Image Processing Tools for Meteorology and Air Quality Modeling and Preliminary Results
Limei Ran Center for Environmental Modeling for Policy Development Institute for the Environment, UNC-Chapel Hill Chapel Hill, NC 27599-6116 Jonathan Pleim and Robert Gilliam Atmospheric Modeling and Analysis Division USEPA/ORD/NERL, RTP, NC
Meteorology and air quality modeling uses image data, such as land cover and satellite data. Image data are stored in pixel-based format and are often called raster data as well. In order to process those image data needed for meteorology and air quality modeling, modelers often have to know geospatial processing and to use a GIS and image processing software system such as arcGIS. Over the years, we have been developing GIS Shapefile processing tools Spatial Allocator (SA) for emission and air quality modeling. We enhanced the SA with raster processing tools in the 3.5 release last year to process 2001 National Land Cover Database (NLCD) and Moderate Resolution Imaging Spectroradiometer (MODIS) land cover data. Since then, we have developed Geostationary Operational Environmental Satellite (GOES) data re-gridding tools. Currently, we are developing MODIS cloud and OMI satellite aerosol and NO2 re-gridding tools. With those raster tools, modelers can just use a script file to define their domain information and input image data for generating modeling domain grid satellite data. Then, computed domain grid satellite data can be used in data assimilation and model output evaluation. Meanwhile, we are developing next generation biogenic emission land cover database (BELD4) using 2001 30m NLCD, 1km MODIS land use data, Forest Analysis Inventory (FIA), National Agricultural Statistics Service (NASS) crop information, and other data. We have incorporated 2001 NLCD and MODIS land use data in the Weather Research and Forecast (WRF) and Community Multiscale Air Quality (CMAQ) modeling for the CONUS and east US 12km domains. Preliminary WRF results show slight improvement and CMAQ runs show largest difference in the bi-directional NH3 surface flux. We believe that these new land cover data should have more effects on both meteorological and air quality model simulations at higher resolution modeling. We will demonstrate those effects in Houston 4 km WRF-CMAQ simulations with comparison to metrological and air quality observations made during the TexAQS II field experiment in 2006. The goal of this presentation is to show the new image processing tools developed and some preliminary results from WRF-CMAQ simulations using the data generated by these tools.