Attachment
USDA Forest Service
Technical Review of the Greene Energy Resource Recovery Project
Prevention of Significant Deterioration (PSD) Plan Approval Application
Background and Project Description
Wellington Development is seeking a Prevention of Significant Deterioration (PSD) permit to construct and operate a waste combustion circulating fluidized bed (CFB) electrical generating facility just east of the Monongahela River, near Nemacolin, PA. The proposed 525 MW facility, Greene Energy, is a resource recovery project that will reclaim 1,500 acres of formerly mined industrial property, providing environmental benefits for local air quality and water resources.
The proposed facility will use two CFB boilers that will combust a fuel blend consisting of 85% reclaimed coal waste (gob) and 15% run-of-mine (ROM) coal. The proposed facility will also install pollution control equipment to control emissions of sulfur dioxide (SO2), particulate matter (PM) and nitrogen oxides (NOx). With the implemented controls, the Greene facility will have the potential to emit 5,649 tons per year (TPY) of SO2, 2,414 TPY of NOx and 1,207 TPY of PM. Greene Energy has the potential to impact air quality and air quality related values (AQRVs) at two class I areas administered by the Monongahela National Forest (MNF); Dolly Sods and Otter Creek Wildernesses. These Wildernesses are the closest class I areas to the project site, with Dolly Sods located slightly more than 100 km SSE and Otter Creek located slightly less than 100 km SSE of Greene Energy.
Air Quality Impacts Analyses
A Class I air quality analysis was included in the Prevention of Significant Deterioration (PSD) Plan Approval Application submitted to the Pennsylvania Department of Environmental Protection (PADEP) July 12, 2004. The applicant conducted the modeling analysis, using the CALPUFF (CALMET/CALPOST) dispersion-modeling package, to assess potential incremental increases in pollution and impacts to air quality and AQRVs in the WV Wildernesses.
We provided comments on an initial modeling analysis early in 2004. On January 22, 2004, our modeling contractor, Dr. Howard Gebhart, sent a letter to the consultant with questions regarding the modeling methodologies. This letter was followed up with additional questions in a second letter dated February 3, 2004. Our review of the most recent modeling runs submitted July 12, 2004 shows that a majority of these concerns have been addressed and corrected by the applicant.
However one major modeling concern remains. EPA’s modeling guidelines for plume dispersion calculations in CALPUFF set MDSIP=3, which employs the standard Pasquill-Gifford-Turner dispersion coefficients. The applicant chose to use MDISP=2 which employs internally calculated micrometeorological variables to estimate dispersion. We find that impacts at class I areas increase when the model is run using the appropriate default (MDISP=3). On January 26, 2005 we notified Mr. Nolan (PADEP)of this deviation from the modeling guidelines. The applicant should provide verification for why the use of this non-default option is appropriate. We have not received any further information from PADEP or the applicant on this issue.
Class I Increment Analyses
The PSD increment modeling results are summarized in Table 1. Modeled concentrations resulting from Greene’s emissions were above the class I significant impact levels (SILs) for PSD increment consumption at both Otter Creek and Dolly Sods. Greene Energy, alone, exceeds the Class I increment SILs for short-term SO2 concentrations (both 3-hour and 24-hour averages) and short term PM10 (24-hour) concentrations. For other pollutants, NOx, and annual mean SO2and PM10concentrations, the maximum impacts are less than their respective SILs. Because the short term SO2and PM10 class I increment SILs were exceeded, the need for a multi-source CALPUFF modeling analysis was triggered. This analysis determines whether or not class I increment has been exceeded and whether the proposed source is a significant contributor to any increment violation.
Table 1. Class I PSD increment modeling results showing maximum impacts reported in micrograms per cubic meter.
Pollutant / Significant Impact Level & PSD Increment / Otter Creek Wilderness / Dolly Sods WildernessGreen EnergyOnly
Sulfur dioxide (SO2)
3-hour / 1.0 / 25 / 3.8620 / 3.3604
24-hour / 0.2 / 5 / 1.1924 / 1.1482
Annual / 0.1 / 2 / 0.0397 / 0.0437
Particulate matter (PM-10)
24-hour / 0.3 / 8 / 0.3485 / 0.3351
Annual / 0.2 / 4 / 0.0204 / 0.0245
Nitrogen dioxide (NO2)
Annual / 0.1 / 2.5 / 0.0113 / 0.0120
Because the cumulative analysis determines whether pollution concentrations in the class I area have exceeded an allowable maximum amount, it is also important in terms of protecting AQRVs. When increment is violated, it must be corrected through emission reductions. Therefore, it is most beneficial to the class I area if the analysis accurately identifies all sources contributing to increment.
The applicant conducted a cumulative increment analysis, however the results may not accurately reflect increment consumption because it is our opinion that the inventory does not include all applicable sources. Our concerns with the inventory used in the multi-source increment consumption analysis, were outlined in a letter provided to PADEP on February 18, 2005. (The February 18 letter is attached as Appendix B of this report.) As a result of FLM comments, PADEP directed the applicant to make several changes to the inventory used in the cumulative analysis, which includes incorporating additional Ohio sources. We have not seen the cumulative modeling results which implement these corrections.
Although the changes requested by PADEP do not address all of the concerns we presented, itrepresents an improvement over the current inventory. We feel that development of a complete increment inventory is the responsibility of the affected states and EPA;however, we would like to continue to work with PADEP and EPA to improve upon the inventories used in cumulative increment analyses.
We also would like to point out that the Greene inventory was used as a basis for the Robinson cumulative increment inventory. We have been told by PADEP that the inventories used in the Robinson Power and Greene Energy cumulative assessments were essentially the same. Given this information, one would conclude that the two analyses should show similar results. However, the results of the Robinson cumulative increment analysis, received March 24, 2005, indicate that that short-term increments are violated at Dolly Sods Wilderness. Greene Energy’s cumulative increment modeling results do not show similar violations. The applicantor PADEPshould elaborate upon any differences in the modeling techniques applied, or differences in the cumulative inventories which would explain these variations in the results.
Air Quality Related Values
The PSD sections (sec. 160-169) in the Clean Air Act Amendments (CAAA) make it clear that the Federal Land Managers (FLMs) are mandated to “protect the Air Quality Related Values within a class I area” (CAA section 165 (d) (2) (B)). This embodies one of the purposes of the PSD program, which is to “preserve, protect, and enhance” the air quality in “national parks” and “wilderness areas”. Congress shed light on the degree to which FLMs are to implement this “affirmative responsibility” by stating that “the FLM should assume an aggressive role in protecting the air quality values of land areas under their jurisdiction. In cases of doubt the land manager should err on the side of protecting the air quality-related values for future generations.” (Senate Report No. 95-127, 95th Congress, 1st Session, 1977) It is with this regulatory framework and Congressional mandate in mind that we review the AQRV analysis in the Greene Energy Plan Approval Application.
Visibility
Of the many AQRVs FLMs are charged with protecting, visibility is the one AQRV that was specifically mandated by Congress. The assessment of visibility impacts to the class I areas is therefore very important. In October of 2003, we informed the applicant that other than lowering the f(RH) cap from 98 to 95%, we would not accept any deviations from the FLAG methods for determining visibility impairment. Results of the visibility analysis using FLAG methodology are shown in Table 2.
Table 2. Class I visibility modeling results reported in the July 2004 Plan Approval Application. This table shows percent change in light extinction over three modeled years using the FLAG method with RHMAX=95%.
Dolly Sods / Otter CreekMaximum % Change in Extinction from Greene / 12.97 / 17.66
Number of Days Over 5% / 34 / 51
Number of Days Over 10% / 6 / 11
An increase of 10% and greater in light extinction is considered to be perceptible by the general public, and changes above 5% are perceptible to a smaller portion of the population. Previously we mentioned that the applicant’s CALPUFF run set one of the parameters differently from model guidance, and that impacts at class I areas increased when the model was run using the appropriate defaults. For 1996 not only did the maximum change in light extinction increase, the number of days over 5% and 10% increased as well. These results are shown in table 3. Given this information, we conclude that the visibility impacts reported by the applicant may be underestimated.
Table 3: Comparative modeling results for year 1996 using the EPA recommended defaults vs. the deviations ENSR applied.
Dolly Sods / Otter CreekENSR Deviations / EPA Recommended Defaults / ENSR Deviations / EPA Recommended Defaults
Maximum % Change in Extinction from Greene / 11.65% / 13.27% / 9.00% / 13.91%
Number of Days Over 5% (1996) / 3 / 16 / 9 / 20
Number of Days Over 10%
(1996) / 1 / 3 / 0 / 2
The applicant made several modifications to the visibility analyses and concluded that emissions from Greene would not result in visibility reductionsgreater than 5%. One of the first steps of their analysis modified the natural background light extinction to account for the presence of sea salt. Another involved eliminating hours from the analysis when “natural obscurations” (precipitation, fog and clouds) contributed to visibility impairment. We do not feel that these are valid modifications for the following reasons.
Sea Salt: The applicant’s method assumes that any sodium present in the background air is associated with sea salt and fails to account for the fact that sodium is also a natural background soil constituent. A significant fraction of the sodium measured in IMPROVE background data can be attributed to natural soil and is already accounted for in the current natural background estimates. As such, there is no valid technical basis to assume that most or all of the background sodium is tied to sea salt concentrations, and the resulting sea salt “correction” inherent in the applicant’s methods is probably overstated, particularly for Dolly Sods and Otter Creek Wildernesses which are removed from coastal influences.
Incorporation of natural obscuration: The applicant’s modified analysis was a targeted adjustment of selected variables aimed only at reducing impairment on the days where Greene’s impacts were above some perceptible threshold. This resulted in reducing all modeled impacts below the 5% visibility thresholds. The process used by the applicant is summarized below.
- Three surface weather stations were considered to be representative of Dolly Sods and Otter Creek wildernesses; Elkins-RANDO, Martinsburg Eastern and Morgantown Hart. If observations at these stations showed precipitation, fog or nighttime clouds for any hour of the day, those hours were considered to have natural obscuration and the visibility impacts due to Greene Energy for that hour were eliminated from analysisin the following way:visibility impairment was calculated for every hour of the day as a ratio of the impairment due Greene and natural background, scaled to a factor of 1. An hourly ratio that is greater than 1 means there is some percent change above natural background. A change ratio equal to 1 means there is no change from natural background. For hours where it was determined natural obscuration had occurred, the visibility ratio was set to 1, thus “eliminating” Greene’s contribution to visibility impairment for that hour. After accounting for all hours determined to be “naturally obscured”, the geometric mean of all 24 hourly ratios was used to calculate the daily change in extinction.
- However, not all weather stations report observations for an entire 24-hour period. In situations where high modeled visibility impairment occurred during an hour that surface weather data were not available, the applicant relied on “meteorological judgment” to determine whether natural obscuration was occurring in the class I areas. In some cases relative humidity greater than 92% at Morgantown Hart was considered an indicator of fog in Elkins and thus Dolly Sods and Otter Creek.
- After accounting for natural obscurations in this manner, the applicant concluded that all days originally predicted to have greater than 5%visibility change “contained sufficient hours of meteorological interference that recalculation of the daily extinction ratio brought the Project’s visibility impact down below the 5% threshold”.
We do not agree with the process used to reduce visibility impacts due to weather obscuration. First, the method used to identify hours obscured relied on weather observations from stations that are not necessarily representative of conditions at Dolly Sods and Otter Creek, especially for localized weather events. The three weather stations used are located at much lower elevations than the majority of the class I areas, ranging from 663-1095 meters below the highest points in Dolly Sods. Additionally, these weather stations, particularly Morgantown and Martinsburg, are located at a significant distance from the class I areas. The applicant has not provided evidence that weather conditions occurring in these alternate locations are representative of conditions in the class I areas.
Second, eliminating Greene’s impacts during hours of “natural obscuration” and then averaging these hours back into the 24-hour change in extinction calculation discounts any otherwise significant impairment during hours where no meteorological interferences occur. In other words it ignores the importance of the short-term “moment in time” visual experience visitors of class I areas may encounter. Whether or not fog occurred during the morning hours is irrelevant to a visitor in the class I area in the afternoon.
In summary, results of the “FLAG” visibility analysis indicate many days with visibility impairment over 5%, and this may be an underestimate. We do not agree with the applicant’s demonstration that visibility is not adversely affected by emissions from Greene Energy. We are far from the national goal for visibility specified in the CAAA and despite improving trends at Dolly Sods and Otter Creek, visibility is still considered impaired. In an effort to achieve the purposes of the PSD program, we must seriously evaluate modeled impacts that show a perceptible degradation in visibility, particularly in class I areas where visibility is already impaired. Therefore, we anticipate that emissions from Greene Energy will cause an adverse impact to visibility at Dolly Sods and Otter Creek Wildernesses.
Acid Deposition
Two additional AQRVs, water and vegetation, could be affected by Greene Energy’s emissions. It is well known that stream water in many parts of West Virginia, but especially Dolly Sods and Otter Creek, is acidic. The current situation is the result of a combination of underlying geology that lacks buffering capacity and historically high sulfur deposition patterns. These same factors are also affecting soils. As soils acidify several changes take place; among these are the potential loss of nutrients needed by vegetation and the mobilization of aluminum that is toxic to roots and soil biota. Leachates from the soil then move into stream water and affect the aquatic ecosystem. The effects of acid deposition on streams, soil and vegetation within a watershed are linked and affect one another.
Annual emission rates were modeled to predict the impact of Greene Energy’s emission on sulfur and nitrogen deposition at the class I areas (Table 3.). The sulfur deposition, from Greene Energy alone, is predicted to produce reductions in acid neutralizing capacity of 0.109 microequivalents/liter in the most sensitive streams in Dolly Sods, and 0.103 microequivalents/liter at Otter Creek. These are measurable reductions that would further degrade streams that have already been affected adversely by acid deposition and are unable to support aquatic life. We consider measurable sulfur and nitrogen deposition impacts to ecosystems already negatively affected by acid deposition to be unacceptable and therefore adverse. Appendix B provides more details on the current status of aquatic and terrestrial resources and our rationale for this finding of adverse impact.
Table 3: Maximum annual sulfur and nitrogen deposition (kg/ha/yr) from Greene Energy’s emissions at the class I areas in West Virginia.
Otter Creek Wilderness / Dolly Sods WildernessMaximum Annual Deposition
Sulfur in kg/ha / 0.0356 / 0.0378
Nitrogen in kg/ha / 0.00846 / 0.00869
Removal of Gob Piles and Mitigation of Class I Impacts
Wellington is proposing to reclaim over 100 million tons of refuse waste coal over the life of the facility. These mining wastes, referred to as “gob” are the by-products or impurities removed from mined coal. The gob piles at the LTV Nemacolin mine dump are the refuse from over 80 years of coal mining. There are significant local environmental impacts resulting from the gob piles; acidic leachates are transported into the nearby Monongahela River degrading water quality and acidifying surrounding soils. Additionally, the gob piles are sporadically igniting and combusting, resulting in uncontrolled pollutant emissions, creating localized air quality problems.