Aurelie Guilhem ppt

choppedspleenMechanics

Feb 22, 2014 (3 years and 7 months ago)

97 views

Introduction


Sediments are transported
and deposited by

density
flow
, not by tractional or
frictional flow.


Bouma sequence:

from conglomerates at the
bottom to shales on the top


Turbidites:

geological formations that have their origins
in turbidity currents deposits, that deposit from a form of
underwater avalanche that are responsible for distributing vast
amounts of clastic sediment into the deep ocean.


Idealised sequence of sedimentary textures and structures


in a classical turbidite, or Bouma sequence (Bouma, 1962).


Introduction


Interest of the off
-
fault paleoseismology


GPS


high degree of certainty, in few years, of the crustal strain
accumulation.. But just for a portion of a cycle..


Earthquake records


not long enough


Onshore paleoseismology


erosion, urban area..


Off
-
fault paleoseismology


Interest of marine turbidite records


Have to prove they are earthquake
-
triggered


Marine records: more continuous, extend further back in time, more
precise in time (datable foraminifera)


Method used



74 piston, gravity cores from channel/canyon systems draining
Northern California


Mapping channels with multibeam sonar (bathymetry, channel
morphology, sedimentation patterns


Sampled all major channel systems between Mendocino and north of
Monterey Bay


Results


Good agreement with shorter land record


Opportunity to investigate long tem earthquake behaviour of North
San Andreas Fault

Piston corer

Piston core

removed

from corer

Split piston

core being

subsampled.

http://oceanworld.tamu.edu/students/forams/forams_piston_coring.htm


4 segments of
SAF:


Santa Cruz
Mountains


Peninsula


North Coast


Offshore


Several onshore
paleoseismic
sites:


Vedanta:

max
slip rate in late
Holocene 24 +/
-

3mm/yr and 210
+/
-

40 years


Fort Ross:

~230
yr


South of the
Golden Gate:

17
mm/yr


How to identify earthquake
-
triggered turbidites


Possible causes

of turbidites:


Storm or tsunami wave loading


Sediment loading


Storm discharges


Earthquakes


Seismically triggered turbidites are
different
:


Wide area extent


Multiple coarse fraction pulses


Variable provenance


Greater depositional volume


Use a
temporal and spatial pattern

of
event correlation
over 320 km

of coastline

Synchronous triggering and

correlative deposition of turbidites


Regional stratigraphic datum
missing


Correlations depend on
stratigraphic correlations of
other datums and radiocarbon
ages


The
Confluence Test
:


If one canyon contains
n

turbidites and a second
canyon also shows
n

turbidites, and if these
n

events have been
independently triggered, the
channel below the confluence
should contain at least 2
n

instead of only
n
.



8 major confluences


3 heavy minerals


Event “fingerprinting”


All cores are scanned, collecting P
-
wave velocity, gamma
-
ray density,
magnetic susceptibility data, imaged with X
-
radio and grain size analyzed

Event “fingerprint”


First, these data were used to correlate
stratigraphy between cores

at a single site


Found that it was possible to correlate unique
physical property signatures of

individual
turbidites

from different sites within the
same channel


Even possible to correlate turbidites

between
different channels

(some of which never met)


The turbidite “fingerprint” = basis of long
-
distance correlations


Event “fingerprint”

Evolution of a
single event
down channel

over a distance
of 74 km

Radiocarbon analysis


Extraction of

planktic foraminifera

from the
hemipelagic sediment below each turbidite


Bioturbation and basal erosion do not biase
14
C
ages


Method:


Determine hemipelagic thickness


Estimate the degree of basal erosion


Observe that differential erosion is most likely
source of variability at any site


Conversion of hemipelagic thickness to time
(using average of sedimentation rate)


Results

Upper section

poorly preserved

Less dated turbidites

Low foram abundance

Both have 22 events

Results: confluence and mineralogy


Good correlation

between these cores
suggests that input mixing at each confluence
has little effect on the stratigraphy of the
turbidites


Synchronous triggering is the

only viable
explanation


Non
-
synchronous triggering should produce an
amalgamated record that increases in complexity
below each confluence, with only partial
correlations for the synchronous events


Strict test of synchroneity

Results: stratigraphic correlation


Regional correlation of turbidite stratigraphy spanning the Holocene

Results: stratigraphic correlation

Noyo canyon is cut by the NSAF and as an epicentral
distance of zero


explains thicker turbidite records

Time series

-
The youngest 15
events have a mean
repeat time of
~200 yr +/ 60 yr

-
~95 yr: minimum
interval

-
~270 yr: maximum
value

-
Values consistent
with previous
paleoseismic data
onshore

-
Same total
number of events
onshore and
offshore

= land
and marine record
the same events

Discussion


Good correspondence

with land paleoseismic dates
(individual matching, total number of events)


Offshore turbidites as paleoseismic indicators for
the NSAF


Mean recurrence interval coherent with onshore


Epicentral distance

is the controlling factor for turbidite
size


Turbidites correlate across channels where the mineralogy is
different, the physiography is different the sediment sources
are different and the underlying geology is different too


Minimum magnitude and triggering distance

from
the earthquake hypocenter : at least M7.4


But observations of turbidites of small events may also be a
function of the resolutions of the observations


Majority of repeat time intervals

between 150 and 250 yr