field studies, analogue models
and numerical simulations. This has permitted considerable improvement of our
knowledge on caldera formation and particularly, on the stress conditions that lead to the
formation of ring faults. Earlier results from numerical modelling of collapse caldera
formation using linear elasticity indicate that in order to encourage the initiation
of sub-vertical, normal ring faults, the stress field must satisfy three conditions
simultaneously: 1) The minimum value of compressive stress (σ3), i.e. the maximum
tension, must occur at the Earth’s surface, 2) The maximum value of shear stress
(σ1 - σ3) must occur at the lateral margins of the magma chamber, and 3) The
maximum tension (minimum σ3) at surface, must peak approximately above the lateral
ends of the magma chamber. In essence, it has been shown that only few geometric
configurations dominated by the relation between source depth and source aspect
ratio yield conditions that fit these “ideal” stress conditions. As a consequence,
caldera-formation should be a rarely achievable phenomenon. However, this conclusion
stands in stark disagreement with abundant field and analogue experimental data of
collapse caldera formation. The latter investigations show that caldera subsidence
does not necessarily require specific geometries of the analogue reservoir to initiate
ring fault formation and collapse. An open question thus remains as to a potential
misinterpretation of numerical models of caldera collapse. The main objective of this work is
to investigate contradictory results from numerical and analogue models as well as from
field constraints. For this, we refer first to mining subsidence studies, to inform
on the most relevant concepts that may help us to understand collapse initiation.
Second, we report on results from a new set of experimental investigations on the first
stages of roof subsidence and associated initial surface structures that have yet
been poorly studied. Finally, we compare the experimental results with results from
dedicated numerical models to inform on collapse initiation. Our results indicate
that numerical collapse models assuming linear elasticity do not allow to predict
the position of ring fault initiation, but rather depict to loci of near-surface ring
fractures delimiting a potential collapse region. Our findings are in agreement with
time-lapse data from recent caldera collapses on Miyakejima island (Japan) and La
Reunion island (France). We conclude that linear elastic models cannot be applied to
determine the geometric conditions for caldera formation since such models are in fact
not applicable to predict the locus of ring-fault formation at the ground surface. |
|
|
|
|
|
|