As far as I can tell, the field is never used. Is there an intended future use for it? More to the point, is there any reason for other codes using the Hector core to check its value when they get one as a return value?

CH4 concentrations are correct but OH lifetime decrease rather than increase.

From November meeting.

Hi @rplzzz @cahartin FYI I've started to add a "0.2" milestone to some issues. These are ones I think should be high priority (important & doable, or easy); feel free to add/opine.

If Hector can't find an input file, e.g. inpdut/hector_rcp45.ini, it reports

Fri Jan  9 13:38:39 2015:NOTICE:parse: Parsing inpdut/hector_rcp45.ini
* Program exception: Parse error in file inpdut/hector_rcp45.ini line -1
* Function parse, file h_reader.cpp, line 49

This is not very informative.

Minor What should go in component constructor, and what in init()? N Ben Open

Minor When model hits unknown component name in ini, should give better error N core.cpp Ben Open

From @bpbond, originally posted in #27 :

I'm building this code for the first time on my machine, and getting a compiler (clang, set to dialect
gnu++11) error at logger.hpp:33, in the LogLevel enum. Seems like it has to be related to the new
namespace, but not a very helpful error message ("expected '}'")

Later followed up with:

Ah, OK, the clang compiler has a default DEBUG preprocessor macro that signals whether the build
target is debug or release, and that was screwing things up. Option 1 - I change the preprocessor
macro. Option 2 - change DEBUG to HDEBUG, or something like that, in the Hector code. Probably the first.

Critical Excel makes shitty line endings, but people will use it; table reader needs to be robust N csv_table_reader.cpp 17 Ben Ben/Pralit Open

Major Sea level rise code needs re-visiting N slr.cpp Ben Corinne Open

main() kicks off by opening the log thus:

  Logger& glog = Logger::getGlobalLogger();
  glog.open( string( MODEL_NAME ), true, Logger::DEBUG );

This should record log messages at level DEBUG or higher; however, none of the log messages recorded at log level NOTICE have been recorded. E.g, this message

   H_LOG( glog, Logger::NOTICE ) << "Running the core." << endl;

appears nowhere in the output:

rpl@WE27653:~/wrk/hector$ source/hector input/hector_rcp45.ini | grep "Running the core"

Run completed. Press [ENTER] to finish.

rpl@WE27653:~/wrk/hector$ 

Sometime before we open the repository to the public we need to add a file with the text of the license we are releasing under to the top-level directory.

And if it doesn't exist, no logs get written and fails silently.

This was raised by Reviewer 1 of the GMD ms.

Although forcing depends on halocarbon, it seems like we should be able to disable/exclude individual gases.

Minor ocean_csys header needs doxygen documentation N ocean_csys.hpp Ben Corinne Open

Components should have a 'stub' or 'data' mode (à la CESM) in which they simply pull data from a specified file, or return fixed values.

Noticeably different than MAGICC (for RF) and CMIP5 (temperature).

Minor INI parser needs to be robust to spaces around csv= N Ben Open

It should be able to give back the single value for any time.

When this is fixed remove this work around line from N2Ocomponent::prepareToRun():
N2O.set(0, N0);

Cosmetic Log interpolation method set N interolator.cpp Ben Open

Good news, everyone! I've got Hector running with GCAM. Here are some plots from the GCAM data viewer. The red line in each is Hector; the blue line is MAGICC. Note also that MAGICC starts ingesting GCAM emissions after 2005, while Hector starts after 2010.

Now, for the question promised by the issue tag. What's the story with the temperatures? They look awfully high. For example, in 2000 the temperature anomaly is nearly 1.25 C. I pulled the outputstream from a stand-alone run and found the same result:

2000,rcp45,0,temperature,Tgav,1.17862,degC
2000,rcp45,0,temperature,Tgaveq,1.39939,degC

(sub-question: what is the difference between T and Teq?)

I just looked back at figure 4 from the paper, and it seems we see it there too. We track pretty well through sometime shortly after 1950, and then we sort of take off abruptly. Do we have any idea what's causing that? Everything else looks good, but that temperature result is bound to generate some push-back from GCAM users.

Major Solver should be aware of (solve) ALL biomes in simpleNbox N simpleNbox.cpp Ben Open

Major The ocean needs more INI parameters (as opposed to hard-coded values in code) N ocean_component.cpp Ben Corinne Open

Disabling 'leaf' components (no one depends on them) should not affect the model's ability to run. The model should not run if non-leaf components are disabled (and no other component is providing their data). Verify this.

Major Move the bugbase to Github's issue tracker N Ben Open

Minor If there are any 'global' pools, should not also be biomes! N simpleNbox.cpp Ben Open

In addition to a license (see issue #28 ) we probably want a "fair use" document, akin to what (I think) CESM has.

Minor Why doesn't const unitval output work? N unitval.hpp Ben Pralit Open

Currently Hector provides only total forcing and forcing from each halocarbon gas. The model calculates forcing from other atmospheric components and from albedo (obviously---it needs them for the total), but it doesn't provide a way to report them separately from the total.

We may want to rename .first and .last in the tseries to .firstDate and .lastDate for clarity. I was using them incorrectly thinking they were giving me back a value instead of a date.

Minor Using fixed NPP in spinup means bad things can happen if other C pools are too low N simpleNbox.cpp Ben Open

Cosmetic The control flow of DependencyFinder::createOrdering could be less confusing N dependency_finder.cpp Pralit Pralit Open

Major BC and OC forcings need reworking N forcing_component.cpp Ben Open

in both the ini files and the code

Minor tseries code needs to be simplified and cleaned up N tseries.hpp Ben Ben/Pralit Open

Right now, the forcing component has to know about the existence of all the individual halocarbons. It would be a LOT more elegant if it could simply (i) register its dependency on ALL halocarbons, and (ii) query halocarbonComponent for a list of all the extant hc's, then get the forcing information from each. That would mean halocarbons would be purely defined via input, no code changes required, and would be backwards-compatible.

Minor Interpolator needs to be sort its own points (if anything except tseries using it) N interolator.cpp Ben Open

Major Fix H_ASSERT macro to allow the + on the string in visual studio N h_exception.hpp Corinne Ben/Pralit Open

Presumably this means that we'll have a file VERSION.TXT, the contents of which the preprocessor will insert (for macro VERSION) into source files before building. Another macro GITBUILD would insert the latest commit number.

See e.g. http://stackoverflow.com/questions/6526451/how-to-include-git-commit-number-into-a-c-executable

Cosmetic Come up with our own formatted time output N logger::getDateTimeStamp Pralit Pralit Open

Major model crashes if the emissions are not specified in hector.ini file N Corinne Open

In particular,

• building and running the model on all platforms
• a demo showing how to use backend R scripts

This has been (rightly) requested by the GMD reviewers.

Major SLR outputstream visitor is a hack! N csv_outputstream_visitor.cpp Ben Ben Open

Minor All constraints should follow identical rules wrt extrapolation (not allowed) N Ben Open

Minor Log level should be (?) a command-line option? N main.cpp Ben Open