Geology Reference
In-Depth Information
Holocene Arabian Gulf pixel counted facies frequency
Holocene Arabian Gulf prediction from temporal model
Lithofacies
Fig. 10.
Predicted frequencies of facies as obtained from the FPV of the temporal regular Markov chain, which models how
facies transit into each other (i.e. how sand turns into rock and attract coral growth) in the present landscape (hatched)
compared with the actual, pixel-counted, sizes of facies in the Arabian Gulf biostromal landscape (pure colour). Differences
are not signifi cant (
p
< 0.05) for (a) recent Arabian Gulf and (b) Miocene Leitha Limestone. Facies can be made compatible
(compare Fig. 7) by lumping in the recent model the dense corals (dead and alive) and juxtaposing them to the Miocene
framestone, the recent sparse corals and corals with algae and juxtaposing them to the Miocene coral rud- and fl oatstones,
the sand and the hardground (which cannot be differentiated in the Leitha Limestone). Recent seagrass and Miocene marls
as well as recent algal lawns and Miocene oyster rudstone and rhodoliths would be equivalent.
where
sq
contains all values of
squared. Finally,
bioherm. This was not the case in the study area,
where only biostromal frameworks (single-gen-
eration frameworks) or non-framework commu-
nities were found. Thus, the ergodic/absorbing
model did not conform to reality and was rejected.
This process demonstrated that simply using the
same vertices as in the spatial adjacency matrix
and adjusting transitions does not automatically
lead to a temporal model that produces plausible
results.
B
=
N
ยท
R
where
N
is the fundamental matrix and
R
the
matrix of transient states gives the likelihood of
absorption (which is known to be 1, because the
transient states reduce to zero in the FPV).
In analogy to what was done to obtain Fig. 10,
the FPV was calculated (Fig. 11g). Compared with
the values of the regular temporal model and
the spatial matrix, the outcome was different.
Due to absorption, no soft substrata remained
and all time was spent in the hardground-coral
loop. This would be equivalent to reef-building
dynamics, in that all free substratum would be
eventually converted into hardground on which
corals would settle, that would then proceed to
grow generation on top of generation to form a
Facies pattern and environmental change -
Landsat image analysis for the Arabian Gulf
If it is predicted that the vertical and/or horizon-
tal transitions can be used to build a model of
sedimentary system functioning, then it should
also be possible to make fore- or hindcasts regard-
ing a system's state in different environments.
