Annex 2.1 Protocol for model simulations based on IPCC intercomparison.
FIXED GASES based on 1998 values: CH4 and N2O
CH4: global mean
= 1745 ppbv, assume NH = 1790 and SH = 1700 ppbv. Choose a reasonable stratospheric
profile based on your model.
N2O: global mean = 314 ppbv (should not impact these calculations)
CO2: global mean = 365 ppmv (should not impact these calculations)
SURFACE DEPOSITION: O3
We are interested in in the modeled O3 distribution, and thus it is recommended that all participants use as-similar-as-possible surface "deposition velocities" for this species:
For O3: Land (0.60 cm/s), Sea (0.00 cm/s), Poleward of 60 dgrees (0.00 cm/s)
Other wet and dry deposition rates up to individual participants.
EMISSIONS-BASED GASES: NOx, CO, NMHC
The year-2000 emissions data set considers separate source categories per component based on existing standards or an extrapolation of EDGAR data (1x1 degree inventory). The interpolation to model grid is the responsibility of the participant.
Summary of Year-2000 emissions of NO [Tg N]:
|Industry = fossil + bio fuels (~30 + 1.8)||31.8|
|BB = savannah + ag-waste burning / deforest (3.2 + 1.2 + 2.7)||7.1|
|Aircraft ANCAT 2000||0.6|
Fossil fuel/industrial and biofuel emissions (IND). These emissions are provides as aggregated 1x1 gridded emissions. Assume no seasonal dependence.
Aircraft emissions (AV). ANCAT emissions updated for 2000 conditions.
Biomass burning emissions (BB). Use the model emissions for seasonal and geographical distributions and normalize to the global-annual value.
Soil emissions. Use the geographical and temporal distribution of soil emissions specified by Yienger and Levy (1995) of 5.5 Tg N for 1990; or use your model emissions scaled to 5.5 Tg(N)/yr.
Lightning. Scale your model lightning NOx to 5 Tg(N)/yr.
Stratospheric influx. Models should use their own current method.
Summary of Year-2000 emissions of CO [Tg CO]:
|Industry = fossil fuel / domestic burning||650|
|BB = deforestation / savannah burning / waste burning||700|
|BIO = vegetation (150) + oceans (50)||200|
(If the model do not account for NMHC use the
following CO production)
|Terpene oxidation||included in BIO|
|Biomass burning NMHC oxidation||30|
|TOTAL (including model derived emissions)||2675|
Fossil fuel/domestic burning (IND). Gridded 1x1 industrial source supplied. Assume no seasonality.
Vegetation & ocean (BIO). Gridded 1x1 emissions are supplied. The vegetation 150 Tg CO includes potential terpene source. The oceanic 50 Tg CO is consistent with Bates.
Deforestation/savannah burning/waste burning (BB). Model emissions scaled to global-annual value.
The other components of the CO budget need to be calculated internally to each model. We include some approximate numbers above for reference. If possible, please try to keep your total CO sources close to this approximate budget. A CO yield of 35 % C/C per molecule of hydrocarbon oxidized is assumed for isoprene and industrial/biomass-burning NMHC in these estimates.
If NMHC are considered in the simulation:
Summary of Year-2000 emissions of NMHC [Tg
|Industry = fossil fuel / domestic burning||161|
|BB = deforestation / savannah + waste burn||34|
Fossil fuel/domestic burning (IND). A 1x1 EDGAR-based set of gridded emissions is provided along with NMHC breakdowns into 24 compounds/classes. It includes sectors such as power generation, industry, fossil fuel production.
Deforestation / savannah burn / waste burning (BB). A 1x1 EDGAR-based set of gridded emissions is provided along with NMHC breakdowns into 24 compounds/classes. Apply same seasonality as for biomass burning NOx.
Isoprene. Total is based on Hauglustaine & Brasseur work. For gridded emissions use the GEIA data of Guenther et al (1995) scaled to 220 Tg C (i.e., by 43.7%).
Terpenes. Use the GEIA recommendations.
Acetone. Rough estimate is based on Singh et al. Models should use their own distributions if this source can be included.
It is impossible to specify a consistent stratosphere
for the range of 3-D models involved in tropospheric studies, and thus
we ask all participants to use their 'best, current stratosphere' and to
keep this unchanged for ALL simulations.
The diagnostics for these simulations are:
Change in CO, O3, and NOx mixing ratio for January and July conditions (surface, 200 mb, and zonal-mean cross sections).
Change in ozone tropospheric column (DU) as a function of season.
Change in globally averaged OH concentration, methane lifetime, ozone burden as a function of the season.
Annex 2.2 CTM descriptions
INCA (Interactions with Chemistry and Aerosols) is an new emission/chemistry model coupled to the LMDz (Laboratoire de Météorologie Dynamique) general circulation model. LMDz-INCA accounts for emissions, transport (resolved and sub-grid scale), photochemical transformations, and scavenging (dry deposition and washout) of chemical tracers interactively in the GCM. Several versions of the INCA model are currently used depending on the envisaged applications. The standard model resolution is 3.75x2.5 degrees with 19 sigma-p hybrid levels. The GCM also offers the possibility to zoom over specific regions, reaching typical horizontal resolutions of 50x50 km2. The numerics used to solve the time evolution of chemical species is based on a pre-processor. The model can be run in a nudged mode, relaxing to ECMWF winds and temperature. An off-line version of the GCM has also been developed in order to minimize the required computing time for transport simulations. This model is still under development and constitutes the atmospheric component of the IPSL coupled atmosphere-ocean-biosphere model.
Four versions of INCA are currently being used
for gas phase chemistry.
INCA.0. This version is used for the transport of "inert" tracers and does not account for chemical transformations. The typical version of INCA.0 accounts for Rn222, Pb210, SF6 and CO2. This version is specifically designated for the transport of radiactive tracers, and CO2.
INCA.1. Version 1 is designed to simulate the time evolution of tracers using fixed oxidant (i.e., OH, O3, NO3, H2O2) distributions. The three-dimensional distributions of oxidants are prescribed at each time-step according to monthly mean distributions pre-calculated with a more complete version of the model including chemistry. The evolution of long-lived greenhouse gases or chlorine/bromine reservoir species like CH4, N2O, CH3CCl3, CFCs, and HCFCs/HFCs can be calculated with INCA.1. By default this version includes 22 tracers and 39 photochemical reactions.
INCA.2. In this version of INCA, a methane oxidation scheme has been implemented in order to calculate interactively the distribution of tropospheric ozone and OH. This version includes the emissions and chemistry of CH4, CO, and NOx. A standard version of INCA.2 includes 43 tracers and about 100 photochemical reactions.
INCA.3. This more complete version of the model accounts for NMHCs. The oxidation schemes of C2H6, C3H8, C2H4, C3H6, C2H2, isoprene, and terpenes (as alpha-pinene) are considered. All other heavier than C4 hydrocarbons are represented as six generic species (ALKANE, ALKENE, AROMATIC, C2H5OH, MEK, and C2H2). A standard version of INCA.3 includes 99 tracers and about 300 photochemical reactions.
Photolysis rates in INCA are determined based on a pre-calculated multiple entry look-up table. This table is generated using version 4.01 of the Tropospheric Ultraviolet and Visible (TUV) model. The pseudo-spherical discrete ordinates method has been used. The photorates are tabulated for 8 solar zenith angles, 7 ozone columns, 4 surface albedos, 3 temperatures at 500mb, and 2 temperatures at 200 mb. The j values are then multi-interpolated on-line depending on local conditions prevailing in the gridcell and corrected for the effect of cloud cover and optical depth as calculated by the GCM.
Surface emissions of precursors and in-situ aircraft emissions are prepared based on emission inventories using the INCAsflx pre-processor. This processor collects the various emissions on different resolutions, interpolates on the GCM horizontal grid, allows for re-scaling of global inventories if needed, and generates a netCDF input file for INCA with monthly mean emissions.
Lightning NOx emissions are calculated in the GCM based on empirical parameterizations. At each time step these emissions are recalculated interactively and show a strong seasonal and diurnal cycles (Jourdain and Hauglustaine, 2001).
Jourdain, L., and D. A. Hauglustaine, 2001 : The global distribution of lightning NOx simulated on-line in a general circulation model, Physics and Chemistry of the Earth, in press.
The OSLO-CTM2 is a global 3-dimensional chemical transport model that uses pre-calculated fields of winds and other physical parameters to simulate the chemical turnover and distribution of chemical species in the troposphere (Sundet, 1997). The meteorological input data for the model have been generated specifically for this model by running a series of 36 hours forecasts, with the ISF model at the ECMWF at T63 resolution for the year 1996. A new forecast is started every 24 hours from the analysis, allowing 12 hours of spin-up. An extensive set of data is sampled every three hours, including convective mass fluxes, which is not a part of the standard ECMWF archives. Also the temporal resolution (3 hours) is better than the standard ECMWF archives which uses 6 hours. The CTM can be run variable resolution up to T63 (1.87° x 1.87°), however, to limit the amount of CPU-time needed, in this study a horizontal resolution of T21 (5.6° x 5.6°) is used. The vertical resolution is also determined by the input data and the current model version includes 19 levels from the surface up to 10 hPa.
The advection of chemical species is calculated
by the Prather scheme, a second order moment method (Prather, 1986). Convection
is based on the Tiedtke (1987) mass flux scheme, where vertical transport
of species is determined by the surplus or deficit of mass flux in a column.
A comprehensive chemical scheme, including NMHC chemistry, is used. It
includes 55 chemical compounds and 120 gas phase reactions in order to
describe the photochemistry of the troposphere (Berntsen and Isaksen, 1997;
Berntsen and Isaksen, 1999). The scheme is solved using the quasi-steady
state approximation (QSSA). Photodissociation rates are calculated on-line,
following the approach described in Wild et al. (2000). NOx emissions from
lightning are coupled on-line to the convection in the model using the
parameterisation proposed by Price and Rind (1993) and the procedure given
by Berntsen and Isaksen (1999). Mixing in the planetary boundary layer
is treated according to the Holtslag K-profile scheme
(Holtslag et al., 1990). Influence of stratospheric ozone is estimated using a synthetic ozone approach where the ozone flux from the stratosphere is prescribed, but the model transport generates an ozone distribution that varies with time and space.
Berntsen, T.K. and I.S.A. Isaksen, 1997: A global 3D chemical transport model for the troposphere: Model description and CO and O3 results. J. Geophys. Res. 102, 21239-21280.
Berntsen, T. and I.S.A. Isaksen, 1999: Effects of lightning and convection on changes in tropospheric ozone due to NOx emissions from aircraft. Tellus 51B, 766-788.
Holtslag, A.A.M., E.I.F. DeBruijn and H.-L Pan, 1990: A high resolution air mass transformation model for short-range weather forecasting. Mon. Wea. Rev. 118, 1561-1575.
Sundet, J., 1997 Model studies with a 3-D global CTM using ECMWF data. PhD thesis, Dep. of Geophysics, Univ. of Oslo, Norway.
Prather, M.J., 1986: Numerical advection by conservation of second order moments. J. Geophys. Res. 91, 6671-6681.
Price, C. and D. Rind, 1993: What determines the cloud to ground fractions in thunderstorms. J. Geophys. Res. 97, 6579-6613.
Tiedke, M., 1989; A comprehensive mass flux scheme for cumulus parameterization in large scale models. Mon. Wea. Rev. 117, 1779-1800.
Wild, O., X. Zhu and M.J. Prather, 2001:Fast-J, Accuarte simulation of in- and below cloud photolysis in global chemical models. J. Atm. Chem., in press.