Tom Clifton: Fluid Flow: Antidunes

Mount St. Helens Anniversary

Tue, 18 May 2004 17:36:46 GMT

24 years ago right now. At 8:22 a.m. on May 18, 1980 the landscape in the photo above was shattered and transformed into something new. [Mac Net Journal] I am pleased that someone else recognized this event. I was in California on May 18th, but experienced one ash fall that summer while out at Willipa Bay. Five years later, I was working in the blast zone and on the Toutle River, looking at the river deposits produced by the subsequent floods. During this work, I found sedimentary structures produced by antidunes, which ultimately led to my dissertation.

The things you find on the web.

Thu, 13 May 2004 10:14:30 GMT

While reviewing the referer logs for this weblog, I came across a MSN search for "antidune experiments" that generated two hits. I went to the search to see if there were any new data on antidunes. Fluid Flow was the third listing. The second listing was the antidune reference page that I maintain. The top listing pointed to a DOE site that I have never heard about. More importantly the title described antidune structures exactly. So off I went, hoping that someone had made similar observations to mine. What I found, however, was a citation for an abstract that I wrote in 1987. I didn't know that it was available on the web. The abstract is important because it provides a model of how antidunes produce internal sedimentary structures. 17 years later, it still works, though I think upstream dipping strata are more common than I did in 1987.

Mon, 19 May 2003 19:07:32 GMT

NASA offers free aero/fluid dynamics software for OS X. NASA Langley Research Center has released Tetrahedral Unstructured Software System, or TetrUSS, for Mac OS X. TetrUSS is used in aerodynamics and fluid dynamics analysis, and has been used on major projects including High Speed Research / High Speed Civil Transport, Hyper-X, Abrupt Wing Stall, Mars Scout, Joint Strike Fighter and more. What's more, the software has been used in the civilian aerospace industry, academy, automotive, biomedical and civil engineering fields. [MacCentral] Using this to work on antidune flow might be interesting.

Mon, 21 Apr 2003 16:18:43 GMT

DeltaGraph coming to Mac OS X [MacCentral] This is great news. I used DeltaGraph 1.5 to general most of the graphs in my dissertation. To date, DeltaGraph remains the best program for generating scientific graphs that I have found. It is the only application I have that regularly forces me to boot into Classic. DeltaGraph has changed hands multiple times in the past 13 years. The fact that it is still viable and still being upgraded attests to its value.

Sediment Transport in Antidune Flow

Mon, 20 Jan 2003 10:20:04 GMT

Last year I combined some of the thoughts and observations on sediment transport in antidune flow that I posted here into as single story. Unfortunately, I never provided a link to that story. So I am doing it now. The discussion provides an overview of sediment transport in antidune flow and provides a basis for understanding the upstream migration and amplification of antidunes.

Thu, 13 Jun 2002 17:30:48 GMT

Abstract The following abstract is from my dissertation, "Sedimentology of Antidune Flow: controls on sediment transport and stratification" Antidunes are bed configurations that form in sediment under fast, shallow flows. These wave-like features are highly unstable and contrast with more common bedforms, like ripples and dunes, by migrating upstream and changing dramatically in morphology over very short periods of time. Antidunes often amplify rapidly, deforming the flow above them until the water surface becomes unstable and collapses, partially or completely destroying the bedform. Because antidunes display such a dynamic behavior, it is difficult to observe interactions between the flow and the bed and collect data from the flows. As a result, our understanding of how antidunes produce preservable sedimentary structures is limited. Without this understanding, the identification of antidune structures in ancient deposits is difficult and often suspect. The present study aims to improve our understanding of antidunes and their sedimentary structures. It relies on field observations and descriptions of small streams that contain antidunes, the sedimentary structures produced by these streams, and similar structures in ancient deposits. Flow data from streams were collected by video taping antidunes and their associated flows. Experiments where stream channels were altered to form antidunes in rapidly aggrading settings augmented the observations and provided a direct link between antidunes and their internal structures. Antidunes, when migrating in rapidly aggrading settings, produce an intricate pattern of stratification consisting of two distinct types of laminae. The most common type of laminae typically dip in a downstream direction at variable angles and are a type of translatent strata that forms as the antidune trough migrates on an aggrading bed. These thin laminae truncate underlying structures and form the bounding surfaces around inversely graded, lenticular packets of sediment. A second type of laminae mark the instantaneous position of an antidune's upstream face and occur within the lenticular packets. These laminae dip upstream, downlap onto the translatent strata and may occur sporadically. The appearance of antidune structures varies dramatically with aggradation rate and degree of stability displayed by the antidunes. As a result, these structures may be useful in interpreting paleoflow conditions.

Mon, 11 Mar 2002 00:08:36 GMT

Antidune Flow Equations In order to publish my dissertation online, I needed to get all the equations converted to gif images. So I spent some time this afternoon back in Word 3, taking screen shots of the equations. These are the equations for Chapter 4 on sediment transport in antidune flow.

Thu, 07 Mar 2002 00:39:44 GMT

Antidune References I recently posted the reference list from my dissertation to the ES Designs site. Feel free to link to them if you ever have a need to. Each reference has a named anchor consisting of first authors last name and the publication year (<a name="clifton1973">), so you can point directly to a reference. Multiples are differentiated by sequential letters at the end of the entries (for example if there were multiple entries from 1973 with Clifton as the first author, the second reference would be named clifton1973a, the third would be named clifton1973b, etc). You can see examples of how you can link the the list in the discussion below.

Wed, 06 Mar 2002 22:40:12 GMT

Patterns of Erosion and Deposition in Antidune Flow The distribution of shear stress along the bed drives sediment transport (Middleton and Southard, 1984), with the sediment discharge in the flow being a function of the bed shear stress. Sediment discharge is greater where bed shear stress is greater, because the higher bed shear stresses can remove more sediment from the bed. McLean and Smith (1986) indicate that changes in bed elevation with time (whether the sediment is being eroded from or deposited to the bed) is a function of sediment discharge along the bed in a two-dimensional flow. Which in turn is directly related to the bed shear stress. If you move along the bed from a zone of high sediment discharge to low sediment discharge, sediment will accumulate with time. Conversely if you move from a zone of low sediment discharge to high sediment discharge, sediment will be removed from the bed with time. Because sediment discharge is directly related to bed shear stress, the points of maximum and minimum bed shear stress on the bed define boundary points between zones of increasing and decreasing sediment discharge. Erosion (removal of sediment from the bed over time) occurs where bed shear stress is increasing along the bed. Deposition (addition of sediment to the bed over time) occurs where bed shear is decreasing along the bed. Now if you look at the patterns of bed shear stress along the bed and the resulting zones of deposition and erosion, you can see why antidunes migrate upstream and why they typically build in amplitude. Starting the crest of the upstream antidune, there is a shear stress minimum just downstream of the crest. From this minimum, bed shear stress increases along the downstream side of the bedform to a maximum just downstream of the trough. Erosion occurs along this interval. Moving on from this bed shear stress maximum, bed shear stress drops along the upstream side of the bedform, as the flow slows down and gets deeper, to the bed shear stress minimum just downstream from the crest. Depostion occurs in this interval of decreasing bed shear stress. The result is that sediment is added to the upstream side of the bedform and removed from the downstream side. As this happens, the bedform moves upstream. Similarly, since the trough is in a zone of erosion and the crest is in zone of deposition. The bedform will build in amplitude over time.

Bed Shear Stress

Mon, 04 Mar 2002 21:35:11 GMT

Bed Shear Stress Sediment transport is driven by bed shear stress. The shear stress (force per unit area) of the flow on the bed is what plucks sand gains off of the bed and entrains them in the flow. If the bed shear stress exceeds a threshold value for a particular grain size and composition, that grain will be lifted from the bed and carried by the flow. If the bed shear stress drops below that threshold value, the entrained grain will drop back to the bed. So the bed shear stress controls whether erosion (removal of grains) or deposition (settling of grains) will occur. The bed shear stress (tau t) is a function of the gradient in the velocity profile of the flow. This gradient is essentially a function of the flow velocity over flow depth (U/d). In antidunes, this gradient is at a maximum in the trough and minimum at the crest since U_T/d_T is greater than U_C/d_C. Since the flow has a mass, there are additional components to the bed shear stress. The flow's inertia causes it to impinge on the bed at trough and separate from the bed at the crest. These effects shift the bed shear stress maximum (t_max) and minimum (t_min) in a downstream direction from the trough and crest as shown in the figure above.

Mon, 04 Mar 2002 01:40:59 GMT

Sediment Transport in Antidune Flow The key to understanding antidunes, their behavior, and the sedimentary structures that they produce requires insight into how sediment is transported in antidune flows. It is remarkable that after nearly 90 years of study, there are no published studies detailing sediment transport in antidune flow. The following discussion provides an overview of sediment transport in antidune flow. A detailed discussion gets into some hairy equations that I haven't looked for over 10 years, so we will avoid that for now. But the general discussion will show how the distribution of flow velocity and depth along the bed sets up the distribution of bed shear stress. The bed shear stress distribution, in turn, controls the sediment flux and determines the zones of deposition and erosion on the bed. Starting with a simple configuration of antidunes with an amplitude (a) and wavelenth (lamda l). Flow over the trough and crest is supercritical, meaning that at both the trough and crest the ratio of flow velocity (U) to the square root of flow depth (d) times gravitational acceleration is greater than one. Supercritical flows respond to changes in bed elevation differently than subcritical flows. If a supercritical flow encounters positive feature on the bed (like a mound sticking up in the flow), it responds by slowing down and increasing in depth (Middleton and Southard, 1984). Similarly, for a negative feature, the flow becomes faster and shallower. With antidunes, the crest is a positive feature on the bed, while the trough is a negative feature. The flow responds so that the flow velocity in the trough (U_T) is greater than the flow velocity at the crest (U_C), while the flow depth in the trough (d_T) is less than the flow depth at the crest (d_C).

Wed, 27 Feb 2002 18:36:56 GMT

Discussing Antidunes I had been thinking of creating a new Manila site to host my discussions on antidunes. With the advent of comments, however, I don't think that is necessary. My concern is that I am posting information that has not gone through a formal peer-review process (beyond my dissertation committee). As such, it has not achieved the same level of standard as a scientific journal article. A Weblog like this, however, with its high degree of linking, has the potential for top listings in web search engines. So people searching for antidunes could see my information first. I don't feel comfortable presenting my data and ideas, without allowing other workers in sediment transport and fluid flow to present their thoughts. This must be a discussion, not a presentation. The comment feature allows a level discussion. It is not perfect and it may not reach peer-review standards. But, it can allow us to advance the science. I welcome your thoughts on the data and interpretations presented here.

Mon, 11 Feb 2002 18:46:53 GMT

Antidune Trivia G.K. Gilbert coined the term "antidune" in a 1914 US Geological Survey Professional Paper entitled "Transportation of debris by running water". The paper reports the results of a series of early flume studies* that Gilbert performed at the University of California in Berkeley. I would love to get a copy of this paper for my library. *flume studies are a method of studying bedforms and channelized flow. An artificial channel is set up in a laboratory and then flows of various speeds and depths are run through the channel to see how the channel bed interacts with the flow.

Fri, 01 Feb 2002 16:36:45 GMT

On the technical details of the image and its line drawing below. In my dissertation, I recognized two distinct types of laminae produced by antidunes. The first, which I call "type 1" lies in sharp erosional contact with the underlying sediment. In my box cores, these thin (<1 mm thick) laminae are defined by fine, magnetite-rich sand with coarser arkosic sand lying gradationally above the laminae. These laminae define the basal portion of a inversely-graded packet of sediment that are commonly lenticular, though can be somewhat tabular in appearance. These laminae typically dip in the downstream direction, though locally show a transition from downstream to upstream dips, giving a sigmoidal or s-shaped appearance to the laminae. The second type of laminae, "type 2", are defined by alternating zones of magnetite-rich and arkosic sand. These laminae show variable thickness between 0.5 and 3 mm. The laminae generally dip in an upstream direction and are flat or concave up along their length. Type 2 laminae are truncated by type 1 laminae at their up-dip ends and merge into type 1 laminae in the down-dip direction.

Thu, 31 Jan 2002 10:49:41 GMT

Last night, I posted a cleaner version of the box core image I put up last week. Apparently Canvas does not do as good a job generating jpegs from tiff images as Adobe ImageReady.

Mon, 28 Jan 2002 19:23:43 GMT

Internal sedimentary structures produced by antidunes. The image is a positive print of an xray-radiograph taken of a box core from a stream that was manipulated to produce antidunes under aggrading* conditions. The scale bar in the line drawing is 13 cm long. * aggrading means sand was accumulating on the the stream bed, so that the stream bed was building up with time. Some amount of aggradation is necessary to preserve sedimentary structures until the sediments become rock.

Sat, 26 Jan 2002 07:13:55 GMT

I received a reprint of the antidune paper in Sedimentology that I mentioned last week. The paper describes sedimentary structures produced by antidunes in a laboratory flume. The study shows many of the features that I described in my dissertation. The one thing that is missing from the study is an explanation of how the internal sedimentary structures that they observe are produced by antidunes. There is not a clear explanation of what is happening on the bed to produce the structures. I will see if I can explain this clearly.

Sat, 19 Jan 2002 01:28:34 GMT

[Macro error: Can't evaluate the expression because the name "116875776" hasn't been defined.] Antidunes in the wild. This picture shows the surface waves breaking above antidunes in San Gregorio Creek, CA. Antidunes often display a behavior of building until the surface wave above them breaks and disrupts the flow over the bedform. The stream is flowing left to right.

Sat, 19 Jan 2002 00:44:31 GMT

[Macro error: Can't evaluate the expression because the name "116875776" hasn't been defined.] This schematic shows basic form of an antidune. The stream bed is deformed into a set of waveforms (bedforms) with a definite crest and trough. The water surface is also deformed into a similar waveform. The water surface waveform is in-phase with the bedform. The water current is flowing from left to right while the bedform migrates from right to left.

Fri, 18 Jan 2002 23:42:25 GMT

Antidune Basics Antidunes are in-phase bedforms that form on a sandy substrate beneath super-critical flows. Antidunes migrate upstream against the current flow, opposite the migration direction of most bedforms (such as dunes and ripples). What does this mean? Antidunes are type of bedform (any regular, wave-like form that occurs on a channel bed). Antidunes are in-phase with the water surface above them, so the crest and trough of the antidune are matched by a crest and trough in the water's surface above them. Antidunes form in fast, shallow flows where the flow is described as super-critical. Most other bedforms, such as ripples and dunes, form in sub-critical flows that are generally deeper and slower. The two most striking features of antidunes, that they are in-phase with the water's surface and migrate upstream, are in contrast to most other bedforms. These features result from the contrasting flow regimes (super-critical vs sub-critical) that antidunes and other bedforms form in. Antidunes commonly occur in small streams that flow across beaches. Antidunes appear where the stream flows down the beach face toward the ocean. You will see trains of waves at the streams surface. If you look beneath the waters surface, you will notice a train of waves in the stream bed that mimics the waters surface. These bedforms are antidunes. As you watch the surface waves carefully, you will notice that they are moving upstream against the current. This upstream movement of the surface wave is caused by the upstream migration of the underlying antidune.

Fri, 18 Jan 2002 16:52:05 GMT

Last night I found the archives of my antidune data, texts, and graphics. Today's challenge will be finding the old applications to access the files. Does anyone have an old copy of Personal Press that I could borrow?

Thu, 17 Jan 2002 22:25:46 GMT

Hey, it looks like people are still working on antidunes. Here is an abstract from an article that is in press in Sedimentology.

Thu, 17 Jan 2002 20:32:28 GMT

Here is a quicktime movie of antidunes in a flume.