Reading Results

After a run* call (Running a Simulation), results live in pam.r, a dict of numpy arrays keyed by output variable. The leading dimensions are always the profile grid (ngridx, ngridy); the remaining dimensions depend on the variable – e.g. output level and angle for passive output, layer for active output, frequency and polarization where applicable. Check pam.r[var].shape and the corresponding Settings/Nml-File (nfreqs, noutlevels, number of hydrometeors in Descriptor File Structure, radar_npol) to make sense of a given array rather than assuming a fixed axis order across variables.

The most commonly used variables:

Variable	Contents
tb	Brightness temperature [K], passive output
Ze	Radar reflectivity [dBZ], active output
Att_hydro	Two-way hydrometeor attenuation [dB]
Att_atmo	Two-way atmospheric gas attenuation [dB]
radar_spectra	Full Doppler spectrum (only if nmlSet["radar_mode"] is "spectrum" or "moments")
radar_vel	Doppler velocity bins for radar_spectra
radar_moments	Radar moments (mean Doppler velocity, skewness, kurtosis, …) derived from the spectrum
radar_slopes	Left/right spectral slopes
emissivity	Surface emissivity actually used

pam.r["nmlSettings"] carries a copy of the nmlSet the run was made with, so a saved result is self-describing.

Writing output

pyPamtra.core.pyPamtra.writeResultsToNetCDF writes pam.r (plus selected profile variables) to a NetCDF file:

pam.writeResultsToNetCDF("output.nc", ncForm="NETCDF4")

pyPamtra.core.pyPamtra.writeResultsToNumpy / pyPamtra.core.pyPamtra.loadResultsFromNumpy round-trip the entire session (pam.r, pam.p, pam.nmlSet, pam.set, and the descriptor file) through a single pickle file instead – lighter-weight than NetCDF and useful for resuming/inspecting a run later, but Python pickles rather than NetCDF, so they aren’t meant for sharing outside Python or across pyPamtra versions. Pass seperateFiles=True to write one plain .npy file per variable into a directory instead, useful for very large result sets.

Quick-look plotting

pyPamtra.plot has two plotting helpers built on pam.r: pyPamtra.plot.plotTB (a 2D brightness-temperature map over the profile grid) and pyPamtra.plot.plotTBLine (brightness temperature along one grid axis, e.g. for a flight track or transect). Both operate directly on pam.r after a run.