Maximum resolution of resource grid

[jacksonlord]

In making a resource grid and using the orographic flow model on a given terrain for further turbine layouts and optimizations, what is the maximum resolution of the grid that is possible in WAsP? It seems that when creating the window to produce the grid, there is no way to create one with more than 10,000 nodes. This can limit the resolution of the resulting wrg in all but the smallest projects to a grid spacing of well over 100m. Is there a way around this to get higher resolution models, and is this even desirable? It seems like resolutions of 50m and lower are common and necessary for micrositing.



Thanks,

Jackson

[Duncan]

While you're configuring the resource grid, the number of nodes is calculated. If it exceeds 10K, the displayed number turns red. But that's just a warning, not an actual constraint. You can still do the calculation.

This is a vestigal feature from version 8. I think the idea was that if you increase the resolution without changing the size of the grid, it's easy accidentally to configure huge calculation task.

Now our computers are more powerful, and people's projects are getting bigger. In WAsP 9, we introduced more intelligent management of the existing resource grid results which make it less of a disaster to make a mistake in your configuration.

The warning colour is probably unnecessary now, I'll add a feature request for the next release of WAsP 10 to remove the threshold. (Or to increase it to some number which is huge by todays standards, not those of 2002!)

[jacksonlord]

Ah ha, I never thought to just try it anyway. Thanks for your help.

What do you think is the relation between the resolution of the orography data or the contour interval to the accuracy of wrg resolution? Is there some rule of thumb on grid spacing for preliminary turbine layouts and further micro-siting? I'm thinking that rolling prairie terrain with a contour interval of 3 meters could produce an accurate 20-50m grid resolution....

[Duncan]

Well, you're asking a modelling question now, and I'm not really competent to advise. Maybe one of the scientists or other WAsP users will chip in with a suggestion.

I can point out, though, that it's quite easy to experiment. You can set up several grids at different resolutions covering a small sample area. Then you can see for yourself what effects (if any)arise from different sampling densities. HTH.

[Niels Gylling Mortensen]

20-50 m resolution sounds about right to me. I often start with a large resolution grid and then divide by 2 a couple of times, say: 800, 400, 200, 100, 50, 25. Makes for fast initial views and WAsP will not need to calculate all grid points all the time, just the new ones. Finest resolutions calculate during lunch break or overnight.

Remember that there should be lots of map around the resource grid: at least 100 times the calculation height in all directions.