Model Simulation Results

SUMMARY: This page contains links to each of the model scenarios simulated for the VISTAS regional haze study. Clicking on the bulleted lists below will bring up detailed descriptions of each scenario along with results and QA plots. January 2002 Results are available now, and other episodes will be available beginning late Sept 2003.

Domain and Scenario Information
Model Sensitivity Run Schedule (Excel file)
Description of Model Performance Metrics (PDF file)
January 2002 Results
July 1999 Results
July 2001 Results
Final Case Results - All 3 Episodes
CAMx Results (July 1999) including Flexinests
Results using CCTM Mass Conservation Patch

VISTAS Modeling Domains

Download plots of domains and vertical layers (PDF file)
CMAQ 36 km domain with nested in the MM5 36 km domain
CMAQ 36 km domain with nested CMAQ 12 km domain
CMAQ 12 km domain
Proposed Layer Averaging for 18L or 19L CMAQ (excel spreadsheet)

VISTAS Simulation Periods

MM5 Sensitivity Cases & Simulation periods (pdf)
CMAQ Simulation periods (DOC)

Results of Model Performance Evaluation

Model vs. Ambient Species Mapping

CMAQ	CMAQ-SOAmods	CMAQ-AIM	CAMx
IMPROVE	IMPROVE	IMPROVE	IMPROVE
CASTNET	same as CMAQ	CASTNet	CASTNet
STN	STN	STN	STN
NADP	same as CMAQ	NADP	NADP
SEARCH daily	SEARCH daily	SEARCH daily	SEARCH daily
SEARCH hourly	SEARCH hourly	SEARCH hourly	SEARCH hourly

July, 1999 Model Results

CMAQ results

CAMx results for July 1999

CAMx Grid Resolution Case Definitions:

pxacm.baseB.cb4.001: 36km Simulation with no 12km nests
pxacm.baseB.cb4.001.e12m12: 36/12 nested simulation with 12km emissions and met.
pxacm.baseB.cb4.001.e12m36: 36/12 nested simulation with 12km emissions and flexinested met.
pxacm.baseB.cb4.001.e36m12: 36/12 nested simulation with flexinested emissions and 12km met.
pxacm.baseB.cb4.001.e36m36: 36/12 nested simulation with flexinested emissions and flexinested met

CAMx Performance Evaluation Results:

CAMx Flexinest: e36m12 versus 12k CAMx e36m36)

January, 2002 Model Results

Tabulated Model Performance Metrics:

CB4, CB4-2002 and SAPRC99 Comparisons

Summary Evaluation Results for 36 km Sensitivity Runs:

Summary Evaluation Results for 36km/12km vs 12 km Simulations:

Pont and click maps for site specific model evaluation results:

36 km Model and IMPROVE data

Evaluation of Spinup Period: Sensitivity case with no spinup (i.e., clean initial conditions) minus base case with 7 day spinup. Differences should trend toward zero as the simulation progresses. Results suggest that a 6 or 7 day spinup period is sufficient for the January 2002 VISTAS episode.

July, 2001 Model Results

Evaluation of Mass Conservation Patch in CMAQ

While implementing DDM in the CMAQ aerosol code Georgia Tech (Prof. Ted Russell's group) found a possible error in mass conservation in the aerosol solver of CMAQ. We have tested a "mass conservation patch" developed by Georgia Tech and modified by Jim Boylan for CB4 in CMAQ. The plots below show the difference between CMAQ with the mass conservation patch compared to a base case simulation. Grey colors in the plots indicate areas with small or no change. Yellows and reds incidate increase in mass, and blues indicate areas with reductions in mass after applying the patch. (We thank Professor Russell for providing a copy of the mass patch code.)

Two changes are noted: there are apparently random, transient variaitons in aerosol species which might be due to numerical error in the solver. There transients are as large as several ug/m3.

There are also smaller changes that occur throughout a subregion and persist for several hours and are likely to be a result of the mass conservation problem. The transient noise occurs for most aerosol species while the more systematic errors are significant only for nitrate which shows changes of approximately 0.5 ug/m3:

Final Case Model Results - All 3 Episodes

CMAQ results

VISTAS Phase I Regional Haze Modeling

Supported by: Visibility Improvement State & Tribal Association of the Southeast