diff --git a/FIGSEXAMPLES/destine_obs_satellite_figs.jpg b/FIGSEXAMPLES/destine_obs_satellite_figs.jpg new file mode 100644 index 0000000000000000000000000000000000000000..50d7798fa69b91b65cb59b4728fc0c1b1f1c6349 Binary files /dev/null and b/FIGSEXAMPLES/destine_obs_satellite_figs.jpg differ diff --git a/README.md b/README.md index db4515f6b9fddfe8738b8c6ee1615a3be8dffa56..5721c865d15d0fb8c54b6ab344d9f142d99da86b 100644 --- a/README.md +++ b/README.md @@ -33,12 +33,18 @@ The OBSALL application is currently being developed as a set of scripts (written - Note, needed input includes: list of radiosounding stations with geographical coordinates, modelled data at stations coordinates at 2 metre height and at 850 hPha pressure level, pre-computed quantiles as a function of time of year and rank histogram bootstrap mean square deviation (MSD) values. - ![Semantic description of image](FIGSEXAMPLES/destine_obs_radsound_figs.jpg "Plots for difference between air temperature at 2 m and 850 hPa pressure level: T2-T850")* Plots for difference in air temperatures at 2 metre and at 850 Ha pressure level (T2-T850) for period of 1 Jan - 31 Dec 2010: **Figure 2a** - Quantiles/ time series data and rank histogram (00 and 12 UTCs) for radiosounding station 51431; **Figure 2b** - P-values for T2-T850 quantiles; **Figure 2c** - Rank histogram summary statistics for 00 and 12 UTCs: normalized p-value frequencies and normalized quantile frequencies* -# Part AMSU-A (satellite observations) - in development +# Part AMSU-A (satellite observations) 1. `main_amsua.sh` - main bash-script to run Apps for observational data for satellite-based AMSU-A measurements & `set_env.sh`- bash-script which defines variables and functions, and required modules to be loaded and used in the workflow. 2. `gsv_mod_data.sh` - bash-script to run pre-processing (using `cdo` operators; in later versions - `polytope`) of 2D on surface (`sfc`) and 3D modelled data on pressure levels (`pl`) extracted with gsv interface over global domain. The `Radiance Simulator` (using `convert_pl2ml.py`) interpolates modelled data from `pl`-levels to “model”-levels (note, workflow will initially be demonstrated on a single selected location i.e., around a point of 70.0N, 45.0E in the Barents Sea; and later geographically expanded using `polytope`) and preprocessed data saved in (temporary) hourly GRIB-files. Required 3D variables are: `t` - temperature (130), `q` - specific humidity (133), `clwc` - specific liquid water content (246); and 2D variables are: `skt` - skin temperature (235), `ci` - sea ice fraction (31), `lsm` - land-sea mask (172),`sp` - surface pressure (134), `10u` and `10v` - U and V components of wind speed at 10 m (165 and 166), `2t` - air temperature at 2 m (167), `2d` - dewpoint temperature at 2 m (168), and `z` - geopotential (for orography) (129). 3. `amsua_obs.sh` - bash-script to prepare a metadata file for computation of AMSU-A model counterparts from preprocessed hourly GSV data. It runs depending on the type of climate simulation: (i) reality following data assimilation-like, and (ii) free running climate simulation. For (i) the one-to-one model counterparts can be produced, so actual metadata from AMSU-A observations in the vicinity of the selected location are extracted from ODB. This is done with `filter_data_v3.py` For (ii) the model output can be compared to observations only in climatological terms, so metadata is written only for the centre point of the selected area. 4. `amsua_mod.sh` - bash-script to run the `Radiance Simulator` tool with pre-described control configuration parameters and to produce a model counterpart (file in netcdf-format), which (using `convert_netcdf2txt.py`) is converted to (temporary) dat-file and added (using `import` with `odb_api`) to ODB. -5. The monitoring for AMSU-A is under development. Initially it was done on Puhti HPC of CSC, and then, to be migrated, tested, and implemented into workflow on Lumi HPC of CSC. The required polytope tool is expected to be released in December 2023. For AMSU-A monitoring, `amsua_stats.sh` bash-script is used to run (additional `bash`, `fortran`, `python` scripts) calculating/producing similar statistics as for TEMP, using daily mean values inside the selected area. +5. `amsua_stats.sh` - bash-script to run (addtional `bash`, `fortran`, `python` scripts) calculating/producing: +- (i) rank histograms for all locations-areas from satellite observations for daily mean values of brightness temperature (with `produce_rank_histograms_all_locations.sh`, using `rank_histograms_one_location.f95` and `sql/import` with `odb_api`), +- (ii) standard plots for each location-area from satellite observations (with `produce_standard_plots_all_locations.sh` using `plot_quantiles_rankhist_amsua.py` and `sql` with `oadb_api`; see an example of such plot in **Figure 3a**), and +- (iii) summary rank histograms for locations-areas from satellite observations (`summary_rank_histograms_all_locations.sh` using `rank_histogram_sumstats_locations.f95`, `plot_p_values_map_amsua.py`, `plot_rankhist_sum_all_locations.py` and `sql` with `odb_api`; see examples of such plots in **Figure 3b** and **Figure 3c**). +- Note, needed input includes: list of locations-areas from satellite observations with geographical coordinates, modelled data over geographical locations-areas coordinates at sfc-, pl-, and o2d-levels, pre-computed quantiles as a function of time of year and rank histogram bootstrap mean square deviation (MSD) values. +- ![Semantic description of image](FIGSEXAMPLES/destine_obs_satellite_figs.jpg "Plots for brightness temperature /AMSU-A Channel 5/ for selected geographical location-area /a00/")* Plots for brightness temperature /AMSU-A Channel 5/ for selected geographical location-area /a00/ for period of 1 Jan - 31 Dec 2000: **Figure 3a** - Quantiles/ time series data and rank histogram for daily mean brightness temperature for location-area /a00/; **Figure 3b** - P-values for brightness temperature quantiles for locations-areas; **Figure 3c** - Rank histogram summary statistics for brightness temperature for locations-areas: normalized p-value frequencies and normalized quantile frequencies* + # Installation **Git**: to copy the "obsall" repository to your local directory via Git using: