HYSPLIT Forum: hysplitbbs.arl.noaa.gov

I have been making ARL grids with the given code used attached using MeteoInfo. Some of my grids have some missing data points in them. The normal way to encode missing data is to make it a large value outside the realm of possibility as a tag. Because of the reliance on the maximum difference in the grid points in the ARL encoding, this ends up smoothing the rest of the grid to one value. It works if I change them to something closer to other data points but this would have to change for each new variable with a different scale.

Is there a better way to encode data points as Missing? This usually happens around the surface when working around mountain ranges and such.