Commit 29d1c14a authored by Ingrid Luijkx's avatar Ingrid Luijkx
Browse files

Example rc file to work with random observation operator

parent 1dbbaae1
! CarbonTracker Data Assimilation Shell (CTDAS) Copyright (C) 2017 Wouter Peters.
! Users are recommended to contact the developers ( to receive
! updates of the code. See also:
! This program is free software: you can redistribute it and/or modify it under the
! terms of the GNU General Public License as published by the Free Software Foundation,
! version 3. This program is distributed in the hope that it will be useful, but
! WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
! FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
! You should have received a copy of the GNU General Public License along with this
! program. If not, see <>.
! author: Wouter Peters
! This is a blueprint for an rc-file used in CTDAS. Feel free to modify it, and please go to the main webpage for further documentation.
! Note that rc-files have the convention that commented lines start with an exclamation mark (!), while special lines start with a hashtag (#).
! When running the script, this /.rc file will be copied to your run directory, and some items will be replaced for you.
! The result will be a nearly ready-to-go rc-file for your assimilation job. The entries and their meaning are explained by the comments below.
! Created on August 20th, 2013 by Wouter Peters
! The time for which to start and end the data assimilation experiment in format YYYY-MM-DD HH:MM:SS
time.start : 2012-01-01 00:00:00
time.finish : 2012-01-15 00:00:00
! Whether to restart the CTDAS system from a previous cycle, or to start the sequence fresh. Valid entries are T/F/True/False/TRUE/FALSE
time.restart : False
! The length of a cycle is given in days, such that the integer 7 denotes the typically used weekly cycle. Valid entries are integers > 1
time.cycle : 1
! The number of cycles of lag to use for a smoother version of CTDAS. CarbonTracker CO2 typically uses 5 weeks of lag. Valid entries are integers > 0
time.nlag : 3
! The directory under which the code, input, and output will be stored. This is the base directory for a run. The word
! '/' will be replaced through the script by a user-specified folder name. DO NOT REPLACE
dir.da_run : /Storage/CO2/naomi036/test_ctdas
! The resources used to complete the data assimilation experiment. This depends on your computing platform.
! The number of cycles per job denotes how many cycles should be completed before starting a new process or job, this
! allows you to complete many cycles before resubmitting a job to the queue and having to wait again for resources.
! Valid entries are integers > 0
da.resources.ncycles_per_job : 1
! The ntasks specifies the number of threads to use for the MPI part of the code, if relevant. Note that the CTDAS code
! itself is not parallelized and the python code underlying CTDAS does not use multiple processors. The chosen observation
! operator though might use many processors, like TM5. Valid entries are integers > 0
da.resources.ntasks : 1
! This specifies the amount of wall-clock time to request for each job. Its value depends on your computing platform and might take
! any form appropriate for your system. Typically, HPC queueing systems allow you a certain number of hours of usage before
! your job is killed, and you are expected to finalize and submit a next job before that time. Valid entries are strings.
da.resources.ntime : 04:00:00
! The resource settings above will cause the creation of a job file in which 2 cycles will be run, and 30 threads
! are asked for a duration of 4 hours
! Info on the DA system used, this depends on your application of CTDAS and might refer to for instance CO2, or CH4 optimizations.
da.system : CarbonTracker
! The specific settings for your system are read from a separate rc-file, which points to the data directories, observations, etc
da.system.rc : da/rc/carbontracker_random.rc
! This flag should probably be moved to the da.system.rc file. It denotes which type of filtering to use in the optimizer
da.system.localization : None
! Info on the observation operator to be used, these keys help to identify the settings for the transport model in this case
da.obsoperator : TM5
! The TM5 transport model is controlled by an rc-file as well. The value below refers to the configuration of the TM5 model to
! be used as observation operator in this experiment.
da.obsoperator.home : ${HOME}/TM5
da.obsoperator.rc : ${da.obsoperator.home}/tm5-ctdas-ei-zoom.rc
! The number of ensemble members used in the experiment. Valid entries are integers > 2
da.optimizer.nmembers : 10
! Finally, info on the archive task, if any. Archive tasks are run after each cycle to ensure that the results of each cycle are
! preserved, even if you run on scratch space or a temporary disk. Since an experiment can take multiple weeks to complete, moving
! your results out of the way, or backing them up, is usually a good idea. Note that the tasks are commented and need to be uncommented
! to use this feature.
! The following key identifies that two archive tasks will be executed, one called 'alldata' and the other 'resultsonly'.
!task.rsync : alldata onlyresults
! The specifics for the first task.
! 1> Which source directories to back up. Valid entry is a list of folders separated by spaces
! 2> Which destination directory to use. Valid entries are a folder name, or server and folder name in rsync format as below
! 3> Which flags to add to the rsync command
! The settings below will result in an rsync command that looks like:
! rsync -auv -e ssh ${dir.da_run}
!task.rsync.alldata.sourcedirs : ${dir.da_run}
!task.rsync.alldata.destinationdir :
!task.rsync.alldata.flags g -auv -e ssh
! Repeated for rsync task 2, note that we only back up the analysis and output dirs here
!task.rsync.onlyresults.sourcedirs : ${dir.da_run}/analysis ${dir.da_run}/output
!task.rsync.onlyresults.destinationdir :
!task.rsync.onlyresults.flags : -auv -e ssh
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment