XES 96-1 Run

Goal: Run 96-1 was carried out in a small (10-cell) prototype of the XES basin. It was the first test of the experimental stratigraphy system.

Reseachers: Chris Paola, Jim Mullin, Chris Ellis, David Mohrig, John Swenson, Gary Parker, Tom Hickson, Paul Heller, Lincoln Pratson, James Syvitski, Ben Sheets, Nikki Strong

Overview:
1. The run was designed to compare shoreline migration and stratigraphy for slow and rapid base-level cycles, where slow and rapid refer to cycle period relative to the basin equilibrium time. Subsidence was bowl-shaped (maximum in the basin center) and constant in time. The supplied sediment comprised 50% quartz sand and 50% anthracite sand by volume. The supply rate was constant in time.

Result:
1. The major finding of the run was that slow base-level cycles produce neither the strong phase lag nor the attenuated shoreline response that had been predicted theoretically. However, both base-level cycles produced a significant overshoot of shoreline relative to its position at the start of the cycle. A set of spectacular growth faults developed in the interval between the cycles.

Movies: (right click to download)
[jt96_vertical_sections.avi: 2.6Mb]
: Series of vertical sections of sedimentary deposit

[jt96_run.mov: 5Mb]
: Fast Fall - Incised valley development during a rapid fall in base level. Key features are labled in the movie. The incised valley that forms in relatively narrow, steep walled, and lengths basinward as delta at channel mouth is exposed by continued base level fall. Growth faults develop during base level fall.
: Slow Fall - In this experiment absolute base level fall takes place at rates similar to basin subsidence. The geometry and rates of subsidence of the basin floor is balanced such that the shoreline progrades from a zone of relative base-level rise (i.e. absolute base level fall is slower than basin subsidence producing a slow relative rise) into a zone of relative base-level fall (i.e. rate of all of absolute base level is faster than basin subsidence). At this point (see Figure), which is where the video starts, a nickpoint rapidly cuts headward back to the source. As the nickpoint retreats, the slight increase in sediment supplied just downstream forces deposition. Thus as the knick point steps to the right so to does the attendant zone of aggradation. Over time the valley widens into a broad basin-scale unconformity. See discussion in Heller et al., (2000) for details.

Publications:
Heller P.L., Steel R., and Paola C., 2001, Geomorphology and sequence stratigraphy due to slow and rapid base-level changes in an experimental subsiding basin (XES96-1), American Association of Petroleum Geologists Bulletin, v.85, no.5, p. 817-838 [PDF]

Paola, C., J. Mullin, C. Ellis, D.C. Mohrig, J.B. Swenson, G. Parker, T. Hickson, P.L. Heller, L. Pratson, J. Syvitski, B. Sheets, and N. Strong, 2001, Experimental stratigraphy, GSA Today, v.11, no.7, p.4-9. [PDF]

XES 99-1 Run

Goal: The main scientific aim of this experiment was to investigate the influences of various allocyclic controls on alluvial architecture. The experiment was divided into four stages, each of which isolated a particular tectonic or climatic scenario. Further, as the first experiment in the full-scale XES facility, this experiment served as a test of the subsidence and data collection instrumentation.

Reseachers: Ben Sheets, Nikki Strong, Tom Hickson, Chris Paola

Result:
1. Laterally (cross-stream) asymmetric subsidence did not produce a measurable signal in the channel stacking patterns-channel density was not higher in regions of rapid subsidence. Though not an intended consequence of the experimental design, this phenomenon was a result of a relatively high wetted fraction of the fluvial surface (30-40%), and of a relatively low water to sediment discharge ratio (Qw/Qs; ~50:1) that led to highly mobile fluvial channels. These factors led to a scenario in which the lateral subsidence variation never produced a topographic expression, and therefore fluvial channels were not attracted to that region (see Hickson et al., in submittal)

2. A decrease in subsidence rate, while other parameters were held constant, led to a decrease in channel stacking density in this experiment. Though a counter-intuitive result, we have come to understand this behavior as a consequence of decreased fluvial transport capacity during the slow-subsidence stage of the experiment. This effect can be accounted for by a transformation from linear spatial coordinates to a sediment extraction coordinate system, which accounts for variation in basin-scale sedimentation patterns (see Strong et al., in press).

3. In this experiment, the bulk of alluvial deposition was accomplished by short-lived flows, as indicated by poor correlation between flow occupation and short-term sedimentation patterns. Established channels acted largely as conduits for sediment, while overbank flow expansions and failed avulsions deposit a disproportionate amount of sediment. This is a phenomenon for which there is field as well as experimental evidence, suggesting that it is a generic feature of channelized flow systems (see Sheets et al., 2002).

4. There is a consistent scale that measures the time required to average individual depositional events into large scale stratal patterns. We term this scale the 'stratigraphic integral scale.' In this experiment, the stratigraphic integral scale is equal to the time necessary for the deposition of several (5-10) scour depths worth of sediment at the average aggradation rate (see Sheets et al., 2002).

5. Estimating the reduction of the dry fraction (i.e., percentage of surface not affected by flooding since some arbitrary time) with time may provide an improved approach to the evaluation of the risk associated with alluvial fan flooding. The reduction of dry fraction can be approximated by a harmonic law, of which the characteristic decay time is proportional to the average cross sectional area of the flow, and inversely proportional to the sediment supply (see Cazanacli et al., 2002).

Movies: (right click to download)
[jt99_surface_flow.avi: 30.3Mb]
: General overview of various important flow processes present in stage 3 of the XES 99-1 experiment, as discussed in Cazanacli et al. 2002, and Sheets et al. 2002. This movie is annotated.

[p1.t2.37-39.mn021-40.mov: 18.5Mb]
: This is an unannotated time-lapse video from approximately runtime 37:21 to 37:40 during stage 1 of the XES 99-1 experiment. Stage 1 was characterized by laterally (cross stream) asymmetric and rapid subsidence, with high water and sediment supply (see Sheets, et al., 2002 for stage description).

[p2.t11.44-46.0-20.mov: 18.6Mb]
: This is an unannotated time-lapse video from approximately runtime 44:00 to 44:20 during stage 2 of the XES 99-1 experiment. Stage 2 was characterized by laterally symmetric, rapid, rigid beam subsidence, with high water and sediment supply (see Sheets, et al., 2002 for stage description).

[s3.t25.88-90.mn021-041.mov: 36.6Mb]
: This is an unannotated time-lapse video from approximately runtime 88:21 to 88:41 during stage 3 of the XES 99-1 experiment. Stage 3 was characterized by laterally symmetric, slow, rigid beam subsidence, with low water and sediment supply (see Sheets, et al., 2002 for stage description).

Publications:
Cazanacli, D.A., Paola, C., and Parker, G., 2002, Experimental steep, braided flow: application to flooding risk on fans: Journal of Hydraulic Engineering, v. 128, p. 322-330 [PDF]

Hickson, T.A., Sheets, B.A., Paola, C., in submittal, Experimental test of tectonic controls on three-dimensional alluvial facies architecture.

Sheets B.A., Hickson T.A., Paola C., 2002, Assembling the stratigraphic record: Depositional patterns and time-scales in an experimental alluvial basin: Basin Research, v.14, no.3, p. 287-301 [PDF]

Strong, N., B.A. Sheets, T.A. Hickson, and C. Paola, A mass-balance framework for quantifying downstream changes in fluvial architecture: Sedimentology, in press [PDF]

XES 02-1 Run

Goal: The XES basin develops strata under clearly defined variations in sediment discharge, rates and geometries of subsidence, and absolute base-level change. The experiment (XES 02-1) designed to investigate the effect of slow, rapid, and superimposed base-level cycles on shoreline migration and stratigraphic response, under conditions of passive margin type subsidence.

Reseachers: Nikki Strong, Wonsuck Kim, Ben Sheets, John Martin, Chris Paola

Overview:
1. Sediment and water were mixed and fed from a single point source. The sediment feed rate was 0.0182 m3/hr, the water discharge was 1.5 m3/hr, and their rates were kept constant throughout the experiment. The sediment mixture was composed of 63 % quartz sand (110 mm), 27 % coal sand (bimodal: 460 and 190 mm), and 10 % kaolinite.

2. Subsidence rates increased linearly downstream in the basin so as to produce simple linear-hinge type subsidence. The rates of subsidence were held constant in time. The maximum subsidence rate in the downstream end of the basin was 3.7 mm/hr. Absolute base-level change (i.e., absolute change of water surface level in the "ocean" part of the basin) of the XES 02-1 experiment included sinusoidal base-level cycles with two time scales. The first slow base-level cycle lasted 108 hours, beginning at runtime 26 hr. It was followed by a rapid base-level cycle lasting 18 hours beginning at runtime 144 hr. Base level was stable before the first slow cycle and between these two cycles; this stable base level is the experimental datum. The second part of the experiment comprised six rapid cycles superimposed on one slow cycle, beginning at runtime 202 hr. The slow component had a duration of 108 hours, and the six rapid base-level cycles were each 18 hours in duration. Absolute base level reached a minimum level of 0.11 m below initial at runtimes 80 and 153 hr on the low and high frequency sinusoidal curves respectively, and a minimum of 0.21 m below initial level at runtimes 247 and 265 hr during the third and fourth superimposed cycles. The maximum rate of the base-level change was 22.3 mm/hr on the falling and rising inflection points of the second superimposed cycle and fifth superimposed cycle, respectively.

Result:
1. In most field settings it would be difficult to obtain the data needed to constrain all the terms in the predictive equation. We measure the degradation of the accuracy of the predicted shoreline dynamics as we reduce the amount of data available by replacing observed time variation of successive terms in the equation with their mean values. By this measure, base level is the most important variable in predicting shoreline migration, followed in turn by sediment supply at the shoreline, geometry of the foreset, and the average subsidence rate across the foreset.

2. Autogenic signals in shoreline trajectory imprinted on the allogenic signatures are generally thought of as local "noise". However, the variability in the experimental shoreline data persists even when the shoreline migration is averaged laterally. The autogenic signal in the shoreline migration rate (i.e., a high-frequency variability of the rate) is strongest during relative base-level rise and weakest during relative base-level fall. Base-level change, which is the only externally imposed time-variable parameter in the experiment, can work either with or against the sediment transport regime and thus can magnify or diminish autogenic processes.

Movies: (right click to download)
[slow cycle movie.avi: 35.6Mb]
: A movie of the isolated slow cycle of base level change, 108 hours of run time, condensed to three minutes.

[rapid cycle movie.avi: 55.7Mb]
: The 18 hour rapid cycle of base level change, condensed to three minutes.

[topography movie.avi: 2.55Mb]
: A movie made from the scanned topographic data.

[avg_strata.mov: 4.69Mb]
: A time series of the reconstructed stratigraphy, sediment surface elevation averaged normal to the stream direction. The yellow dots are the shoreline positions at each scan time and the yellow line is the shoreline trajectory.

[meanshl.mov: 177Mb]
: A movie showing changes in the shoreline and the mean shoreline (white horizontal line in the movie) throughout the XES 02 experiment, created by every 10th minute overhead images.

Publications:
Kim, W., Paola, C., Voller, R.V., and Swenson, B.J., 2006, Experimental measurement of the relative importance of controls on shoreline migration: Journal of Sedimentary Research, v. 76 [PDF]

Low Froude Number Experiment

Goal: To study the effects of Froude number on experimental fan geomorphology and stratigraphy.

Reseachers: John Martin, Ben Sheets, Chris Paola, Michael Kelberer

Experiment Overview:
1. Alluvial fans constructed under experimental conditions tend to have a high sediment supply and steep slopes, which drive the Froude Number above 1 (supercritical). Since this condition is rare in natural channelized flows, this experiment was undertaken to address the sensitivity of experimental geomorphology and stratigraphic architecture to Froude Number.

2. This experiment was run over an eight-month period and accumulated an unprecedented 2400 hours of run time. The long run time was due to an extremely low (4-6 g/min) sediment discharge that was needed to maintain very large water-sediment discharge ratios (103 - 104). At the termination of the experiment fan thickness at the shoreline was 5.5 cm.

Result:
1. Geomorphology: Subcritical flow (~ 0.29 - 0.55) was achieved on average only after we set the water-sediment (bedload) discharge ratio to approximately 10000 (a ratio approaching/proportional to that of natural channel systems). Progressive decrease in the slope was not proportional to increase in the water-sediment discharge ratio, an indication that the efficiency of sediment transport across the fan actually decreased as more water was introduced to the system. This is most likely due to the tendency of the flow to spread laterally vs. cut deeper channels in our non-cohesive bed material.

2. Stratigraphy: Preliminary analysis of the experimental stratigraphy shows us that the large-scale deposit architecture, namely channel and sheet deposits, is qualitatively identical to previous supercritical experiment strata. A major difference is the presence of ripple-derived cross laminations, which provide us with two scales of stratigraphic information (e.g. bounding surfaces). Also, due to the extremely slow fan aggradation rate (~ 30 m/hr) stratal preservation is unlike other deposits from the Delta Basin and Jurassic Tank, with a heavy bias towards channel deposits.

Continuing Analysis:

1. The very low sediment supply slowed the channel avulsion frequency down considerably - we'd like to know 1) how much and 2) how closely related sediment supply and avulsion frequency are.

2. Using bounding surface concepts to evaluate how organized the hierarchy of channel deposits is preserved in the depositional record.

3. Estimating preservation and the ambiguous notion of stratigraphic completeness given the drastically different preservation style of this experiment.

4. Constraining the formative sediment transport from the depositional record.

Movies: (right click to download)
[r2003-small-3-2-2004.mov: 286.2Mb]

Publications:

Riparian Vegetation and Braided Stream Dynamics

Goal:
1. To study and quantify the interactions between riparian vegetation, channel morphology, and flow dynamics.

2. To investigate how river systems self-organize as a result of these interactions.

3. To investigate spatial and dynamic scaling in braided rivers with and without vegetation.

Reseachers: Michal Tal, Chris Paola, Elizabeth Tilman (Water Resources, Univ. of MN), Efi Foufoula-Georgiou (Civil Engineering, Univ. of MN)

Result:
Ongoing experiments at the St. Anthony Falls Laboratory are designed to isolate the effects of vegetation on braided stream dynamics. These experiments show how a fully braided stream with a noncohesive bed transitions to a single-thread (meandering) system when continuously forced with vegetation. Time-lapse photography and measurements of bed topography, flow depth, sediment output, and flow velocities enable us to study and quantify the morphodynamics of the system associated with this change.

Movies: (link to website)
[http://www.geo.umn.edu/people/grads/talx0001/movies.htm]

Publications:
Tal, M., Gran, K., Murray, A. B., Paola, C., Hicks, D. M., 2004, Riparian vegetation as a primary control on channel characteristics in multi-thread rivers, in Riparian Vegetation and Fluvial Geomorphology: Hydraulic, Hydrologic, and Geotechnical Interaction, Sean J. Bennett and Andrew Simon, Eds., American Geophysical Union Monograph. [PDF]

Murray, A. B., and Paola, C., 2003, Modeling the effect of vegetation on channel pattern in bedload rivers: Earth Surface Processes and Landforms v. 28, p. 131-143. [PDF]

Gran, K. and Paola, C., 2001, Riparian vegetation controls on braided stream dynamics: Water Resources Research v. 37, no. 12, p. 3275-3283. [PDF]

Sapozhnikov, V. and Foufoula-Georgiou, E., 1997, Experimental evidence of dynamic scaling and indications of self-organized criticality in braided rivers: Water Resources Research v. 33, no. 8, p. 1983-1991.