CMAQv5.1 In-line Calculation of Photolysis Rates

Brief Description

Cloud distribution and their optical properties are more consistent clouds described in meteorological input files for CMAQ simulations.

• In previous versions of CMAQ, a vertical column could have one uniform cloud layer composed of water droplets with a fixed radius. Relative humidity determined properties of the cloud layer.

Aerosol extinction and scattering are revised regarding spectral dependence, refractive indices and new options for how to calculate their optical properties. Minor changes address the two areas.

• Variables are added to diagnostic files to improve interpreting prediction.
• Source code is restructured to improve its transparency and computation performance.

Cloud Updates

Clouds and their condensates have vertical distribution and diversity based predictions from the meteorological model (WRF).

• Clouds have vertically varying densities of condensed water (hydrometeors)
• Multiple types of hydrometeors cause scattering and absorption
  • Liquid Cloud droplets, Ice particles, rain, snowflakes, and graupel
• Scattering and absorption include effects from resolved and sub-grid clouds
• Resolved clouds based on Hydrometer Concentrations predicted by WRF
  • 3D cloud fractions diagnosed from WRF hydrometeor concentrations if not available in meteorological input files
  • Hydrometer sizes computed based routines taken from WRF version 3.5 and Community Atmosphere Model (CAM) 3.0 models
  • Cloud layering accounted by simple power law expression (Voulgarakis et al., 2009)
• Sub-grid Clouds are based on results from acm_ae6 or acm_ae6_mp cloud modules in CMAQ
  • cloud fractions and hydrometeor concentrations dependent on acm convective parameterization scheme

Optical Properties account for a greater number and variability in cloud condensates.

• Cloud Droplets have variable effective radius
  • Diagnosed based on temperature, land use category and snow coverage at the surface
    • Method adapted from WRF RRTMG radiation module
  • Hu and Stamnes (1983) parameterization determine extinction and scattering properties
• Ice particles have variable effective diameter
  • values based on air temperature
    • Method also based on RRTMG radiation module in WRF version 3.5
  • Extinction and scattering properties computed using Fu (1996)
• Rain Droplets affect radiation in and below clouds
  • Optical properties computed with simple expressions from the GSF radiation module in WRF version 3.5 (Chou and Suarez, 1999)
• Snow and Graupel effects are based on equivalent concentration of ice particles
  • Snowflake and graupel effective size determines their ice equivalent concentration
    • Method adapted from the RRTMG radiation module in WRF version 3.5
  • Effective diameters computed based on concentration assuming an exponential size distribution
    • Method adapted from CAM version 3

Aerosol Changes

Refractive Indices now wavelength dependent

• Water values from Segelstein (1981)
• Elemental Carbon (soot values) from three sources
  • Bond (2012), Bond and Bergstrom (2006) and Chang et al. (1990)
• OPAC Data base for remaining species
  • Uses aqueous solute values for sulfate, nitrate, ammonium, and sea salt species
  • Uses dust values for trace metals, organic carbon, and unidentified species

The refractive index values are (from v5.1 PHOT_OPTICS.dat, replaces similar information from v5.0.2 aero_photdata.F):

Effective Wavelength (nm)	Water	Aq. Solute	Dust	Seasalt	Soot (EC)
294.590	1.373 + 4.4522E-09 i	1.530 + 1.0374E-02 i	1.530 + 1.0374E-02 i	1.510 + 2.3237E-06 i	1.850 + 7.1000E-01 i
303.151	1.370 + 3.9350E-09 i	1.530 + 7.8622E-03 i	1.530 + 8.0609E-03 i	1.510 + 1.8973E-06 i	1.850 + 7.1000E-01 i
310.007	1.368 + 3.5394E-09 i	1.530 + 7.3992E-03 i	1.530 + 8.0000E-03 i	1.510 + 1.6643E-06 i	1.850 + 7.1000E-01 i
316.434	1.366 + 3.2873E-09 i	1.530 + 7.0130E-03 i	1.530 + 8.0000E-03 i	1.510 + 1.4486E-06 i	1.850 + 7.1000E-01 i
333.076	1.362 + 2.9125E-09 i	1.530 + 6.0069E-03 i	1.530 + 8.0000E-03 i	1.510 + 8.8650E-07 i	1.850 + 7.1000E-01 i
381.997	1.353 + 1.8931E-09 i	1.530 + 5.0081E-03 i	1.530 + 8.0000E-03 i	1.504 + 1.4358E-07 i	1.850 + 7.1000E-01 i
607.723	1.333 + 5.2523E-08 i	1.528 + 7.0012E-03 i	1.528 + 8.0000E-03 i	1.492 + 1.5223E-06 i	1.855 + 7.1381E-01 i

Optical surrogates are assigned in AERO_DATA.F (aerospc%optic_surr).

New runtime option for Mixing Model used to determine optical properties

• Uniform Sphere (default option; the preceding environment variables set false or undefined)
  • Refractive Index is a volume average over aerosol components
• Core-Shell or Stratified Aerosol Structure
  • Core: elemental (black) carbon
  • Shell: uniform mixture of remaining components
  • To use this option, the run-script should set the CORESHELL_OPTICS environment variable

         setenv CORESHELL_OPTICS T # Core-Shell or Stratified Sphere using Mie Scattering Theory

Runtime options for how to compute extinction and scattering properties

• Core-Shell or Stratified Sphere using Mie Scattering Theory
  • Setting the environment variable CORESHELL_OPTICS to true requires using a unique solution to Mie Scattering Theory
• Fast Optics (default option; the preceding environment variables set false or undefined)
  • Uses approximations to Mie Scattering Theory for uniform sphere
• Mie Scattering Theory for uniform sphere
  • To use this option, the run-script should set the MIE_OPTICS environment variable

         setenv MIE_OPTICS       T # Mie Scattering Theory for uniform sphere

If CORESHELL_OPTICS and MIE_OPTICS equal true and false, respectively, Fast Optics computes optical properties for an aerosol mode when the mode does not contain elemental carbon or when the core has a radius less a one thousandth of the total radius.

Changes to Diagnostic Files

New outputs describe resolved and sub-grid clouds.

• Effective two dimensional fraction and total liquid content over a vertical column as well as total optical depth from clouds.

Three dimensional outputs are supplemented.

• Resolved cloud fraction, actinic flux, optical depth, aerosol optical depth, single scattering albedo and asymmetry factor.

Transmission and reflection coefficients (Liou, 1990) are given at the surface and the top of the atmosphere, respectively, averaged over all wavebands.

• Both direct and diffuse coefficients.
• Values from radiative transfer solutions that do and do not include cloud effects.

Significance and Impact

Preliminary results showed that the bias for ozone increased over the July 2011 while it decreased over January 2011 over the continental US with 12X12 km² grid cells. For aerosol sulfate, the magnitude of model bias decrease over both simulation periods.

Preliminary results showed that model run times decreased by approximately 10% when using revised photolysis rate calculation for the above periods and model domain. The reductions were gained from two types of changes in the source code. Comparisons between character variables were converted to comparisons between integers. Recalculating data was reduced when variables determining them did not change.

Affected files

Modified files

• phot.F
• CSQY_DATA.F
• opphot.F
• PHOT_MOD.F
• all CSQY_DATA files under the photochemical mechanisms

New files

• AERO_PHOTDATA.F calculates aerosol scattering and extinction properties per layer.
• PHOTOLYSIS_ALBEDO.F calculates surface albedo over domain.
• CLOUD_OPTICS.F calculates cloud scattering and extinction properties per layer.
• OMI.dat contains input data describing the total ozone density.
• PHOT_MET_DATA.F contains routines and data that read and calculate meteorological and physical variables.
• twoway_rrtmg_aero_optics.F90 calculates aerosol scattering and extinction properties per layer if Mie scattering or Core-Shell mixing model requested at model run-time.
• PHOT_OPTICS.dat contains input data needed to calculate the solar flux, surface albedo and optical properties of aerosols and clouds.
  • a new environment variable defines the input file in the CMAQ run script as shown below.

setenv  OPTICS_DATA  ${NML}/PHOT_OPTICS.dat

References

Fu (1996), An accurate parameterization of the solar radiative properties of cirrus ice suitable for climate models, J. of Climate, vol. 9, pp. 2058-2082.

Hong et al. (1998), Implementation of Prognostic Cloud Scheme for a Regional Spectral Model. Monhtly Weather Review, vol. 126, 2621-2639.

Hu and Stamnes (1993), An accurate parameterization of the radiative properties of water clouds suitable for use in climate models, J. of Climate, vol. 6, pp. 728-742.

Kristjässon, Edwards, and Mitchell (1999), A new parameterization scheme for the optical properties of ice crystals for use in general circulation models of the atmosphere, Phys. Chem. Earth, B24, 231–236.

Liou, K.N. (1990), An Introduction to Atmospheric Radiation, Academic Press Inc., New York, 181 p.

Randall, D. A., (1995), Parameterizing fractional cloudiness produced by cumulus entrainment. Preprints, Workshop on Cloud Microphysics Parameterizations in Global Atmospheric Circulation Models, Kananaskis, AB, Canada, WMO, 1–16.

Segelstein (1981), The Complex Refractive Index of Water, M.S. Thesis, University of Missouri, Kansas City, MO.

Voulgarakis, A., Savage, N. H., Wild, O., Carver, G. D., Clemitshaw, K. C., and Pyle, J. A. (2009), Upgrading photolysis in the p-TOMCAT CTM: model evaluation and assessment of the role of clouds, Geosci. Model Dev., 2, 59–72.

Wainwright et. al (2014) J. of Appl. Meteo. Climat., vol. 53. 2072-2090.

Contact

William T. Hutzell, National Exposure Research Laboratory, U.S. EPA