Source code for ppcpy.qc.pollySaturationDetect

import numpy as np
import logging
from scipy import ndimage
import numpy as np



def pollySaturationDetect(data_cube, rfill=250, sigSaturateThresh=500):
    """detect the bins which are fully saturated by the clouds.

    INPUTS:
        data: dict
            Data dictionary. See documentation for format.

    KEYWORDS:
        hfill: float 
            Minimum range gap to fill (m). Default: 250
        sigSaturateThresh: float
            Threshold of saturated signal (photon count). Default: 500

    OUTPUTS:
        flag: boolean ndarray
            True indicates current range bin is saturated by clouds.

    HISTORY:
        - 2018-12-21: First Edition by Zhenping
        - 2019-07-08: Fix the bug of converting signal to PCR.
        - 2025-05-14: translated and changed the algorithm to use scipy.ndimage
    """
    logging.info('Saturation detection')
    
    # assumption that PCR is calculated beforehand
    PCR = data_cube.retrievals_highres['PCR_slice']

    nChannels = data_cube.num_of_channels
    height=data_cube.retrievals_highres['range']
    hres = height[1] - height[0]
    hfill_bins = int(rfill/hres)

    flagSaturation = np.full(PCR.shape, False, dtype=bool)
    for iChannel in range(nChannels):
        flag = PCR[:,:,iChannel] > sigSaturateThresh
        # manually unmask the first 3 range gates as they might only be affected by straylight
        flag[:,:3] = False 
        flag = ndimage.binary_closing(flag, structure=np.ones((2,hfill_bins)))
        flagSaturation[:,:,iChannel] = flag

    return flagSaturation