PLOTO FMI data long-term storage and availability description
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%


Data is available in CSC long-term object storage system allas as described below.

Background document: Deliverable report D3.3.


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%


SULINA (use case A) data from
/fmi/scratch/project_2001816/palm_cases/PLOTO/SULINA/ is found in
allas through

https://a3s.fi/PLOTO_SULINA/PLOTO_SULINA.tgz
or
s3://PLOTO_SULINA
by s3 protocol.

It is about 30 GB.

This directory contains the PLOTO Use case A (only the Sulina area)
LES run directiories WD_??? with essential parts of the result data
and the final channel-masked data in subdirectories
CHANNEL_MASKED_RESULT_DATA/WD_???/ used in the downscaling, and
precursor run setups, and symbolic links to the input data and
postprocessing scripts stored permanently in
/fmi/projappl/project_2001816/palm_cases/PLOTO/SULINA/

Also numerous analysis results and other files (mostly small files) are
stored here.

The principal post-processing scripts for long-term analyses are:
- preProcessRCMdata.py  (only needed on projappl to preprocess the RCM-data)
- LES_ReferencingWind.py
- downScaleWind3D.py

All these scripts are found also in allas in
/fmi/projappl/project_2001816/palm_cases/PLOTO/ see description below.

Two additional local scripts are needed for the operational short-term
downscaled wind perdiction system WindPredict:
- wp_esik_data_sulina.py
- wp_esik_skaalaus_sulina.py


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%


Budapest (use case B) data from
/fmi/scratch/project_2001816/palm_cases/PLOTO/BUDAPEST/ is found in
allas through

https://a3s.fi/PLOTO_BUDAPEST/PLOTO_BUDAPEST.tgz
or
s3://PLOTO_BUDAPEST
by s3 protocol.

It is about 30 GB.

This directory contains the PLOTO Use case B LES run directiories
WD_??? with essential parts of the result data and precursor run
setups, and symbolic links to the input data and postprocessing
scripts stored permanently in
/fmi/projappl/project_2001816/palm_cases/PLOTO/BUDAPEST/

Also numerous analysis result and other files (mostly small files) are:
stored here.

The principal post-processing scripts for long-term analyses are
- preProcessRCMdata.py  (only needed on projappl to preprocess the RCM-data)
- LES_ReferencingWind.py
- downScaleWind3D.py

All these scripts are found also in allas in
/fmi/projappl/project_2001816/palm_cases/PLOTO/ see description below.

Two additional local scripts are needed for the operational short-term
downscaled wind perdiction system WindPredict:
- wp_esik_data_budapest.py
- wp_esik_skaalaus_budapest.py


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%


Wallonie (use case C) data from
/fmi/scratch/project_2001816/palm_cases/PLOTO/WALLONIE/ is found in
allas through

https://https://a3s.fi/PLOTO_WALLONIE/PLOTO_WALLONIE.tgz
or
s3://PLOTO_WALLONIE
by s3 protocol.

It is about 275 GB.

This directory contains the PLOTO use case C LES run directiories
RESULTS/WD_??? with the run setups (but no raw results), and symbolic
links to the input data stored in
/fmi/projappl/project_2001816/palm_cases/PLOTO/WALLONIE/, and the
final river-masked data in subdirectories
RIVER_MASKED_RESULT_DATA/WD_???/ including the final result files
River_uv_2D.nc (for wave analysis) and River_wswd_Gust2D.nc (for gust
analysis) used in the downscaling and also their precursor files
COLLOCATED_N??_M?? (almost raw results, only transformed to collocated grid).

The principal postprocessing scripts for long-term analyses are:
- LES_ReferencingWind.py
- validateWind.py (for validation)
- riverMask2DMdata.py (for wave analysis)
- riverMask3DGdata.py (for gust analysis)
- downScaleWind2DMRiver.py (for wave analysis)
- downScaleWindGustRiver.py (for gust analysis)
are stored locally here under the subdirectory WALLONIE and also in
/fmi/projappl/project_2001816/palm_cases/PLOTO/WALLONIE/ see description below.

Also numerous analysis result and other files (mostly small files) are
stored here. All these scripts are found also in allas in
/fmi/projappl/project_2001816/palm_cases/PLOTO/ see description below.

Three additional local scripts are needed for the operational
short-term downscaled wind perdiction system WindPredict and stored
under RIVER_MASKED_RESULT_DATA:
- wp_esik_data_wallonia.py
- wp_esik_data_wallonia_1.py
- wp_esik_data_wallonia_2.py


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%


Input-data files and postprocessing and analysis scripts etc stored in
/fmi/projappl/project_2001816/palm_cases/PLOTO/ are found also in
allas through

https://a3s.fi/PLOTO_INPUT_DATA_AND_SCRIPTS_ALL_USECASES/PLOTO_INPUT_DATA_AND_SCRIPTS_ALL_USECASES.tgz
or
s3://PLOTO_INPUT_DATA_AND_SCRIPTS_ALL_USECASES
by s3-protocol.

It is about 7 GB.

This package contains subdirectories for all the studied use cases:
SULINA, BUDAPEST and WALLONIE (Liege). For the Romanian use case (A),
the wind studies were only conducted for SULINA area.

These use-case subdirectories in turn contain the following subdirectories:

1) subdirectories: EUROCORDEX_WIND_SULINA, EUROCORDEX_WIND_Budapest
and EUROCORDEX_WIND_Liege for the original EUROCORDEX climate-model
wind data including several GCM/RCM combinations for scenarios:
RCP2.6, RCP4.5 and RCP8.5 for 2006-2100(2099 for some models), and
historical data 1971-2005. The wind data is at 10 m AGL in the points
reported in D3.3. Data format is NetCDF. The wind parameters stored
are: daily maximum 10 min averaged wind speed (sfcWindmax), Daily
maximum gust wind speed based on parameterized gust wind speed
(wsgsmax), six-hourly instantaneous u- and v-wind velocity components
(uas and vas). For Sulina, there are two sets of this data, one for
Sulina port in Sulina town (Sulina_port) and another for channel
opening to Black sea (Sulina_channel_delta); see report D3.3.

2) Subdirectories preprocessedRCMdata contain EUROCORDEX climate-model
wind data preprocessed by the python script preProcessRCMdata.py also
stored here which reads the given RCM wind data files to extract the
time series of the wind parameters sfcWindmax, wsgsmax, uas, and vas
introduced above for each given GCM/RCM pairs and RCP-scenarios and
estimates the wind direction of the daily maximum 10 min averaged wind
based on the six-hourly wind vector data. This approximation has to be
made because the wind direction corresponding to the daily maximum 10
min averaged wind speed is unfortunately not stored in the EURO-CORDEX
database. The resulting data is written in ASCII text-formatted files
for each case (GCM/RCM-pair and RCP/historical) in subdirectory
preprocessedRCMdata to be used as input data for downscaling.

3) Subdirectories PIDS contain the PALM input topography and
vegetation data files (so-called static drivers) for all domains
including nested domains.

4) Subdirectories parin_setup contain PALM parameter input file setups
(ASCII text files) for all modelling domains including nested
domains. And subdirectories PARIN contain root domain parameter input
files for all wind sectors. Nested-domain parameter input files do not
depend on the wind sector.

There are the following essential python scripts in each use-case
subdirectory:

1) preProcessRCMdata.py already described above

2) LES_ReferencingWind.py which extracts the referencing wind
information for downscaling from each LES result.

3) channelMask3D.py postprocesses the LES output files for
downscaling. These postprocessing actions depend on the use case and
may involve e.g. masking and combination of LES output files.

4) downScaleWind3D.py is the actual downscaling script.

5) wind_predict_esik.py is needed to prepare data for the short-term
operational downscaled wind prediction system windPredict.

6) wind_predict_esik_skaalaus.py (the script name varies a little
between use cases) is needed to prepare data for the short-term
operational downscaled wind prediction system windPredict.

In addition to these, there are some other less essential
subdirectories and files depending on the use case.


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

