What

1
Whats New in MCIP2?

• Tanya L. Otte
• Atmospheric Modeling Division
• U.S. EPA ORD National Exposure Research
  Laboratory
• Research Triangle Park, NC
• On assignment from NOAA Air Resources
  Laboratory.

Disclaimer The research presented here was
performed under the Memorandum of Understanding
between the U.S. Environmental Protection Agency
(EPA) and the U.S. Department of Commerces
National Oceanic and Atmospheric Administration
(NOAA) and under agreement number DW13921548.
Although it has been reviewed by EPA and NOAA and
approved for publication, it does not necessarily
reflect their views or policies.
2
New Release MCIPv2.3

• Passed on to CMAS for release.
• Contains several minor modifications.
• Includes community input.
• Should upgrade MCIP with new CMAQ.
• Tested on IBM and Linux (PG Compiler)
• Details in CHANGES file and in preprint.
• Most notable changes discussed here.

3
(1) Added Graupel to Output

• Accommodates advanced mixed-phase schemes in MM5
  Reisner II, GSFC, Schultz
• If available, QG is passed through.
• Requires new CMAQ to use in chemistry.
• Omitting graupel (if in MM5 output) can be
  significant.

4
Graupel vs. Microphysical Species
5
(2) P-X VEGFRC Work-Around

• NCAR bug in MM5v3.5 fix in MM5v3.6.2
• In solve.F, VEGFRC updated every 24 h after init
  to unrealistic value for P-X
• Using bad P-X VEGFRC leads to negative dry
  deposition velocities one hour per day
• Work-around uses 23rd h values for 24th h

6
Example of Bad VEGFRC in P-X
7
(3) Full Jacobian to Output

• Fields initially removed in MCIPv2.2.
• Fields now needed for correction to CMAQ vertical
  diffusion algorithm.
• Adds JACOBF (levels 1 NLAYS) to METCRO3D.
• Adds JACOBS (level 0 ground) to METCRO2D.

8
(4) P-X Variables for BEIS 3.12

• Fields initially removed in MCIPv2.2.
• Fields used, if available, for BEIS 3.12.
• Restored SOIT1, SOIT2, SOIM1, SOIM2, and SLTYP to
  METCRO2D, if P-X in MM5.

9
(5) GRIDDESC Formatting

• GRIDDESC from MCIP2 had minor error.
• Added a line containing to top of output.
• Also attempted to generalize number formats for
  improved portability.
• On some machines, numbers with E format in
  FORTRAN appear in scientific notation, which
  cannot be read successfully.
• Can edit old GRIDDESC files manually, if needed.

10
(6) Restored GRIDBDY2D

• File initially removed in MCIPv2.2.
• Users initializing CMAQ from other models (e.g.,
  global apps) need lat/lon coords of lateral
  boundaries.
• GRIDBDY2D contains lat/lon land use.
• File is very small (10s of KB).

11
(7) Error-Checking on Layers

• Added in MCIPv2.2 to prevent users from defining
  top and bottom layers thinner than in MM5.
• Error-checking not robust for same as MM5, due to
  real-number round-off.
• Introduced robust error-checking to avoid
  real-number comparisons.

12
(8) 10-m Winds

• Users interested in meteorology verification from
  MCIP output.
• Added 10-m wind speed and direction for
  comparison with surface observations.
• Diagnosed (NOT passed through).
• Similarity theory for speed.
• Assume no directional shear for direction.
• WSPD10 and WDIR10 in METCRO2D.

13
(9) 2-m Temperature

• Most popular PBL schemes in MM5 now directly
  output 2-m temperature.
• New field TEMP2 in METCRO2D.
• Has not been tested in SMOKE.
• Should be used in place of both TEMP1P5 and
  TEMP10. Do not use with TEMP10.

14
TEMP2 vs. TEMP1P5
15
(10) Optional GRID Files

• Newer emphasis on annual simulations.
• GRID files are static in time.
• Namelist logical (makegrid) via script variable
  MKGRID.
• makegrid .F. shuts off all GRID files.
• May need to create files seasonally for
  non-hydrostatic base state in MM5.

16
(11) Explicit Geometry Calculations

• Old dot-point latitude, longitude, and map-scale
  factors bilinearly interpolated from
  cross-points.
• Added explicit calculation.
• Slight differences in contravariant winds.
  (O10-7)
• More significant differences expected for large
  domains, and far from projection center.
• Need to modify earth radius in const_mod.F if
  changed in MM5.

17
(12) Eliminate Fields of Missing

• Old Output all microphysics fields and P-X
  fields, regardless of presence in MM5
• Can lead to serious waste of space, especially
  with new QG field (now 5 3D Q)
• Change to code is transparent to user

18
(13) Land Use Updates

• Changed DLUSE in GRIDCRO2D to be USGS-based MM5
  land use, if not rediagnosing PBL and not RADM
  dry dep
• No impact, unless using field outside std code
• Removed 5 unsupported land use options
• Land use can only be passed from MM5
• Fractional land use not directly supported
• LUPROC output not supported (fixed proj)
• Other deleted options were developmental

19
(14) Polar Stereographic/Mercator

• Old I/O API headers for Mercator and polar
  stereographic were incorrect
• Should not have impacted CMAQ processing, but
  caused problems in PAVE
• Could use M3EDHDR to update ?, ?, ?
• New release corrects ?, ?, ? settings

20
(15) Minimum PBL Height (Eta)

• MM5 implementation of M-Y Eta PBL scheme can have
  negative PBL heights
• Can lead to negative WSTAR, sometimes can crash
  CMAQ
• New minimum of top of lowest full layer to
  prevent very small and negative values

21
Summary

• New release, MCIPv2.3
• No serious bug fixes, many minor updates
• Should consider upgrading to MCIPv2.3 because of
  new features
• Will need to upgrade MCIP and CMAQ concurrently

22
Looking Ahead

• MCIP for WRF
• Based on U. Houston IMAQS WCIP
• WRF-EM initially
• MCIP for toxics
• Mercury, chlorine, others
• To be user option in standard release
• Checks of I/O API libraries against code using
  IOPARMS3 (in I/O API version 3)

23
Credits

• Bret Anderson
• Santosh Chandru
• Carlie Coats
• Rafik Djouad
• Robert Gilliam
• Éric Giroux
• Andy Hawkins
• Jinyou Liang
• Rohit Mathur
• Luca Delle Monache
• Jon Pleim
• Todd Plessel

• Maria Prodanova-Boiadjieva
• Pedro Sanhueza
• Donna Schwede
• Jeff Stehr
• Narisara Thongboonchoo
• Brian Timin
• Shaheen Tonse
• Zion Wang
• David Wong
• Aijun Xiu
• Jeff Young
• Alexis Zubrow