Transcript Experimental results: source to sink Source to sink AGU Chapman conference Chris Paola SAFL Saint Anthony Falls Lab NCED Quantitative stratigraphy group University of Minnesota National Center for Earth-Surface Dynamics.

Experimental results: source to sink
Source to sink
AGU Chapman conference
Chris Paola
SAFL
Saint Anthony
Falls Lab
NCED
Quantitative stratigraphy group
University of Minnesota
National Center for
Earth-Surface Dynamics
Particular thanks to…
John Martin, Exxon Mobil Wonsuck Kim, UT Austin
Key ideas
Source-to-sink thinking becomes increasingly important with
increasing time scale
These ideas are readily seen in small-scale experiments
because time scale is directly related to system size
On source to sink scales sedimentary environments are
process domains linked via moving boundaries
On source to sink scales, mass balance is a first-order control
on sedimentary facies
Signal transmission is strongly influenced by sediment storage
& release
Depositional steady state
Steady States
Grade: no mass loss or
gain
Erosional: mass gain
(erosion) balances uplift
Depositional: mass loss
(deposition) balances
subsidence
qs


 qs

x
Moving boundaries: dynamic process domains
linked by internal boundary conditions
John B. Swenson &
Vaughan Voller
Poster M-28 by Matt Wolinsky has a
complete S2S example
DO NOT PANIC.
This talk contains images and data from laboratoryscale experiments
These experiments are not miniature analogs of
natural systems
They are experiments, not models. Their relevance to
field scales comes from scale independence, not
classical scaling
Experimental Earthscape (XES) system
6m
3m
Time is greatly compressed
Subsidence-surface interaction on
accessible time scales
Sink In a box!
Quantifying mass balance: fractional sediment extraction
define a dimensionless distance  in terms of mass loss down
the depositional system:
called Ain earlier
papers
 ( x) 
x

rate of deposition
rT ( x)dx
0
NB: the interval ∆T is
chosen to be long
enough to average out
flow-controlled
fluctuations
qs0
sediment supply
e.g.  = 0.3 means the distance over which 30% of the sediment is
extracted from the system.
Strong et al., 2005, IAS Fluvial Sedimentology 7
Quantifying mass balance: fractional sediment extraction
Using mass extraction as a
measure lets us compare
basins of different shape and
size on a consistent basis
1

0
0
x
L
1
Provides a quantitative way of
expressing proximal – distal

0
0
x
L
1
We can think of the point
 = 0.5 as the
“depositional midpoint”
of the basin

0
0
x
L
1

0
0
x
L
Quantifying mass balance: bypass ratio
Bypass ratio  is the ratio between deposition and bypass:
local avg. unit
sediment flux
qs   L dqs 
( x) 


rL  qs dx 
rate of deposition
1
 d 
 1    
 dx 
basin length
The two measures are directly related
1
Applying the chi transformation to stratigraphy
x = 2.4 m
Note: consistently
lower channel
density for slow
subsidence stage
x = 3.58 m
0
20
cm
40
Strong et al., 2005, IAS Fluvial Sedimentology 7
Applying the chi transformation to stratigraphy
 = 0.4
X=2.4 m
X=1.64 m
X=2.44 m
X=2.4 m
At 40% mass
extraction, the deposit
is still channel
dominated
= =0.7
0.75
But by 70% extraction,
0
predominant
20
depositional element iscm
40
sheets (extensive, thin
lobes)
Strong et al., 2005, IAS Fluvial Sedimentology 7
X=3.58 m
X=2.4 m
X=3.58 m
X=3.58 m
Why should mass balance affect stacking?
• channel fraction & stacking density depend on
rate of channel mobility relative to rate of
deposition
– high mobility rel. to deposition  high channel density
• channel mobility  bed-material flux
• thus high values of flux/deposition (bypass ratio)
 more frequent + more active channels 
increased channel density
Application to turbidite mini-basins
Brazos-Trinity System, offshore Gulf of Mexico
From Beaubouef
and
Friedmann 2000
Paola & Martin, in limbo
Basin 4: From Beaubouef et al. 2003
XES 01 turbidity currents in a mini-basin
Violet et al. 2005 JSR
East Breaks Minibasin
XES 01 turbidity currents in a mini-basin
crater
point of
maximum
subsidence
Violet et al. 2005
Violet et al. 2005 JSR
XES01 vs. Brazos-Trinity System
From Beaubouef et al
2003
XES01 vs. Brazos-Trinity System
Chi = 0.23
Channelized
Chi = 0.1
1.70 m
Beaubouef et al 2003
XES01 vs. Brazos-Trinity System
Chi = 0.05
1.3 m
Chi = 0.1
Beaubouef et al 2003
XES01 vs. Brazos-Trinity System
Chi = 0.5
Lobe switching
Chi = 0.61
2.25 m
Beaubouef et al 2003
XES01 vs. Brazos-Trinity System
Chi = 0.81
3.10 m
Chi > 0.95
4.40 m
Chi = 0.86
Beaubouef et al 2003
Bed curvature statistics
XES 01
East Breaks Minibasin
Curvature: channels vs expansion deposits
Similar changes with increasing
mass extraction in unconfined
turbidites and fluvial deposits
Mass-balance effects: experimental
half-graben basin
Modified from Leeder and Gawthorpe (1987) and Mack and Seager (1990)
Sean Connell (UNM), Wonsuck Kim, Gary Smith (UNM),
Chris Paola
XES 06 plan view setup
Wonsuck Kim
XES06-1: Cross Section Profile
Initial Conditions Stage 0b (0 hrs)
Axial-Dominant Stage 1b (80 hours)
Footwall-Dominant Stage 2 (123 hours)
Axial-Dominant Stage 3 (180 hours)
Hanging-wall Stage 4 (225 hours)
Hanging-wall Stage 4 (225 hours)
These are dynamic
moving boundaries
analogous in some
ways to shorelines
Kim et al. 2011 Geology, in review
Eustatic sediment pumping: general idea
Sediment is transferred offshore during RSL falls
But it is preferentially retained in the fluvial system
during RSL rise
So what is the net effect of eustatic cycling on sediment
delivery to the deep ocean, and in particular, is there
net ‘pumping’ effect associated with repeated eustatic
cycling?
XES 02 experiment
base level curve
run basics
0
-50
base level (mm)
Goal:
measure the stratigraphic effects
of isolated & superposed eustatic
cycles
-100
-150
-200
-250
-300
slow cycle
-350
symmetrical
amplitude: 11cm
duration: 108 hours
-400
0
50
100
150
200
250
300
run time (hrs)
rapid cycle
superposed cycle
6 rapid cycles on one slow cycle
final basement
0
flow direction
1.15 m
initial basement
symmetrical
amplitude: 11cm
duration: 18 hours
5.7 m
XES 02
data collection and preparation
90 usable scans of the entire experimental
surface
89 isopach maps
1 cm-resolution stratigraphic images
474 strike images
125 full dip sections
basin volume
2 cut patterns
dip volume
strike volume
Time-dependent cumulative marine fraction
The model:
● constantgeometry
● mass
conserving
Kim et al. 2009 SEPM Spec Pub 92
● 2 moving
boundaries
Time-dependent cumulative marine fraction
To quantify the
effect of eustatic
pumping, we
need a reference
case: clinoform
progradation
with constant
eustatic sea
level (ESL)
with ESL
cycles
slow cycle: no net effect
overall net pumping
rapid cycle net pumping
no ESL
cycles
Time-dependent cumulative marine fraction
simple ESL
cycles, constant
water
displacement
Compare the
case as run with
superposed ESL
cycles with the
same scenario
but with simple
monofrequency
ESL cycles,
same water
displacement
superposed ESL
cycles
net increment from
superposed cycles
no ESL
cycles
Preserved cumulative marine fraction
with ESL cycles
no ESL cycles
slow cycle: no net effect
overall net pumping
rapid cycle net pumping
The effects are
similar if you
look at final
preserved
marine fraction
rather than
marine fraction
through time
Summary of pumping effect
Little net pumping effect from ESL cycles that do not
create net fluvial erosion – fluvial loss during fall is
compensated exactly by fluvial gain during rise
Net effect including all slow and rapid cycles: increase final
marine fraction from 0.35 to 0.49
Net effect of adding superposed high-frequency cycles:
increase final marine fraction from 0.45 to 0.49
Net pumping effects become strong when sediment supply
is phase-shifted relative to ESL (as originally proposed
by Perlmutter et al.)
Obliteration of supply signals by stick-slip
sediment transport
Key idea: threshold-dominated transport leads to
sediment storage and release (stick-slip transport)
Jerolmack & Paola 2010 GRL
43
Storage and release of sediment under steady conditions
0.12
14
12
0.1
10
0.08
8
0.06
6
0.04
4
0.02
2
0
44
0
0.5
1
1.5
2
2.5
3
4
x 10
0
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
4
x 10
Thresholds and randomness
Deposit of a rice pile
is constructed from
this output
(avalanching and
stick-slip)
Steady input
Intermittent output
14
12
10
8
6
Steps of all sizes form in profile (storage)
Threshold exceedance causes failure (release)
4
45
2
0
0
0.5
1
1.5
2
2.5
3
4
x 10
Numerical Rice Pile [Frette, 1993]
t
t+1
A simple, threshold-based
toppling transport model
h
1
4
250
200
8
1
4
8
x
MODEL
Height profile
EXPERIMENT
Height profile
150
100
50
46
Numerical rice pile - results
14
20
EXPERIMENT – output flux
MODEL – output flux
18
12
16
10
14
12
8
10
6
8
6
4
4
2
2
0
0
23
1
10
3.5 2
3 4
4
4.5
5
6
5 7
8 5.5
9
0
10
6
0
0.5
1
1.5
2
2.5
4
3
4
x 10
x 10
Fluctuations over a wide range of scales
1
10
1/ t x
0
10
Variability saturates at t = tx
-2
-1
10
h
h
-2
10
MODEL
Power Spectrum
of output flux
-3
10
Lh 
tx ~
qsin
-4
10
-4
10
-3
10
-2
10
-1
10
frequency
0
10
1
10
L
47
Stick-slip transport obliterates high f sediment cycles, but…
250
2
10
Qs
T < tx
1
200
10
0
10
150
-1
10
100
1/ t x
-2
time
10
50
-3
10
0
60
65
70
75
80
85
90
95
100
-4
10
-4
-3
10
-2
10
-1
10
0
10
1
10
10
1/time
Ensemble-averaged Power spectra of flux data
0
10
-1
10
-2
10
spectral density
Qs
T < tx
time
-3
10
-4
10
-5
1/ t x
10
-6
10
-7
10
-5
10
-4
10
-3
10
-2
10
frequency
1/time
-1
10
48
0
10
1
10
Cycles with period larger than largest avalanche are preserved
5
1.2
Correlation coeff.
Spectral density
10
4
10
3
10
0.8
2
10
0.6
1
1/ t x
10
0
10
0.4
-1
-2
-4
10
T > tx
0.2
10
10
1
-3
-2
10
-1
10
0
10
-0.2
1
10
0
10
0
0.5
1
1.5
-2
10
-3
4
4.5
5
4
x 10
1
0.8
correlation coefficient
10
spectral density
3.5
1.2
Correlation coeff.
Spectral density
3
Autocorrelation of flux data
Ensemble-averaged Power spectra of flux data
10
-4
10
-5
1/ t x
10
-6
10
-7
-8
-5
10
-4
10
0.6
0.4
T > tx
0.2
10
10
2.5
Lag [time]
1/time
-1
2
-3
10
-2
10
frequency
1/time
-1
10
0
10
1
10
0
-0.2
0
1
2
3
4
lag [time]
5
6
Lag [time]
7
498
4
x 10
Summary: S2S ideas
• Mass balance as first-order control on
deposit architecture across the sink
• Mass balance and moving boundaries
explain domains fed by multiple inputs
• Weak net offshore pumping from base
level cycles under steady sediment
supply
• Signal shredding by stick-slip transport
How is fluvial sediment mass balance
influenced by offshore conditions?
John B. Swenson1, Jeré A. Mohr1, Chris Paola2,3, & Lincoln F. Pratson4
(1) Department of Geological Sciences, University of Minnesota Duluth, Duluth, MN, USA
(2) Department of Geology & Geophysics, University of Minnesota, Minneapolis, MN, USA
(3) St. Anthony Falls Laboratory, University of Minnesota, Minneapolis, MN, USA
(4) Division of Earth & Ocean Sciences, Duke University, Durham, NC, USA
Ask a fluvial geomorphologist what
controls erosion and deposition in the
fluvial system, and you hear things
like:
•
•
•
•
•
Water discharge
Sediment supply
The ratio of the above
Slope
Grain size
The answer involves local fluvial variables
Let’s look at the problem another way…
Fluvial system is one part of linked depositional system
What role do non-eustatic, downstream processes play in
controlling large-scale fluvial sedimentation?
Choke Points – A Conceptual Model
Motivation: Fluviodeltaic clinoforms migrate as approximately
self-similar waveforms.
Choke Points - Limiting Cases
Mechanisms for Affecting Flux at the Foreset Toe (Qst):
Pre-existing basin
geometry
Clinoform toe “feels”
underlying topography
Steckler et al. (1999)
Alongshore transport
High wave energy can ‘smear’
fluvial sediment flux laterally,
effectively un-choking toe
Turbidity currents
Sustained turbidity currents can reduce
foreset slope (Kostic et al., 2002) and
affect how foreset toe interacts with
underlying topography
Un-choking the clinoform system with a combination of
underlying topography and sustained turbidity currents:
Supporting flume experiments (J. Mohr):
Ramp angle ~ 26º ( ~ 20% < angle of repose)
Silt (40 mm) fed once clinoform toe reaches ramp
Experimental Results – Sustained Turbidity
Currents
No turbidity currents
Turbidity currents
Results: Sensitivity to
concentration of suspended
silt (Csilt)
Results: Fluvial aggradation and shoreline progradation
Fluvial aggradation:
For Csilt > 2%, reduction of
foreset angle stalls system,
resulting in fluvial bypass and
incision
Shoreline response:
For Csilt > 2%, reduced foreset
angle un-chokes clinoform toe,
thereby arresting progradation
Stratigraphic implications:
Conclusions:
• Clinoform toe is a critical point (a ‘choke point’) in
the linked depositional system
• Flux discontinuity across foreset controls
shoreline progradation and large-scale fluvial
sedimentation
• Turbidity currents in combination with basement
geometry can ‘un-choke’ the clinoform system
• Un-choking is a mechanism for sediment transfer
to deep-marine environments