| Message/Author |
|
|
|
|
Hello,
5.3 beta is now up and running. Here are a few notes on the main differences between it and 5.1. There are few differences visually (a couple of extra boxes) but quite a few changes under the bonnet so to speak...
1. Flow routing.
This version now includes code to prevent discharge from being routed back upon itself. This allowed flow to be 'double' counted (or more) when it flowed back and forth along channels in consecutive scans. This allows a much more accurate calcualtion of the discharge that passes through a point (hampered severely by the double counting) which allows different methods of Tau Calculation. THese changes are invisible and not on a new check box etc..
2. Tau calculation
Previuosly this was carried out using Bed slope. This led to problems with lateral erosion (point bars made large bed slopes causing them to be eroded). This led to the developemnt of water surface slope (not recommended), and finally using velocity to calc Tau. This seems a much better option, and required better calc of Q, in order to calc the velocity (see above). There is a check box in the numerical section that allows you to choose any of these options. I would reccomend using bedslope for catchment runs, and Velocity(which may work fine for catchment too - especially finer DEMS) for reach simulations. Velocity must be used for runs with lateral 2 scheme.
3. Erosion routing.
Because there are different ways of caclulating Tau, there needed to be different ways of routing the sediment. If you calculate Tau using velocity then route using bedslope, then you can end up with situations where the velocity is high and bedslope in the wrong direction giving some unusual land forms. Therefore, when velocity for Tau is selected, sediment is routed according to discharge distribtion - (in effect velocity) as in according to the distribution of discharges from a cell. If lateral2 is selected, then part of the sediment is also routed according to the lateral slope.
When bedslope for Tau is selected, then sediment is routed according to bedslope. Jorge found that when eroding with bedslope, but distributing according to Q (as above) then the model was quite capable of creating its own dams!
4. Pit filling
To do with the new Q calculation, there is a pit filling algorithm that operates continuously. It does not 'fill' pits with sediment, but with 'stationary' water, so discharge flows over them as if they had been filled with sediment, but the depth is still retained for velocity calcualtion etc..
5. Lateral erosion
This has changed quite a bit (lateral scheme 2 that is). This is still subject to many changes, and is something I am actively working on. If you want to use this (in reach mode esp) then I recommend you get in touch with me before trying. It is quite stable, but also there are many mroe parameters to play around with.
6. Vegetation
Different implementation, but should act in a similar way. Best to check new version against old for this, if it is important for your simulations.
7. Bedrock
Now it works. For some reason in the development 5.3 versions it had been removed/commented out!
8. Output file errors
I hope these have gone. Many people have commented on there being spikes in the output data. I hope that these have now been resolved in this version. Please let me know if there are still problems.
9. Config files.. Old config files will work on this version, but will not update boxes that are new on this version. New config files will not work on older versions, only backwards comatible I'm afraid.
10. Tracer - I have not tested the tracer components in the new version (its happening at the moment) so best check against old if you are using them.
Thanks, and please post any comments/findings here,
Tom |
|
|
| Back to top |
|
