Data

The satellite and station data for this case study can be downloaded here:

Read netCDF files (satellite data)

The MSG data is provided in netcdf-Format and can be read into python via:

import xarray as xr

satellite_data = xr.open_dataset("path/to/file.nc")

The data is structured in a DataSet with multiple DataArrays, each representing one of the MSG SEVIRI bands. A cloud mask is also provided here. However, we only want to use this cloud mask for validation purposes and not for our training.

satellite_data
<xarray.Dataset>
Dimensions:   (time: 1, x: 1024, y: 768)
Coordinates:
  * time      (time) datetime64[ns] 2005-03-15T15:00:09.743000
  * x         (x) float64 -1.184e+06 -1.181e+06 ... 1.883e+06 1.886e+06
  * y         (y) float64 5.405e+06 5.402e+06 5.399e+06 ... 3.107e+06 3.104e+06
    acq_time  (y) datetime64[ns] 2005-03-15T15:12:40.449000 ... 2005-03-15T15...
Data variables:
    IR_016    (time, y, x) float32 ...
    IR_039    (time, y, x) float32 ...
    IR_087    (time, y, x) float32 ...
    IR_097    (time, y, x) float32 ...
    IR_108    (time, y, x) float32 ...
    IR_120    (time, y, x) float32 ...
    IR_134    (time, y, x) float32 ...
    VIS006    (time, y, x) float32 ...
    VIS008    (time, y, x) float32 ...
    WV_062    (time, y, x) float32 ...
    WV_073    (time, y, x) float32 ...
    cmask     (time, y, x) uint8 ...
Attributes:
    comment:      This data set was clipped to the LCRS-Domain and preprocess...
    instrument:   seviri
    history:      Created by pytroll/satpy on 2021-06-24 11:15:03.165717
    Conventions:  CF-1.7

Read GeoTiffs (DEM)

A digital elevation model (DEM) is provided as GeoTiff and can be opened via:

dem = xr.open_dataset("path/to/file.tif",engine="rasterio").band_data[0]

Read CSVs (station data)

METAR station measurements and meta data (with station location information) are provided in csv-files which can be opened via:

import pandas as pd

station_data = pd.read_csv("path/to/file.csv",parse_dates=["time"])
station_meta_data = pd.read_csv("path/to/file.csv")
station_data
icao time cloudcover cloud_altitude
0 EBAW 2005-01-15 00:00:00 3 2400
1 EBBR 2005-01-15 00:00:00 3 3600
2 EBCI 2005-01-15 00:00:00 3 2200
3 EBLG 2005-01-15 00:00:00 1 -999
4 EBOS 2005-01-15 00:00:00 3 2600
... ... ... ... ...
52007 LZKZ 2006-12-15 21:00:00 3 3300
52008 LZPP 2006-12-15 21:00:00 3 3500
52009 LZSL 2006-12-15 21:00:00 3 3300
52010 LZTT 2006-12-15 21:00:00 1 -999
52011 LZZI 2006-12-15 21:00:00 3 2400

52012 rows × 4 columns

Columns are encoded as follows:

  • icao: ID of METAR station

  • time: Time of measurement (YYYYmmddHHMM)

  • cloudcover: Cloud cover (1: cloud free, 2: cloud contaminated, 3 cloud covered)

  • cloud_altitude: Cloud altitude in meters above ground (missing data value: -999)

station_meta_data
icao x y
0 EBAW 2.920539e+05 4.611705e+06
1 EBBR 2.956507e+05 4.596070e+06
2 EBCI 2.967015e+05 4.571825e+06
3 EBLG 3.608589e+05 4.580852e+06
4 EBOS 1.875154e+05 4.613160e+06
... ... ... ...
268 LZKZ 1.430620e+06 4.434616e+06
269 LZPP 1.212834e+06 4.443047e+06
270 LZSL 1.296992e+06 4.439913e+06
271 LZTT 1.354894e+06 4.461921e+06
272 LZZI 1.246986e+06 4.476102e+06

273 rows × 3 columns

Columns are encoded as follows:

  • icao: ID of METAR station

  • x: x-Coordinate of station location in GEOS projection (MSG)

  • y: y-Coordinate of station location in GEOS projection (MSG)

Task

  • Download all data sets and make yourselves familiar with the data