Ozone Transport Assessment Group Modeling Summary

In the Ozone Transport Assessment Group Round 3 modeling evaluation, the entire Commonwealth of Virginia has been included in the northeast emission control region (Zone III). Since this region also includes the Northeast Corridor nonattainment areas, control requirements are likely to be greatest in this zone.

Virginia disagrees with the assignment of any part of the Commonwealth in Zone 3 other than the Northern Virginia portion of the Washington, DC, nonattainment area and the Ozone Transport Region (OTR). The Virginia Department of Environmental Quality has previously requested that the non-OTR portion of Virginia be reassigned to the southeast control region (Zone IV), but this request has not yet been acted on.

To support this request, the DEQ has performed a modeling evaluation to assess the impact such a change in boundaries and corresponding decrease in control levels would have on ozone concentrations in the Northeast. In addition, the effect of applying increased controls in the revised Zone IV (Southeast including the non-OTR portion of Virginia) was evaluated to examine the total impact of the revised Zone IV on the Northeast Corridor.

The results of the DEQ modeling clearly show that (1) the reassignment of the non-OTR portion of Virginia to Zone IV and (2) the implementation of more-stringent point source NOx controls in all of the revised Zone IV has very little impact on predicted ozone concentrations in the Northeast. As a result, DEQ is again seeking the support of OTAG to approve the requested reassignment of the non-OTR portion of Virginia to the southeast region.

Background

During the OTAG Round 3 modeling evaluation, the entire Commonwealth of Virginia has been included in northeast control zone (Zone III) where the most stringent controls are likely to be applied. Virginia does not agree with this zone assignment for the Round 3 model runs, and has requested that only the six Northern Virginia localities that make up the Virginia portion of the Washington, DC, nonattainment area and the Ozone Transport Region (OTR) be included in the Zone III, with the remainder of Virginia reassigned to the southeast control zone (Zone IV).

The justification for this is that there have been no modeling results produced to date that support the inclusion of Virginia in the northeast zone. Additionally, the rest of Virginia is not part of OTR as defined by Clean Air Act. Thus far, Virginia's request has not been acted on, and the proposed zone boundaries with all of Virginia in the northeast control zone have been used for most of the Round 3 modeling.

In order to address the issue of where Virginia should be placed, DEQ has conducted a UAM-V modeling study using the 1995 OTAG modeling database. Several geographical sensitivity analyses have been performed by DEQ in addition to the OTAG geographical runs to examine the impact of utility NOx controls in Virginia on ozone concentrations in Northeast nonattainment areas.

Methodology

The model and domain structure setups were the same as that used in the OTAG UAM-V application. The 1995 episode (July 7-18, 1995) was selected for the modeling study because the 1995 episode was believed to be the most representative episode for regional ozone transport analysis. In addition, as part of the OTAG Northeast Modeling and Analysis Center (NEMAC), DEQ has been performing OTAG Round 2 and Round 3 modeling for this episode, so that all input data for the episode are available online at DEQ

The analysis was performed using the same OTAG meteorology, air quality and biogenic emission input files. The anthropogenic emission files were generated at DEQ using OTAG Round 3 emissions data derived from 2007 Base 1c emission scenario. The control zones used for this study were based on the Round 3 runs (Run A - I) with modified Zone III and Zone IV boundaries.

The zone map for OTAG 9 Round 3 runs is shown in Figure 1a, and the revised zone map for this study is shown in Figure 1b. The non-OTR portion of Virginia was assigned to Zone IV (southeast) in the revised map instead of Zone III (northeast) as in the OTAG map. The model runs by DEQ are based on the revised zone boundaries.

The emission scenarios along with the utility point source control levels for each zone are listed in Table 1. The control levels are the same as those defined by OTAG. The area and non-road mobile emissions controls were at level 2, and on-road mobile emissions controls were set at level 1.3.

Table 2 summarizes the utility emission control levels for several Round 3 runs. The results of these runs were compared to the Virginia runs. As shown in Tables 1 and 2, Run V3 is designed to simulate OTAG Run C with the revised Zone III and IV boundaries. The difference between the two runs is that level 1 controls (55 percent reduction) are applied in the non-OTR portion of Virginia instead of level 2b (75 percent reduction) in Run V3. As was the case in most Round 3 runs, level 2b NOx controls were applied in the serious/severe nonattainment areas (Northeast Corridor, Atlanta and Lake Michigan).

Table 1: Virginia geographical sensitivity runs

Table 2: Round 3 geographical runs

Results

Comparison of Run V3 with Run C indicates the effect of applying level 1 vs. level 2b point source controls in the non-OTR portion of Virginia. The spatial plot of layer 1 daily maximum ozone differences (greatest episode-composite decrease) between Run C and V3 with Base 1c (2007) ozone concentrations greater than 100 ppb (threshold = 100 ppb) is shown in Figure 2.

The plot indicates that the more-stringent 2b controls in the non-OTR portion of Virginia have little positive impact on ozone concentrations in the Northeast Corridor on the high ozone days. In fact, the predicted ozone decreases occur primarily within Virginia. There was only one cell in Maryland where the predicted ozone decrease ranged from 2 to 6 ppb (actual value was 2.5 ppb) when base case ozone was greater than 100 ppb.

A detailed review of the model results on a day-to-day basis revealed that the predicted 2.5 ppb ozone decrease in Maryland occurred on only one day (July 17). A review of ambient air quality data found that only one monitor recorded a marginal exceedance within the Northeast Corridor on that day.

The use of Base 1c as threshold concentrations is the OTAG modeling center's convention. However, there are no OTC MOU or level 2b controls applied in the Northeast for this baseline scenario. In order to evaluate the transported ozone benefits beyond local controls in the Northeast, Strategy A (so-called reference run for Round 3) should be used for threshold concentrations. Strategy A applies level 2b controls in the Northeast Corridor.

Figure 3 illustrates the same daily maximum ozone difference plot, this time with Run A concentrations greater than 100 ppb as the threshold. The plot clearly indicates that there is no ozone benefit more than 2 ppb in Northeast Corridor where the ozone concentrations are greater than 100 ppb when assuming level 2b controls in Northeast Corridor.

Point source NOx emissions in Virginia, especially utility emissions, are relatively small compared to both the Northeast and the Southeast states. The difference between the NOx reductions in Virginia from level 2b to 1 is only 71 tons/day based on the current OTAG emission inventory. Therefore, the small predicted ozone response to this small emission difference should be expected. The analysis of OTAG alternative geographical runs, which were presented at the recent OTAG meeting in Philadelphia and Washington, DC, provided the same evidence.

While the model results show that Virginia alone has little impact on the Northeast as discussed above, it could be argued that Virginia along with other states may collectively have a significant impact on the Northeast. To this end, the effect of additional point source NOx controls in all of Zone IV including the non-OTR portion of Virginia also was investigated.

Figure 4 shows spatial plot of the greatest episode composite decreases in daily maximum ozone between Strategy H and V4 with Run A concentrations greater than 100 ppb. Run H applies level 2b controls throughout the OTAG fine grid, including Virginia and Zone IV. Run V4 applies level 1 controls in the non-OTR portion of Virginia and the Southeast (revised Zone IV). This plot clearly illustrates that there is very little ozone benefit (less than 2 ppb) in the Northeast during the high ozone days from more-stringent point source NOx controls (level 2b vs. level 1) applied to the entire revised Zone IV.

A reduction of more than 14 ppb in ozone concentrations occurred in the western part of Zone IV. In some areas, the ozone benefits extend farther into other zones. However, the effect of NOx controls in Zone IV on ozone in the areas other than Northeast Corridor are outside the scope of this study and not evaluated further since Virginia has no impact on these areas.

Several numerical matrices have been used to further evaluate the relative effectiveness of various strategies in different zones on the southern and central northeast subdomains as defined by the OTAG modeling group. This analysis is primarily based on grid-cell hours greater than 100 ppb, which provide an indication of temporal duration and spatial extent of high ozone concentrations.

Figure 5 presents the bar chart of total relative ozone improvements (benefits) in percentage from Strategy Base 1c to Strategy I for southern-corridor (top chart) and central corridor (bottom chart). Strategy I is considered maximum controls; therefore, its relative ozone improvement is 100 percent of the total ozone benefit compared to Base 1c. A comparison of the relative ozone benefits between Strategy V3 and Strategy C suggests that more-stringent NOx controls in the non-OTR portion of Virginia would provide virtually no additional benefit in either northeast subdomain. Similarly, a comparison of the relative ozone benefits between Strategy V4 and Strategy H indicates that more-stringent NOx controls in entire revised Zone IV provide no significant benefits in the Northeast Corridor as well.

Inspection of Figure 5 also suggests that Strategy A has achieved 70 percent to 80 percent of the total ozone reductions in both the southern and central Northeast Corridor subdomains. Run A is a combined strategy that includes both local and regional NOx and VOC reductions from area/non-road, mobile and point sources.

Conclusions

The results of the DEQ modeling clearly demonstrate that the application of more-stringent NOx controls (beyond level 1) for utilities in the non-OTR portion of Virginia provides little benefit to the Northeast Corridor.

Additional NOx controls (beyond level 1) in both the non-OTR portion of Virginia and the rest of Zone IV provide little transported ozone benefit to the Northeast Corridor. Therefore, DEQ urges the Policy Group to reassign the non-OTR portion of Virginia to Zone IV for policy-related decision making purposes.