SpaceMika

You know what makes a thesis-cruncher ridiculously happy? Getting her references sorted out!

The glory that is WorldCat allows me to fill in any missing information in my entries (and fix accent-errors induced by my fights with RefWorks, unlinked because it is unloved). The equally glorious JabRef provides a pretty, clean interface for seeing all my references at once, editing any misplaced data (for instance, all my book chapters were classified as “miscellaneous” instead of “in book”, and tagging entries as sane names (say, LastnameYear instead of refworksEntryNumber) so I can actually cite from memory instead of looking up the unique identifier name every time. The clear and complete NatBib Reference Sheet lets me automatically insert references in absolutely any format my little heart can dream of. Finally, the continuing and pervasive joy of LaTeX Wikibooks (this time, the Accent Reference Sheet so I can spell all the foreign places and authors correctly) keeps my thesis beautifully formatted without me needing to fight with it.

…this is the sound of girl well-satisfied with her references.

While reading thesis-writing advice, I came across a gem:

Engineers invent; scientists discover.

When writing a literature review, one of two things is supposed to happen:

1. You know nothing and learn everything. You start off with a mental structure of what information you need to learn, and systematically review the literature to fill in the details (citing references as you go). You end up with an orderly, complete review of the literature and a solid understanding of the field.
2. You know everything and just need to write it out. You start off a master of the material and already know what you’re going to write, and either already know or can easily find the applicable references to support your claims. You end up with an orderly, complete review of the literature that provides a solid understanding of the field for your readers.

My extensive association with graduate students is leading me to believe that both these instances are incredibly rare. What typically seems to happen is more like:

3. You know just enough to know that you don’t know anything. You don’t know enough to structure a systematic search, but you try anyway and write in a disjointed manner with each new reference sending you off on tangents, and when you have enough built that you just barely see the structure you should’ve been aiming for all along, you realize that either you missed a big gaping hole and frantically go searching for the missing key foundational reference, or that something you read for recreational side-reading is actually key evidence and you can’t remember for the life of you what the paper was. You end up with a wobbly, mostly-coherent review and have your fingers crossed that you aren’t missing anything too vital.

Although I’ve found a surprising number of landslide-bloggers (my favourite is Dave’s), google searches on the DAN-W and DAN3D software packages seem to drop people here. I’m fairly regularly getting comments asking how to go about modeling particular landslides, or acquire the software, or related queries. To speed up response time, I’ve developed an F.A.Q.

1. How can I get DAN-W or DAN3D?
DAN-W is owned by Oldrich Hungr. Please see his software website and contact him directly with any inquires about acquiring the software. DAN3D was developed by Scott McDougall as part of his phD thesis. To the best of my knowledge, it is for research purposes only and not currently available commercially, but again, Oldrich Hungr knows for sure.

2. What information do I need about a landslide to model its runout?
For DAN-W, you need a profile of the travel path (including entrainment zones) and the source area, and the width of the path. This can be either a list of coordinates or a to-scale sketch which you can enter directly into the software. For DAN3D, you need digital terrain models (topography) of the area before and after the landslide. You will need to format this as ASCII grid files of the path, the source area, and any entrainment.

3. What rheology should I use?
If you’re doing a back-analysis, you use whatever rheology and parameters provide the appropriate runout distance, debris distribution, and velocity profile. If you’re doing a forward prediction, you can follow the suggestions in my thesis (currently TBA, sorry!), or back-analyze cases similar to your target and use that range of parameters in your prediction.

4. Tell me more about a particular landslide.
If I’ve personally modeled a landslide, it should be floating around this site somewhere. Most are linked off the Thesis page, although the latest versions haven’t been translated from thesis-formatting to website-formatting and hopped online yet.

5. What about modeling this specific landslide not on your website?
If you’re working on modeling a landslide I haven’t seen before, I’m curious. Tell me about it!

It is 5pm, and I have less than 700 hours to complete my thesis. Disjointed thoughts:

- I finished wading through stacks of notebooks with all the data I’ve collected and transcribed it; too much of it is “almost” for me to sleep soundly tonight. For the first time the Thesis page has links to posts you can’t see — I’m sorry about that, but in an effort to transcribe the notebooks some things are in raw list format but I wanted to store those lists somewhere logical.

- I need to do yet one more sort through the hardcopy files of landslide cases to add missing references to my RefWorks database. (Ditto for going through the digital files, but that’s a smaller task.) I should probably pair it with scanning any hardcopy reference that I don’t have digitalized so I can (eventually) toss the hardcopies without losing data. I still dream of the day I can burn CDs (DVDs) of neat little landslide folders containing all the information necessary to model the event — profiles, grds, references, photos, and the final DAN-W or DAN-3D file I used with the results of the best-fit parameters.

- Laptops without external monitors are designed to induce hunched shoulders — despite my rather wired life I’ve never spent enough unending hours that I was unable to realign before the next session until now.

- I need a standard opening sentence for my relic landslides — my “In/On year/date, blah blah” doesn’t work without a year or date!

The target audience of my thesis is me, back when I first started this degree. It is the scientist with only limited geoscience exposure, and who is trying to figure out what’s important and what isn’t in landslide papers and is more than slightly confused by it all. Although the style will be technical and I will be using appropriate vocabulary (“entrain” really can’t be replaced with “scooped up and mixed in” even if it has more accurate denotations), what I choose to include and how I lay it out will be guided by this philosophy.

I spent a recent bus journey meditating on how to express this balance of elegant simplicity while trying to find yet variation of “the debris went down the slope, across the valley, then spread,” I realized something phenomenal: in science, being repetitive isn’t dull, it’s consistent!

I have all my landslides in a database. My lovely CodeMonkey has written a little program to suck the data out and slip it into a standard, consistent paragraph. I am going to hand-edit, but the structure will be written by machine.

Awesome.

For debris flows, the Voellmy model typically produces better simulations, particularly regarding velocities along the path. Based on back-analyses of debris flows worldwide using DAN, calibrated values of f typically range between 0.07 and 0.2, while values of ξ range between 100 and 600 m/s2 (Hungr et al. 1998; Ayotte and Hungr 2000; Jakob et al. 2000; Hürlimann et al. 2003; Revellino et al. 2004; Bertolo and Wieczorek 2005). In the cases presented here, debris flows can be simulated well using the Voellmy rheology, although the parameters are less constrained than avalanches with an f ranging from 0.08 to 0.1 and ξ ranging from 200 to 1,000 m/s².

From Geohazards IV paper:

The frictional rheology can be used for dry friction in the source area, with a transition to the Voellmy model where significant entrainment of saturated soil begins. A similar scheme was used for the coal waste flow slides. A switch from frictional to Voellmy models is often required to produce satisfactory simulation when events initiate on open slopes and then become channelized (Ayotte and Hungr 2000, Hungr and Evans 2004).

From Geohazards IV paper:

For rock avalanches, the frictional resistance model typically produces reasonable simulations of the observed runout distance. Hungr and Evans (1996) and Pirulli (2005) used DAN-W to back-analyze 34 different rock avalanches using the frictional resistance model. The calibrated φb values ranged between 8º and 23º with a mean of 16º. For the Canadian cases, the best result with the frictional rheology falls within this range with and φb of 20º.

Hungr and Evans (1996) also noted that the Voellmy rheology produced consistently good debris distribution, velocity profile, and runout distance for f values between 0.03 and 0.24 and ξ between 100 and 1000 m/s2. Only events involving runout across a glacier or substantial entrainment combined with channelization had calibrated coefficients less than 0.1. For the Canadian cases, the best results were obtained with Voellmy as the dominant rheology, with f between 0.02-0.15 and ξ between 250-500 m/s².

Taken from Geohazards IV paper, requires re-write & expansion for thesis

Rheology

Both DAN models use simple homogeneous hypothetical materials that simulate the bulk behaviour of complex heterogeneous real landslide materials (Hungr 1995). The properties of the hypothetical materials must be assessed through back-analyses of real cases. For our back-analyses of rapid landslides we use frictional, Voellmy, and Bingham rheologies.

The Frictional rheology

In the frictional rheology, the basal shear stress τzx opposing motion is expressed as:

τzx = – σz tanφb [1]

where σz is the total bed-normal stress at the base of the flow and φb is the bulk basal friction angle (with tan φb = (1-ru) tan φ, where ru is the pore pressure ratio and φ is the dynamic basal friction angle). Overestimation of velocities and often unrealistic forward-tapering deposits are characteristics of the frictional model. (McDougall 2006)

The Voellmy rheology

The Voellmy rheology combines frictional and turbulent models such that

τzx = σz f + ρgν²/ξ [2]

where f is the frictional coefficient, ρ is the material density, g is gravitational acceleration, ν is the depth-averaged flow velocity, and ξ is the turbulence term. In comparison to the frictional model, the Voellmy model typically produces better simulations of velocity and deposit distribution.

The Bingham rheology

The Bingham resistance model combines plastic and viscous behaviour. The shear resistance is determined by solving the following cubic equation:

τzx³ + 3 (τyeild/2 + μBingham νxh) τzx² – τyeild³/2 = 0 [3]

where τyeild is the Bingham yield stress and μBingham is the Bingham viscosity. The Bingham model may produce better simulations of events involving clayey or highly plasticity materials.