1.0 REGULATORY REQUIREMENTS The RHR requires that each state develop periodic progress reports describing the progress towards reasonable progress goals as outlined in the state’s RHR SIP. SIPs were originally due in December 2007, but the majority of SIPs were not submitted by this deadline, and some are currently still pending approvals and/or full or partial Federal Implementation Plan (FIP) revisions and resubmittals. The fully approved SIPs or FIPs establish an implementation strategy to reach natural conditions by the year 2064, or an alternate natural conditions goal if a state can demonstrate that an alternate rate of improvement is reasonable. Because not all SIPs and FIPs have been approved, long-term progress goals have not been defined for each CIA. The WRAP RHRPR work will focus on characterizing the difference between baseline conditions and progress period conditions. Specific regulatory questions addressed by the RHRPR work include: What are the current visibility conditions for the most impaired and least impaired days? What is the difference between current visibility conditions and baseline conditions for the most impaired and least impaired days? What are the current visibility trends? What is the change in emissions that occurred between the baseline period and the progress period? Regional, states and CIA specific analysis summaries will be designed to help States with their evaluation of the current or proposed implementation plan elements and strategies. Separate from the RHRPR, individual states will evaluate specific regulatory questions also required in the individual states’ reasonable progress reports – these include: 1.1 What is the status of implementation of all measures included in the implementation plan? What emissions reductions have been achieved through implementation of these measures? What emissions from within or outside of the state have limited or impeded progress in reducing pollutant emission and improving visibility? Are current implementation plan elements and strategies sufficient to enable the state or other states with mandatory Federal CIAs affected by the State, to meet all established reasonable progress goals? RHR PROGRESS RULE CONSIDERATIONS The basic premise of the RHR is to ensure that visibility on the 20% worst days continues to improve, and that visibility on the 20% best days does not get worse, as measured in units of deciview (DV) calculated using data measured at IMPROVE monitoring sites. Progress is measured in discreet 5-year average increments, beginning with the 2000-2004 baseline average, 1 and proceeding with each subsequent 5-year average (e.g. 2005-2009, 2010-2014, etc.). Some of the more subtle, but important, considerations for RHR calculations are described below. 1.1.1 Seasonality of 20% Worst Days To determine the 5-year average of the 20% best and worst days, the highest and lowest 20% of days for each complete year are first selected and then averaged, and a 5-year average is calculated from annual averages. Because of the annual variability of pollutants, especially the regional and sporadic nature of wildfires in the west, the selection of the 20% best and worst days introduces the seasonality of components as a variable. Figure 1-1 presents daily average aerosol pollution for the BALD1 in Arizona measured during the 2005-2009 progress period. Days selected as 20% worst are highlighted with red boxes under the respective days. Note that the days selected as the 20% worst days are largely dependent on high Particulate Organic Matter (POM) measurements, which are often associated with wildfire. Some of the highest days occurred during the winter, but many of the highest POM days occur during the summer, depending on the year. Figures 1-2 and 1-3 and present the monthly averages of pollutants measured at the BALD1 site averaged over the 2000-2004 baseline period, and the 2005-2009 progress period, respectively. Figure 1-4 presents the percent of worst days by season and year. POM and ammonium sulfate averaged highest during these periods. During the baseline years, the highest POM concentrations occurred during the summer months, averaging highest in June, and during the 2005-2009 progress period, POM averaged highest in November. Ammonium sulfate was more consistent during both periods, averaging highest between July and August. Because POM is one of the primary contributors to regional haze at the BALD1 site, the highest POM days influence the selection of the 20% worst visibility days. Consequentially, any days selected during the summer months would consequentially have higher sulfate than days selected during the winter months, although overall sulfate levels may remain the same or be lower. 2 Figure 1-1. Daily Aerosol Extinction measured by the Mount Baldy Wilderness Area BALD1 IMPROVE monitor during the 2005-2009 progress period. 3 Mount Baldy WA (BALD1 Site) Monthly Average Aerosol Extinction Baseline Period (2000-2004) 14 Aerosol Extinction (Mm-1) 12 Amm. Sulfate 10 Amm. Nitrate POM 8 LAC 6 Soil 4 CM 2 Sea Salt Figure 1-2. Dec. Nov. Oct. Sep. Aug. Jul. Jun. May Apr. Mar. Feb. Jan. 0 Monthly Average Aerosol Extinction measured by the Mount Baldy Wilderness Area BALD1 IMPROVE monitor during the 2000-2004 baseline period. Mount Baldy WA (BALD1 Site) Monthly Average Aerosol Extinction First Progress Period (2005-2009) 14 Aerosol Extinction (Mm-1) 12 Amm. Sulfate Amm. Nitrate POM LAC Soil CM Sea Salt 10 8 6 4 2 Figure 1-3. Dec. Nov. Oct. Sep. Aug. Jul. Jun. May Apr. Mar. Feb. Jan. 0 Monthly Average Aerosol Extinction measured by the Mount Baldy Wilderness Area BALD1 IMPROVE monitor during the 2005-2009 progress period. Mount Baldy WA (BALD1 Site) Seasonal Variability of 20% Worst Days 80% Winter (Jan. Mar.) 60% Spring (Apr. - Jun.) 40% Summer (Jul. - Sep.) Fall (Oct. - Dec.) 20% Figure 1-4. 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 0% 2000 Percent of 20% Worst Days 100% Seasonal variation of 20% worst Aerosol Extinction days measured by the Mount Baldy Wilderness Area BALD1 IMPROVE monitor. 4 1.1.2 Discreet 5-year Averages vs. Trends The RHR prescribes that progress be measured using discreet 5-year average increments, and EPA’s Guidance for Tracking Progress Under the Regional Haze Rule RHR (EPA, 2003) states that determining trends for all the individual species that contribute to haze is especially helpful in tracking progress. Individual high or low years can affect the 5-year averages, so looking at annual trends can also be instructive in determining the long term behavior of a pollutant. When comparing the baseline period to the first progress period, there are some instances where the 5-year averages and the calculated trends show different characteristics. Figure 1-5 presents an example of an increase in 5-year average DV for the 20% worst days at the Anaconda-Pintlar Wilderness Area SULA1 IMPROVE site, which had a decreasing annual DV trend. Trends for this report are calculated using Kendall-Theil statistics, which are often used in environmental applications because these statistics are resistant to outliers. The annual averages indicate that the 2007 DV year was high, but otherwise averages were decreasing. For all sites, both 5-year average differences and trends will be reported, and any differing characteristics will be noted and described. Anaconda-Pintlar WA and Selway-Bitterroot WA (SULA1) Annual Averages, Period Averages and Trends Deciview Extinction (DV) 25 20 15 20% Worst Days Annual Average 20% Worst Days Period Average Trend; slope = -0.2 (p-value = 0.2) 17.0 13.4 All Days Annual Average Trend; slope = -0.2 (p-value = 0.09) 10 5 3.0 2.5 Figure 1-5. 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 0 20% Best Days Annual Average 20% Best Days Period Average Trend; slope = -0.1 (p-value = 0.03) Annual averages, period averages and trend statistics for SULA1 IMPROVE site. 1.1.3 Averaging Considerations for DV Calculations The RHR metric, as defined using deciviews, does not provide information regarding the relative contributions of individual species to overall visibility. DV extinction is logarithmically related to the total extinction (bext), e.g. DV=10ln(bext/10), where bext is the sum of extinction as calculated from individual species mass measurements. Looking at individual species extinction is useful for RHR considerations because some species originate from largely anthropogenic sources, while others from a mixture of both anthropogenic and natural sources. Because of the logarithmic nature of DV, it is not possible separate species in terms of DV, so total extinction in bext is a useful parameter when determining individual species behavior. 5 EPA’s Guidance for Tracking Progress Under the Regional Haze Rule ((EPA, 2003) specifies that the 5-year average DV value is calculated as an average of annual DV values, which are in turn calculated as averages of daily DV values. This means that for annual and 5year averages, The average of DV value is not logarithmically related to the average of bext values, (e.g. DV(avg.) ≠ 10ln(bext(avg.)/10). In some cases, characterizing the changes in extinction between the baseline and progress period using both DV and bext units can exhibit different behavior. Figure 1-6 shows an example of DV and total extinction as calculated for the SULA1 monitor. The figure indicates that, while the calculated DV haze index decreases slightly, total extinction as measured in Mm-1 actually increases. Yellowstone NP, Grand Teton NP, Red Rock Lakes WA and Teton WA (YELL2) Annual Average and Period Averages 50 20 34.5 36.1 11.8 11.5 40 15 30 Figure 1-6. DV Annual Average DV Period Average Total Extinction Annual Average Total Extinction Period Average 2010 2009 2008 2007 2006 2005 0 2004 0 2003 10 2002 5 2001 20 2000 10 Total Extinction (Mm-1) Haze Index (DV) 25 Annual and period average aerosol extinction measured by the YELL2 IMPROVE monitor. Decreasing DV, but increasing bext for the 20% worst days was observed at nine WRAP CIA sites, including SYCA1 in Arizona, DOME1, PINN1 and TRIN1 in California, HECA1 in Oregon, GAMO1 in Montana, WHPA1 in Washington and BRID1 and YELL2 in Wyoming. Slightly increasing DV associated with decreasing bext averages was observed at the CRLA1 site in Oregon. 6 2.0 2.1 WRAP REGIONAL SUMMARIES MONITORING DATA The basic premise of the RHR is to ensure that visibility on the 20% worst days continues to improve, and that visibility on the 20% best days does not get worse, as measured in units of deciview DV calculated from data at IMPROVE monitoring sites. Progress is measured in 5-year average increments beginning with the 2000-04 baseline average, and proceeding with each subsequent 5-year average (e.g. 2005-09, 2010-14, etc.). Subsection 501.308(g)(3)(ii) of the RHR requires that states assess the difference between current visibility conditions for the most impaired and least impaired days and baseline visibility conditions. Figures 2-1 and 2-2 present the difference between the 2000-2004 average baseline period and the 2005-2009 first progress period in DV for the 20% worst and 20% best days, respectively, for CIA IMPROVE sites in the WRAP region. The maps indicate that 5-year average extinction on the 20% worst days decreased at most sites, but showed some increases at several sites. The map for the 20% best days indicates that best days did not get worse, and actually improved, at all but one site. The one exception was very slight increase of 0.1 DV at the KALM1 site in Western Oregon. 7 Figure 2-1. Change in DV extinction between baseline period average (2000-2004) and the first progress period average (2005-2009) for 20% worst visibility days. Figure 2-2. Change in DV extinction between baseline period average (2000-2004) and the first progress period average (2005-2009) for 20% best visibility days. 8 The RHR metrics, as defined using deciviews, do not provide information regarding the relative contributions of individual species to overall visibility. Apportioning haze to the various species contributors is important in assessing which pollutants can be controlled, as some species originate from largely anthropogenic sources, while others from a mixture of both anthropogenic and natural sources. Figures 2-3 and 2-4 presents a regional map of average aerosol extinction for the baseline period (2000-04), and the first progress period average (2005-09) for the IMPROVE monitors which represent Federal CIAs in the WRAP region. The size of the pie chart is related to the magnitude of visibility impairment, and colors represent the relative contribution of the pollutants measured by the IMPROVE network to quantify visibility impacts. The maps indicate that particulate organic matter (green), which is often related to wildfire activity, is a large factor in visibility reduction in the west. Visibility impairment in western CIAs that are directly adjacent to more populated areas in the West is influenced more by ammonium nitrate (red), which is commonly associated with combustion activities, especially industrial activities and vehicles. Ammonium sulfate (yellow) represents most of the visibility impairment at the Hawaii sites, and up to one third of the impairment in the contiguous U.S. The largest contributor to ammonium sulfate concentrations in the contiguous U.S. and Alaska is generally industrial activities such as coal burning power plants, while natural volcanic activity contributes to the high measured ammonium sulfate at Hawaii sites. Figures 2-5 and 2-6 present the individual components of haze that have either decreased (Figure 2-5) between the 2000-2004 baseline period and the 2005-2009 progress period, or increased (Figure 2-6). To relate these changes to 5-year DV average differences, decreasing aerosol extinction components associated a decrease of more than 0.1 DV are highlighted in blue in Figure 2-5, and increasing aerosol extinction components associated with an increase of more than 0.1 DV are highlighted in red in Figure 2-6. Most of the decreases in DV values were associated with decreasing ammonium nitrate or decreasing POM. Decreases in California, eastern Oregon and Idaho were were largely due to ammonium nitrate reductions, while decreases in northern Washington and Montana, Colorado, New Mexico and Arizona were largely due to decreasing POM. Some sulfate reductions were also measured in western Washington and northwestern Oregon. Most of the increased in DV values were associated with increasing POM in California, Idaho, Montana and Utah. Ammonium sulfate contributed to increases in Alaska, Hawaii, and a few of the sites in the contiguous states. 9 Figure 2-3. Regional average of aerosol extinction by pollutant for baseline period average (2000 – 2004) for 20% worst visibility days. Figure 2-4. Regional average of aerosol extinction by pollutant for the first progress period average (2005 – 2009) for 20% worst visibility days. 10 Figure 2-5. Magnitude of aerosol extinction components that have decreased between the baseline average (2000-2004) and the first progress period average (2005-2009) for the 20% worst visibility days. Figure 2-6. Magnitude of aerosol extinction components that have increased between the baseline average (2000-2004) and the first progress period average (2005-2009) for the 20% worst visibility days. 11 2.1.1 Annual Trends Subsection 501.308(g)(3)(iii) of the RHR requires that states present an analysis of the change in visibility for the most impaired and least impaired days over the past 5 years. RHR guidance further states that determining trends for all the individual species that contribute to haze is especially helpful in tracking progress, as the implementation of the haze regulation can only be accomplished by reducing the concentrations of the particulate species that are responsible for the man-made portion of the worst haze days. In the long term, tracking trends of species contributions to haze provides information that can be useful in determining whether implemented emissions controls are having the expected effects. As described in Section 1.1.2, trend analysis may help characterize long term pollutant behavior, in part because the trend statistics used here are more resistant to individual high and low years than 5-year averages. The RHR progress period only reflects the 2005-2009 data, but IMPROVE data are currently available through 2010 and trends are shown here for the years 2000-2010. Because the first of the regional haze SIPs was approved in 2008, use of the most recently available data may better reflect any implementation of new control measures proposed in the Long-Term Strategies adopted by the States. Ten year extinction trends for the period 2000-2010 were computed for both 20% worst days, and for all sampled days at each IMPROVE CIA site that had at least five years of complete data. Theil slopes were calculated to determine the magnitude of the trend, and pvalues were calculated using Mann-Kendall trend analysis to determine the significance of each slope. Lower p-values indicate higher confidence levels in the computed slopes, where a p-value less than or equal to 0.20 (80% confidence level) is used here as an indicator of significance. Figures 2-7 through 2-20 present maps indicating trends for 20% worst days, and for all sampled days, for each parameter. The RHR specifically targets the 20% worst days, but looking at trends for all days may help identify seasonality effects on 20% worst days selection. Consistency between worst day and all day trends adds confidence to the characterization of the trend, and differences may suggest a seasonality affect that will be investigated in the individual state and CIA sections. Specific observations by parameter are listed below: Figures 2-7 and 2-8 indicate decreasing ammonium nitrate trends at nearly all sites. Slight increases are shown in both worst days and all days averages at the Denali site in Alaska. Figures 2-9 and 2-10 indicate most sites show decreasing ammonium sulfate trends, with consistent worst and all day trends in Hawaii and western Montana/eastern North Dakota. Increasing trends in northwest California and southwest Oregon are only significant for the 20% worst days, but not all days, indicating that the seasonality of the worst days may be influencing trends at these sites. 12 Figures 2-11 and 2-12 indicate that most POM trends are either decreasing or insignificant, although most increasing DV values were due to increases in 5-year average POM. Figures 2-13 and 2-14 indicate that LAC is also generally trending down, but values are low compared to other parameters. Figures 2-15 and 2-16 indicate that trends in soil are mostly insignificant, and any significant trends are small compared to other parameters. Figures 2-17 and 2-18 indicate that Coarse Mass has variable trends, but 5-year average DV values indicated that increasing CM was only affecting increasing DV averages at a couple of sites (e.g. THSI1 in Oregon and CANY1 in Utah). Figures 2-19 and 2-20 indicate that Sea trends are mostly insignificant, with the largest significantly increasing trends measured on the pacific coast for the worst days. 13 Figure 2-7. 10-year annual average ammonium nitrate extinction trends for 20% worst days at CIA IMPROVE sites in the WRAP region. Figure 2-8. 10-year annual average ammonium nitrate extinction trends for all measured days at CIA IMPROVE sites in the WRAP region. 14 Figure 2-9. 10-year annual average ammonium sulfate extinction trends for 20% worst days at CIA IMPROVE sites in the WRAP region. Figure 2-10. 10-year annual average ammonium sulfate extinction trends for all measured days at CIA IMPROVE sites in the WRAP region. 15 Figure 2-11. 10-year annual average particulate organic matter extinction trends for 20% worst days at CIA IMPROVE sites in the WRAP region. Figure 2-12. 10-year annual average particulate organic matter extinction trends for all measured days at CIA IMPROVE sites in the WRAP region. 16 Figure 2-13. 10-year annual average light absorbing carbon extinction trends for 20% worst days at CIA IMPROVE sites in the WRAP region. Figure 2-14. 10-year annual average light absorbing carbon extinction trends for all measured days at CIA IMPROVE sites in the WRAP region. 17 Figure 2-15. 10-year annual average soil extinction trends for 20% worst days at CIA IMPROVE sites in the WRAP region. Figure 2-16. 10-year annual average soil extinction trends for all measured days at CIA IMPROVE sites in the WRAP region. 18 Figure 2-17. 10-year annual average coarse mass extinction trends for 20% worst days at CIA IMPROVE sites in the WRAP region. Figure 2-18. 10-year annual average coarse mass extinction trends for all measured days at CIA IMPROVE sites in the WRAP region. 19 Figure 2-19. 10-year annual average sea salt extinction trends for 20% worst days at CIA IMPROVE sites in the WRAP region. Figure 2-20. 10-year annual average sea salt extinction trends for all measured days at CIA IMPROVE sites in the WRAP region. 20