| Message/Author |
|
|
|
|
| Below is a string of emails from Jennifer describing the problems and a solution we had to not getting much output from CAESAR |
|
|
|
|
Here is the first few lines of my Bala100+0.dat file, and as you can see time in first column, water (which does come in further down) and then something in the third column.
1 0.000 0.003 0.0 0.000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
2 0.000 0.003 0.0 0.000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
3 0.000 0.003 0.0 0.000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
4 0.000 0.003 0.0 0.000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
5 0.000 0.003 0.0 0.000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
6 0.000 0.003 0.0 0.000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
7 0.000 0.003 0.0 0.000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
8 0.000 0.003 0.0 0.000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000 0.000000
I have attached the xml file which got these results as well,
Thanks you very much,
Jennifer |
|
|
|
|
Jennifer - this is fine,
the zero's, instead of numbers mean that there is no sediment coming out at the right hand edge of your catchment - I dont know why.. may be because there are not large enough floods - or some other reason.
Our server has blocked your attatchment - try zipping it then sending it. It might be best if you put your dem, rainfall file and the config file all in one zip file and sent them over then I can have a look proper,.
Tom |
|
|
|
|
Hi Jennifer
Looks OK to me - but I wonder if the problems may be with your DEM? One process I go through with the DEM's that I use is to carve out an intitial channel using a set of free tools called ArcHyrdo, that you can download and use within Arc GIS 8 or 9. These identify the stream network and allow you to cut out an intitial channel - it gives the model somewhere to start if you like. I have attatched a very simple set of instructions of how I use it to the email... Otherwise, all your paramters look fine to me...
Tom |
|
|
|
|
Hi Tom,
So I have followed you instructions and carved out the channel ( But still no sediment is coming through.
I know it’s a long shot, but do you have any other ideas where I might be going wrong? Or shall I put it down to problems within the DEM?
Thanks again,
Jennifer |
|
|
|
|
Firstly, it works...
You've done a really good job on the DEM - thats something that often causes people lots of problems.
OK, what I have done/found is...
1. Your DEM is 5m, but one value (Min Q for depth calculation - in the numerical tab) was a value more appropriate for a 50m dem. I would change it from 0.2 to 0.02
2. Your M value is fairly high for a small catchment (though not massively) so it takes a fair sized rainfall event to make a flood. I have tweaked (added a few valuies in the first 30 lines) your rainfall file to make some more rainfall at the beginning. Sometimes it takes quite alot of rain to get the hyrdological model going - as it starts off in a completely dry state - imagine running on a parched landscape, it takes a while to get things going.
3. Your sediment sizes are the defualt - which are based on the Waitaki braided river in new zealand - with much larger grain sizes in general. So your small floods are not causing much erosion or deposition! If you use a distribition with more smaller grainsizes you will get more erosion. Ideally, you should be able to measure these from your catchment (measure the stream bed grainsize).
4. I used the latest version of CAESAR that is now up on the website. It shouldnt make too much of a difference but might.
5. Finally - might be worth remembering that your bedrock file is probably based on the old DEM, not the new one with the river channel cut out.
I've zipped and attatched the congfig files and the other files I used. DOwnload the latest version of the code and these should run fine,
Cheers, and let me know how you get on,
Tom |
|
|
| Back to top |
|
