HR: 1340h
AN: A23C-1472 [Abstracts]
TI: A New Direct Coupled Regional-scale Meteorology and Chemistry Model
AU: * Li, J
EM: lijing@rias.atm.nacu.edu.tw
AF: The Institute of Atmospheric Physics
National Central University, No.300, Jhongda Rd., Jhongli, 320, Taiwan
AU: Hsu, S
EM: garyhsu@rias.atm.ncu.edu.tw
AF: The Institute of Atmospheric Physics
National Central University, No.300, Jhongda Rd., Jhongli, 320, Taiwan
AU: Liu, T
EM: tliu@cc.ncu.edu.tw
AF: The Institute of Atmospheric Physics
National Central University, No.300, Jhongda Rd., Jhongli, 320, Taiwan
AU: Chiang, C
EM: chiang@rias.atm.ncu.edu.tw
AF: The Institute of Atmospheric Physics
National Central University, No.300, Jhongda Rd., Jhongli, 320, Taiwan
AU: Chang, J
EM: julius@ncu.edu.tw
AF: The Institute of Atmospheric Physics
National Central University, No.300, Jhongda Rd., Jhongli, 320, Taiwan
AB:
WRF/Chem was first developed in the US and generously made available to the international research
community a short time ago. Starting from this, many groups have contributed new components and subroutines
to this model. Based on WRF/Chem, a new online integrated model system named WRF/ChemT was
established in Taiwan. It is significantly different from WRF/Chem in the following important aspects.
For an online model, all chemical species emission must be direct coupled to WRF meteorology. All publicly
available versions of WRF/Chem do not have this fundamental coupling. For these WRF/Chem models all
emission data must first be preprocessed by SMOKE or other emission models driven by MM5 or WRF
meteorologies in offline manner. WRF/ChemT has a self-consistent online emission process. We replaced the
old emission driver with NCU driver, the plume rise of point sources and biogenic VOCs emission are calculated
online. So that meteorology model, emission model and chemistry transport model are coupled directly in
WRF/ChemT.
Cloud impact on actinic flux should be consistent with WRF cloud-aerosol submodel used, not just moisture
parameterization. Photolysis rates in WRF/ChemT are self consistent in every sub modules. New dry deposition
routines were developed including addition of a vertical mixing scheme named the Asymmetrical Convective
Model (ACM) which is used in CMAQ. The advantage of using ACM submodel had been demonstrated in earlier
studies.
Computational inefficiency has been a lingering problem for WRF/Chem. We have worked on this aspect of
WRF/Chem development and by using a new chemical solver and also reorganizing the operator splitting
computational algorithm we have made significant computational speed gain. WRF/chemT is about a factor of 4
faster in the chemistry solver and a factor of 2 faster in chemical species transport. When added together it is
about a factor of 2 faster than WRF/Chem(version 2.1.2), i. e. gas-phase chemistry and meteorology are now
equally fast.
WRF/ChemT was evaluated and applied in regional air quality research in Taiwan. The comparison with
WRF/Chem and selected current applications will be discussed in this report.
DE: 0345 Pollution: urban and regional (0305, 0478, 4251)
DE: 0478 Pollution: urban, regional and global (0345, 4251)
DE: 3307 Boundary layer processes
DE: 3314 Convective processes
DE: 3355 Regional modeling
SC: Atmospheric Sciences [A]
MN: 2007 Fall Meeting