small steps and giant leaps? - purdue...
TRANSCRIPT
Small steps and giant leaps? (Thoughts for the future based on a decade of progress in
© John Wainwright: [email protected]
(Thoughts for the future based on a decade of progress in understanding and modelling hydrologic and erosion processes)
John WainwrightDepartment of Geography, University of Sheffield(Durham University from 1/10/11)
ISELE September 20th 2011 Anchorage
…with a little help from:
• Tony Parsons
• Eva Müller
• Laura Turnbull
• Richard Brazier
• James Cooper
• Peter Biró
• James Millington
• Irantzu LexartzaArtza
• Tristan Ibrahim
• Becky Briant
© John Wainwright: [email protected]
• Ben Kitchener
• Graham Hargrave
• Ed Long
• Mark Powell
• Greg Okin
• Jill Stewart
• Banti Fenti
• Roger Simanton
• Yuichi Onda
• Tomomi Furukawa
• Eiichiro Obana
• Becky Briant
• Gareth Mottram
• Brandon Bestelmeyer
• Ed Fredrickson
• Bill Schlesinger
• Athol Abrahams
• Andy Baird
• Nikki Dodd
• Sarah Dunn
• …
The problem
© John Wainwright: [email protected]
Overall framework
space travel distance
Lajeunesse
et al. (2010) WRR
Geomorphology
© John Wainwright: [email protected]
time
space-time(process)
methodology?
virtual velocity
connectivity
Wainwright et al. (2011) Geomorphology
A small step backwards ...
• Why model? Epstein’s Sixteen Reasons Other Than Prediction to Build Models (JASSS vol. 11, no. 4 12):
1. Explain (very distinct from predict)
10. Demonstrate tradeoffs / suggest efficiencies
© John Wainwright: [email protected]
predict)
2. Guide data collection
3. Illuminate core dynamics
4. Suggest dynamical analogies
5. Discover new questions
6. Promote a scientific habit of mind
7. Bound (bracket) outcomes to plausible ranges
8. Illuminate core uncertainties.
9. Offer crisis options in near-real time
suggest efficiencies
11. Challenge the robustness of prevailing theory through perturbations
12. Expose prevailing wisdom as incompatible with available data
13. Train practitioners
14. Discipline the policy dialogue
15. Educate the general public
16. Reveal the apparently simple (complex) to be complex (simple)
A simple exercise in data collection? (A historical step backwards)
A journey of a thousand
© John Wainwright: [email protected]
A
B
C
a thousand miles begins with a single step
Lao-Tzu
What do we measure when we measure “erosion”?
© John Wainwright: [email protected]
Conventional (implicit) assumption
Assuming limited transport distances
fluxyield
Assuming exponential transport distances(and uniform slope conditions)
0.8
0.9
1
sedim
ent-deliv
ery
ratio 4 2
1 0.5
λ =
“
© John Wainwright: [email protected]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 20 40 60 80 100slope length m
sedim
ent-deliv
ery
ratio
1 0.5
0.25 0.1
0.05 0.01
(effective)
“
(Wainwright et al., 2001)
dueKLxu
xLx
x
9/42
)(
1)( −+
∫=+ϕ
-1
1 mm
2 mm
∫+
=+Lx
x
dELx ξξϕ .)()(
Analytical scaling based on sediment flux
© John Wainwright: [email protected]
distance m
sedim
ent flux g s
-1 2 mm
4 mm
Theoretical curves for interrill erosion based on travel distance (Parsons et al., 2004)
Instrument housing
© John Wainwright: [email protected]
Rain gauge
2 m2 plot reservoir
21 m2 plot reservoirs
housing
Flume
sedim
ent flux g s
-1
1 mm
2 mm
dueKLxu
xLx
x
9/42
)(
1)( −+
∫=+ϕ
0.6
0.7
0.8
∫+
=+Lx
x
dELx ξξϕ .)()(
Analytical scaling based on sediment flux
© John Wainwright: [email protected]
distance m
sedim
ent flux g s
4 mm
0
0.1
0.2
0.3
0.4
0.5
0.6
0 5 10 15 20 25 30
Measured on interrill plots at Walnut Gulch (Parsons et al., 2006)
Theoretical curves for interrill erosion based on travel distance (Parsons et al., 2004)
time silt & clay sand
min Measured Modelled Measured Modelled
0 - 4 32 % 28 % 68 % 72 %
5 - 9 53 % 26 % 47 % 74 %
10 - 20 27 % 28 % 73 % 72 %
© John Wainwright: [email protected]
Wainwright et al. (2008b) ESPL
5
6
7
8
9
modelle
d s
edim
ent k
g
equation 19
equation 20b
1:1
© John Wainwright: [email protected]
0
1
2
3
4
0 1 2 3 4 5 6 7 8 9
observed sediment kg
modelle
d s
edim
ent k
g
Wainwright et al. (2008c) ESPL
Wainwright et al. (2008c) ESPL© John Wainwright: [email protected]
Back to the data ...© John Wainwright: [email protected]
MAHLERAN - Marker in Cell
Cooper et al. (forthcoming)
© John Wainwright: [email protected]
flow depth marker particles
Cooper
© John Wainwright: [email protected]
Magnetite study 16 years later...
total erosion 0·697 kg m−1
depth of soil loss 0·033 mm a−1
© John Wainwright: [email protected]
Parsons et al. (2010)
background
Simulation of markers moving
away from
seedline
MAHLERAN - Marker in Cell
© John Wainwright: [email protected]
seedline
Cooper et al. (forthcoming)
New techniques
Dynamics of Runoff And Erosion Modelling (DRÆM)
© John Wainwright: [email protected]
www.shef.ac.uk/draem
200 – 212 μm dry sand
2.5 mm droplet impacting at 5.94 m s-1
Dynamics of Runoff And Erosion Modelling (DRÆM)ti
me
© John Wainwright: [email protected]
3.5 mm droplet impacting at 6.13 m s-1
200 – 212 μm dry sand
Image sequence at 3000 fpsSequence duration 0.11 s
Trajectories of splashed particlestim
e
260
280
300
320
340
0 100 200 300 400 500 600 700 800 900
260
280
300
320
340
360
380
Interaction of single water droplet with a flat surface – Particle TrackingR
aw
Im
ag
e
(pix
els
)
Ce
ntr
e l
oca
tio
ns
of
ide
nti
fie
d
pa
rtic
les
Dynamics of Runoff And Erosion Modelling (DRÆM)
© John Wainwright: [email protected]
0 100 200 300 400 500 600 700 800 900
360
380
Ce
ntr
e l
oca
tio
ns
of
ide
nti
fie
d
pa
rtic
les
Long et al. (forthcoming)
200 250 300 350 400 450 500 550 600 650 700
260
280
300
320
340
360
380
Velocity (m s-1)
Displacement (mm)
Dis
pla
cem
en
t (m
m)
3000 fps, 1024 x 512 pixels (64.6 mm x 32.4 mm), 0.063 mm/pixel resolution
Water diameter = 2 mm, Sand Dia. < 212 µm, Height 0.5 m, Impact velocity = 2.8 m s-1
Pa
rtic
le t
rack
s a
nd
ass
oci
ate
d v
elo
city
pro
file
Dynamics of Runoff And Erosion Modelling (DRÆM)
© John Wainwright: [email protected]
Surface-profile measurement taken using
calibrated images of the sand surface Long et al. (forthcoming)
Red
Dynamics of Runoff And Erosion Modelling (DRÆM)
Further developments:
tracking of multiple particle
sizes simultaneously in 3D
© John Wainwright: [email protected]
Green
BlueColoured particles 0.2 – 0.6 mm
...with neutral buoyancy
particles to track flow fields
RFID tagging of particles
© John Wainwright: [email protected]
Changes in measured water and sediment flux as
0
2000
4000
6000
8000
10000
12000
0 10 20 30 40 50 60
Qt (l)
Plot 1: grassPlot 2: grass shrubPlot 3: shrub grassPlot 4: shrub
Connectivity : joined-up thinking?(1)
54.5%bare
(2)57.0%
© John Wainwright: [email protected]
flux as vegetation structure changes from grass to shrub
0 10 20 30 40 50 60
ER (mm)
0
2000
4000
6000
8000
10000
12000
0 10 20 30 40 50
Sedim
ent y
ield
(g)
ER (mm)
Plot 1: grass
Plot 2: grass shrub
Increasing connectivity
Turnbull et al. (2010) Hydol Proc(2010) Ecosystems
(3)73.8%
(4)76.7%
Tarbush
Grass
GrassTarbush
800%
1000%
1200%
Water800%
1000%
1200%
Grass
downslope of shrubs:
Shrubs
downslope of grass:
The importance of place
© John Wainwright: [email protected]
Grass
300 m 140 m upslope 140 m downslope
0%
200%
400%
600%
800%
-140 -90 -40 10 60 110
distance from ecotone m
change in flu
x
Water
Sediment
Ammonium
0%
200%
400%
600%
800%
-140 -90 -40 10 60 110
change in
flu
x
distance from ecotone m
Water Water
Sediment Sediment
Ammonium Ammonium
Müller et al. (2008) Ecological Modelling
Complexity and evolution
Chaos is merely order waiting to be deciphered
José Saramago
Top-down (Newtonian*)• Grid- or element-based (usually fixed)• Imposed model structure across entire domain
. (2006) WRR
© John Wainwright: [email protected]
domain• Assumes system “knows” about all other parts• Represents dynamics but in a static way
Bottom-up (Darwinian*)• Cell-, particle- or individual-based (and may change through time)• Model based on local interactions or rules• Only interacting elements “know” about each other• Can represent evolutionary dynamics
*terms of Newman et al. (2006)
Longer time scales
© John Wainwright: [email protected]
Relative temperature change over 742 ka based on EPICA DOME C core of Augustin et al. (2004)
0 1 2
0
1
2Small changes in catchment structure
Produce complex patterns in
© John Wainwright: [email protected]
0
1
2
Wainwright (2006) Catena
Produce complex patterns in catchment runoff and
sediment flux as a result of changing climate
... which are even more complex when rainfall and temperature
change are asynchronous
Comparison against sedimentary data
© John Wainwright: [email protected]
Briant et al. (forthcoming)
CybErosion
© John Wainwright: [email protected]
• Incorporates dynamic behaviour of animals and
vegetation and effects of human decision-makingWainwright (2008); Wainwright and Millington (2010)
Closing Thoughts
• Models have various uses, so too must data
• What are the inbuilt assumptions of existing models (and thus data)?
© John Wainwright: [email protected]
thus data)?
• Parameterization based on physical principles rather than empirical convenience ...
An example of limited parameterization data
• The threshold for flow detachment for mixtures of particle sizes can be γ
θ
β
β1 β2
A
© John Wainwright: [email protected]
particle sizes can be demonstrated to vary significantly about the threshold for a uniform particle size ...
• ... so threshold parameters cannot be independent of each other (and must change dynamically)
θ
0
0.05
0.1
0.15
0.2
0.25
0.05 0.15 0.25 0.35 0.44 0.54 0.64 0.74 0.84 0.93 1.03 1.13
critical threshold velocity m s-1
frequency
125 µm
2 mm
125 µµµµm
500 µµµµm
4 mm
Wainwright et al. (2003)
Closing Thoughts• Models have various uses, so too must data
• What are the inbuilt assumptions of existing models (and thus data)?
© John Wainwright: [email protected]
• ... and thus also be dynamic
• ... and reflect uniqueness of place
• What do models produce? And is it the same with data?
• So what are we actually doing when we compare models with data?
• Parameterization based on physical principles rather than empirical convenience ...
Closing Thoughts
• How “valid” are “validation” data?
• Most data have implicit scales which need to be recognized when parameterizing models
© John Wainwright: [email protected]
recognized when parameterizing models
• Calibration is NEVER the last step of a study (nor is application of a calibrated model)
• Connectivity means seeing the environment as a whole
• ... which means also speaking to other environmental (including social) scientists (disciplinary connectivity)
• Complex response means we need to test the limits of existing reductionist approaches
ReferencesParsons, AJ, J Wainwright and AD Abrahams 1993 ‘Tracing sediment movement on semi-arid
grassland using magnetic susceptibility’, Earth Surface Processes and Landforms 18, 721–732.
Wainwright, J, AJ Parsons, DM Powell and RE Brazier 2001 ‘A new conceptual framework for understanding and predicting erosion by water from hillslopes and catchments’, in JC Ascough II and DC Flanagan (eds) Soil Erosion Research for the 21st Century. Proceedings of the International Symposium, 607–610, American Society of Agricultural Engineers, St
Joseph, MI.
Brazier, RE, DM Powell, AJ Parsons and J Wainwright 2003 ‘Monitoring and modelling runoff in semi-arid areas from the hillslope to the watershed scale’, in KG Renard, SA McElroy, WJ Gburek, HE Canfield and RL Scott (eds) First Interagency Conference on Research in the Watersheds (Benson, AZ, October 27–30, 2003), 101–106, USDA-ARS, Washington DC.
Powell, DM, RE Brazier, J Wainwright and AJ Parsons 2003 ‘Stream bed scour and fill in low-order ephemeral stream channels’, in KG Renard, SA McElroy, WJ Gburek, HE Canfield and
Müller, EN, J Wainwright and AJ Parsons 2007 ‘The stability of vegetation boundaries and the propagation of desertification in the American Southwest: a modelling approach’, Ecological Modelling 208, 91–101.
Parsons, AJ, J Wainwright, RE Brazier and DM Powell 2008 ‘Is sediment delivery a fallacy? Reply’, Earth Surface Processes and Landforms 33, 1360–1361. DOI: 10.1002/esp.1627.
Parsons, AJ, J Wainwright, RE Brazier and DM Powell 2008 ‘Scale relationships in hillsloperunoff and erosion. Reply’, Earth Surface Processes and Landforms 33, 1637–1638. DOI: 10.1002/esp.1628.
Tatard, L, O Planchon, J Wainwright, G Nord, D Favis-Mortlock, N Silvera, O Ribolzi, M Estevesand C-H Huang 2008 ‘Measurement and modelling of high-resolution flow-velocity data under simulated rainfall on a low-slope sandy soil’, Journal of Hydrology 348, 1–12. doi: 10.1016/j.jhydrol.2007.07.016.
Wainwright, J, AJ Parsons, EN Müller, RE Brazier, DM Powell and B Fenti 2008 ‘A transport-distance approach to scaling erosion rates: 1. background and model development’, Earth Surface Processes and Landforms 33, 813–826. DOI: 10.1002/esp.1624
Wainwright, J, AJ Parsons, EN Müller, RE Brazier, DM Powell and B Fenti 2008 ‘A transport-distance approach to scaling erosion rates: 2. Sensitivity and evaluation of MAHLERAN’, Earth Surface Processes and Landforms 33, 962–984. DOI: 10.1002/esp.1623
© John Wainwright: [email protected]
order ephemeral stream channels’, in KG Renard, SA McElroy, WJ Gburek, HE Canfield and RL Scott (eds) First Interagency Conference on Research in the Watersheds (Benson, Az, October 27–30, 2003), 180–185, USDA-ARS, Washington DC.
Parsons, AJ, J Wainwright, DM Powell, J Kaduk and RE Brazier 2004 ‘A conceptual model for understanding and predicting erosion by water’, Earth Surface Processes and Landforms 29 (10), 1293–1302.
Powell, DM, RE Brazier, J Wainwright, AJ Parsons and J Kaduk 2005 ‘Stream-bed scour and fill in low-order dryland channels’, Water Resources Research 41, W05019, doi:10.1029/2004WR003662.
Parsons, AJ, J Wainwright, RE Brazier and DM Powell 2006 ‘Is sediment delivery a fallacy?’, Earth Surface Processes and Landforms 31, 1325–1328.
Parsons, AJ, RE Brazier and J Wainwright 2006 ‘A conceptual model for sediment and nutrient fluxes from rural land’, International Journal of Biodiversity Science and Management 2, 1–3.
Parsons, AJ, RE Brazier, J Wainwright and DM Powell 2006 ‘Scale relationships in hillsloperunoff and erosion’, Earth Surface Processes and Landforms 31, 1384–1393.
Powell, DM, RE Brazier, J Wainwright, AJ Parsons and M Nichols 2006 ‘Spatial patterns of scour and fill in dryland sand bed streams’, Water Resources Research 42, W08412, doi:10.1029/2005WR004516.
Powell, DM, RE Brazier, AJ Parsons, J Wainwright and M Nichols 2007 ‘Sediment transfer and storage in dryland headwater streams’, Geomorphology88, 152–166.
Brazier, RE, AJ Parsons, J Wainwright, DM Powell and WH Schlesinger 2007 ‘Upscalingunderstanding of nutrient dynamics associated with overland flow in a semi-arid environment’, Biogeochemistry 82, 265–278.Doi: 10.1007/s10533-007-9070-x
Müller, EN, J Wainwright and AJ Parsons 2007 ‘The impact of connectivity on the modelling of water fluxes in semi-arid shrubland environments’, Water Resources Research 43, W09412, doi:10.1029/2006WR005006.
Surface Processes and Landforms 33, 962–984. DOI: 10.1002/esp.1623
Wainwright, J, AJ Parsons, EN Müller, RE Brazier, DM Powell and B Fenti 2008 ‘A transport-distance approach to scaling erosion rates: 3. Evaluating scaling characteristics of MAHLERAN’, Earth Surface Processes and Landforms 33, 1113–1128, DOI: 10.1002/esp.1622
Wainwright, J, AJ Parsons, EN Müller, RE Brazier and DM Powell 2009 ‘Response to Hairsine’sand Sander’s “Comment on ‘A transport-distance based approach to scaling erosion rates:’ Parts 1, 2 and 3 by Wainwright et al.”’, Earth Surface Processes and Landforms 34, 886–890, DOI: 10.1002/esp.1781.
Wainwright, J, AJ Parsons, EN Müller, RE Brazier and DM Powell 2009 ‘Response to Kinnell’s“Comment on ‘A transport-distance approach to scaling erosion rates: 3. Evaluating scaling characteristics of MAHLERAN’”’, Earth Surface Processes and Landforms 34, 1320–1321.
Brazier, RE, J Wainwright and AJ Parsons 2010 ‘Scaling soil-erosion models in space and time', in RPC Morgan and MA Nearing (eds) Handbook of Erosion Modelling, 98–116, Wiley-Blackwell, Chichester.
Parsons, AJ, J Wainwright, T Furukawa and Y Onda 2010 ‘Using sediment travel distance to estimate medium-term erosion rates: a sixteen-year record’. Earth Surface Processes and Landforms 35, 1694–1700. DOI: 10.1002/esp.2011.
Turnbull, L, J Wainwright and RE Brazier 2010 ‘Hydrology, erosion and nutrient transfers over a transition from semi-arid grassland to shrubland in the South-Western USA: a modelling assessment’. Journal of Hydrology 388, 258–272, DOI:10.1016/j.jhydrol.2010.05.005.
Wainwright, J, AJ Parsons, EN Müller, RE Brazier and DM Powell 2010 ‘Standing proud: a response to “Soil-erosion models: where do we really stand?” by Smith et al.’, Earth Surface Processes and Landforms 35, 1349–1356. DOI: 10.1002/esp.2047
Turnbull, L, J Wainwright and RE Brazier 2011 ‘Nitrogen and phosphorus dynamics during runoff events over a transition from grassland to shrubland in the south-western United States’. Hydrological Processes 25, 1–17. DOI: 10.1002/hyp.7806