NCEP Operational Air Quality Forecast Change Log


AQF Operational Run Forecasts


Date

Model Run

Change
May 13, 2008      12Z Run
  • Point, area and mobile emissions will be upgraded based upon recent EPA National Emissions Inventory (NEI,  2005) and then projected for the current year.  EGU sources will use 2006 CEM data projected for 2008.  
  • For mobile sources, the  EPA Office of  Transportation and Air Quality (OTAQ ) estimates will be used in addition to 2005 NEI v1 emission data sets.  Use of OTAQ on-road emission estimates is a departure from the temperature dependent regression approach used in previous years.
March 31, 2008 12Z Run   Changes to the NAM WRF-NMM weather forecast model implemented.  Impact on AQ specific weather and ozone predictions are shown here.
  • Expanded the Computational domain by 18%
  • Gravity wave mountain drag parameterization implemented
  • Modified horizontal advection for improved mass conservation
  • Improved Surface Longwave radiation calculation
  • Upgraded GSI data assimilation with NMM based backgroun error covariances
  • Improved assimilation of AIRS radiances, GOES single field of view radiances, MODIS and surface Mesonet winds
  • Use of NAM forecast precip to drive soil moisture calculations outside CONUS where precip observations are lacking
  • Increased terrain smoothing
September 18, 2007      12Z run   Experimental CONUS CMAQ gas-phase only forecast system moved to NCO Operational slot:
  • Expanded domain (5X) to cover Continental U.S.
  • Common WRF-NMM hybrid sigma-P vertical coordinate
  • NAM clear sky radiation prediction used to scale CMAQ photolysis
  • Asymmetric Convective Model used to drive CMAQ moist convection mixing
  • Asymmetric Convective Model-2 used to drive CMAQ PBL mixing
  • Constant, static Ozone Lateral Boundary conditions for all CMAQ levels (turned off use of GFS ozone at CMAQ Lateral boundary top)
  • Updates to CMAQ V4.6 with bug corrections to deposition, optimized advection scheme and plume rise
  • Special California NEI 2002 Non-Road mobile emissions

Eastern U.S. Operational Run Change Log (2005-Sept. 2007)
July 3, 2007     12Z Run CMAQ Point source emission files updated with 2007 projections.
May 16, 2007      00Z Run CMAQ static  area emissions updated with 2007 projections.
September 5, 2006     12Z Run CMAQ now driven by updates to NAM model (additional NAM-Y changes implemented) that include increase to horizontal diffusion and corrections to SST initial conditions.
       August 22, 2006     18Z Run CRISIS Fix:  PREMAQ array bounds corrections implemented. Very small improvement in ozone predictions.
August 15, 2006     12 Z Run CMAQ now driven by NAM-Y meteorology  which was implemented into NAM slot.  NAM-Y had corrections to vertical and horizontal diffusion and lake SST initialization which resulted in more realistic warmer,dryer boundary layers that increase production of ozone.  The NAM-Y impact on CMAQ forecasts can be seen from Tsidulko analysis.
       August 01, 2006     12Z Run CRISIS Fix:  Some Point  Source Emissions were still configured for 2005 .  These emissions files were updated for 2006 projections.
July 28, 2006     18Z Run CRISIS FIX:  The WRF post code  interpolates WRF outputs from the WRF hybrid vertical  coordinate systems to CMAQ sigma levels and  uses simple interpolation for all  elements except temperature.  For temperature,  to be consistent  with hydrostatic balance, it first computes geopotential heights  for upper and lower boundaries (using the hydrostatic equation) and then computes temperature at desired sigma level using the hydrostatic equation again. This algorithm doesn't work in the first model layer, and in this case temperature is extrapolated from the upper layers.  In case of an inversion, since the algorithm involves several upper layers, the bottom sigma temperature happens to be higher then the first model layer temperature. To fix the problem for   loose coupling,. temperature is simply interpolated  (just like all other elements) for the lowest model layer, while using hydrostatically derived temperature for all the upper layers. Impact of these changes can be seen here.
June 20, 2006    12Z Run Operational CMAQ Eastern U.S. ozone run driven by operational NAM-WRF.
12 km CMAQ gas w/ reduced CB4 gas  phase chemistry forecasts run twice/day(06 and 12Z) to 48 hours.  WRF outputs interpolated to CMAQ sigma verfical coordinate for loose vertical coupling.
  • 6 hour CMAQ cycling for initial conditions
  • Lateral Boundaries are static except at CMAQ top layer where GFS ozone is specified
  • Above-cloud downward convective mixing turned off
  •  Point/Area Source Emissions  for 2006
April 4, 2006
   18 Z cycle
  Eastern U.S. (3X) run initiated with NAM-WRF parallel run.  WRF outputs interpolated to CMAQ sigma verfical coordinate for loose vertical coupling for 6 hour cycling.  Operational CMAQ NAM-Eta and NAM-WRF 3X domain runs are viewed here 

   August 31, 2005
   00 Z cycle
  NE operational forecasts replaced with Eastern U.S. (3X)  Experimental runs.
Click here for more info on Eastern U.S. configuration
The 3X operational forecasts were the same as the NE U.S. runs except:
  • Larger CMAQ domain: 268x259x22
  • Lateral Boundaries are static except at CMAQ top layer where GFS ozone is specified
  • Above-cloud downward convective mixing turned off
N.E. U.S. Operational Run Change Log (2004-2005)
   June 10, 2005
   00 Z Run
  A bug was corrected in the interpolation  of ozone from the new GFS T382 to the CMAQ 12 km grid.  This has helped reduce ozone overpredictions found in runs from  May 31-June 10, 2005.
  May 31, 2005
   12 Z GFS Upgrade
 The NCEP Global Forecast System (GFS) was upgraded to T382 (~35 km).
In addition, the following implementations were changed:
  • A new land/sea ice physics
  • Reduced vertical diffusion
  • Enhanced mountain blocking
  • assimilation of AIRS satellite data
GFS ozone is used to prescribe the CMAQ lateral Boundary Conditions above 6 km.
  May 3, 2005
   12 Z NAM-12 CMAQ Operational (1X domain)
   Upgraded NAM-12.   Upgraded CMAQ  :
   PREMAQ updated to use NAM-12 1 km landuse data
    Minor CMAQ updates to version 4.4
    2002 Point/Area Source Emissions projected for 2005
    Updated Mobile 6 source emissions
    GFS ozone for CMAQ LBCs above 6 km
 
2004
  September 15, 2004
    00 Z
  N.E. U.S. Run declared operational

May 1, 2004

12 Z Eta-12

CMAQ Experimental

Implemented Experimental System

2002  Point/Area Source Emissions Projected for 2004

Updated Mobile 6 mobile source emissions

6 hour initial condition cycling

Use of GFS ozone predictions for CMAQ upper Lateral Boundary Conditions

March 2004 Eta-12 implementation

48 hour forecasts for both 06 and 12 UTC runs

May 5, 2004

12 Z Experimental

Correction to 6 hour Cycling system. Now turned on.

May 17, 2004

12 Z Experimental

Correction to GFS ozone LBC ingest.  Now working.

May 26, 2004

12z Experimental

Sfc ozone Grib files now available on /com/aqm/prod

July 20, 2004  

06z   

1. Limited  GFS LBC  ozone ingest to above 6 km AGL (CMAQ sigma lvl 18)
2. Specification of  a minimum mixing coefficiant (Kz) in CMAQ pbl routines
   *  Allows for lower Kz in rural areas than urban at night :  reduces  nightime O3 overprediction in rural  areas
   * Prevents precursor concentrations in urban areas becoming too large at night and
     thereby reduces overtitration in urban areas (more mixing in urban areas at nght)


 

NCEP FVS verification system

 

Date

System Component

Change
June 1, 2005
  Grid2obs VSDB records
  VSDB exceedence records (FHO) now being produced for daily maximum 1 and 8 hour ozone predictions.
March 25, 2005
 Grid2obs VSDB
  VSDB file generation turned on for 2005 Ozone season:
Backward averaged convention used for both Airnow bufr and cmaq forecasts
L1L2 and FHO statistics now computed (except daily maximums)
 2004 FVS VSDB File changes

May 31, 2004

Grid2obs VSDB records  

VSDB records (L1,L2) begun for NE domain on 146 grid using forward averaged BUFR observations
June 9-June 20
 BUFR generation
 BUFR software changed at EPA to produce backward averages. 
VSDB records not produced
June 21, 2004
 BUFR Generation
 BUFR softward changed at EPA to produce foreward averages for NCO data tanks.
Bufr tanks and VSDB records for June 9-20 recreated with foreward avging
July 12, 2004
 VSDB records
 VSDB records are now computed for East US domain on Grid 146, 142
Subdomain records computed for all grids (eg: NEC, SEC,APL...)

July 22, 2004     

3X VSDB Records      

VSDB records computed for developmental 3x domain (Grid 142) over full domain, and subregions, and over 1 x domain
August 1, 2004 VSDB records VSDB subdomains updated to new FVS regions.  MDW, GMC and  LMV now
replace NMW, SMW (midwest) subregions.
August 1, 2004
 FHO VSDB records
 Began creating  1h & 8 h VSDB FHO records for events :
1hr ozone > 20, 40, 65, 85, 105, 125 ppb
8h avg ozone > 20, 40, 65, 85 ppb
 


 



Summer 2003 North East Domain Forecasts

 

Date

Model Run

Change

June 25, 2003

12 Z Eta-12 – CMAQ Experimental

Implemented Experimental System with 24 hour ozone initialization cycling.

Used Spring 2003 Eta-12 implementation

July 23-24

12 Z Experimental

CMAQ Archives lost

July 31- August 4

12 Z Experimental

Cold Start Runs…No cycling

August 5

12 Z NCO run

 

Transfer to NCO

Cold Start…no cycling

August 5

06Z Experimental

06Z implemented

6 hour cycling (6z only)

August 15

06 Z Experimental

Cold Start

August 20

06 Z NCO

12 Z NCO

Transfer of 6 hour cycling to NCO for 6 and 12 Z runs.

Cold Start

September 9

06 Z, 12 Z NCO

Eta-Post:  landuse, soiltype variable packing fixed.  Previously landuse values incorrectly packed to equal 0 (Ocean) or 10.