Settings/Nml-File

This file defines general settings of the PAMTRA run (e.g., whether active or passive simulations, just radar moments or full spectrum, paths for in- and output, etc.). These settings are stored in a dictionary nmlSet of the pyPamtra object. In addition, some low level settings, like verbosity of FORTRAN and python, are stored in the set dictionary. The full information of available options can be found in settings.f90 in the PAMTRA source directory.

nmlSet() settings

Variable	Values	Default	Description
active	bool	True	Activate radar simulator
add_obs_height_to_layer	bool	False	If observation heights for the output are selected, these can be added as additional atmospheric layer boudnaries. In case the vertical grid of observation layers is very fine, it can happen that several heights are within the same layer and therefore give the same result. For cases where the observation height is very far away from the next atmospheric model layer, it might not be 100% representative and it could be beneficial as well.
conserve_mass_rescale_dsd	bool	True	In case the mass mixing ratio for an hydrometeor calculated integrating the drop-size-distribution (DSD) doesn’t correspond to the input value, rescale the DSD to account for the mass loss.
creator	str	Pamtrauser	Netcdf file creator
data_path	str	$PAMTRA_DATADIR	Path for emissivity files and other data. If value is $PAMTRA_DATADIR, the corresponding environment variable is used.
emissivity	positive float [0,1]	0.6	Surface emissivity used for both polarizations
file_desc	str	“”	In pure FORTRAN mode and netCDF output, this string is used as an extension to the output file name. For sensitivity studies this might be helpful.
gas_mod	L93, R98	R98	Model for gas absorption. Either ROSENKRANZ (R98) or LIEBE (L93)
hydro_adaptive_grid	bool	True
hydro_fullspec	bool	False	For pyPamtra only: Do not estimate particle diameter, mass, area, number concentration, rho and aspect ratio directly from the descriptor file but pass them directly from python to PAMTRA using numpy arrays. See also addFullSpectra() of pyPamtra’s descriptorFile class.
hydro_includehydroinrhoair	bool	True	Include hydrometeors when estimating the density of wet air. Different models use different conventions here.
hydro_limit_density_area	bool	True	Change mass, cross section area and density of particles in case it is larger or smaller than possible. Min density is hydro_softsphere_min_density, max density is 917 kg/m3. max area is D2
hydro_softsphere_min_density	positive float	10.0	If hydro_limit_density_area=True, limit minimal density to this value.
hydro_threshold	positive float	1e-10	minimum required hydrometeor concentration kg/m3.
lgas_extinction	bool	True	gas extinction desired
lhyd_extinction	bool	True	hydrometeor extinction desired
liq_mod	str	Ell
obs_height	positive float	833000.0	upper level output height [m] (> 100000. for satellite)
outpol	str	VH
passive	bool	True	estimate brightness temperatures
radar_allow_negative_dD_dU	bool	False	allow that particle velocity is decreasing with size. Should be usually set to false.
radar_airmotion	boolean	False	Consider air motion in direction of radar beam.
radar_airmotion_linear_steps	positive integer	30	For linear function: number of discrete intervals.
radar_airmotion_model	constant, linear, step	step	Model to describe vertical air motion: Either constant velocity, linear change from vmin to vmax or abrupt change using a step function.
radar_airmotion_step_vmin	positive float	0.5	For step function: volume ratio between vmin and vmax.
radar_airmotion_vmin	float	-4 m/s	Minimal air motion of for step and linear function. Also used for constant air motion.
radar_airmotion_vmax	float	4 m/s	Maximal air motion of for step and linear function.
radar_aliasing_nyquist_interv	positive integer	1	Consider aliasing effects for overspending the nyquist range radar_aliasing_nyquist_interv times.
radar_attenuation	disabled, bottom-up, top-down	disabled	Attenuate radar spectrum and Z_e depending on measurement geometry (bottom-up for upward looking, top-down for downward-looking).
radar_convolution_fft	boolean	True	Use FFT for convolution. FFt is much faster, but can have numerical issues in rare cases.
radar_fwhr_beamwidth_deg	float*	0.3	radar full width half radiation beamwidth (required for spectral broadening estimation)
radar_integration_time	float*	1.4	radar beamwidth (required for spectral broadening estimation)
radar_K2 ( |K_w^2| )	positive float*	0.93	Dielectric factor of water used to estimate radr reflectivity.
radar_max_v ( v_nyq )	float*	-7.885 m/s	Maximum Nyquist velocity (usually radar_min_V = -radar_max_V)
radar_min_v ( v_nyq )	float*	7.885 m/s	Minimum Nyquist velocity
radar_peak_min_bins	int*	2	Minimum peak width
radar_peak_min_snr	float*	-10 dB	Minimal required SNR reqired for a peak. See radar_peak_min_snr for defintion
radar_peak_snr_definition	specLin | log	log	log: radar_peak_min_snr describes snr of peak in dB. linSpec: radar_peak_min_snr descibes mean signal+noise to noise ratio (available for historical reasons)
radar_mode	simple, spectrum, moments	simple	Use “simple” radar simulator provides only Z_e by integrating over D. The advanced “spectrum” simulator simulates the complete radar Doppler spectrum and estimates all moments from the spectrum. “moments” is identical to “spectrum” but the full Doppler spectrum is discarded to save memory.
radar_nfft ( N_fft )	positive integer	256	Number of FFT points in the Doppler spectrum
radar_no_Ave ( Nave )	positive integer*	150	Number of spectral averages
radar_noise_distance_factor	positive float*	2.0	Required distance of the peak edge to the noise level. If radar_noise_distance_factor<0 and radar_use_hildebrand, then noise_max from Hildebrand is used for peak edge determination. Sometimes, lower SNR values can be achieved with radar_noise_distance_factor instead of noise_max
radar_npeaks	1	1	Number of detected peaks in the Doppler spectrum. As of today fixed to 1.
radar_pnoise0 ( N_1000 )	float*	-32.23 dBz	Radar noise at 1km in same unit as reflectivity Z_e
radar_polarisation	NN, HV, VH, VV, HH	NN	Radar polarisation. NN: no polarisation, HV: horizontal transmit, vertical receive, etc.. Can be a comma separated list.
radar_receiver_miscalibration	float*	0.0 dB	Radar calibration error
radar_receiver_uncertainty_std	positive float*	0.0	Add Gaussian noise to radar noise level to simulate unstable receivers
radar_save_noise_corrected_spectra	boolean	False	For debugging purposes: Save radar Doppler spectrum after noise is removed
radar_smooth_spectrum	boolean	True	smooth spectrum before estimating moments
radar_use_hildebrand	boolean	False	Derive N_P not from radar_pnoise0 but using the method of citet{hildebrand:1974a}. Set radar_noise_distance_factor<0 to use also noise_max from hildebrand for determination od the peak edge. Sometimes, lower SNR values can be achieved with radar_noise_distance_factor instead of noise_max
radar_use_wider_peak	boolean	False	Include the found peak edge (if peak edge is still larger than mean noise) into the peak which is used for moment estimation.
randomseed	integer	0	0 is real noise, -1 means that the seed is created from latitude and longitude, other value gives always the same random numbers
read_turbulence_ascii	bool	False	If .true. turbulence need to be included in the ascii input_file, rightmost column. Not relevant for pyPamtra and for passive simulations.
salinity	float	33.0	sea surface salinity
save_psd	boolean	False	also saves the PSDs used for radiative transfer
save_ssp	boolean	False	also saves the single scattering properties used for radiative transfer
tmatrix_db	none or file	none	use data base to cache T-Matrix calculations
tmatrix_db_path	str	database/	path to T-Matrix data base
write_nc	bool	True	write netcdf or ascii output

* These variables can be also provided as list to account for different instrument specifications. In this case, each entry corresponds to one frequency.

set() settings

Variable	Values	Default	Description
verbose	positive integer	0	Verbosity of the FORTRAN routines
pyVerbose	positive integer	0	Verbosity of the pyPamtra python modules
namelist_file	str	TMPFILE	path and name of the FORTRAN namelist file
freqs	list of float	empty	list of frequencies, set automatically at program start

Other default settings

Variable	Values	Default	Description
sfc_refl	S,L,F	S	Specular, Lambertian, or Fresnel