CAESAR discussion board >> Edge?


Edge?

CAESAR discussion board > Model development >

Message/Author

Clare (Clare) posted on Monday, February 20, 2006 - 10:38 am

Hi Tom,

I've been looking through your new code and notice that you use an array 'Edge' to alter values of tau. Please could you tell me what 'edge' is and what it does?

Thanks!
Clare

Tom Coulthard (Tom) posted on Monday, February 20, 2006 - 11:22 am

Hello Clare,
Edge[x,y] is a new array that contains the 'edge' gradient - that is the curvature of bank cells of a bend - +ve being outside, -ve being inside. This is then interpolated across the river channel to provide a cross stream graident (of sorts) which is then used to route a portion of the sediment being moved.

I'm a little concerned, as It shouldnt be alter ing Tau! What line is this happening at - I'll investigate!
Thanks,
Tom

Clare (Clare) posted on Monday, February 20, 2006 - 02:10 pm

Its in erode_w, line 9681, and 9701 (version 5.1d).

Don't worry too much - it's commented out! I just wondered why it might be there.

Thanks, Clare

Tom Coulthard (Tom) posted on Monday, February 20, 2006 - 02:28 pm

Ah!
THese were experimental lines (the lateral erosion bit has been a long time in the making) whereby if the cell was on the outside of a bend, Tau was increased - if onthe inside, decreased.. As you said - now commented out! I guess thats the problem with developing the code, all the rubbish I tried out gets left there! I should have a good purge on these lines some day!
Tom

Clare (Clare) posted on Monday, February 20, 2006 - 10:17 pm

Good - that makes sense!

A couple more questions I'm afraid...:

Further down in erode_w, the edge array is used to calculate a second bedslope that is combined with a cross_stream_grad_factor to amend where sediment is routed to. Is this still needed?

And - on the next little bit of code in erode_w, to distribute sediment to surrounding cells, why do the array indices change e.g. from bedslope(1) to dist(2,n)?

Hope those questions make sense! Thanks, Clare

Tom Coulthard (Tom) posted on Wednesday, February 22, 2006 - 09:40 am

Hello Clare,
Yes, that is needed, this is what moves sediment across the channel to build up point bars etc..

And the array indices change, because dist was created about 9 or 10 years ago, and bedslope 1, and for some reason they use different numbers to indicate the direction, e.g one starts with 1 at the N direction, and the other 1 is NW etc.. Just a bit of code latency.. I guess I should tidy things up one day!!

Tom