SPEEDY-IER: READ.me File
Sylvia Dee <sylvia.dee@rice.edu>

Modified 01/07/2020

SPEEDY-IER is a fast‐physics atmospheric circulation model suitable for long ensemble integrations: 
the efficient AGCM Simplified Parameterizations, Primitive Equation Dynamics (SPEEDY), with 
newly added water isotope physics. The model (SPEEDY‐isotope‐enabled reconstructions (IER)) 
simulates the hydrological cycle and isotope ratios in atmospheric water at a fraction of 
the computational cost of Intergovernmental Panel on Climate Change (IPCC)‐class GCMs. 

Publication Link: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/2014JD022194


Compilation and run instructions for SPEEDY-IER:


To compile, change directory (cd) into 'run' and use the command:

>> ./run_speedy.bash T30 XXX 0

Those three commands, as specified in the bash script, are: the resolution 
(T30/T21), experiment name (3 characters only), and restart (0 for no, 1 for yes).

This bash script will compile the model, searching for all the source code in the 'source' 
file (this is where the fortran 'makefile' is). 

To run the model, >> 'run_speedy-ier.pbs' 

This PBS file should appear in the temp folder created
upon compilation with the run name (e.g. TMP_XXX). 

When you run the model, it will send a call to open the input files, which are in v40.input. 
This initialization file includes time step information, physics parameters, etc. 

When you run the model it should ask you four questions. 
One is "would you like to modify the time stepping parameters?"

-- You can say "y" to this and then change the start date depending on what input files you are using. 

For example, to run SPEEDY-IER with historical SSTs, IYEAR=1980. 
Then change the variable 'nmonths' to reflect the number of years you want it to run for. 

Example:
'IYEAR = 1980' and 'NMONTHS = 12' 

You can also modify these fields directly in the ver40.input folder.

The "data" folder contains the forcing files. 
You can reconfigure the file "inpfiles.s" which is in ver40.input to change which forcing files are used. 

SD Known issues: bias in upper mid-latitude water isotopes in vapor due to convection physics. Bug fix in progress.