HGT_M=0

[zlebo]

We are running real simulations over Wyoming, nesting down to 100 m. At 300 m, we found an issue where the LU index is set to 17, which I presume is ocean, and the terrain height is set to 0. I attached a screenshot showing the output of geogrid.exe for a domain with a grid spacing of 300 m. You will see that the min HGT_M value is 0 (this domain is almost entirely over WY, so the minimum height should be greater than 1000 m. And, you can also see that it occurs at individual grid points (you may have to look closely to see it but it is there in the bottom right). This occurs at every 100th grid point in the y-direction (forming a line of individual points at sea level).

We have exhausted options to remove this other than to write a simple post-processing script to find these locations and replace them with adjacent values.

I should note that this occurs regardless of the averaging or smoothing scheme selected. Here, I show you the results for no smoothing and nearest neighbor because it makes it easy to see. I suspect that this is happening at the coarser resolutions but the smoothing and averaging cause the points to be undetectable.

[davegill]

@zlebo
Would you please attach your namelist.wps file, what version of the code that you are running, and let us know if you changed the default settings in the GEOGRID.TBL file.

[zlebo]

@davegill Attached. We changed GEOGRID.TBL only to isolate the issue to these individual points (removing smoothing and no averaging of the input data). The points show up using the default GEOGRID.TBL, but they are smoothed. This is version 3.9.1; however, I downloaded the newest version of WPS and get the same issu
namelist.wps.txt
e.

[cross85]

@zlebo what are you using as input? sometimes DEM files has holes in areas with no-data in places with water or heavy shadow. So, if you are using some high resolution input, as SRTM data, check if that is the problem, then you can change and use other data source or fix the data.