|
1
|
|
|
2
|
|
|
3
|
|
|
4
|
|
|
5
|
|
|
6
|
|
|
7
|
|
|
8
|
|
|
9
|
|
|
10
|
- “Why prediction of grain behavior is difficult in geophysical granular
systems”
- “…predictive capabilities are, and will remain, relatively independent
of our knowledge of fundamental granular physics”
- “…the constitution of debris flows can be expected to vary accordingly,
so that there is no universal constitutive description of this
phenomenon as there is for hydraulics”
- the variability of granular agglomerations is so large that fundamental
physics is not capable of accurately describing the system and its
variations
- P. Haff (Powders and Grains ’97)
|
|
11
|
- We can do an adequate job in validating modeling/numerical and some
lab/field results. We don’t do so well in other aspects
- Details matter. A lot.
- Predictive capabilities – certainly
for field geophysics – must account for that variability
|
|
12
|
|
|
13
|
- Characteristic length scales (mm to km)
- e.g for Mount St. Helens (mudflow –1985)
- Runout distance »
31,000 m
- Descent height »
2,150 m
- Flow length(L) »
100-2,000
- Flow thickness(H) »
1-10 m
- Mean diameter of sediment material 10-3-10 m
- (data from Iverson 1995, Iverson & Denlinger 2001)
|
|
14
|
|
|
15
|
- Depth average
- the continuity equation:
- where
|
|
16
|
|
|
17
|
|
|
18
|
- Hyperbolic system of balance laws
- Simulation environment TITAN2D
- high order, slope-limiting, upwinding, two dimensional Godunov solver
without splitting
- GIS integration
- several approximate Riemann solvers examined
- parallel, adaptive mesh refinement
|
|
19
|
- Refinement criteria include:
- Regions of significant mass and mass flow
- Regions of large gradients or fluxes in the solution – residual based
error estimators
- Regions of abrupt changes in topography
- Regions of high interest for hazard assessment (hospitals, bridges,
etc.)
- For dynamic problems, also need to un-refine.
- Incorporate unrefinement beyond the original grid
|
|
20
|
- GRASS-Geographic Resources Analysis Support System: open source software
for data management, image processing, graphics production, spatial
modeling and visualization of data
- Required elevation data is obtained dynamically at the scale of the
grid. The other information (slopes, curvatures) are computed on the fly
|
|
21
|
- variation in measured friction angles, depending on measurement
technique
- initial packing of a sample influences initial dynamics, even if that
initial data is ultimately washed out of the system
|
|
22
|
|
|
23
|
|
|
24
|
|
|
25
|
|
|
26
|
|
|
27
|
|
|
28
|
|
|
29
|
|
|
30
|
|
|
31
|
|
|
32
|
|
|
33
|
|
|
34
|
|
|
35
|
|
|
36
|
- Variability in flow
- 35 deg internal friction
- convenient c.o.m. reference
- angles near critical
- basal c.o.m. variance
- .97 .195
- .64 .129
- -.63 .127
- -.64 .126
|
|
37
|
|
|
38
|
|
|
39
|
|
|
40
|
- Tungurahua volcano, Central Ecuador, 5023 m.a.s.l.
- Up to 3000 m of relief over surrounding landscape.
- Deposits from previous activity in 1886 and 1916‑1918 record
highly explosive activity.
- Historical records show Baños may have been affected, and potentially
inundated, by either debris flows or pyroclastic flows as far back as
the mid 1700’s.
- The deposits in the Vazcún Valley record at least two periods of
activity, the ages of which are uncertain but may correlate with the
activity of 1886 and 1916‑1918.
|
|
41
|
|
|
42
|
|
|
43
|
|
|
44
|
- Current 10x10 m DEM for Tungurahua does not accurately represent
topography.
- GPS data show significant differences to current DEM.
- Location of DEM-derived river offset by up to 300 m to the west.
- Simulated flows effectively bypass Baños as a result.
|
|
45
|
|
|
46
|
|
|
47
|
|
|
48
|
|
|
49
|
|
|
50
|
|
|
51
|
|
|
52
|
|
Notes
