This program is used to process and output statistics for the inter-comparison of for example PPS results and CloudSat/CALIPSO observations.

• The main running program is: process_atrain_match.py which will do matchups or process_master.py will do matchups and retrieve result foreach case.

• The program truth_imager_match.py is the main library for matching and truth_imager_make_statistics(_lib).py is the main libaries for statistics.

• The compile_stat.py accumulates statistics uses the module statistics to accumulate statistics (only run if process_master have been used and there exist statistics).

• Program is updated wo be able to use CALIOP-CALIPSO, CPR (CloudSat), AMSR_E, Synop
  or CATS (ISS) as truth. Modules used to handle the truths (read, reshape, etc): cloudsat.py calipso.py amsr.py iss.py synop.py mora.py

• Program can read satellite data from: PPS, CCI and MAIA etc. When satllite data comes from PPS-MODIS also modis lvl-2 data can be matched. Files to read imager satellite data: read_pps.py
  read_maia.py read_cci.py
  read_oca.py read_patmosx.py
  read_modis_products.py

• Format of matchup files, and reading and writing ot these can be found in matchobject_io.py.

• Imager cloud top height datasets have been re-calculated to heights above mean sea level using CloudSat and CALIPSO elevation data

• The MODIS cloud flag has been added to the extracted CALIPSO dataset. This enables direct comparisons to the MODIS cloud mask! Consequently, corresponding MODIS Cloud Mask statistics are calculated and printed.

• The Vertical Feature Mask parameter in the CALIPSO dataset has been used to subdivide results into three cloud groups: Low, Medium and High. This has enabled an evaluation of PPS Cloud Type results and a further sub-division of Cloud Top Height results

• The National Snow and Ice Data Center (NSIDC) ice and snow mapping results have been added to the extracted Calipso parameters. Together with the IGBP land use classification it is then possible to isolate the study to focus on one of the several surface categories.

• Modules: reshaped_files_scr example script to plot things from reshaped files.

• Configuration in the etc/atrain_match.cfg. There are some enironment variables that need to be set:

  ATRAIN_MATCH_CONFIG_FILE # Name of atrain_match config file default atrain_match.cfg
  ATRAINMATCH_CONFIG_DIR # path to atrain_match.cfg
  AREA_CONFIG_FILE       # only for plotting
  VALIDATION_RESULTS_DIR # path to validation main directory
  ATRAIN_RESOLUTION      # 1 or 5

Run unit test with:

• python -m unittest atrain_match/tests/init.py