Merge branch 'feature-STILT'

ctdas-stilt.jb

+#!/bin/sh
+#$ das.py
+#$ co2
+#$ nserial 1
+#$ 06:30:00
+#$ /bin/sh
+
+echo "All output piped to file template.log"
+#source /usr/local/Modules/3.2.8/init/sh
+#source /opt/intel/bin/ifortvars.sh intel64
+export HOST='daint'
+#module load python
+export icycle_in_job=999
+python ctdas-stilt.py rc=ctdas-stilt.rc -v $1 >& ctdas-stilt.log &

ctdas-stilt.log

+Traceback (most recent call last):
+  File "ctdas-stilt.py", line 30, in <module>
+    from da.platform.cartesius import CartesiusPlatform
+  File "/nfs/home5/awoude/CTDAS/da/platform/cartesius.py", line 135
+    print 'output', output
+                 ^
+SyntaxError: Missing parentheses in call to 'print'. Did you mean print('output', output)?

ctdas-stilt.py

+"""CarbonTracker Data Assimilation Shell (CTDAS) Copyright (C) 2017 Wouter Peters. 
+Users are recommended to contact the developers (wouter.peters@wur.nl) to receive
+updates of the code. See also: http://www.carbontracker.eu. 
+
+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 <http://www.gnu.org/licenses/>."""
+#!/usr/bin/env python
+
+#################################################################################################
+# First order of business is always to make all other python modules accessible through the path
+#################################################################################################
+
+import sys
+import os
+import logging
+sys.path.append(os.getcwd())
+
+#################################################################################################
+# Next, import the tools needed to initialize a data assimilation cycle
+#################################################################################################
+
+from da.tools.initexit import start_logger, validate_opts_args, parse_options, CycleControl
+from da.tools.pipeline import forward_pipeline, header, footer
+from da.platform.cartesius import CartesiusPlatform
+from da.baseclasses.dasystem import DaSystem 
+from da.baseclasses.statevector import StateVector 
+from da.carbondioxide.obspack_globalviewplus2 import ObsPackObservations 
+from da.carbondioxide.optimizer import CO2Optimizer
+from da.baseclasses.observationoperator import ObservationOperator
+#from da.analysis.expand_fluxes import save_weekly_avg_1x1_data, save_weekly_avg_state_data, save_weekly_avg_tc_data, save_weekly_avg_ext_tc_data
+#from da.analysis.expand_molefractions import write_mole_fractions
+
+
+#################################################################################################
+# Parse and validate the command line options, start logging
+#################################################################################################
+
+start_logger()
+opts, args = parse_options()
+opts, args = validate_opts_args(opts, args)
+
+#################################################################################################
+# Create the Cycle Control object for this job    
+#################################################################################################
+
+dacycle = CycleControl(opts, args)
+
+
+platform = MaunaloaPlatform()
+dasystem = DaSystem(dacycle['da.system.rc'])
+obsoperator = ObservationOperator(dacycle['da.obsoperator.rc'])
+samples     = ObsPackObservations()
+statevector = StateVector()
+optimizer = CO2Optimizer()
+
+##########################################################################################
+################### ENTER THE PIPELINE WITH THE OBJECTS PASSED BY THE USER ###############
+##########################################################################################
+
+
+logging.info(header + "Entering Pipeline " + footer) 
+
+#ensemble_smoother_pipeline(dacycle, platform, dasystem, samples, statevector, obsoperator, optimizer)
+forward_pipeline(dacycle, platform, dasystem, samples, statevector, obsoperator, optimizer)
+
+##########################################################################################
+################### All done, extra stuff can be added next, such as analysis
+##########################################################################################
+
+#logging.info(header + "Starting analysis" + footer) 
+#sys.exit(0)
+#
+#save_weekly_avg_1x1_data(dacycle, statevector)
+#save_weekly_avg_state_data(dacycle, statevector)
+#save_weekly_avg_tc_data(dacycle, statevector)
+#save_weekly_avg_ext_tc_data(dacycle)
+#write_mole_fractions(dacycle)
+#
+#sys.exit(0)
+
+

ctdas-stilt.rc

+! CarbonTracker Data Assimilation Shell (CTDAS) Copyright (C) 2017 Wouter Peters. 
+! Users are recommended to contact the developers (wouter.peters@wur.nl) to receive
+! updates of the code. See also: http://www.carbontracker.eu. 
+!
+! 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 <http://www.gnu.org/licenses/>. 
+
+! 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 start_ctdas.sh, 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.
+!
+!
+! HISTORY:
+!
+! 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          : 2016-01-10 00:00:00
+time.finish         : 2016-01-15 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          : 7
+
+! 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 start_ctdas.sh script by a user-specified folder name. DO NOT REPLACE
+
+dir.da_run          : /projects/0/ctdas/awoude
+
+! 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         : STILT
+
+!
+! 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    : da/stilt
+da.obsoperator.rc      : ${da.obsoperator.home}/stilt.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} you@yourserver.com:/yourfolder/
+!
+
+!task.rsync.alldata.sourcedirs : ${dir.da_run}
+!task.rsync.alldata.destinationdir : you@yourserver.com:/yourfolder/
+!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 : you@yourserver.com:/yourfolder/
+!task.rsync.onlyresults.flags : -auv -e ssh
+