Describe processes represented by the model
|
The bed is represented by a 2-D matrix. At this time the bed is 250 x 250. Each block (i,j) is taken to be a slab of sediment 10cm x 10cm x 1cm deep.
A second matrix represents the flow. This is the same everywhere in the domain at each time point, except for a random spatial fluctuation representing turbulence.
The user-defined flow picks up (or puts down) sediment according to a transport law. Three transport laws have been tested: Bailard (1981), Ribberink (1998) or simple rules. The simple rules are simply thresholds: (if flow greater than 70cm/sec pick up one block).
Once sand block have been picked up, they are moved forward with the flow. Generally, I have used a fraction of the distance that the water would travel (jump_frac). So, with a flow of 100cm/sec, in one second that water goes 100 cm. The sand in this case would go 50 cm (half the distance). At the extremes, the model is sensitive to this parameter, but at intermediate values, it is not.
Tested flows have consisted of combined sinusoidal flow+steady flow, purely osc, purely steady, and natural flow time series taken from current meter measurements. All flows have a random spatial fluctuation added at each time point.
Once bedforms are generated, feedback rules are employed by altering the flow over the bedform. Once a bedform gets tall, the flow over the top accelerates, increasing transport at that location. In the steep lee of a bedform, a shadow zone forms where flow goes to ~zero, therefore transport goes to zero. These mechanisms destroy or build bedforms.
|
Describe key physical parameters and equations
|
"Key" parameters:
- the transport thresholds (if you use the simple rules) (it is cool, model results are similar with all the transport models that have been tested)
- the flow magnitudes and turbulence magnitude
- Jump_frac
- all the little parameters in the transport equations
Most parameters act to speed or slow transport, which in turn speeds or slows the growth of bedforms.
|