Updates on processing HKO data

[Ma Ho Long Perry]

Dear TEMIR developers and users,

In order to utilize HKO data in the simulation, I have written a tool in R that helps processing HKO data for simulation in TEMIR.

What's new:

1. Including solar radiation input
2. A separated script for pre-processing HKO data from .txt/.xlsx format to .csv format
3. A technical note describing the processing procedures: developers are encouraged to read it so you can modify my code easily when other data comes in

The folder "HKO_processing" containing the R scripts and the technical note is placed under "TGABI/Ma/" in Dropbox and you are advised to put the entire folder in code_v1.1 under your own TEMIR folder.

The processed data can be accessed through /project/TGABI/TEMIR/TEMIR_inputs in "Charles"

Next step(s): We may need to think of how to convert the direct and diffuse solar radiation data to the corresponding PAR...

For details, please refer to the User's guide at the top of the script. Feel free to contact me if there are any problems or issues when using the tool.

-----

Other issue for TEMIR:

Please consider typing the following after applying f_simulate_ij if you are using FLUXNET/HKO data.

==============================================================================

rm(list = input_data_df['TEMIR_var_name'])
# rm(list = self_input_data_df['TEMIR_var_name']) # HKO data only

==============================================================================

like this:

Reason: Whenever there is unreliable data from either of them, the data of the previous timestep are not replaced if that variable does not belong to any one of GC input. It causes the usage of met field at previous time step in the simulation instead. In order to prevent this, these lines should be added.

Best regards,
Perry Ma