U1 The DEIR identifies the project’s potential inconsistency with the regional air quality plan as a significant effect, which belies the commenter’s claim that the DEIR minimizes the impact of the project on regional air quality.

U2 As stated on page 4.5-61, the DEIR indicates that increases in emissions would be significant if they would result in significant increases in local criteria air pollutant concentrations, in significant increases in health risks in the vicinities of the plants, or in significant increases relative to emissions projections used in regional air quality plans. The first two types of impacts (criteria air pollutants and health risks) were evaluated using standard thresholds rather than any percentage-, or ratio-, type threshold. The third type of impact (comparison with plan projections) did use a one-percent test as an indication of emissions estimates that would be notably out-of-sync with those developed for the regional air quality plan. It should be noted that the comparison with power plant projections was the basis for a project-specific impact determination, not a cumulative impact determination. Using the one-percent test, the DEIR concludes that emissions increases would be a significant effect of the project. Ironically, the one-percent test or "ratio" method, to which the commenter objects, is the only significance criterion by which air quality impacts due to the project were found to be significant in the DEIR. It is noted also that, in developing the one-percent criterion, the EIR consultant conferred with staff at BAAQMD (the agency responsible for developing emissions projections used in the regional air quality plan), who agreed that this approach was both reasonable and appropriate in this context (Guy, 1998). Please also see response to Comment U13. For additional information regarding cumulative impacts on regional ozone and PM-10 emissions, see responses to Comments U14 and U16.

Reference:

U3 The choice of the endpoints does not fail to account for any substantial changes in operations which may occur at these plants. As discussed in Chapter 3 and Attachment C, the difference in incentives and opportunities between PG&E and any new owners will diminish, particularly starting in 2002. At that date, PG&E can enter the direct access market, just as the new owners. In addition, PG&E must add increasingly stringent emission control measures at its plants in the Bay Area. As a result, total NO_x emissions will fall after 1999 in any case assuming a constant behavioral change (the only assumption possible in this type of analysis). Any increase in electricity demand is likely to be relatively small in the three-year period from 1999 to 2001, and, with the expected addition of substantial new generation resources in the next five years (see Section 5.2.2 (page 5-3), most of this increased demand will be met by other facilities.

The EIR does not ignore potential air quality effects in the years between 1999 and 2005. The air quality section includes an comparison of power plant emissions under the Baseline scenario (with continued PG&E ownership), and A-Max scenario (new ownership) with projected power plant emissions contained in the ’97 Clean Air Plan. The years of analysis are 2000 and 2003, because these are the years for which emission estimates are provided in the ’97 Clean Air Plan. These analyses are summarized in the DEIR in Table 4.5-35 (on page 4.5-78), in Table 4.5-36 (on page 4.5-79 and in Table 4.5-37 (on page 4.5-80). The DEIR determined that if the plants were operated at the A-Max levels, there would be a significant, unavoidable but temporary effect. The impact would be significant in the year 2000, but with the implementation of Mitigation Measure 4.5-5, it would be less than significant by the year 2003.

Finally, the EIR does not minimize 2005 cumulative impacts associated with new owners of the plants simply because it does not compare the new owners’ operations with PG&E’s projected 2005 operations if PG&E were to retain the plants. Indeed, the EIR employs a conservative approach by attributing any changes in plant operations between 1999 and 2005 to the project itself. In any event, the DEIR does analyze PG&E’s projected operations in 2005 as Alternative 1, the No Project Alternative (beginning on page 6-9). Thus, the EIR does acknowledge continued differences between PG&E and new owners in 2005.

U4 The capacity factors modeled in the DEIR as reported in Table 3.1 represent an annualized capacity factor for each generating unit. Actual operations vary during the course of a year, based on the demand at any time and the operational characteristics of a unit. Therefore, it does not make sense to use the capacity factors as strict operational limits on the power plants because, under a given set of circumstances, a given plant may need to be operated temporarily at higher levels to meet the immediate energy demand. Nor is it true that absent such limits the DEIR has failed to properly analyze the adverse impacts of the project. As discussed in Chapter 3 of the DEIR, the Analytical Maximum scenario represents the highest capacities at which the plants realistically could operate, taking into account a variety of limiting factors. As noted in the DEIR (page 3-12, second complete paragraph), the Analytical Maximum scenario is extremely unlikely to reflect a true operating scenario. A number of unlikely assumptions were made in running the SERASYM™ model so that the Analytical Maximum scenario would represent worst-case operating levels for purposes of environmental analysis. For example, the scenario assumes that all three fossil-fueled power plants would have an unlimited supply of natural gas at a price 25 percent below the least expensive supply of gas assumed to be available to fuel California power plants. Given the very conservative nature of this assumption, if circumstances were to change, as suggested by the commenter, it is much more likely that the natural gas prices paid by the plant owners would be higher, not lower, than the prices modeled, which would tend to suppress rather than boost operating capacities. Regarding other inputs, such as transmission system capability and operating procedures, these factors are more likely to favor new plants, not the plants being divested. For example, if existing transmission constraints were removed, it would make it more likely that generation would be boosted at more efficient plants, not at the less efficient Bay Area plants. Please also see the responses to Comments F53 and F54.

Please note that, in addition to analyzing the annual capacity factor changes, the DEIR also assesses the adverse effects of maximum 1-hour, 8-hour, and 24-hour operations of the power plants at substantially higher operating levels than the annual capacity factors. The results of this analysis are summarized in the tables addressing air quality concentrations. The concentration estimates for the Potrero, Contra Costa, and Pittsburg Power Plants are presented in Tables 4.5-29 (page 4.5-63), 4.5-31 (page 4.5-66), and 4.5-32 (page 4.5-68), respectively.

U5 BAAQMD developed Regulation 9, Rule 11 as a means of implementing Best Available Retrofit Control Technology (BARCT) for the source category of utility boilers. It should be noted that the power plants proposed for divestiture are existing permitted sources, which are subject to BARCT, rather than Best Available Control Technology (BACT), which applies to new sources. Under BAAQMD Regulation 9, Rule 11, the steam boilers at the three fossil-fueled power plants to be divested are subject to increasingly stringent standards for NO_x. Mitigation Measure 4.5-5 requires either permit modifications or changes to Regulation 9, Rule 11, either of which would assure that these increasingly stringent standards will apply to PG&E or to new owners. (The DEIR lists modification of Regulation 9, Rule 11, or equivalent permit revisions, as Mitigation Measure 4.5-5.) Moreover, to meet these standards, PG&E or the new owners would likely have two main options: (1) install more effective NO_x reduction technology, or (2) decrease the use of one or more steam boilers.

Thus, by 2005 and with implementation of Mitigation Measure 4.5-5, NO_x emissions, even under the A-Max condition, would be substantially less than under existing conditions (see Table 4.5-26 on page 4.5-57 of the DEIR). Therefore, two of the types of mitigation called for by the commenter (more effective emissions control technology and reduced capacity factors) are precisely the types of options that PG&E or new owners have to meet the requirements of Regulation 9, Rule 11, so no further mitigation would be needed.

The commenter also cites offsets as a possible mitigation measure; however, offsets in the form of emissions credits are not generally considered CEQA mitigation since they are not contemporaneous emissions reductions. With regard to offsets, see also response to Comment U17.

U6 Air quality significance criterion #1, "violation of an ambient air quality standard or substantial contribution to a projected violation of an ambient air quality standard," is an appropriate criterion to use in evaluating both project-specific and cumulative impacts and was used to evaluate both types of impacts in the DEIR. DEIR Tables 4.5-29, 4.5-31, and 4.5-32 were formatted specifically to allow for easy comparison with the concentration standards associated with significance criterion #1. The distinction cited by the commenter between how cumulative effects are to be examined for purposes of an Initial Study versus how they are to be examined for an EIR does not reflect the current CEQA Guidelines. Current CEQA Guidelines Section 15130 (a) extends the concept of "cumulatively considerable," which has been the guidance for Initial Studies, to the discussion of cumulative effects in an EIR. The EIR’s analysis of cumulative impacts comports with CEQA requirements. Please see responses to Comments U14 and U16.

U7 The Prevention of Significant Deterioration (PSD) regulation is used by BAAQMD when conducting a new source review. The facilities in this project are not new sources that would be subject to PSD or that would require emissions offsets. Thus, the PSD regulation would not normally apply to these plants. However, in order to conservatively judge whether the project would substantially contribute to violations of air quality standards, the EIR embraced and fully applied the PSD regulation. The 24-hour concentration change of 5 micrograms per cubic meter and the annual average change of 1 microgram per cubic meter, that are stated in criterion #1, are not actually the ultimate PSD standards contained in the PSD regulation. They are concentration increases that are defined in the PSD regulation as threshold levels, below which there would be no substantial contributions to a projected air quality concentration change. Under the PSD regulation, if a new source would exceed these very low threshold levels, then further analysis is required and additional standards apply. However, as indicated in Section 4.5 of the DEIR, the project emissions would fall below even these threshold levels. Therefore, the project’s impact is less than significant.

U8 The DEIR does not imply that there would be no health effects for levels below those identified in the referenced studies. Instead, it uses the levels of 20 micrograms per cubic meter and 10 micrograms per cubic meter to determine if the total plants (existing emissions plus emissions from divestiture) are major causes of respiratory problems at the maximum receptor, similar to the methods used by EPA when establishing new PM-2.5 standards (see responses to Comments F62 and F74). When considering the impacts from divestiture alone, the increases in concentrations of the 1999 A-Max and 2005 A-Max concentrations over the Baseline concentrations were compared to the more stringent thresholds of 5 micrograms per cubic meter and 1 microgram per cubic meter.

U9 Although the study cited in the comment suggests a relationship between low levels of PM-10 and health effects, there is considerable uncertainty in the magnitudes of the health effects and relative risks at lower levels, as was stated in the EPA 1996 staff assessment report on PM-10 and PM-2.5 (see response to Comment F74). The data contained in the study that are cited in the comment show a higher relative risk for the San Francisco area than for other cities. However, a plot in the study shows a much higher error bound for the San Francisco data than for the other cities. The error bound for San Francisco extends to levels that are equivalent to or lower than other cities in the study. Thus, it cannot be concluded that the relative risks for San Francisco are actually higher.

All of the relative risks included in the plot were for the same exposure level of 100 micrograms per cubic meter, well above the maximum PM-10 level (57 micrograms per cubic meter) that was measured for the San Francisco area in the past three years. The maximum 24-hour average concentration reported for San Francisco in the cited study was 139 micrograms per cubic meter, occurring between 1981 and 1990, which again is well above the maximum level of recent years. Because of major changes in the types of fuels used in the PG&E boilers and because of the elimination of lead in gasoline in the mid-1980s, PM-10 concentration levels in the area have decreased significantly. Therefore, some of the conclusions in the cited study may not apply to present and future conditions.

U10 Conclusions cannot be drawn from the Thurston study (summarized in documents attached by the commenter) that there is increased vulnerability in the Bayview-Hunters Point area over other metropolitan areas. The relative risk coefficients described in response to Comment U9 show error bounds for the San Francisco data that are much greater than the other cities. Therefore, there is much greater uncertainty in describing the effects of a 100 microgram per cubic meter change (the normalized exposure level) for San Francisco than for other cities. Also, the study was carried out for the period from 1981-1990 when PM-10 levels were much higher than present-day levels in San Francisco. The highest 24-hour average level in San Francisco that was used in the study was 139 micrograms per cubic meter, which is much higher than the maximum level monitored in the last three years in the vicinity of the Potrero facility (57 micrograms per cubic meter). PM-10 ambient air concentrations in the Bay Area in the last few years have decreased for several reasons, including the elimination of leaded gasoline in automobiles, which were sources of fine particulate matter in the form of lead oxide, and the ceasing of fuel oil burning, which also is a source of particulate matter. Much of the particulate matter that was released into the atmosphere in the region in the 1980s contained toxic substances, such as lead from auto exhaust and polynuclear aromatic hydrocarbons from burning residual fuel oil in steam boilers. The PM-10 monitoring data at the Arkansas station in the Bayview Hunters Point area shows maximum ambient air levels that are lower than levels in other parts of the Bay Area (see Tables 4.5-29, 4.5-31 and 4.5-32).

The studies regarding higher incidences of respiratory related illnesses in the Bayview Hunters Point area that are cited on page 4.5-31 of the DEIR do not identify the causes of the respiratory related illnesses, nor do they relate hospitalizations to elevated exposure levels of particulate matter. On that page, the DEIR states that a detailed study would be necessary to better determine the cause(s) of the health effects. Such a study is beyond the scope of this EIR. In addition, there is no inference in the earlier study that the power plants are a major factor in the respiratory illnesses in the Bayview Hunters Point area, and the contributions of the power plants to existing and future local concentrations of criteria air pollutants in the area are minimal (see DEIR, Table 4.5-29, page 4.5-63).

Volume III of the EPA Air Quality Criteria Document for Particulate Matter (USEPA, 1996a) indicates that a number of factors and confounding parameters besides ambient air PM-10 levels can considerably affect respiratory related health effects, such as particle size and the composition of the particulate matter. With regard to the composition of particulate matter from natural gas combustion at the PG&E plants, there is very little information on the health effects from exposure to these substances. In other words, not all particulate matter is the same with regard to health effects. It is very important to include both composition and particle size when assessing health effects. Because this information was not available, the DEIR conservatively assumed that all particulate matter was the same with regard to health effects.

There is very little information on the composition of particulate matter emissions from gas-fired boilers, such as those used at the PG&E plants, mainly because the emissions of total particulate matter are so low that the levels of any toxic substances that may be contained in particulate matter are usually below the detection limits. Particulate matter from the combustion of fossil fuel, including natural gas, residual fuel oil and distillate fuel, consists of a carbon core with other substances adsorbed to the surface of the carbon core. The combustion of residual fuel oil and distillate fuel produces other products of incomplete combustion that are absorbed onto the carbon core. These products of incomplete combustion consist mainly of multi-chain carbon organic fragments, such as polycyclic aromatic hydrocarbons (PAHs). PAHs are a group of compounds that have been considered by USEPA and by the California Office of Environmental Health and Hazard Assessment (OEHHA) to be carcinogens. For oil, wood, and coal combustion, PAHs are considered to be one of the major toxic substances of concern. Particulate matter emissions from fossil fuel combustion may also include toxic metals that are both carcinogens and non-carcinogens. Most of these toxic metals are contained in the fuel initially.

With respect to natural gas combustion, there is little opportunity to form long-chain PAHs, because the starting fuel consists mainly of methane, which contains only one carbon atom per molecule. Therefore, the particles emitted from natural gas combustion contain mostly carbon. To verify this assumption, the measurement results for natural gas combustion were compared with measured emissions from the combustion of residual fuel oil. The comparisons are summarized in Table U10-1.

The emissions reported in the table are based on measurements for oil-fired and gas-fired electric utility steam boilers, which are reported in the updated EPA publication AP-42 (Version 5 sections 1.3 & 1.4 USEPA, 1998). Emission factors for the two fuels were converted to the common units (pounds of pollutant per million BTU of fuel combusted). For natural gas combustion, measurements were carried out for a number of PAHs but the levels were below the detection limits of the instrumentation. Although it is unclear whether these substances are actually present in the particulate matter, these substances were included in the analysis at one half the detection limit.

Using the conservative approach of including PAHs at one half the detection limit for natural gas combustion, Table U10-1. shows that total PAHs are about 5.7 percent of the levels measured for residual fuel oil combustion.

A similar comparison was conducted for the emissions of toxic metals from the combustion of the two fuels. The results in Table U10-1 show that total toxic metal emissions from combustion of natural gas were approximately 1.1 percent of the total toxic metal emissions from the combustion of residual fuel oil. These results show that the toxic component of particulate matter is significantly lower than particulate matter from residual fuel oil combustion. Table U10-1 also shows that total particulate matter emission from gas fired boilers are about 20 percent of the levels reported for oil fired boilers, for the same heat input.

Another important factor in the health effects from inhaling particulate matter relates to the deposition rates in the respiratory tract. A principal factor in affecting deposition rates is the particle size. Studies reported in the EPA Criteria Document for Particulate Matter (USEPA 1996) and in other literature (Raabe, 1984) show that deposition rates in the respiratory tract are higher in the size range between 1.0 micron and 2.5 microns and for particles less than 0.2 microns. Relative deposition rates for the size range between 0.2 microns and 1.0 micron are lower (see also response to Comment F74).

The cumulative particle size distribution for utility boilers firing residual oil indicates that 71 percent of the particles are less than 10 microns, 52 percent are less than 2.5 microns, 39 percent are less than 1 micron, and 20 percent are less than 0.625 microns. Since a considerable portion of the emissions are less than 2.5 microns, they can deposit in the respiratory tract.

Particulate matter emissions for natural gas combustion are usually less than 1 micron in size, with a considerable portion being between 0.1 microns and 1.0 micron (USEPA, 1998). Since a greater fraction of particles emitted from natural gas combustion appear to be in this size range, with lower deposition than emissions from fuel oil combustion, it can be inferred that a greater fraction from gas-fired emissions would be expired upon exhalation and would not be deposited in the respiratory tract. In the fine particle size range less than 0.1 micron, deposition in the respiratory tract particles would again increase.

References:

SOURCE: EPA's AP-42, Version 5, Sections 1.3 and 1.4.
a Emission factors for natural gas that were identified as being less than method detection thresholds were assumed as one-half of the detection threshold and are identified above with an asterick.
b AP-42 did not provide a #6 fuel oil emission factor for benzo(a)pyrene. Therefore, it was assumed to be the same as natural gas.
c The particulate matter emission factor for #6 fuel oil is based on a sulfur content of 0.28.

U11 The comment attempts to relate annual emissions changes for the three PG&E fossil-fueled plants under the 1999 and 2005 A-Max scenarios directly to expected health effects, without addressing the actual ambient air concentration changes. It is concentration changes in the ambient air that are related to health effects. The DEIR shows on Tables 4.5-29 through 4.5-32 that the estimated maximum 24-hour average increases in ambient air PM-10 concentrations from divestiture would range from 0.5 to 3.6 micrograms per cubic meter at the maximum receptor. These maximum short-term increases are for localized areas and are not representative of increases for the entire region around the plants. Typical maximum short-term concentration increases in the area for population exposure, based on the modeling, would be considerably lower (less than 1.0 microgram per cubic meter). Most of the studies relating health effects from increases in exposure to particulate matter observed health effects when there were concentration increases of 20 to 25 micrograms per cubic meter for PM-2.5 and increases of 50 micrograms per cubic meter for PM-10. Other, more recent studies have reported observed health effects for increases of 10 micrograms per cubic meter, although with much greater uncertainty. None of the studies reported observed health effects for small increases in PM-10 concentrations (i.e., increases less than about 5 micrograms per cubic meter).

The EPA Criteria Document for Particulate Matter, which is cited in the response to Comment F74, shows a wide range in relative risk coefficients for concentration increases of 10 micrograms per cubic meter, varying by a factor of five. For greater increases, such as 20 to 50 micrograms per cubic meter, the Criteria Document indicates that health effects were observed and that relative risk coefficients were more certain. There is no indication in the document that there would be any health effects for increases less than 1 microgram per cubic meter. In fact, the EPA Criteria Document on Particulate Matter does not rule out a threshold below which no health effects would occur.

The risks at low levels, as identified in Exhibit C of the comment, were derived by extrapolating the relative risk plots for higher concentration increases to no concentration increase (zero concentration increase), even though there is no evidence that there would be any health effects at extremely low levels, especially for moderate background levels of short-term concentrations (30 micrograms per cubic meter). The studies do not report observed health effects at these small increases with moderate background levels.

The analysis that is referenced as Exhibit C in the comment extrapolates the relative risk coefficients that were derived from increases of 10 micrograms per cubic meter, and estimates relative risks for extremely small increases in concentration of about 0.02 micrograms per cubic meter. These very small increases were then used in Exhibit C to estimate increased mortality in the Bay Area. However, there is no indication that increases in particulate matter concentrations of 0.02 micrograms per cubic meter would cause any health effects at all, especially mortality. It is stated in Exhibit C that

In fact, the studies show an association for only moderate increases in concentrations (20 to 50 micrograms per cubic meter), but do not show any association for extremely small increases in concentrations (0.02 to 1.0 micrograms per cubic meter). Therefore, the estimates in mortality identified in Exhibit C have no scientific basis.

The comment then estimates mortality for the proposed project by scaling the annual emissions changes for all three fossil-fueled plants to emissions for the proposed plant in Exhibit C (the San Francisco Energy Facility). The comment does not relate exposure levels to health effects. The comment builds on unfounded risk estimates for extremely low concentration increases identified in Exhibit C to estimate correspondingly unrealistic risk estimates for the divestiture project. Based on the small increases in concentrations from the project, there is no indication that there would be any increase in mortality or morbidity in the region. The DEIR does not suggest that high mortality and morbidity levels identified in the comment are insignificant. Instead, the DEIR states that the small increases of PM-10 concentrations from the project (under the A-Max scenarios) pose less than significant health effects.

U12 The methods that were used to assess the project impacts where background levels have already exceeded the ambient air standards are consistent with the approaches used by BAAQMD, in which the more restrictive thresholds described in Significance Criterion #1 were used. For further explanation on how this significance criterion was used, see response to Comment F74. With regard to particulate matter emissions, BACT is already used at the PG&E plants, which involves using natural gas, the cleanest of all fossil fuels. With regard to ozone, the precursor nitrogen oxides will be controlled by 90 percent over 1997 emission levels by installing selective catalytic reduction (SCR), which is BACT for these units. Offsets would be required only if the residual impacts after using BACT are significant (measurable). Since the impacts from divestiture did not exceed these levels, offsets are not needed.

U13 Significance criterion #5, the one-percent test, attempts to distinguish important differences between emissions forecasts developed for a given project and forecasts used in the regional air quality plan. Criterion #5 was modeled after a significance criterion developed by BAAQMD and recommended in its former CEQA Guidelines document (BAAQMD, 1985). It is acknowledged that the current BAAQMD CEQA Guidelines document (April 1996) no longer includes the one-percent test as a significance criterion. However, the nature of this project, which involves the sale of an existing emissions source, is so different from that of a typical development project, such as a new subdivision or office park, that it is not unreasonable to include significance criteria in addition to (i.e., to supplement, but not replace) those recommended in the current BAAQMD CEQA Guidelines. Unlike most emissions sources, power plants are a separate line-item on the regional emissions inventories that are used in regional air quality plans, which provide the basis for the type of comparison provided in the DEIR. It is noted also that, in developing this criterion, the EIR preparers conferred with staff at BAAQMD, who agreed that this approach was both reasonable and appropriate in this context (Guy, Bill, Bay Area Air Quality Management District, telephone conversation, July 29, 1998). Please also see response to Comment U16.

U14 The commenter indicates that the cumulative air quality analysis in the DEIR used the judicially discredited "ratio" theory to determine that the project’s emissions would not be significant because they would comprise less than one percent of the region’s emissions. The commenter is incorrect as to the standards and methodology of the DEIR.

While the one-percent test helps to characterize the severity of an impact, it is acknowledged that it alone is not a sufficient basis for concluding that a cumulative effect would be less than significant. In this case, the conclusions drawn in the DEIR are supported on two bases. First, since the emissions sources that are the subject of this project are covered under Air District permits and since the emissions estimates (including both predicted emissions increases and decreases) would be consistent with those permits, the changes in power plant emissions (regardless of year or whether they are project-specific or cumulative) are presumed to be less than significant under CEQA Guidelines 15064(i) and the BAAQMD CEQA Guidelines. As explained on page 4.5-61 of the DEIR, while the emissions changes themselves were deemed less than significant on a regulatory basis, other types of impacts that flow from changes in emissions were subject to further evaluation, such as changes in local concentrations and health risks and consistency with regional emissions projections (see Impacts 4.5-2, 4.5-3, and 4-5-5, respectively). Second, the estimated power plant emissions of ozone precursors and PM-10 (and its precursors) would be less under cumulative 2005 and 2015 conditions than under 1999 Baseline conditions and, therefore, they do not contribute to the cumulative regional effect of increased emissions from Bay Area growth and development. The necessary text revisions are provided below.

The following text and table is hereby added after the first paragraph on page 4.5-59 of the DEIR:

U16 The commenter raises numerous issues, which are addressed individually below.

With respect to a regional PM-10 plan, it is acknowledged that no such plan has been developed to address the region’s nonattainment designation with respect to the state PM-10 standard since none is required under the California Clean Air Act. The ’97 Clean Air Plan states that, while the plan does not address PM-10 specifically, several of the control measures in the plan would reduce PM-10 concentrations (BAAQMD, 1997). Specifically, the ’97 Clean Air Plan cites the control measures reducing vehicle-miles-traveled and NO_x emissions as sources of PM-10 reductions by reducing two substantial components of ambient PM-10 concentrations, i.e., road dust and nitrates, respectively.

With respect to the federally mandated air quality plan, the DEIR indicates on page 4.5-16 that a revised State Implementation Plan (SIP) will be required due to U.S. EPA’s decision to change the designation of the Bay Area back to "nonattainment" for the national ozone standard.

With respect to the regional plan addressing the state ozone standard, it is acknowledged that this plan does not predict attainment of the state ozone standard at all places and at all time in the Bay Area for the foreseeable future. However, it does predict continued improvement in regional ozone concentrations. Emissions of both ozone precursors (ROG and NO_x) are expected to decrease between 1997 and 2003, and a reduction in the precursors would logically result in lower ozone concentrations. With implementation of the measures included in the ’97 Clean Air Plan, basin-wide ROG emissions are expected to decline from 488 to 373 tons per summer day between 1997 and 2003, and basin-wide NO_x emissions are expected to decline from 632 to 480 tons per summer day over the same period (BAAQMD, 1997). Such emissions estimates as those prepared for the ’97 Clean Air Plan take into account expected growth and development in the Bay Area.

With respect to the approach to cumulative air quality analysis, the commenter essentially advocates a significance threshold of "one additional molecule" of a nonattainment pollutant in a nonattainment area. Such an approach is not required or supported under the CEQA Guidelines. The CEQA Guidelines refer to a substantial contribution to existing or projected air quality violation as a basis for determining significant effect, not just any contribution. In addition, there is a distinction to be noted between existing conditions and project effects. The commenter cites a sentence from the setting section; however, the project-specific and cumulative analyses both focus on the net environmental change related to the project. Therefore, the question is not whether power plant emissions contribute "nonattainment" pollutants to the atmosphere, but whether power plant emissions would increase or decrease relative to a baseline value. For cumulative analysis, the 1999 baseline scenario represents the "baseline" value to which cumulative scenarios are compared in the EIR. This is the appropriate approach to cumulative impact analysis in an EIR. The response to comment U14 examines the regional cumulative emissions changes of the two pollutants for which the Bay Area is in "nonattainment," ozone and PM-10, and concludes that the net changes in power plant emissions of ozone precursors and PM-10 (and its precursors) would be negative compared to the 1999 Baseline case and, as such, would not contribute to the cumulative effect. For local cumulative concentration effects, please see the column on the far right-hand side of DEIR Tables 4.5-29, 4.5-31, and 4.4-32.

U17 Though how much the divested plant will operate is uncertain, it is very improbable that the Potrero Power Plant would operate more than the level resulting from the operating assumptions associated with the Analytical Maximum case (see responses to Comments F53 and F54). Although the DEIR forecasts Analytical Maximum annual capacities for the Potrero plant to be 44 percent in 1999 and 40 percent in 2005, for these Analytical Maximum scenarios, Potrero Unit 3 (the steam generating unit) is forecast to operate at a 75.6 percent capacity factor in 1999 (see DEIR Table G-4) and at a 63.5 percent capacity factor in 2005 (see DEIR Table G-6). These are unprecedented levels of operation since a steam generating unit typically averages no higher than about a 50 percent capacity factor.

In preparing this Responses To Comments Document, the project team analyzed a minor change to the 1999 heat rate for Potrero Unit 3, a steam boiler. The resulting changes in forecasts from the SERASYM model are analyzed below in detail so that the commenter can see how changes in the capacity factors would affect air quality conclusions. In general, it would seem that only Impact 4.5-5 is sensitive to minor (or even moderate) changes in capacity factors. Impact 4.5-5 was identified as a significant, unavoidable, but temporary impact of the project (ending in 2003). With increases greater than those identified below, Impact 4.5-5 would still be significant in 2003. However, no other modifications are known that would cause any such increases, and the Analytical Maximum results are designed to capture maximum possible operations. The other three units at the Potrero plant are all distillate-fired combustion turbine (CT) peaking units that are limited by BAAQMD rules to operating no more than 10 percent of the year. They are used for reliability purposes by the ISO and it is inconceivable that they would be employed much more heavily than projected unless they were both converted to natural gas fuel to reduce variable costs of operations and they were exempted from BAAQMD operating time limits. Assuming either occurrence would be speculative. Thus, since the Analytical Maximum case represents an extremely high (and unlikely) level of operations for the plant, consideration of the emission levels resulting from still higher levels of operations is not warranted.

As described above, new information from PG&E's 1998 Title V submittal, which is part of the on-going air quality permitting process, indicates a change in the heat rate (a direct measure of unit efficiency) for Potrero Unit 3 at this unit's higher levels of generation. The 1999 Baseline and Analytical Maximum Scenarios have been rerun to reflect the updated Potrero heat rates. No other scenarios need updating, because the updated heat rates for Potrero Unit 3 were already included in the 2005 model runs in the DEIR. The modified results forecast that the overall annual capacity factor for the Potrero Power Plant under the 1999 Baseline Scenario (continued PG&E ownership) would be about 1.5 percent higher than reported in the DEIR, increasing from 24.6 to 26.1 percent. The 1999 Analytical Maximum scenario (with the revised heat rate) would increase 4.5 percent above the value reported in the DEIR, from 44.1 to 48.6 percent. These changes would raise the Baseline scenario for expected PG&E operations of the Potrero plant in 1999 and, although the changes would raise the Analytical Maximum capacity factors for the Potrero plant a slightly higher percentage, upon review, the consequences of these changes are minimal and as analyzed below, the changes do not affect the conclusions of the DEIR. Were these changes in operations to affect any topical area, the area of concern would be air quality. A brief summary of the effects of these changes on air quality (for 1999) are provided below. Because the 1999 Baseline would increase at the Potrero Power Plant, the incremental change between the 1999 Baseline and the 2005 Analytical Maximum would be slightly reduced in all cases, because, as stated above, the 2005 Analytical Maximum already includes the more efficient heat rate for Potrero Unit 3.

U18 The Analytical Maximum Potrero Power Plant annual capacity factor would be 40 percent in 2005 (see Table 3.1 and Table G-6 in the DEIR). A model run from the 2005 Analytical Maximum was reviewed day by day, and the daily plant capacity factors ranged from 0 percent to 67 percent. The majority of the days had capacity factors between 30 and 60 percent. These daily levels were reviewed closely in the DEIR in the analysis of ambient air quality contaminant concentrations near the Potrero Power Plant (see Table 4.5-29 on page 4.5-63) and, as shown in the two right-hand columns of Table 4.5-29, the effect of the emissions from the project are minimal in comparison to all of the ambient air quality 1-hour and 24-hour standards.

U19 The basic modeling assumptions used to derive the capacity factors are set forth in the DEIR at pages 3-8 through 3-13. The commenter is specifically referred to the first paragraph of page 3-9 of the DEIR, which describes briefly the qualifications of SERA in running the model for more than a decade and also notes that an expanded list of modeling assumptions and discussion of the modeling is presented in Attachment G to the DEIR. Attachment G has 16 pages of text and 20 data tables that address model assumptions and results. This level of information is meant to enable the public and decisions-makers to make an independent, reasoned judgement.

For further clarification, an assumption used in deriving the capacity factors that was not fully discussed in Chapter 3 or Attachment G of the DEIR is that a generating unit’s maximum net capacity value may vary during the course of a year, particularly for a combustion turbine because its output capability is affected by the surrounding ambient air temperature. In such instances, the DEIR analysis used an average capacity value in the annual capacity factor calculation. Such seasonal capacity variations are present at Hunters Point 1, Potrero 4, Potrero 5, and Potrero 6 combustion turbine units, which have the following monthly varying ratings:

Month	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
MW	56	55	54	53	50	48	49	49	48	49	54	56

The rating of each steam boiler generating unit is invariant during the course of a year and those of the units being offered for divestiture are presented in the tables found in Attachment G to the DEIR.

U20 The cited value of 44 percent relates to the estimated annual capacity factor for the Potrero Power Plant in 1999 under Analytical Maximum operating assumptions. Given the conservative assumptions underlying the Analytical Maximum scenario, it is not foreseeable that the plant would operate at an annual capacity factor higher than that identified under the Analytical Maximum (see responses to Comments F53 and F54), and since DEIR Table 4.5-29 shows that annual-average concentrations in the vicinity of the Potrero power plant would not exceed the corresponding ambient standards (even under Analytical Maximum conditions), there is no basis upon which to impose a plant-specific annual cap on generation. In contrast to the annual-average concentrations, which reflect the 44 percent capacity factor cited above, the DEIR analysis of concentrations for averaging periods of 24-hours and less reflects maximum daily emissions rates from the Potrero units. Thus, the potential adverse impacts on both annual averages and lesser averaging periods have been fully evaluated in the DEIR.

U21 The BAAQMD’s Arkansas Street station data is presented since it is the closest monitoring station to the Potrero Power Plant. There is no monitoring station in the immediate vicinity of the plant. The analytical approach for impact assessment assumed that the data from the monitoring station on Arkansas Street was entirely unaffected by the plant, but is representative of background conditions (i.e., not including a power plant increment) in the project vicinity. Use of monitoring data from the closest station as the basis for background concentrations is the conventional approach in developing concentration estimates from a facility (transportation, industrial, etc.) that can be compared with ambient air quality standards. Worst-case incremental concentrations from the power plant (shown in Table 4.5-29 of the DEIR) were added to background concentrations that were derived from data collected at the Arkansas Street station. This is a conservative approach since it is possible that the measurements at that Arkansas Street station include some incremental contribution from the Potrero Power Plant, and if so, then the DEIR double-counts that increment by assuming no such contribution.

U22 It is true that a portion of NO_x emissions will eventually convert to secondary nitrate particulate matter (secondary aerosols), although reaction times can vary considerably. Conversion rates can range from less than 1 percent per hour (Randerson, 1984) to as high as 10 to 30 percent per hour (California Air Resources Board, 1998). However, a recent study indicated that in power plant plumes, nitrate particulate matter does not begin to form until at least 20 to 40 minutes after nitrogen oxides are released to the atmosphere (Seigneur, 1998). For typical wind speeds in the area, conversion to nitrate would not begin until the plume travels three to six miles from the emission source. The study (Seigneur, 1998) indicates that maximum conversion would occur after an early morning release (7:00 a.m.) of nitrogen oxides on a day with relatively high ozone levels in the atmosphere. For emission releases occurring at 5:00 p.m., nitrate aerosols would not begin to form until about five hours of travel time. A key factor in the conversion of nitrogen oxides to nitrate aerosol involves a series of complex reactions in which atmospheric ozone reacts with nitrogen dioxide (NO₂) to form the nitrate (NO³) radical. However, the high initial concentration of nitric oxide (NO) in the power plat plume tends to scavenge atmospheric ozone, limiting the availability of atmospheric ozone to be available to react with nitrogen dioxide, an important step in forming nitrate aerosol.

Since the receptors with the maximum contributions of particulate matter are near the power plants (approximately one mile away), there is very little opportunity for nitrate aerosol to form in that short time period and nitrate would not contribute to particulate matter levels at the maximum receptors. With regard to larger distances from the plant, a portion of the nitrogen oxides emissions would eventually convert to nitrate particulate matter. However, there will be less secondary aerosol formed from the plant emissions in the future because of large reductions in NO_x emissions to meet BAAQMD Regulation 9, Rule 11. Thus, there will be reduced nitrate aerosol contributions to regional particulate matter levels. Please refer to response to Comment U14.

References:

California Air Resources Board, telephone communication, November 1998.

Randerson, D., Power Production and the Atmosphere, Chapter 1 of Atmospheric Science and Power Production, DOE/TIC – 27601, U.S. Department of Energy, 1984.

Seigneur, C., P. Karamchandani, and A. Koo, Reduced Gas-Phase Kinetic Mechanism for Atmospheric Plume Chemistry, ES&T, Vol. 32, No. 11, 1998

U23 Atmospheric dispersion modeling conducted for this DEIR and, subsequently, in response to Comment B6 was limited in scope, and was performed only to refine the air quality impact analysis and supplement the health risk assessment (HRA). As discussed on page 4.5-42 of the DEIR, the HRA for air quality was initially based on work performed by PG&E in 1992. Modeling data and analysis from these PG&E 1992 HRA studies for the Potrero, Contra Costa, and Pittsburg Power Plants were utilized to evaluate impacts of the divestiture project. The DEIR analysis was accomplished by updating air pollutant emission information from the 1992 PG&E studies for the three divestiture project analysis scenarios (1999 baseline, 1999 A-Max, and 2005 A-Max).

The 1992 PG&E HRA studies were of varying detail. The Potrero HRA study provided very detailed information on sources, impacts, and locations of impacts. The PG&E HRA studies for the Pittsburg and Contra Costa Power Plants, while detailed, were based on "screening" level analyses which were imprecise on locations of plant impacts, and were conservative in nature. During the analysis of these data, the conservative nature of the Pittsburg Power Plant’s 1992 HRA screening data appeared in need of refinement for this DEIR. Similarly, air quality issues with the Potrero and Hunters Point Power Plants (Hunters Point Power Plant was initially part of the DEIR analysis but was later removed from A.98-01-008 in July 1998) indicated that further refinement of the 1992 PG&E studies were needed. Conversely, analysis of the Contra Costa Power Plant for the DEIR indicated that, although the PG&E 1992 HRA study data provided conservative results, local impacts were sufficiently less than significant to not require further refinement for the purposes of this DEIR.

PG&E provided model input source parameter data for the Pittsburg and Potrero Power Plants, meteorological data from PG&E-operated monitoring stations at both plants, and grids of receptors networks suitable for use at each plant. (Receptors are points where an atmospheric dispersion model predicts impacts from pollution sources being simulated by the model.) Included in these receptor grids were the locations of sensitive receptors close to both the Pittsburg and Potrero Power Plants, as identified by PG&E. The receptor network for the Pittsburg Power Plant is an area representing the City of Pittsburg. For the Potrero plant, the receptor network covered sensitive receptors in the immediate vicinity west of the Potrero plant and south to a distance of about 3.5 km (2 miles) covering the northern elevated portion of the Hunters Point/Bayview area. Both of these analyses considered the local topography. The Pittsburg plant analysis was performed with meteorological data for 1994, while the Potrero plant analysis utilized data for the period 9/28/91 to 9/28/92.

The dispersion model analysis was conducted using the Industrial Source Complex Short Term 3 Model (ISC3), Version 97363. This model was developed and approved for use by the U.S. Environmental Protection Agency and represents the state of the art in atmospheric dispersion modeling. The model was used assuming standard regulatory default options and BAAQMD modeling guidelines.

Air pollutant emissions used in the analysis presented in the DEIR were derived from data presented in Section 4.5 of the DEIR, as well as data presented in Attachment G. Short-term (24-hour average) PM-10 emissions for the Pittsburg Power Plant analysis were derived from a special simulation model analysis discussed in Section 3.2 of Attachment G.

Related information to this modeling discussion in contained in response to Comment 1-9.

U24 Atmospheric dispersion modeling has been routinely utilized since the 1970s as an analysis tool to evaluate the local fate and transport of air pollutants. A suite of mathematical models have been developed and refined by the U.S. EPA and others for use by analysts and regulators. Lists of these EPA approved models and methods of application of the models are provided at 40 Code of Federal Regulations Part 51, Appendix W. Many carefully conducted model prediction versus ambient monitoring data studies have been conducted over the years. To summarize these studies is well beyond the scope of this EIR, however, in general, it can be said that when the appropriate EPA approved model is properly utilized, the model-predicted impacts can be expected to be at least within a factor of 2 of actual field measurements or better (40CFR51, Appendix W). This factor of 2 is accepted by the EPA as a reasonable measure of model performance. Thus, from the perspective of regulators, the atmospheric dispersions models recommended for use by the EPA do and have been shown to "predict present effects from present pollution source conditions," within an accepted level of accuracy.

The following new reference is hereby added to Section 4.5:

Federal Register, July 1, 1997.

U25 With respect to model assumptions, please see response to Comment U24. Naturally, any changes in input data will affect the results of the model. Since the conservative Analytical Maximum capacity factors were used as input model data for the EIR, there is no need to further demonstrate how changes in such input data would affect the outcome.

U26 Ozone is normally characterized as a regional pollutant and is, therefore, appropriately the subject of a regional air quality planning effort. The DEIR addresses the project impact in the context of this regional air quality planning effort beginning on page 4.5-77 under Impact 4.5-5. The discussion of cumulative power plant emissions of ozone precursors has been revised. Please see the response to Comment U14.

U27 The DEIR does not dismiss the significance of the project’s ozone precursor emissions. Page 4.5-18 of the DEIR shows a table in the text of NO_x emissions limits, per million Btu of energy consumption, that are allowed for PG&E’s boilers between 1997 and 2005. The table shows that NO_x emissions for the intermediate year of 1999 will decrease by about 39 percent per million Btu of energy consumption compared to 1997 levels. It is these reductions on a per million BTU basis that will reduce precursor effects on ozone formation. Since, on a given day, the maximum operating rates under the 1999 A-Max scenario are not expected to change from the 1999 Baseline, the daily NO_x emissions will not change from the 1999 Baseline emissions, even though total annual emissions can be higher under divestiture for 1999.

In 2005, short-term and annual emissions will be reduced considerably compared to the 1999 Baseline. While it is possible that emissions from the plants would be less in 2005 under PG&E’s continued ownership than with divestiture (and they may not, depending on how PG&E chose to operate the plants at that time), the environmental impacts of the project are properly judged in comparison with the baseline, the existing physical setting in 1999. Furthermore, the analysis of Alternative 1, the No Project Alternative (beginning on page 6-9 of the DEIR), indicates how the environmental impacts of the project in 2005 would differ from PG&E’s continued ownership of the plants. See also Table G-2 of the DEIR for 2005 emissions data for Alternative 1, which can be compared to Table G-6 concerning 2005 emissions data for the project.

U28 The significance threshold is based on emissions inventories and forecasts developed for the ’97 Clean Air Plan (’97 CAP), which is the state-mandated air quality plan, not the SIP, which the federally mandated air quality plan. The state standard, nonattainment of which is the trigger for the ’97 CAP, is more stringent than the corresponding national standard, nonattainment of which is the trigger for the SIP. Because of this, the ’97 CAP invariably includes more control measures than the SIP. For instance, Regulation 9, Rule 11 is a control measure identified in the ’97 CAP that has not been incorporated into the SIP. The DEIR recognizes the seriousness of the potential increase in emissions under the project relative to those assumed for the ’97 CAP and identifies that effect as significant.

U29 As discussed in response to Comment U14, by 2005, power plant emissions of ozone precursors and PM-10 (and its precursors) would be less than under 1999 Baseline conditions. Therefore, power plant emissions would not contribute to the cumulative effect of increased emissions from additional growth and development in the Bay Area. However, prior to 2003, the DEIR predicts that Bay Area power plant emissions would be significantly greater than those foreseen in the ’97 Clean Air Plan. The essence of this project-specific impact is that power plant emissions may well increase relative to 1999 Baseline conditions, at least during the first couple of years after divestiture (1999 and 2000), when the ’97 Clean Air Plan anticipated a decrease, and may not decrease at the rate assumed in the ’97 Clean Air Plan after that time. The latter effect, i.e., decreases that do not match the rate that is anticipated, is referred to by the commenter as "minimization of reduction." (The DEIR describes this effect beginning with the last paragraph on page 4.5-80.) The commenter’s conclusion that such effects would be significant is consistent with the DEIR, which also concludes that the effect would be significant, at least through 2002. The one main difference is that the commenter characterizes the effect as "cumulative," while the DEIR identifies it as a project effect. Please see response to comment U16 for information concerning the regional air quality plans and their connection to future trends in ozone and PM-10 levels. Please also see response to Comment U27 concerning comparison of PG&E’s continued ownership beyond 1999 to the proposed sale of the plants.

U30 The DEIR notes on page 4.5-16, second full paragraph, the recent decision by U.S. EPA to re-designate the Bay Area back to "nonattainment" for the national one-hour average ozone standard, which triggers the need for a revised SIP. The revised SIP must be adopted by the Association of Bay Area Governments, BAAQMD, and the Metropolitan Transportation Commission and submitted to U.S. EPA, through the California Air Resources Board, by June 15, 1999. The DEIR identifies the increase in power plant NO_x emissions as a significant, unavoidable impact of the project (see Impact 4.5-5 on page 4.5-77 of the DEIR). The discussion of cumulative power plant emissions of ozone precursors has been revised; please see the response to Comment U14. Also, please refer to the discussion of the distinction and application of Best Available Retrofit Control Technology (BARCT) (existing sources) and Best Available Control Technology (BACT) (new sources) in response to Comment U5.

U31 This apparent inconsistency can be explained through reference to the different averaging periods of the various ambient air quality standards. To calculate worst-case concentrations of carbon monoxide, nitrogen dioxide, and sulfur dioxide for averaging periods from one-hour to 24-hours, the maximum hourly emissions rates from Potrero Power Plant sources were used in combination with actual meteorological data compiled at the plant. This essentially assumes that on a maximum day the power plant would operate at full capacity for periods including 24-hours, and since the maximum emissions rate at full capacity would be the same under all scenarios in a given year (such as 1999 Baseline and 1999 A-Max), then it follows that the corresponding maximum concentrations would also be the same. It should be noted that Table 4.5-29 indicates that maximum concentrations of these pollutants for these averaging periods (i.e., including the power plant increment and the background increment) would be less than the corresponding ambient air quality standards, and the impact would therefore not be significant.

In contrast, the concentration modeling method used to estimate annual averages took into account the different annual capacity factors, which were also used in developing the annual emissions estimates shown in Table 4.5-23 of the DEIR. Thus, the changes in estimated annual-average concentrations of nitrogen dioxide and sulfur dioxide from the Potrero plant, as shown in Table 4.5-29 (column Titled "1999 Analytical Maximum") of the DEIR, are roughly proportional to the annual emissions changes shown in Table 4.5-23. The resulting maximum annual concentrations of these pollutants would be less than the corresponding standards, and the impact would therefore not be significant.

U32 As explained in response to Comment U31, the changes in annual emissions estimates shown in Table 4.5-23 of the DEIR are reflected in changes in annual-average concentration estimates shown in Table 4.5-29 (not for lesser averaging periods).

U33 See response to Comment F29.

U34 Although there is no standard significance threshold for cancer risk, the California Air Pollution Control Officers Association (CAPCOA) has established the level of 10 in a million for the Air Toxics "Hot Spots" legislation (AB2588) as a conservative significance threshold for assessing health effects. This level was identified in the DEIR under Significance Criterion #3 on page 4.5-50. The DEIR takes into consideration the changes in cancer risks from the plants in future years under divestiture, and it states on pages 4.5-71 and 72 that cancer risks would change in proportion to the amount of fuel use that is caused by increases in energy generation. The changes in carcinogenic risks that are reported in the DEIR take into consideration the additional fuel use. The incremental risks from the PG&E plant emissions under divestiture are shown on Table 4.5-34 of the DEIR. These incremental risks are no greater than 0.1 in a million. The cumulative health risks which are reported in response to comment B6 show that the health risks, including other proposed plants near the Pittsburg and Contra Costa facilities, are also less than one in a million. These minimal incremental health risks are not cumulatively considerable (even in light of the existing setting of concern) since they are de minimus conditions, i.e., the environmental conditions would be essentially the same with or without the project.

The comment refers to studies that indicate higher cancer levels for certain population sectors. However, none of the studies indicate the causes of these increased rates, nor do they establish a linkage between ambient air pollution levels in the region and health effect endpoints.

U35 Table 4.5-10 of the DEIR shows that in 1992 and earlier, measurable emissions of metals and PAHs, as well as benzene and formaldehyde, occur as a result of burning fuel oil and natural gas in the steam boilers. Annual emissions of these substances in 1987 are the greatest, because of greater use of fuel oil. In 1990 and 1992, annual emissions of metals and PAHs decrease, and for some pollutants, the emissions are below measurable levels, which is indicated by the dashed line in the table entry. The levels that are below measurable levels vary by pollutant, partly because of different measurement techniques for each pollutant. However, any trace quantities below measurable levels are so low that, even as a total, they would not contribute to the exposure level risks in any measurable amount, which is less than 0.1 in a million. These emissions again would occur as a result of burning fuel oil for part of the year and would not contribute to chronic exposure levels. For 1995, the table shows emissions of benzene and formaldehyde, because only natural gas was burned in the boilers, due to BAAQMD Regulation 9, Rule 11. Annual emissions from the standby turbines at Potrero are immeasurably small and do not significantly affect total annual emissions from the plant. Thus, after 1995, the toxic pollutants of concern for chronic exposure are benzene and formaldehyde.

The DEIR (pages 4.5-71 and 4.5-72) states that annual emissions of benzene and formaldehyde for 1999 and 2005 will increase because of increases in annual natural gas usage. The health risk changes from these emissions changes in 1999 and 2005 are shown in Table 4.5-34. The table shows that at the Potrero plant, maximum cancer risks would increase from 0.17 in a million in 1999 Baseline to 0.23 in a million for 1999 A-Max. Under the 2005 A-Max scenario, the maximum risks are estimated to increase to 0.28 in a million. These maximum risk estimates are well below the significance threshold of 10 in a million.

U36 Table 5.1 is a list of existing, proposed, and planned projects that would be located within a one-mile radius of each of the power plants, including the Potrero plant. In the case of the Potrero plant, the information was provided by the San Francisco Planning Department and the Port of San Francisco (see response to Comment F51 regarding additions to Table 5.1). Because no site has been identified and no specific project proposal for a new facility has been received, it was not appropriate to include such a speculative project in Table 5.1. However, the DEIR does assess the cumulative environmental impacts of a new power plant in San Francisco. The 2005 Cumulative Analytical Maximum Scenario assumes construction and operation in or near San Francisco of new generation facilities totaling 480 MW. Section 5.3.3 (commencing on page 5-38) examines the cumulative impacts of Variant 1, which includes construction of a new 240 MW power plant in northern San Mateo County or within the City and County of San Francisco. The discussion in Section 5.3.3 acknowledges that the plant could be located on the same site as, or adjacent to, the Potrero Power Plant, and could be considered an expansion of that plant.

If the Potrero Power Plant were repowered, for cumulative analysis purposes, the impacts would be essentially the same as the impacts identified for construction of a new 480 MW at or near the Potrero plant. See response to Comment U37 for additional information. Project-specific impacts of repowering the Potrero plant or constructing new generating facilities would be evaluated in the project-specific environmental review that would be required under CEQA at the time such a project were proposed.

U37 As noted in response to Comment U36, the DEIR does in fact anticipate that the additional generating capacity could be built adjacent to or on the Potrero Power Plant site. The commenter correctly cites the DEIR text regarding a new owner’s interest in repowering the Potrero Power Plant. Based on projected demand, the DEIR forecasts that an additional 480 MW of generation would be needed by 2005 to support the closure of the Hunters Point Power Plant. While the new owners of the Potrero Power Plant might elect to expand capacity at the plant, they would only do so if construction of any other proposed new plant supplying 480 MW did not proceed. There are several reasons for this. From a regulatory standpoint, public agencies with jurisdiction over new generating facilities, such as the Bay Area Air Quality Management District (BAAQMD) and the CEC, have demonstrated a reluctance to approve excess generating capacity for environmental and other reasons. For example, if the existing Potrero Power Plant were to continue operating, a new 480 MW facility were constructed, and the Potrero Power Plant Units 1 and 2 were repowered, San Francisco would have generating capacity beyond its anticipated demand in 2005. In the unlikely event that significant excess generating capacity were to be approved in San Francisco, the Potrero Power Plant owner would be precluded from exporting excess power to other markets due to the same transmission constraints that limit the amount of power that can be imported into San Francisco. Within the San Francisco market, since the lowest bidder would prevail, power generated at the Potrero Power Plant’s older, less efficient (and hence more costly) units would be unable to compete with newer generating facilities. Thus, even absent the regulatory impediments, there would be strong economic disincentives for the owner of the Potrero Power Plant to expand its capacity if other new 480 MW facilities were also built.

The DEIR examined the potential 2005 cumulative environmental effects from continued operation of the Potrero Power Plant along with the addition of a new 480 MW power plant. The resulting modeled capacity factors are presented in Table 5.2 on page 5-17. For the reasons discussed above and in response to Comment U36, this analysis also covers the potential environmental effects that would result from repowering of the Potrero Power Plant.

U38 The cleanup of potentially contaminated soils at the plant sites, a beneficial environmental impact, could be advanced by the project. No specific remediation plan has been proposed at this time and the nature and extent of any remediation have not been determined. Soil remediation activities have, in general, the potential to affect surface water quality since the work typically includes disturbing soils. However, as explained on page 4.4-14, soil remediation activities are highly regulated and it is expected that the regulations for erosion and water quality control described in Section 4.3 and the regulations governing hazardous wastes described in Section 4.9.1 would be sufficient to protect water quality so that any impacts would be at a less-than-significant level. Please also see responses to Comments F40 and F41.

U39 The EIR does not claim to substitute the NPDES permit for an environmental analysis. It does, however, assume that the RWQCB issued the permit conditions such that no significant impact to water quality would occur from operation of the facilities at the maximum generating capacity of each plant. Therefore, the continued compliance with those conditions by the new owners, regardless of the plant operational levels, would result in less than significant impacts.

U40 The plant operating conditions reasonably foreseen and analyzed in this EIR project less-than-significant impacts on water quality. There is no reason to believe that new owners of the plants would have any increased tendency (over PG&E) to violate applicable permits. If the RWQCB decides at any point in the future that the operating conditions are detrimental to the environment or that there is a reason to suspect that the plant will not be able to meet discharge requirements that adequately protect marine resources, the RWQCB would modify its permit conditions. The RWQCB, as a trustee and regulatory agency, is properly entrusted with this role by the state.

U41 The DEIR does conclude that air quality impacts would be significant. Specifically, the DEIR concludes that the project would result in power plant NO_x emissions that are sufficiently different from those included in the regional air quality plan, and therefore, the project can be determined to be inconsistent with that plan. The DEIR further concludes that this impact would be unavoidable (albeit temporary) (page 4.5-81). The commenter is correct in noting that increasingly stringent NO_x controls would likely be installed, over time, to reduce overall power plant NO_x emissions with or without the project, but the commenter calls for an accelerated schedule for installation of these controls. However, an implementation schedule has already been established through BAAQMD Regulation 9, Rule 11, which presumably represents the most aggressive, and yet feasible, schedule possible considering such factors as cost and reliability issues.

U42 Please see responses to Comments F30 and 1-12. If the Potrero plant were repowered, the repowered facility would be subject to BAAQMD Rules and Regulations, which may require implementation of Best Available Control Technology (BACT). The EIR’s use of the Analytical Maximum capacity factors for the plants under new ownership ensures that emissions estimates were not minimized, and instead are likely overstated. The air quality analysis presents and analyzes data in a straightforward manner that meets or exceeds the requirements of CEQA and pertinent case law.

 

TOP	Return to Table of Contents