satpy.readers.gms.gms5_vissr_l1b module

Reader for GMS-5 VISSR Level 1B data.

Introduction

The gms5_vissr_l1b reader can decode, navigate and calibrate Level 1B data from the Visible and Infrared Spin Scan Radiometer (VISSR) in VISSR archive format. Corresponding platforms are GMS-5 (Japanese Geostationary Meteorological Satellite) and GOES-09 (2003-2006 backup after MTSAT-1 launch failure).

VISSR has four channels, each stored in a separate file:

VISSR_20020101_0031_IR1.A.IMG
VISSR_20020101_0031_IR2.A.IMG
VISSR_20020101_0031_IR3.A.IMG
VISSR_20020101_0031_VIS.A.IMG

This is how to read them with Satpy:

from satpy import Scene
import glob

filenames = glob.glob(""/data/VISSR*")
scene = Scene(filenames, reader="gms5-vissr_l1b")
scene.load(["VIS", "IR1"])

References:

Details about platform, instrument and data format can be found in the following references:

• VISSR Format Description

• GMS User Guide

Compression

Gzip-compressed VISSR files can be decompressed on the fly using FSFile:

import fsspec
from satpy import Scene
from satpy.readers import FSFile

filename = "VISSR_19960217_2331_IR1.A.IMG.gz"
open_file = fsspec.open(filename, compression="gzip")
fs_file = FSFile(open_file)
scene = Scene([fs_file], reader="gms5-vissr_l1b")
scene.load(["IR1"])

Calibration

Sensor counts are calibrated by looking up reflectance/temperature values in the calibration tables included in each file. See section 2.2 in the VISSR user guide.

Navigation

VISSR images are oversampled and not rectified.

Oversampling

VISSR oversamples the viewed scene in E-W direction by a factor of ~1.46: IR/VIS pixels are 14/3.5 urad on a side, but the instrument samples every 9.57/2.39 urad in E-W direction. That means pixels are actually overlapping on the ground.

This cannot be represented by a pyresample area definition, so each dataset is accompanied by 2-dimensional longitude and latitude coordinates. For resampling purpose a full disc area definition with uniform sampling is provided via

scene[dataset].attrs["area_def_uniform_sampling"]

Rectification

VISSR images are not rectified. That means lon/lat coordinates are different

1. for all channels of the same repeat cycle, even if their spatial resolution is identical (IR channels)

2. for different repeat cycles, even if the channel is identical

However, the above area definition is using the nominal subsatellite point as projection center. As this rarely changes, the area definition is pretty constant.

Performance

Navigation of VISSR images is computationally expensive, because for each pixel the view vector of the (rotating) instrument needs to be intersected with the earth, including interpolation of attitude and orbit prediction. For IR channels this takes about 10 seconds, for VIS channels about 160 seconds.

Space Pixels

VISSR produces data for pixels outside the Earth disk (i.e. atmospheric limb or deep space pixels). By default, these pixels are masked out as they contain data of limited or no value, but some applications do require these pixels. To turn off masking, set mask_space=False upon scene creation:

import satpy
import glob

filenames = glob.glob("VISSR*.IMG")
scene = satpy.Scene(filenames,
                    reader="gms5-vissr_l1b",
                    reader_kwargs={"mask_space": False})
scene.load(["VIS", "IR1])

Metadata

Dataset attributes include metadata such as time and orbital parameters, see Metadata.

Partial Scans

Between 2001 and 2003 VISSR also recorded partial scans of the northern hemisphere. On demand a special Typhoon schedule would be activated between 03:00 and 05:00 UTC.

class satpy.readers.gms.gms5_vissr_l1b.AreaDefEstimator(coord_conv_params, metadata)

Bases: object

Estimate area definition for VISSR images.

Initialize the area definition estimator.

Parameters:

• coord_conv_params – Coordinate conversion parameters

• metadata – VISSR file metadata

_get_name_dict(dataset_id)

_get_proj4_dict()

_get_proj_dict(dataset_id)

_get_shape_dict(dataset_id)

full_disk_size = {'IR': 2366, 'VIS': 9464}

get_area_def_uniform_sampling(dataset_id)

Get full disk area definition with uniform sampling.

Parameters:

dataset_id – ID of the corresponding dataset.

class satpy.readers.gms.gms5_vissr_l1b.Calibrator(calib_table)

Bases: object

Calibrate VISSR data to reflectance or brightness temperature.

Reference: Section 2.2 in the VISSR User Guide.

Initialize the calibrator.

Parameters:

calib_table – Calibration table

_calibrate(counts)

_convert_to_percent(res)

_lookup_calib_table(counts, calib_table)

_make_data_array(interp, counts)

_postproc(res, calibration)

calibrate(counts, calibration)

Transform counts to given calibration level.

class satpy.readers.gms.gms5_vissr_l1b.GMS5VISSRFileHandler(filename, filename_info, filetype_info, mask_space=True)

Bases: BaseFileHandler

File handler for GMS-5 VISSR data in VISSR archive format.

Initialize the file handler.

Parameters:

• filename – Name of file to be read

• filename_info – Information obtained from filename

• filetype_info – Information about file type

• mask_space – Mask space pixels.

_attach_lons_lats(dataset, dataset_id)

_calibrate(counts, dataset_id)

static _concat_orbit_prediction(orb_pred_1, orb_pred_2)

Concatenate orbit prediction data.

It is split over two image parameter blocks in the header.

property _coord_conv

_get_acq_time(dask_array)

_get_actual_shape()

_get_area_def_uniform_sampling(dataset_id)

_get_attitude_prediction()

_get_calibration_table(dataset_id)

static _get_channel_type(parameter_block_size)

_get_counts(image_data)

_get_earth_ellipsoid()

_get_frame_parameters_key()

_get_image_coords(data)

_get_image_data()

_get_image_data_type_specs()

_get_image_offset(dataset_id)

_get_line_number(dask_array)

_get_lons_lats(dataset, dataset_id)

_get_mda()

_get_navigation_parameters(dataset_id)

_get_nominal_shape()

_get_orbit_prediction()

_get_orbital_parameters()

_get_predicted_navigation_params()

Get predictions of time-dependent navigation parameters.

_get_proj_params(dataset_id)

_get_scanning_angles(dataset_id)

_get_static_navigation_params(dataset_id)

Get static navigation parameters.

Note that, “central_line_number_of_vissr_frame” is different for each channel, even if their spatial resolution is identical. For example:

VIS: 5513.0 IR1: 1378.5 IR2: 1378.7 IR3: 1379.1001

_get_time_parameters()

_make_counts_data_array(image_data)

_make_lons_lats_data_array(lons, lats)

_mask_space_pixels(dataset, space_masker)

property _mode_block

_read_control_block(file_obj)

_read_header(filename)

_read_image_data()

static _read_image_param(file_obj, param, channel_type)

Read a single image parameter block from the header.

_read_image_params(file_obj, channel_type)

Read image parameters from the header.

_update_attrs(dataset, dataset_id, ds_info)

property end_time

Nominal end time of the dataset.

get_dataset(dataset_id, ds_info)

Get dataset from file.

property start_time

Nominal start time of the dataset.

class satpy.readers.gms.gms5_vissr_l1b.SpaceMasker(image_data, channel)

Bases: object

Mask pixels outside the earth disk.

Initialize the space masker.

Parameters:

• image_data – Image data

• channel – Channel name

_correct_vis_edges(edges)

Correct VIS edges.

VIS data contains earth edges of IR channel. Compensate for that by scaling with a factor of 4 (1 IR pixel ~ 4 VIS pixels).

_fill_value = -1

_get_earth_edges()

_get_earth_edges_per_scan_line(cardinal)

_get_earth_mask()

mask_space(dataset)

Mask space pixels in the given dataset.

satpy.readers.gms.gms5_vissr_l1b._get_alternative_channel_name(dataset_id)

satpy.readers.gms.gms5_vissr_l1b._recarr2dict(arr, preserve=None)

satpy.readers.gms.gms5_vissr_l1b.get_earth_mask(shape, earth_edges, fill_value=-1)

Get binary mask where 1/0 indicates earth/space.

Parameters:

• shape – Image shape

• earth_edges – First and last earth pixel in each scanline

• fill_value – Fill value for scanlines not intersecting the earth.

satpy.readers.gms.gms5_vissr_l1b.is_vis_channel(channel_name)

Check if it’s the visible channel.

satpy.readers.gms.gms5_vissr_l1b.read_from_file_obj(file_obj, dtype, count, offset=0)

Read data from file object.

Parameters:

• file_obj – An open file object.

• dtype – Data type to be read.

• count – Number of elements to be read.

• offset – Byte offset where to start reading.