Spatial Resolution Impacts on National Meteorological Center Nested Grid Model Simulations

David D. Houghton UCAR Visiting Scientist Program, National Meteorological Center, Washington, D.C.

Search for other papers by David D. Houghton in
Current site
Google Scholar
PubMed
Close
,
Ralph A. Petersen National Meteorological Center, Washington. D.C

Search for other papers by Ralph A. Petersen in
Current site
Google Scholar
PubMed
Close
, and
Richard L. Wobus General Sciences Corp., Laurel, Maryland

Search for other papers by Richard L. Wobus in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Forecasts from different resolution versions of the National Meteorological Center Nested Grid Model (NGM) are compared for two case studies to assess an optimal ratio of model vertical and horizontal resolutions. Four combinations are considered: 1) 16 layers and 80-km horizontal grid over the United States (the operational version of the model), 2) 32 layers and 80-km horizontal grid, 3) 16 layers and 40-km horizontal grid, and 4) 32 layers and 40-km horizontal grid. Resolution impacts are evaluated for a number of weather system components such as extratropical cyclone evolution, baroclinic and frontal zone structure, jet-stream blow, moisture fields, and precipitation.

Resolution impacts for this limited sample are relatively small for synoptic-scale features such as the position of the extratropical cyclone and main jet-stream flows. Larger impacts are noted for smaller-scale horizontal structure and gradients, frontal zone associated circulations and hydrological cycle components. Vertical resolution enhancement effects on the NGM, which already has added resolution near the lower boundary, are less dramatic in the lower troposphere than those for horizontal resolution, but are important for defining upper-level frontal structures and circulations where the NGM's vertical structure is coarser. Conclusions concerning consistency of horizontal and vertical resolution impacts on baroclinic zone structure and spurious noise generation found in earlier studies with simpler models are confirmed and brought into perspective for comprehensive numerical models and operational weather prediction model applications for the two cases discussed. The effects of the improvements in small-scale forecast accuracy, however, are difficult either to generalize due to the limited number of case studies or to assess because of the lack of high-resolution verification information and evaluation techniques.

Abstract

Forecasts from different resolution versions of the National Meteorological Center Nested Grid Model (NGM) are compared for two case studies to assess an optimal ratio of model vertical and horizontal resolutions. Four combinations are considered: 1) 16 layers and 80-km horizontal grid over the United States (the operational version of the model), 2) 32 layers and 80-km horizontal grid, 3) 16 layers and 40-km horizontal grid, and 4) 32 layers and 40-km horizontal grid. Resolution impacts are evaluated for a number of weather system components such as extratropical cyclone evolution, baroclinic and frontal zone structure, jet-stream blow, moisture fields, and precipitation.

Resolution impacts for this limited sample are relatively small for synoptic-scale features such as the position of the extratropical cyclone and main jet-stream flows. Larger impacts are noted for smaller-scale horizontal structure and gradients, frontal zone associated circulations and hydrological cycle components. Vertical resolution enhancement effects on the NGM, which already has added resolution near the lower boundary, are less dramatic in the lower troposphere than those for horizontal resolution, but are important for defining upper-level frontal structures and circulations where the NGM's vertical structure is coarser. Conclusions concerning consistency of horizontal and vertical resolution impacts on baroclinic zone structure and spurious noise generation found in earlier studies with simpler models are confirmed and brought into perspective for comprehensive numerical models and operational weather prediction model applications for the two cases discussed. The effects of the improvements in small-scale forecast accuracy, however, are difficult either to generalize due to the limited number of case studies or to assess because of the lack of high-resolution verification information and evaluation techniques.

Save