Spatial Discretization and Leap Frogging

[Stephen Barabas]

Recently through my Mesoscale class, I have become interested in this category concerning the use of nested grids to analyze and forecast mesoscale phenomena. I know a bit about it, being just a beginner on the subject I lack what others know. Fellow classmates and myself are engaging in a research project concerning the use of the Taylor series and hydrostatic equilibrium for the momentum equation, using it to solve leap frog and the Courant Friedrich Lewy scheme and to model out a forecast zone for temperatures one hour in advanced. (Big deal)

In any regard, if anyone has knowledge on this subject please submit whatever you can. I, as I'm sure anyone else interested, would appreciate reading what YOU have got to say! I'll submit what I can myself over time.

(Thanks to Donald J. Perkey for making it comprehensible in my Mesoscale Meteorology and Forecasting text.)

 

[Stephen Barabas]

And. . . I should have posted this under the advanced category. My apologies.

 

[Robert Edmonds]

Agreed that this would probably be best in the advanced category.

While, I am relatively new to modeling I can share some of what I have stumbled across while modeling. While higher order schemes can help with stability and accuracy, this is not always the case. With the density current model (discussed in the advanced catagory), when I was having problems with it about a half a year ago, I tried applying higher and higher order methods. While in time it seemed to help, in space it seemed to create more issues than it solved. Remember with higher order methods in space, you are sampling cells farther out, so far out that in real life the information may not be able to travel to the box you're doing the derivative for. I think though it is largely dependent on what your trying to solve. Best of luck...

Again, I'm new to this too. So, there are probably others who might be able to shed more light on this.