Source code for ppcpy.retrievals.highres




import numpy as np
import ppcpy.qc.transCor as transCor
import ppcpy.retrievals.depolarization as depolarization
import ppcpy.misc.helper as helper
import logging

from scipy.interpolate import interp1d


[docs] def attbsc_2d(data_cube, nr:bool=True, collect_debug:bool=False): """Attenuated Backscatter. Parameters ---------- data_cube : object Main PicassoProc object. nr : bool, optional If Ture, calculate the attbsc for FR and NR channels. Default is True. collect_debug : bool, optional If True, collects debug information. Default is False. Attributes ---------- data_cube.retrievals_highres[f"attBsc_{channel}"] : ndarray Time-height attenuated backscatter per channel. Notes ----- - If ´nr=True´ also yeild the attenuated backscatter of the near-range channels. - In addition to the standard far-range channels the function also yeilds the attenuated backscatter of the overlap corrected far-range channels. ..TODO:: is it correct to transmission correct the overlap corrected signals before calculating their attenuated backscatter? Are not the OL signals already transmission corrected? """ rgs = data_cube.retrievals_highres['range'] time = data_cube.retrievals_highres['time64'] ranges_squared = rgs**2 ranges2d = np.repeat(ranges_squared[np.newaxis, :], time.shape[0], axis=0) channels = [(355, 'total', 'FR'), (387, 'total', 'FR'), (532, 'total', 'FR'), (607, 'total', 'FR'), (1064, 'total', 'FR')] if nr: channels += [(532, 'total', 'NR'), (607, 'total', 'NR'), (355, 'total', 'NR'), (387, 'total', 'NR')] for wv, t, tel in channels: channel = f"{wv}_{t}_{tel}" sig = np.squeeze( data_cube.retrievals_highres[f'sigTCor'][:, :, data_cube.gf(wv, t, tel)]) if channel not in data_cube.LCused.keys(): logging.info(f'{channel} skipped at attbsc_2d') continue attBsc = sig * ranges2d / data_cube.LCused[channel]['LC'] attBsc[data_cube.retrievals_highres['depCalMask'], :] = np.nan data_cube.retrievals_highres[f"attBsc_{channel}"] = attBsc # experimental, the calibration constant requires the OL corrected signal if 'sigOLCor' in data_cube.retrievals_highres: logging.warning(f"Exprimental, attenuated backscatter solution for {channel}") sigOLTCor, _ = transCor.transCorGHK_cube(data_cube, signal='OLCor') channels = [(355, 'total', 'FR'), (532, 'total', 'FR'), (1064, 'total', 'FR')] for wv, t, tel in channels: channel = f"{wv}_{t}_{tel}" #sig = np.squeeze( # data_cube.retrievals_highres[f'sigOLCor'][:, :, data_cube.gf(wv, t, tel)]) sig = np.squeeze(sigOLTCor[:, :, data_cube.gf(wv, t, tel)]) if channel not in data_cube.LCused.keys(): logging.info(f'{channel} skipped at attbsc_2d OL') continue attBsc = sig * ranges2d / data_cube.LCused[channel]['LC'] attBsc[data_cube.retrievals_highres['depCalMask'], :] = np.nan data_cube.retrievals_highres[f"attBsc_{wv}_{t}_OC"] = attBsc
[docs] def voldepol_2d(data_cube): """Calculate the volume depolarisation ratio. Parameters ---------- data_cube : object Main PicassoProc object. Attributes ---------- data_cube.retrievals_highres[f"voldepol_{wv}_total_{tel}"] : ndarray Time-hight volume depolarization ratio at wavelengths 353 nm, 532 nm, 1064 nm, and 532 nm DFOV. """ config_dict = data_cube.polly_config_dict channels = [ (532, 'FR'), (355, 'FR'), (1064, 'FR')] if '532_DFOV' in data_cube.etaused: channels += [(532, 'DFOV')] logging.info("voldepol also for DFOV") for wv, tel in channels: if tel == 'DFOV': flagt = data_cube.gf(wv, 'total', 'NR') else: flagt = data_cube.gf(wv, 'total', tel) flagc = data_cube.gf(wv, 'cross', tel) if np.any(flagt) and np.any(flagc): sigt = np.squeeze( data_cube.retrievals_highres[f'sigBGCor'][:, :, flagt]) sigc = np.squeeze( data_cube.retrievals_highres[f'sigBGCor'][:, :, flagc]) vdr, vdrStd = depolarization.calc_profile_vdr( sigt=sigt, sigc=sigc, Gt=config_dict['G'][flagt], Gr=config_dict['G'][flagc], Ht=config_dict['H'][flagt], Hr=config_dict['H'][flagc], eta=data_cube.etaused[f'{wv}_{tel}']['eta'], voldepol_error=config_dict[f'voldepol_error_{wv}'], window=1 ) vdr[data_cube.retrievals_highres['depCalMask'], :] = np.nan data_cube.retrievals_highres[f"voldepol_{wv}_total_{tel}"] = vdr
[docs] def wvmr_2d(data_cube): """Water Vapor Mixing Ratio. Parameters ---------- data_cube : object Main PicassoProc object. Attributes ---------- data_cube.retrievals_highres[f"wvmr_407_FR"] : ndarray Time-height water vapor mixing ratio at 407 nm FR. Notes ----- .. TODO:: Save highres wvmr data to .nc file **History** - 2026-09-11: First edition by Jakob """ height = data_cube.retrievals_highres['range'] config_dict = data_cube.polly_config_dict tel = 'FR' # Only applied for far-range channels. # interpolation molExt_387 = interp1d( data_cube.mol_2d['time'].values.astype('datetime64[s]').astype(int), data_cube.mol_2d['mExt_387'].values, axis=0)( data_cube.retrievals_highres['time64'].astype('datetime64[s]').astype(int)) molExt_407 = interp1d( data_cube.mol_2d['time'].values.astype('datetime64[s]').astype(int), data_cube.mol_2d['mExt_407'].values, axis=0)( data_cube.retrievals_highres['time64'].astype('datetime64[s]').astype(int)) # transmission correction molOD_387 = np.nancumsum(molExt_387 * np.concatenate(([height[0]], np.diff(height))), axis=1) molOD_407 = np.nancumsum(molExt_407 * np.concatenate(([height[0]], np.diff(height))), axis=1) trans_387 = np.exp(-2 * molOD_387) trans_407 = np.exp(-2 * molOD_407) # apply smoothing (same as for quasi) flag387 = data_cube.gf('387', 'total', tel) flag407 = data_cube.gf('407', 'total', tel) sig387 = np.squeeze( data_cube.retrievals_highres['sigBGCor'][:, :, flag387]) sig407 = np.squeeze( data_cube.retrievals_highres['sigBGCor'][:, :, flag407]) smooth_t = int(np.array(config_dict['quasi_smooth_t'])[flag407][0] / 2) smooth_h = int(np.array(config_dict['quasi_smooth_h'])[flag407][0] / 2) sig387 = helper.smooth2a(sig387, smooth_t, smooth_h) sig407 = helper.smooth2a(sig407, smooth_t, smooth_h) wvmr_raw = (sig407 / sig387) * (trans_387 / trans_407) # apply wv_const wvmr = wvmr_raw * data_cube.WVCused[f"407_{tel}"]['WVC'] # quality mask if config_dict['flagOnlyUseValidQuasiData']: quality_mask_387 = np.squeeze(data_cube.retrievals_highres['quality_mask'][:, :, flag387]) quality_mask_407 = np.squeeze(data_cube.retrievals_highres['quality_mask'][:, :, flag407]) wvmr[(quality_mask_387 != 0) | (quality_mask_407 != 0)] = np.nan data_cube.retrievals_highres[f"wvmr_407_{tel}"] = wvmr