Bret Schichtel and Rudolf Husar, (Washington U.), 8/08/96 Summary Draft
The source region of influence is the area around a source that it is most likely to impact. The region of influence is dependent upon a pollutant's lifetime and the transport direction and speed from the source. A receptor is only impacted by emissions from a source if the emissions are transported to the receptor before the pollutants are removed by transformation or deposition. In addition to determining likely regions of source impacts, properties of the source region of influence can be used to characterize transport directions and speeds.
There are numerous means for estimating the source region of influence. In this paper a qualitative technique based upon meteorology is presented. This technique allows for relative comparisons to be made in space and time, identifying regions and times more conducive to pollutant transport than other regions and times. This technique is applied to the Eastern US to characterize the potential source region of influence and transport during the summers (June - August) from 1991 - 1995. The source region of influence is then estimated for periods of high and low ozone concentrations in a New England, South East, and Industrial Midwest regions.
A source region of influence is calculated from source plumes over selected periods of time. A source plume identifies the location of all air parcels at a specific time released from the source prior to the time of interest. Figure 1, presents four examples of plumes from St. Louis Missouri. These plumes were created by releasing three particles every two hours for five days prior to the time shown in each plot. The age of the particles is indicated their color, with red particles being less than 6 hours old, and blue particles being about 5 days old. The plume in each plot is impacting different parts of the Eastern US, for example during 7/15/1992 02:00 (Figure 1A) the plume is impacting a thin region from St. Louis to Maine, however, at 8/16/1992 12 (Figure 1C) the plume impacts a thin region from St. Louis to south eastern Oklahoma, and a region cover most of Texas. One general characteristic of all plumes is that the younger particles tend to be closer to the source than the older ones, due to the shorter time for them to be transport away from the source.
Multiple plumes from the same source at different times can be grouped together. For example, the St. Louis MO plumes every two hours from July - September 1992 (1104 plumes) are combined in Figure 2. The particles have been colored based upon their percent remaining mass, which was calculated using simple decay kinetics. A decay rate of approximately 2%/hr was used, which relates to a pollutant with a characteristic lifetime of two days. Both the particle density and percent remaining mass decrease with distance from the source. The particle density decreases because the further distance away from the source, the less likely the wind will blow in that direction. The percent remaining mass decreases due to increased travel time with distance from the source.
Another means of presenting the information in Figure 2 is in the form of a transfer matrix (Figure 3). A transfer matrix is a probability density function describing the likelihood of source emissions impacting a receptor. It can be calculated from the ratio of the sum of the remaining mass from all particles in a given area by the source's total ambient mass, the sum of the remaining mass of all particles released from the source for all plumes composing the transfer matrix. The transfer matrix in Figure 3 has a sharp decreasing gradient with distance from the source. This is the result of the decreasing remaining mass and particle density with distance from the source. The transfer matrix for a secondary species, such as ozone, will peak way from the source, since there is a time lag between the emissions of the precursor species and the formation of the secondary species.
The source region of influence is calculated from a source's transfer matrix. It is defined as the smallest region along a line of constant TM encompassing approximately 63% of the remaining mass (Figure 3). The source region of influence presented in Figure 3 is for a primary species. However, the boundary defining this region is approximately the same for a secondary species with the same characteristic lifetime.
Examples of source regions of influence for St. Louis MO, during Q3 1992 and 1995 are presented in Figure 4 for three different pollutant lifetimes. It is evident from these plots that the region of influence is highly dependent on the pollutant lifetime. During 1992, a pollutant with a one day lifetime has a region of influence that extends to Indiana, but for a three day lifetime it extends to the Atlantic coast.
The influence of the wind direction and speed can also be seen in these plots. The distances from the source to the source region of influence boundary is indicative of the characteristic wind speed. The greater the distances, the greater the characteristic wind speed.
The direction of transport can be seen from the elongation of
the regions of influence, where the primarily result of the elongation
is a higher frequency of transport in the direction of the elongation.
As shown in Figure 4, the predominate wind direction during both
years is to the east, but during 1995 the winds tended to have
a more southerly component than during 1992.
An airmass history database containing source plumes for 504 sources evenly distributed over most of North America (Figure 5) was created for the years 1991 - 1995. The database contains five day plumes created by releasing three particles from each source every two hours over the 5 year time period. The airmass history database was generated using the CAPITA Monte Carlo Model (Schichtel, 1995) driven by wind fields from the National Meteorological Center's Nested Grid Model. From this database, transfer matrices were created for each source and each 24 hour period. These transfer matrices were then selectively aggregated to create source regions of influence.
The source regions of influence are highly dependent upon the pollutant lifetime. The lifetime of ozone is not well known, but thought to be between 1 and 4 days, and most certainly varies with space and time. Due to this uncertainty in the lifetime, it is not possible to interpret the source regions of influence in a quantitative manner for ozone. Instead, it will be used in a qualitative manner by examining the relative differences in the characteristic transport from sources over the Eastern US under varying conditions.
The daily transfer matrices were aggregated together for the summer months (June - August) 1991 - 1995, from which source regions of influence for pollutants with one and two day lifetimes were created. Figure 6, presents these regions of influence for six sources over the Eastern US. This Figure shows that on average there are two characteristic transport regemes over the Eastern US. One is from Texas north then east to the Atlantic. The other regeme is in the the deep South. The source region of influence in the South are smaller than the rest of the East for a given age. For example, the Atlanta Georgia source region of influence is about 40% smaller than St. Louis'. Consequently, the wind speeds tend to be lower in the south then the rest of the East.
Figure 8, exhibits an alternative way of presenting the transport information contained in the source regions of influence. Each vector in this figure presents the characteristic transport direction for each source. The vectors are based upon the elongation of the source region of influence. The more oblong the source region of influence the longer the vector. A region containing long vectors pointing in the same direction is indicative of a well defined ventilating flow, while a region containing short vectors with varying directions is indicative of a stagnant flow. The general transport direction from Texas around the South and to the East is evident in Figure 8. Also, in the south the vectors are shorter and have more variable directions then the rest of the East.
The transport conditions for high and low ozone condition in a New England, Industrial Midwest, and South Eastern region were examined by creating source regions of influence for the 20% of the highest and lowest daily maximum ozone concentrations over each region during June - August 1991 - 1995. Figure 8, displays the domain of each in which the daily maximum ozone was average to distinguish the high and low ozone values. The source regions of influence and characteristic transport directions for the high and low ozone condition in each region are presented in Figures 9 -14.
Figures 9 and 10 show that the high ozone concentrations occur in the South in conjunction with stagnant airmasses. The source regions of influence are more than twice as small during high ozone concentrations then on average (Figure 6). The South East vectors in the characteristic transport direction plots (Figure 10) are very short and point in all directions. Low ozone concentrations occur in the South East when the wind speeds are higher with airflow coming from either the Gulf of Mexico or the Atlantic Ocean (Figure 9).
High ozone concentration in the Industrial Midwest also occur during stagnant airflow (Figure 11 and 12). As shown in Figure 12, the transport direction vectors are very short and point in varing directions. In the rest of the Eastern US, the characteristic transport shows a clockwise motion around the Industrial Midwest indicative of a high pressure system centered over the Industrial Midwest. The low ozone concentrations tend to occur inconjuntion with ventilating conditions, higher wind speed and persistent air flow from the northwest and southeast.
In New England the characteristic transport during the high ozone
periods occur under ventilating conditions as opposed to the stagnant
conditions for the other two regions. As shown in Figure 13 and
14, the high ozone occurs when there is high speed persistent
flow from the Industrial Midwest to New England. The lowest ozone
concentrations occur when the airflow is primarily from the northwest.
The flow during these times is less defined and slower then during
the high ozone concentrations. Also, low ozone in New England
is associated with more stagnant meandering flow in the Industrial
Midwest. lectively aggregated to create source regions of influence.
Figure 1. Forward plumes from a St. Louis source. The red particles have ages less than 6 hours while the blue particles are approximately 5 days old.
Figure 2. Five day St. Louis Missouri plumes for all of Q3 (July - September) 1992. Simple decay kinetics was added to each plume to simulate the loss of pollutants from an air parcel as it ages. The rate of decay was approximately 2%/hr, the equivalent of a two day pollutant lifetime.
Figure 3. A St. Louis Missouri transfer matrix for Q3, 1992. The transfer matrix was created assuming simple first order decay kinetics with an approximate decay rate of 2%/hr.
Figure 4. The source regions of influence for a St. Louis MO source during Q3 1992 and 1995. The source regions of influence were calculated for a pollutant with a one, two, and three day lifetime.
Figure 5. The location of all of the sources used to create the airmass history database.
Figure 6. Source regions of influence for six sites located through out the Eastern US for pollutants with one and two day lifetimes during the summer months of 1991 - 1995.
Figure 7. The characteristic transport direction over the Eastern US during the summer months from 1991 - 1995 for a pollutant with a one day lifetime.
Figure 9. Source regions of influence for six sites located through out the Eastern US for pollutants with one and two day lifetimes for the days associated with the A) highest 20% and B) lowest 20% of the Daily average maximum ozone over the South Eastern US.
Figure 10. The characteristic transport direction over the Eastern US for a pollutant with a one day lifetime for the days associated with the A) highest 20% and B) lowest 20% of the Daily average maximum ozone over the South Eastern US.
Figure 11. Source regions of influence for six sites located through out the Eastern US for pollutants with one and two day lifetimes for the days associated with the A) highest 20% and B) lowest 20% of the Daily average maximum ozone over the Industrial Midwest.
Figure 12. The characteristic transport direction over the Eastern US for a pollutant with a one day lifetime for the days associated with the A) highest 20% and B) lowest 20% of the Daily average maximum ozone over the Industrial Midwest.
Figure 13. Source regions of influence for six sites located through out the Eastern US for pollutants with one and two day lifetimes for the days associated with the A) highest 20% and B) lowest 20% of the Daily average maximum ozone over New England.
Figure 14. The characteristic transport direction over the Eastern US for a pollutant with a one day lifetime for the days associated with the A) highest 20% and B) lowest 20% of the Daily average maximum ozone over New England.