Comparison: ARB's 2001 vs 2000 Risk Maps

This page last reviewed August 17, 2010

Return to Maps- Page


Risk comparison: 2000 versus 2001 maps   Top of Page

Overall, the currently posted risk maps based on the year 2001 emissions are consistent with the previously posted maps based on the 2000 emissions. The most notable changes in the appearance of the maps are the inclusion of the Northern California region, along with an increase in map resolution, in some cases down to the 1 kilometer level. The viewer will also notice certain urban areas (e.g., San Francisco and Central Los Angeles) have shown a small, but visible increase in total risk. There are a number of reasons for this, that vary somewhat by the location. For San Francisco, much of this is due to a redistribution of offroad mobile source diesel emissions by the use of adjusted demographic data for the year 2001 run, as is explained below, in the section titled: Changes to the offroad diesel distribution.

For the Los Angeles region, much of this increase is due to the use of the 1999 meteorology files, in place of the 1996 meteorology files used for the previously posted 2000 baseline year, as is explained below, in the section titled: Improved meteorological data. In addition, there were small increases in industrial source diesel risk in certain urban areas, as well as decreases in mainly rural and agricultural areas, as is explained below, in the section titled: Industrial diesel sources.

What has changed between the 2000 and the 2001 year runs?  Top of Page

  • Implementation of emission controls:  Each year, reductions in cancer-causing pollutants are achieved from emissions control programs that have come into place. Important contributors to these reductions include the Diesel Risk Reduction Program, and the  Mobile Source Program. The Mobile Source Program includes aspects of the Diesel Risk Reduction Program, along with other programs that significantly reduce toxic pollutants, such as benzene and 1,3-butadiene.
  • All California modeled:  The year 2000 version of the maps did not include Northern California. The currently posted year 2001 version now models all of California.
  • Year 2000 census tract data:  The year 2000 version of the maps used the 1990 census tract data. The currently posted maps were modeled and processed using year 2000 census tract data.
  • Improved meteorological data:  The ARB uses the meteorological files developed by the U.S.EPA for their implementation of ASPEN. The U.S.EPA staff provided ARB staff with their latest 1999-year meteorological files. These files have additional weather stations, which improves the modeling. For Los Angeles, this resulted in an increase in the risk estimate averaged across the county between 1996 and 1999. While this is an interesting anomaly, it is not a vital message to be derived from these maps. Inhalation cancer risk is estimated over the period of a lifetime (about 70 years), and, therefore, it is important not to become focused on year to year meteorological differences for lifetime risk modeling.
  • Improved geographical distribution and resolution of data:  Top of Page
    The geographical distribution for nonpoint sources was improved in several areas. Point sources are provided with specific geographic locations. Nonpoint sources have to be distributed using other methods. Improvements in the distribution data for nonpoint sources allowed model inputs in some cases to be at a 1 km resolution. Note that in some cases this may cause the year 2001 maps to show apparent increases, or decreases in risk for local areas. This is because, in some cases the year 2001 maps are being averaged over 1 square kilometer. For the year 2000 maps the 2 to 4 kilometer resolutions that were used meant that the maps were being averaged over 4 square kilometers for some sources, and up to 16 square kilometers for others. It is important to note, that in many cases the actual underlying resolution of the data for the year 2001 maps is unchanged from the year 2000 maps, and may still be a relatively coarse resolution. In addition, because ASPEN is run using census tracts, rural regions will still tend to have very coarse resolutions, due to the large rural census tracts. It is in the urban regions, with small census tracts, that the greatest improvements in resolution will be observed between the year 2000 and year 2001 maps. Another geographical distribution improvement was the the adjustment of the traffic-area-zone onroad mobile source data to reduce the tendency to overcluster emissions.
  • Changes to the offroad diesel distribution:  Top of Page
    For the year 2001 risk maps, a new set of demographic distribution data were used to distribute the non-point sources. The new distributions caused certain offroad mobile diesel emissions source categories to tend to show stronger clustering in many urban areas. This resulted in higher modeled risk in these clusters. San Francisco showed the most pronounced risk clustering increase due to these changes to the distribution of the offroad mobile source diesel emissions. This does NOT mean that there was a large increase in actual diesel risk for offroad sources in these areas. Offroad mobile emissions are difficult to model, specifically because they are mobile and usually do not follow established roadways. Because of that, surrogates for offroad mobile emissions distributions, such as various types of industrial and commercial activity are used. These give only a rough picture of the possible distribution. Any changes to these assumptions can cause shifting of the offroad emissions estimates, and do not indicate a large change in the total amount of offroad emissions, but, rather, a rethinking of the geographical distribution.
  • Industrial diesel sources:  Top of Page
    For San Francisco, Central Los Angeles, Sacramento, Orange, San Diego (central coast of county especially), Stockton, and others, there were small increases in industrial diesel source risk. The industrial diesel risk in the southern San Joaquin Valley decreased.
  • Onroad emissions:  The previously posted year 2000 maps used the EMFAC2001 onroad model. The currently posted 2001 maps are based on EMFAC2002.
  • Improved point source location data:   Top of Page
    ARB emissions inventory staff have worked with air pollution control district staff to dramatically improve the point source location data. However, much work remains to be done in this area.
  • Improved point source stack parameter data:  In addition to point source locations, the modeling of point source emissions relies upon stack parameters, such as exhaust stack height, temperature, diameter and exhaust gas velocity. ARB emissions inventory staff have worked with air pollution control district staff to dramatically improve these data. As with the point source location data, much work still remains to be done to improve this data set.

Please send questions or comments to: