Todolist.txt

This is a list of additions, enhancements, corrections, etc., that I am
currently working on or thinking about, which hopefully will be included
in the next release of the GLOW model, anticipated for 2018.

Anybody that wants to volunteer to work on one of these, feel free.

LBH thresholding
Update SUNCOR
Correct solar irradiance for Sun-Earth distance
Green line recombination emission
O2 A-bands emission
Ion chemistry re-examination
Option for parameterized auroral proton ionization rate
Update magnetic field used in FIELDM and GEOMAG
Re-install conjugate photoelectrons as an option
Update IRI
WACCM-X interface for driver program
Fix electron heating rate
Add He, He, He+, and H+
Add cusp and drizzle fluxes from TGCMs
Re-write EXSECT for table-based electron impact cross-sections

Discussion:

LBH thresholding:
This means putting in different cross sections for each v' upper state,
each of which has a different threshold.  The approach will be to use a total
cross section for N2(a) in the energetic electron redistribution array, but
individual cross sections for each v' in the excitation calculation.  That will
be done in the new subroutine BANDS, which also could handle collisional/cascade
re-distribution.

Update SUNCOR:
SUNCOR needs to be updated for a permanent Y2K fix, and J2000
coordinates.  It's pretty accurate as it is, at least through 2050.  Also add
Sun-Earth distance, which is part of the same algorithm but not currently
calculated.

Correct solar irradiance for Sun-Earth distance:
Nobody ever thought we knew solar EUV to anything close to 7% accuracy, so
I never worried about this before.  Now, with improved measurements,
it seems worth bothering to put it in.  Also note that a comment in the
NRLMSISE-00 source code indicates that the value of F10.7 input to MSIS should
actually be the measured value at Earth, not the 1 AU value, which means that
it should be corrected for Sun-Earth distance.

Green line recombination emission:
This has been on the list for decades.  Just haven't gotten around to it,
because it isn't important above ~100 km, for projects I've been working
on.  But it needs to be done.  Could be done by solving for all the O2
Hertzberg upper states, or as a specific recombination rate, skipping the
details, which brings us to...

O2 A-bands emission:
Also long overdue.  All the pieces for the thermospheric components are
there, but need to deal with recombination, and O3 if we go into the
mesosphere.

Ion chemistry re-examination:
I would like to bring GLOW into general agreement with Richards [2011]
who did a great job with this.  The only reason it hasn't happened yet is that
ion chemistry is not the prime mission here, but it can be important,
especially for aurora, and daytime E-region.

Option for parameterized auroral proton ionization rate:
This is in progress.  Immediate goal is to have a mechanism for calculating
electron density profiles and conductivities, not to produce secondaries and
emission features.

Update magnetic field used in FIELDM and GEOMAG:
This is overdue for an overhaul, but lower priority because FIELDM's only job
is to calculate the dip angle, and electron transport is weakly dependent on
dip angle.  GEOMAG now only handles coordinate transforms for the NOEM nitric
oxide model, which is based on SNOE data that were analyzed with a simple
offset-tilted-dipole model to begin with.  Nevertheless, I'd like to
consolidate these using IGRF and Apex coordinates, including secular
variation, using the standard TGCM method.  This would also make the
conjugate electron calculation (currently deprecated) more accurate.

Re-install conjugate photoelectrons as an option:
This is easily recovered from v. 0.97, but I took it out because of the above,
because it is a large performance hit for global calculations, and because it
wasn't really in use.

Update IRI:
Many might wonder why GLOW still uses IRI-90 as its default empirical a-priori
ionosphere.  The reasons are that GLOW generally only uses a-priori electron
density above 200 km, and the F-region in IRI hasn't changed much.  Also, the
new versions of IRI come with their own input files, imbedded MSIS calls,
and many options that aren't needed by GLOW.  I need to strip out all the
paraphernalia, and create a simple subroutine interface, similar to what is
now used for IRI-90.  It's a lot of messing around with OPC and I'm hoping
somebody else will do it.

WACCM-X interface for driver program:
This will be similar to the READTGCM module.  Any help appreciated!

Fix electron heating rate:
Electron heating rates are calculated, but approximate, because there isn't
enough energy bin resolution below ~1 eV and because of the way the
the bottom bin is treated.

He and H are available from empirical and/or numerical models and can be added
easily, ionization rates and charge exchange chemistry could then be
incorporated.  I anticipate that they would be treated as minor species, i.e.,
neglecting contributions to absorption and photoelectron generation, as is
currently done with N and NO.

Add cusp and drizzle fluxes from TGCMs:
These can be specified on TGCM secondary histories, and are included in
readtgcm; just need a simple interface to generate the appropriate energetic
electrons into ETRANS.

Re-write EXSECT for table-based cross-section input:
This isn't really a low priority, in fact it might be the highest priority,
especially because some people have complained about how inflexible the 
heritage cross section method is.  At the same time, I need to re-evaluate
cross sections, and break out the lumped states, espcially the N2 singlets
because of the importance of LBH, but also the N2 triplets and O2
Hertzbergs.  The reason this is at the bottom of the list is that it is a lot of
work, in part because the secondary electron production is bound up in the
analytic ionization cross-section method, and also because EXSECT has some of
the oldest and messiest of all the GLOW code.