THE HYSPLIT4 GRS CHEMISTRY MODULE
Last Revised: 29 August 2003


ABSTRACT

	The Generic Reaction Set (GRS) model by Azzi (1993,1994) is 
a semi-empirical approach, but it is written in terms of a non-steady
state chemical mechanism using the reaction rate equations for the 
seven species. The GRS is not limited by the photo-stationary 
steady-state assumption of the IER model. Unlike the IER model, the 
implementation of the GRS model within the framework of a 
meteorological transport and dispersion model is non-linear and 
requires adding the precursor contributions from all sources before 
integration. The GRS model has been evaluated by Venkatram (1994) and 
Tonnesen and Jeffries (1994). The core subroutines used in Hysplit to
compute ozone formation using the GRS were developed by Martin Cope, 
Victoria EPA, and later modified by G.D. Hess, BMRC, BoM, Melbourne, 
Australia.  The key to properly linking a Lagrangian model with 
Eulerian chemistry is the conversion of particle mass to concentration
and then back to particle mass after the chemistry has been applied.
The method of Stein (2000) is used in the Hysplit-GRS module.

Azzi, M.; Johnson, G.M. Airtrak: new developments,
Clean Air 1993, 27, 191-193.

Azzi, M.; Johnson, G.M.; Cope, M. An introduction to the generic 
reaction set photochemical smog mechanism, In Proceedings of the 
International Conference of Clean Air Soc. of Aust. and N.Z. 
1992, 2, 451-462.

Stein, A. Incorporation of detailed chemistry into a three-dimensional
Lagrangian-Eulerian hybrid model, Atm. Environ. 34: 4361-4372.

Tonnesen, S.; Jeffries, H.E. Inhibition of odd oxygen production 
in the carbon bond four and generic reaction set mechanisms,
Atmos. Environ. 1994, 28, 1339-1349.

Venkatram, A.; Karamchandani, P.; Pai, P.; Goldstein, R.
The development and application of a simplified ozone modeling 
system (SOMS), Atmos. Environ. 1994, 28, 3665-3678.

 

PC VERSION

	The model is run by executing the batch file "run_model",
which runs the executable (hysp_grs.exe) in the \Exec directory.
When the simulation completes, the output file "cavrg" will have been
created containing the hourly average concentration fields for all
species.  Executing "run_plot" will create an extract file (xdump)
with just the hourly ozone concentrations and then a Postscript file
"concplot.ps" with maps of hourly concentrations. The output files 
are compatible with all standard Hysplit plotting programs. A 
Postscript file viewer or printer is required. The model is 
pre-configured to create output for a three-day test period, June 
24-27, 2003, over the Northeast states, using EDAS archive meteorology.
All required files are provided in the \Fixed directory. The 
configuration (CFG) and CONTROL files have been pre-set to run this 
example case. Other simulations can be run if the appropriate 
meteorological data have been downloaded and the control and 
configuration files modified accordingly.  A familiarity with the 
regular PC version of Hysplit is strongly suggested prior to trying 
run this version of the model.  More detail is provided below about 
the format of some of the input configuration files as well as the 
emissions inventory.  The use of different inventories may require 
reformatting. 


CONTROL file

	The CONTROL file cannot be edited through the standard HYSPLIT
GUI. Manual editing is required. Point source emissions are not 
supported. A gridded emission inventory is required. The emission 
inputs in the CONTROL file have slightly different meanings when using
a gridded inventory.  The emission duration is the same, but the 
emission rate is a multiplier to the gridded emission value.  Normally
it would be set to 1.0. Gridded emissions can produce many particles 
each time step. To minimize the number of particles, one can set the 
emission duration in the control file to a very small value (0.1 hrs)
and set the emission rate multiplyer in the control to 10, thus the 
total emissions are conserved.

	The top of the snapshot concentration grid should match the top
of the model domain to avoid the transport of particles that are not 
chemically transformed. All concentration grids should cover the same
spatial domain. Grid #1 and Grid #3 (concentrations after GRS 
chemistry) should always be defined as the snapshot grids. The 
reactions are determined by the species concentrations on the 
snapshot grid.

	The emission domain is determined by the source lat/lon limits 
set in the CONTROL file.  Note that the source domain must lie entirely within 
the meteorological grid limits, otherwise the results may be un-
predictable. The release height represents the height from the ground 
through which pollutants will be initially distributed. 

Seven species must be defined for GRS chemistry.  Only two are defined
in the gridded emissions file.    

7
NO2
10.0
0.1
00 00 00 00 00
NO#
10.0
0.1
00 00 00 00 00
O3
0.0
0.0
00 00 00 00 00
SGN   
0.0
0.0
00 00 00 00 00
SNGN 
0.0        
0.0 
00 00 00 00 00
VOC  
10.0        
0.1   
00 00 00 00 00 
ISOP 
10.0        
0.1   
00 00 00 00 00 


The model requires a gridded emissions inventory for NMHC 
and NOx, in units of kg/hour.  Model output units will be ppm. 
The following abbreviations are used to identify species:

#  ID	Mol WT	

1: NO2	46	- NO2 (initial value from inventory [10% of NOx])
2: NO#	30	- NO (intial value from inventory [90% of NOx]) 
3: O3	48	- ozone
4: SGN	50	- stable gasous nitrate
5: SNGN	50	- stable non-gaseous nitrate 
6: VOC 	13.7	- initial hydrocarbon (from inventory)               
7: ISOP	13.6	- initial isoprene (computed)

Note that the simulation is configured for NO deposition. A sample
file (control.dep) is provided with default settings for deposition.
The results are very sensitive to the deposition settings and 
substantially lower concentrations are predicted when the default
settings (shown in the User's Guide) for NO2, NO, and O3 are used.
Note the negative molecular weight is used to flag the units conversion
from mass/volume to volume/volume.

0.0 0.0 0.0               NO2
0.0 -46.0 0.1 1.6 0.01
0.01 0.0 0.0
0.0
0.0
0.0 0.0 0.0               NO#
0.0 -30. 0.0 1.3 0.003
0.0019 0.0 0.0
0.0
0.0
0.0 0.0 0.0               O3
0.0 -48. 1.0 1.6 0.01
0.013 0.0 0.0



SETUP.CFG NAMELIST FILE

	The standard setup.cfg results in the release of one 
particle at each grid point per time step.  If more particles are
to be released (at substantial computational cost) then the NUMPAR
parameter should to increased to the appropriate multiple of the
initial number of non-zero emission grid points as noted in the 
MESSAGE file. As noted above in the discussion of the control 
file, the model is configured to release all particles in one
time step, each hour (hence qcycle=1.0).  About 350000 particles
will be created over the 3 day test simulation.  Longer duration
simulations will result in particles moving off the domain.

INITD=0 		run in 3D particle mode 
KHMAX=9999 		don't drop any old particles
NUMPAR=500 		one particle per point if < number grid pts
MAXPAR=350000           max number of particles in simulation
MAXDIM=7  		number of species on one particle
QCYCLE=1.0 		new emission cycle each hour
DELT=0.0 		automatic time step computation
ICHEM=4			forces conc grid = meteo grid



EMISSIONS INVENTORY

	The file "emission.txt" contains all the information
that is required to interpred the data in the gridded emission
inventory file. The file's first record contains information 
about the internal grid cell size that is used by the dispersion
model to accumulate the file's emissions. The emission file 
defines the emissions at latititude-longitude points, the values 
at these points are accumulated in an internal grid, the size of 
which is defined on the first record.  This value can be 
arbitrarily changed according to the desired resolution of the 
simulation.  The pollutant puffs are released with an initial 
size comparable to the accumulation cell size.  Because the 
emission file data are remapped to an internal grid, the file 
can consist of emissions data on a regular grid or just a 
collection of individual cells.  The emission rate in the Control
file is used as an additional multiplication factor for the data 
in the emission file.  Also note that previously discussed
particle number restrictions still apply.  The initial number of 
particles are spread out over the duration of the emission and the 
number of grid cells that are defined in the emission domain. 
The format of the emission.txt file is given below:

Record #1
        I4      - Number (n) of pollutant species in file
        I4      - Number of emissions defined for each 24 hour period
        F10.4   - Conversion factor: file units to model units/hour
        2F10.4  - Accumulation cell size (latitude & longitude)

Record #2
        nA4     - Pollutant character identification each pollutant

Record #3
        A       - the /directory/filename of the emission data file


The 1985 NAPAP emissions inventory file was provided by Stu McKeen
(stu@al.noaa.gov) from NOAA's Aeronomy Laboratory. The file 
NOx_VOC_SO2 contains NOx, Volatile Organic Carbon, and SO2 
emissions. The file is an ASCII formatted file od hourly emissions 
for 24472 1/6 deg latitude by 1/4 degree longitude grids between
24.5 and 59.5 degrees north and 52-135 degrees west longitude.

The records were written with the following FORTRAN write statements:
	
	WRITE(10,'(2I4,2F10.4)')II,JJ,XLON,XLAT
	WRITE(10,'(1P,12E10.3)')(NOX(IH),IH=1,24)
	WRITE(10,'(1P,12E10.3)')(VOC(IH),IH=1,24)
        WRITE(10,'(1P,12E10.3)')(SO2(IH),IH=1,24)

II and JJ are indexes used in the programming. XLON (negative is 
west longitude), and XLAT are the longitude and latitude of the 
south-west corner of each grid square. The emissions in British 
tons per hour per grid cell. The indexing of NOX, VOC, and SO2
are:

	 1 = 0000-0100 (GMT) average
 	 2 = 0100-0200 (GMT) average
	....
	24 = 2300-2400 (GMT) average.

Only grids for which nonzero emissions of either NOX, VOC, or SO2
are within the NAPAP inventory are in the file.  All point, mobile
and nonmobile emissions are included, with no distinction in height 
of emissions from point sources.  The NAPAP85 summertime-weekday 
scenario is used. VOC emissions represent all nonmethane hydrocarbons.
That is, CO and CH4 are not included in the emissions.  VOC emissions
are for 80 degrees F.



COMPILATION (for reference only ... source code not available)

	The Hysplit GRS program is a conditional compilation 
variation of the standard model.  The main concentration program, 
source code, hymodelc.F, is compiled with the parameter -DGRS to 
create the executable hysp_grs.

The following subroutines are required in addition to the standard
Hysplit library routines:

ierno2.f - returns NO2 photoloysis rate coefficient from the downward
           shortwave flux and solar angle 

iervoc.f - computes the isoprene emissions at each grid point based
	   upon the shortwave flux and temperature

grscon.f - called at the end of particle advection and goes through
           each concentration grid cell and solves the GRS chemistry

grseqn.f - chemistry integration subroutines (from Cope and Hess)

grsdst.f - routine to map change in concentration of each grid cell
           to change in particle mass (from Stein)

emsgrd.f - in the standard Hysplit library, uncomment the mass to
           volume conversion for ppm output. Conversion  is initiated
           by the specification of molecular weight in the control file

