#metpy - NEXRAD L3 NCR data scaling

Question

I have been trying to determine how the NEXRAD NCR data is scaled. I see from the product header that the max value is 63 but after reading in the datadict, the max value in the data variable is 11. I don't see any info in the Federal Meteorological Handbook that discusses scaling. I can plot the data using metpy, no problem but the scale is off. When I plot using the NCEI toolkit the scale is correct. So what's under the hood?

Thanks!

Mary Ellen

Problem resolved. Dug into metpy code to figure it out. – Mary Ellen Craddock Feb 12 '20 at 14:42 — Mary Ellen Craddock, Feb 12 '20 at 14:42

DopplerShift · Answer 1 · 2020-02-13T00:19:24.387

0

So for the benefit of future people looking for the answer, the NCR product, or Composite Reflectivity, is a 16-level product. This means at each grid point in the product, the value is set to the nearest value of the 16 levels; the level value itself isn't set, but rather a value between 0 and 15. These levels can be found in the .thresholds attribute on Level3File.

The quick way to convert the grid of levels to actual physical values is to use .map_data to convert the data array to the appropriate values:

import numpy as np
from metpy.io import Level3File

f = Level3File(filename)
physical_values = f.map_data(f.sym_block[0][0]['data'])

edited Feb 13 '20 at 00:19

answered Feb 12 '20 at 16:49

DopplerShift

5,472
1
21
20

Thanks! I found a work around using map_data(). f = Level3File(fname) datadict = f.sym_block[0][0] data = f.map_data(datadict['data']) data = ma.array(data) data[np.isnan(data)] = ma.masked – Mary Ellen Craddock Feb 12 '20 at 18:01
Sorry for the formatting. Curious if you agree with the approach I took? – Mary Ellen Craddock Feb 12 '20 at 18:04
I completely forgot I added map_data. That's even better to use. I've adjusted my answer to reflect that. – DopplerShift Feb 13 '20 at 00:19

#metpy - NEXRAD L3 NCR data scaling

1 Answers1