SCERP Project Number: A-1
Principal Investigator: Dr. J. E. Geisler
Department of Meteorology, University of Utah
The deserts of the Southwestern United States and Northern Mexico cover a large area of mainly undeveloped lands dotted with locally intensive industrial, commercial, and population centers. Within this region, the principal anthropogenic inputs of air pollution can be approximated as point sources for the purpose of regional transport models. During different synoptic (3-5 day) conditions, the regional transport of pollutants can be quite variable, and can depend strongly on the interaction of the flow with the complex terrain in this region. Topography generates atmospheric flow details which have first order influences upon the transport and dispersion of atmospheric contaminants, particularly below ridge height, where most of the anthropogenic input occurs. Air transport models utilized by the EPA for environmental impact statements are highly simplified, because more complete transport and diffusion models would require long computation times. In recent years, however, computer expense and observational verification problems have diminished. This is mainly due to advent of powerful workstations which can be economically dedicated to local simulations, as well as to deployment of wind profilers near the Mexican-US border. These technical advances will in principal allow testing of advanced computer simulations, which have potential advantages relative to techniques currently used by the EPA.
Such an advanced computer model developed previously in the Department of Meteorology at the University of Utah was in the course of this research project modified to be suitable for the specific application to the dispersion of a passive tracer from point sources located in the border region. The domain of the model was such as to embrace the southwestern US together with a region of comparable size within Mexico. The topography of this entire region was realistically incorporated into the model.
Our initial simulations were done with no ambient synoptic (that is, regional) scale flow present in the model. That is to say that the only mechanism for dispersal of the tracer from one (or more) point sources present was the air motion forced by diurnal heating and cooling in the presence of topography. The results of these simulations suggested that this mechanism alone can provide significant dispersion as far north as the Utah/Arizona border on the time scale of a few days.
The dispersion model was then modified to higher spatial resolution. In particular, we employed the model with a fine spatial resolution in a domain centered on El Paso in order to simulate the effect of diurnal, topographically generated, local circulation on the dispersion of a passive tracer from a single point source there. The results indicated that the topography is such that there is little net topographically induced circulation acting to transport the tracer out of this particular local region.
We then turned to the adaptation of the dispersion model to include the effect of prescribed synoptic (regional) scale flow patterns on the model- simulated dispersion of a passive tracer from sources in the border region. We recognized the need to identify and select synoptic scale flow patterns on the basis of their potential to produce undesirable environmental effects. For example, synoptic scale flows directed toward the United States that are accompanied by strong synoptic scale inversions would constitute the extreme end of the scale. At the same time, we began the task of adapting the dispersion model to accommodate selected synoptic scale flow patterns. We undertook the examination of a large data base to identify and classify over an extended period of several years the synoptic scale flow patterns and vertical temperature structures in the border region. The objective was to identify southerly flow regimes that are accompanied by temperature inversions (the extreme case alluded to above).
The examination of the data led to the development of an index characterizing the low-level atmospheric stability averaged over an are centered along the U.S.-Mexico border. We evaluated the index for each day from our 10-year set of synoptic meteorological data and isolated 11 episodes during that period that had very strong low-level inversions.We then prepared a composite of these episodes. Our regional dispersion model was run to give a forecast of the dispersion of pollutants from the border region under the conditions specified by the composite.
At this stage we undertook and completed the task of developing software to display our model simulations in color animation mode. These successfully showed the evolution of the pattern of a passive tracer released at three sites along the border during several cases with very stable atmospheric conditions (that is, a strong inversion). One of these cases was characterized by a synoptic (regional) scale flow from the southeast.
This brings us to the end of the second year of the project.
Last updated 7/1/99