FEDERAL LAND MANAGERS' AIR QUALITY RELATED VALUES WORKGROUP (FLAG)
PHASE I REPORT
(December 2000)

D. 2. Visibility

a. Introduction

This chapter describes methods for analyzing the impacts on visibility from new or modified air pollution sources. This includes sources that fall under the purview of the Prevention of Significant Deterioration (PSD) regulations and sources that are being analyzed for Environmental Impact Statements under the National Environmental Policy Act (NEPA). The basis for some of the decisions outlined in this chapter is section 169A of the Clean Air Act. The opening statement of this section states: "Congress hereby declares as a national goal the prevention of any future, and the remedying of any existing, impairment of visibility in mandatory class I Federal areas which impairment results from manmade air pollution." Under the regulations promulgated for visibility protection (40 CFR §51.301 (x)) visibility impairment is defined as "…any humanly perceptible change in visibility (visual range, contrast, coloration) from that which would have existed under natural conditions." The remainder of this chapter describes methods that allow for new source growth to be analyzed against the constraint of preventing visibility impairment as defined in 40 CFR §51.301 (x), that is, new source growth should not allow any humanly perceptible change in visibility as compared against natural conditions.

There are three different visibility impact thresholds that are discussed: levels of concern, analysis thresholds, and decision thresholds. These are all interrelated. The levels of concern are visibility impact levels that would alert the FLM to a need for closer scrutiny. The analysis thresholds parallel these levels of concern in that if visibility impacts approach the levels of concern, the FLM would need to see further analyses to make an informed judgement about those impacts. The decision thresholds correspond to the visibility impacts, below which the FLM is not likely to object to an increase in visibility impairing pollutants. It is important to note that the decision thresholds can not be absolute; the FLM is required to make a determination on a "…case-by-case basis taking into account the geographic extent, intensity, duration, frequency and time of visibility impairments…" (40 CFR §51.301 (a)). However, the decision thresholds should be useful as an initial benchmark for analysts to judge whether visibility impacts would likely cause the FLM to object to a proposed action.

Natural Conditions

Comparing the impacts of new source growth against natural conditions implies that natural conditions are defined. At the time of this writing (December 2000) the EPA is working on defining natural conditions in support of their visibility regulations, but that work has not yet been completed. An estimate of natural conditions has been made (NAPAP, 1990). These estimates are only differentiated by the broad categories of the eastern and western United States. FLAG has adopted the appropriate aerosol concentrations from the NAPAP as estimates of natural conditions for each Class I area (Appendix 2.B). These estimates are a surrogate to be used until more definitive values for natural conditions are established.

Visibility Impairment

Before proceeding with the discussion, it is useful to identify the ways that visibility impairment can manifest itself. First, the pollutant loading of a section of the atmosphere can become visible, by the contrast or color difference between a layer or plume and a viewed background, such as a landscape feature or the sky. The second way that visibility is impaired is a general alteration in the appearance of landscape features or the sky, changing the color or the contrast between landscape features or causing features of a view to disappear. The first phenomenon is commonly referred to as plume impairment, whereas the second phenomenon is sometimes referred to as uniform haze impairment. As plumes are transported within a stable atmospheric layer, they may become a layered haze. As plumes and other more diffuse emission sources are transported and become well mixed in the atmosphere, they may develop into a uniform haze.

Visibility Parameters

The analysis methods for new source growth, described in this chapter, only deal with the visibility effects of discrete plumes and the aggregation of discrete plumes. The difference in these phenomena, as treated in this chapter, is whether the visibility effect is primarily seen as a section of the atmosphere which exhibits a change in contrast or color as compared with a viewed background, or whether the effect is due to an alteration of the appearance of the background features themselves. For the first situation, the contrast (C) and color difference index (ΔE) of the plume and the viewing background are calculated. For the second situation, the change in atmospheric light extinction (Δb_ext), relative to natural conditions, is calculated. An approximation for which situation applies is the distance from the point of emission. (Distance serves as an indicator of where steady state conditions may apply.) The visibility impairment from sources within 50 kilometers of a view is usually calculated using contrast and color difference, where visibility impairment from sources greater than 50 kilometers from a view, or the aggregation of a number of plumes, regardless of distance, is usually calculated using the change in light extinction. The distance approximation is useful for distinguishing these two phenomena; the terms "near field" and "distant/multi-source" are sometimes used in the remainder of this document to make this distinction. More information on visibility parameters can be found in Appendix 2.A.

Levels of Concern

The crucial level of concern for visibility impairment is whether it is humanly perceptible as compared against natural conditions (40 CFR §51.301(x)). As noted above, different visibility parameters are applied for different phenomena. A summary of the thresholds of perceptibility for the case of a plume viewed against a background indicates that contrast values (C) of ±0.01 to ±0.05 (note that the sign denotes whether the plume is brighter (+) or darker (-) than the background) are perceptible (NAPAP, 1990). A change in the color difference index (ΔE) of less than 1 to 4 has been identified as the range of perceptibility for this parameter. The Workbook for Plume Visual Impact Screening and Analysis (USEPA, 1992a) suggests that a level of 0.05 for the absolute value of contrast (|C| = 0.05) and ΔE = 2 be used as thresholds in screening analyses; these levels were set in the mid-range of the perceptibility thresholds, in part, because of the conservative nature of the screening calculations. These levels also constitute the FLM's level of concern for screening analyses of plumes viewed against a background. Under circumstances of a more refined analysis, |C| = 0.02 and ΔE = 1 are the levels of concern (USEPA, 1992b). These levels are usually applied for near field analyses where single sources are locating within 50 kilometers of a view.

For the case of visibility impairment which changes the appearance of a viewed background feature, thresholds of perceptibility, where a just noticeable change occurs in the scene, have been found to correspond to a change in extinction (Δb_ext) as low as 2% under ideal conditions, up to 20% (NAPAP, 1990; Pitchford and Malm, 1994). A Δb_ext of 5% will evoke a just noticeable change in most landscapes (NAPAP, 1990). The FLMs are concerned about situations where a change in extinction from new source growth is greater than 5% as compared against natural conditions. Changes in extinction greater than 10% are generally considered unacceptable by the FLMs and will likely raise objections to further pollutant loading without mitigation. These levels are usually applied for distant/multi-source analyses where sources are located more than 50 kilometers from a view or for analyzing the visibility impairment from an aggregation of plumes from multiple sources, regardless of distance.

Cumulative Analyses

A cumulative effects analysis of new source growth (defined as all PSD increment-consuming sources) on visibility impairment should be performed. The change in extinction (Δb_ext) will usually be the visibility parameter examined. The FLMs recognize that cumulative analyses of the effects of new source growth on visibility impairment have only rarely been carried out. Until cumulative analyses are performed for an area, the FLMs are suggesting some analysis thresholds to either trigger a cumulative analysis or allow a source to be permitted if its impact is below certain prescribed levels.

If a cumulative analysis has already been performed for the area, or if other considerations (i.e., NEPA, PSD increments, or other AQRV analyses) require that a cumulative analysis be performed for the proposed source, then the visibility impacts of the source are expected to be considered as part of the cumulative visibility impairment, as compared against natural conditions. When these conditions are met, the inclusion of the proposed source is expected regardless of the predicted visibility impairment of the source, unless its impacts are considered below de minimis.

Analysis Thresholds for New Cumulative Analyses

The analysis thresholds outlined here are interim levels to be used until such time as cumulative analyses are conducted for an area. Change in extinction (Δb_ext) is usually the visibility parameter analyzed for a cumulative analysis. If the visibility impact of a proposed project is below 0.4%* change in extinction, the impacts would be considered below de minimis and would not require further analysis. For situations where a cumulative visibility analysis has not been done or is not required because of other considerations, the following analysis thresholds will apply. If the visibility impact of a proposed source is less than a 5% change in extinction a cumulative analysis would not be expected. For visibility impairment predicted to be above 5%, but less than 10%, change in extinction from a proposed source, a cumulative analysis is expected. If the visibility impairment is predicted to be greater than 10% from a proposed source, the FLM is likely to object to the project regardless of other source growth, unless there is mitigation.

*The de minimis level of 0.4% is defined as 4% of the unacceptable change in extinction (i.e. 10%), paralleling the discussion of significant impact levels in the proposed new source review modifications. (FR 61 38291-38293)

---

Decision Thresholds

Each determination of whether the impacts from a new source or major modification will be considered adverse must, by regulation, be made on a case-by-case basis (40 CFR §51.301(a)). Therefore, the decision thresholds specified here are strictly a guideline. More refined visibility analyses may indicate that the visibility parameters used (i.e., C, ΔE, Δb_ext) do not adequately characterize the visibility for a particular situation; the FLMs will consider such information in making their decision. The decision thresholds parallel the FLM levels of concern. For near field situations where a section of the atmosphere is polluted and is viewed against a scenic background, screening analysis values of contrast with an absolute value less than 0.05 (|C| < 0.05) would not likely result in an objection by the FLM. Similarly, a value of ΔE < 2 from a screening analysis would not likely result in an objection. If a refined near field analysis is performed, values of |C| < 0.02 or ΔE < 1 would not likely result in an objection by the FLM.

For distant or multi-source situations, if a cumulative visibility analysis has not previously been conducted and is not required for other analyses, a single-source change in extinction less than 5% would not generally trigger a need for a cumulative analysis. Under those circumstances, the FLM would not likely object to the proposed action. If the forecast single-source contribution to extinction is between 5% and 10%, or if a cumulative analyses is required or already exists, a special decision threshold applies. If the visibility impairment from the proposed action, in combination with cumulative new source growth, is less than a change in extinction of 10% for all time periods, the FLMs will not likely object to the proposed action. If the visibility impairment from the proposed action, in combination with cumulative new source growth, is greater than or equal to a change in extinction of 10% for any time periods, the FLMs will likely object to the proposed action, unless the contribution from the proposed action is less than a de minimis value of 0.4% for these time periods.

Relationship to Regional Haze Rule

The FLAG recommendations are complimentary to the regional haze rule. However, the visibility recommendations of FLAG are intended for new source review and NEPA type applications, whereas the regional haze rule addresses the effects of existing sources of visibility impairment in conjunction with new source review. The FLAG recommendation is designed to prevent new sources from causing visibility impairment, and the criteria for developing these recommendations do not necessarily apply to existing sources. At the time of this writing, new source review is an ongoing effort, but it will be several years before State Implementation Plans (SIPs) under the regional haze rule are submitted. If the new visibility SIPs adequately account for new source growth, the FLMs may reconsider the FLAG recommendations.

The visibility parameters for cumulative impact analysis, outlined here, are related to those in the regional haze rule. However, an assumption inherent in regional haze is that the pollution is fairly evenly distributed over a broad geographic extent. By contrast, the analysis techniques, described herein, at most deal with the aggregation of a subset of the plumes that might affect regional haze, but do not meet the criteria of being a regional haze.

The levels of concern and de minimis levels described in this document were arrived at, in part, with the knowledge they apply to a limited number of sources under new source review and that the analyses are always compared to natural conditions. The de minimis levels described here should not be used for determining whether emissions from an existing source are reasonably anticipated to cause or contribute to visibility impairment. Those criteria have been laid out in the regulations (40 CFR §51 Subpart P Protection of Visibility) and through interpretations of those regulations by EPA and courts.

While there are some distinct differences between this document and the regional haze rule, there are also some similarities. One of these is the need for conducting a cumulative assessment of visibility impairment. This will include the need for evaluating the effects of sources beyond an individual state's boundaries. Therefore, it is anticipated that when modeling centers are established for SIP development work, the tools they use may be applicable to analyzing both existing impairment as well as the potential impacts of new source growth.

b. Analysis Techniques

There are two fundamentally different approaches one could adopt to determine visibility impairment. One is a technically rigorous, complex, and situation-specific method, while the other is a more generalized approach. The more rigorous approach requires determination of particle concentrations and size distributions, calculation of particle growth dynamics, and application of Mie Theory to determine the optical characteristics of the aerosol distribution. Sophisticated radiative transfer models are then applied, using aerosol optical characteristics, lighting and scene characteristics, and spatial distribution of the pollutants to calculate the path and wavelength of image-forming and non-image-forming light that reaches a specific observer from all points in the scene being viewed.

While such a detailed analysis may be useful for assessing specific cases, it is usually impractical for situations in which visibility could be experienced in a nearly infinite variety of circumstances. Practical limitations frequently dictate that it is more reasonable to use a generalized approach to determine the change in extinction by using bulk-averaged aerosol-specific extinction efficiencies rather than trying to reproduce the complex optical phenomena that may occur in the atmosphere.

Consequently, FLAG recommends the generalized approach for determining the effects on visibility from a proposed new source's emissions. The procedure is to estimate the atmospheric concentrations of visibility impairing pollutants, apply representative visibility parameters, calculate the change from specified reference levels, and compare this change with prescribed threshold values.

FLAG is using estimates of natural conditions as reference levels for Class I visibility analyses. Comparison with natural conditions will help ensure that those conditions will not be impaired in keeping with Section 169A of the CAA. Because of the different requirements of the two modeling approaches discussed below, natural conditions must be expressed using two different metrics:

• Standard visual range (visual range adjusted to a Rayleigh condition of 10 Mm^-1), for near field modeling. Present EPA guideline visibility models traditionally accept visibility conditions expressed in these terms.
• Extinction, for distant/multi-source modeling. Visibility conditions should be expressed in terms of the averaged extinction efficiencies of the individual atmospheric constituents that comprise the total extinction. The relative humidity effects of the hygroscopic particles must be accounted for when the change in extinction is calculated.

Information needed to calculate the above indices is provided in Appendix 2.B for all 156 Class I areas for which visibility is an important attribute. If estimates are needed for Class II areas, the FLM can provide them.

c. Air Quality Models and Visibility Assessment Procedures

The modeling discussion will be divided into two parts to address the very different requirements for 1) near field modeling where plumes or layers are compared against a viewing background and 2) distant/multi-source modeling for plumes and aggregations of plumes that affect the general appearance of a scene. Note that both of the above analyses might apply depending on the source's proximity to all portions of the Class I area or multiple Class I areas.

Near Field Analysis Technique for Analyzing Plumes or Layers Viewed Against a Background

The Model (Near Field - Steady State Conditions Applicable)

EPA has recommended a methodology to assess impacts due to coherent plumes. A guideline, for when these steady state conditions apply, is the distance from the source to the view of concern. This technique is usually applied for sources locating less than 50 km from a Class I area. Applicants must model their potential plume impacts using the screening model, VISCREEN (USEPA, 1992a), or, if the next level of analysis is called for, PLUVUE II (USEPA 1992b and 1996c). Both of these models use steady-state, gaussian-based plume dispersion techniques to calculate one-hour concentrations within an elevated plume. These two models calculate the change in the color difference index (ΔE) and contrast between the plume and the viewing background. Values of ΔE and plume contrast are based on the concentrations of fine primary particulates (including sulfates), nitrogen dioxide (NO₂), and the geometry of the observer, target, plume, and the position of the sun. PLUVUE II also allows consideration of the effects of secondarily formed sulfates. Plume contrast results from an increase or decrease in light transmitted from the viewing background through the plume to the observer. The specifics of the emission scenarios and plume/observer geometries for modeling should be selected in consultation with the appropriate FLM. At the present time there is no recommended procedure for conducting analyses of multiple sources with these modeling tools, so multiple coherent plumes must be treated individually, or combined into a representative single source if reasonable.

The Recommended Prescription (Near Field - Steady State Conditions)

Until better modeling tools are available, FLAG recommends using the present EPA techniques for plume visual impact screening analyses (USEPA 1992a). However, unlike those procedures, which suggest the use of current average annual visibility conditions, FLAG recommends that the visual range corresponding to natural conditions be used to generate the hourly estimates of ΔE and plume contrast. FLAG recommends this change in order for the analysis technique to be consistent with the national visibility goal. For screening-level analyses, FLAG recommends the use of the annual average reconstructed natural conditions given in the last column in Table 2.B-1 in Appendix 2.B. The table entry gives the specified reference level (including the effects of relative humidity) expressed in Mm^-1. The conversion to standard visual range can be made using Equation 1 in Appendix 2.A. For the refined analyses, the reconstructed natural condition is derived from the relative humidity used in the modeling, the corresponding relative humidity adjustment factor (Table 2.A-1), and estimated natural aerosol concentrations (Table 2.B-1).

If a screening analysis of a new or modified source can demonstrate that its emissions will not cause a plume with any hourly estimates of ΔE greater than or equal to 2.0, or the absolute value of the contrast values (|C|) greater than or equal to 0.05, the FLM is not likely to object to the issuance of the PSD permit based on near field visibility impacts and no further near field visibility analyses will be requested. More refined analyses (i.e., PLUVUE II) would be undertaken if the above conditions are not met and would be compared against lower levels of concern; the FLM would not likely object if ΔE < 1.0 and |C| < 0.02.

If the estimated plume parameters exceed the aforementioned values, the FLM would rely on a case-by-case effects-based test (NPS 1993), taking into account magnitude, frequency, duration, and other factors, to decide whether to make an adverse impact determination.

Distant/Multi-Source Techniques for Analyzing Whether a Plume or an Aggregation of Plumes Alters the General Appearance of a Scene

This application is generally more complex than the near field, coherent plume modeling analyses and the guidance from EPA is less definitive, though it is evolving. The modeling system must include the capability to assess single and multiple sources in a temporally and spatially varying meteorological domain, accommodate modeling domains measuring hundreds of kilometers, include rough and complex terrain, provide pollutant concentration estimates for averaging times from one-hour to annual, and address inert and secondarily formed pollutants and dry and wet deposition. In the early 1990s the FLMs and the EPA recognized the need for a consistent, technically credible technique to estimate contributions to air quality of multiple new sources locating more than 50 km from Class I areas. Toward that end, the Interagency Workgroup on Air Quality Modeling (IWAQM) was established to develop a modeling protocol for this application. FLAG proposes to rely on the IWAQM recommendations and modeling guidance for long range pollutant transport (present guidance, USEPA 1998^*). This technique is usually applied when sources are located more than 50 kilometers from portions of a Class I area, when an aggregation of plumes may impact an area, or when the assumptions inherent in steady state visibility models do not apply.

The Model (Distant/Multi-Source)

Revised IWAQM guidance (USEPA 1998^*) recommends non-steady state air quality modeling systems for screening and refined analyses. The IWAQM recommendations are adaptations and refinements of the CALPUFF dispersion modeling system, including the CALMET meteorological model (USEPA 1996a, http://www.src.com/calpuff/calpuff1.htm). This modeling system consists of diagnostic meteorological models, a gaussian puff dispersion model with algorithms for chemical transformation, wet and dry deposition, and complex terrain, and a post processor (CALPOST) for calculating concentration and deposition fields and visibility impacts.

The modeling systems/techniques outlined in this recommendation provide ground level concentrations of visibility impairing pollutants. These concentrations can then be used to calculate the extinction due to these pollutants, using the relationships outlined in Appendix 2.A. The results should be compared against a reference level derived from aerosol information (relative humidity adjusted hygroscopic and non-hygroscopic concentrations plus Rayleigh extinction) given in Appendix 2.B for each Class I area. This reference level is a function of relative humidity. To achieve the best temporal and spatial resolution, relative humidity data included in the meteorological data base of the air quality model and the data provided in Table 2.A-1 is the preferred basis for making the necessary calculation of the relative humidity adjustment term f(RH) for refined visibility analyses. The approach, for screening level analyses, is to use the quarterly averaged reference levels given in Table 2.B-1 that are based on spatially interpolated seasonal relative humidity values and empirically derived f(RH) adjustment factors (IMPROVE 2000). In either approach, the same relative humidity adjustment factor, f(RH), is applied to determine both the reference level and the effect of the incremental increase associated with the new source(s). An example model application is given in Appendix 2.C.

For the purposes of the following prescription, FLAG recommends basing the analyses on block 24-hour averages (i.e., daily) of modeled visibility. The 24-hour average was selected over the 1-hour average time because:

• Our confidence in model performance for 24-hours is higher than for shorter time periods.
• The combined visibility effect of emissions from multiple sources transported over long distances is better represented over 24-hours than for shorter time periods.
• It avoids detailed day/night visibility considerations.
• It avoids developing and implementing site-specific, complex visibility analytical methods that are not available at this time (see discussion under Analysis Techniques).

---

^*At the time of this writing, USEPA is considering similar procedures for incorporation into the Guideline on Air Quality Models (40 CFR§51 Appendix W). This should be consulted for the latest information.

The Recommended Prescription (Distant/Multi-Source)

The FLMs are concerned with the cumulative effects of new source growth on visibility; cumulative analyses need to be conducted. The FLMs recognize, however, that few cumulative visibility analyses have been done, therefore, the following prescription is suggested. If a cumulative analysis has not been performed for an area and if a single project's visibility impairment, compared against natural conditions, is below certain analysis thresholds, then the FLMs are not likely to object to the project or ask that a cumulative analysis be performed before the project proceeds. If a cumulative analysis has already been done or if a cumulative analysis is required because of other considerations (i.e., increment consumption, NEPA, or other AQRVs), or if the analysis thresholds are exceeded, then the impacts of the proposed project are expected to be considered as part of a cumulative visibility analysis.

The prescription is as follows:

1. Calculate the single-source contribution. Compare results with the distant/multi-source Decision Threshold.

• Determine whether a cumulative analysis has been done for the Class I area(s) in question, and if it has been done, use the input files from the cumulative analysis to perform this step.
• If the estimated increase in visibility impairment attributed to the proposed project is <10%, compared against natural conditions, for at least one modeled day, then the FLM will consider the magnitude, frequency, duration, and other factors to assess the impact, but is likely to object to the issuance of the permit.
• If the estimated increase in visibility impairment attributed to the proposed project is <10%, then the analysis should

2. If a cumulative analysis does not exist, compare the single-source contribution with distant/multi-source Analysis Threshold and assess the need for a cumulative analysis.

• If a cumulative analysis does not exist, and if there are no other requirements for a cumulative analysis, and if a new or modified source can demonstrate that its contribution to a change in extinction is <5.0%, compared against natural conditions, for all days, then the FLM is not likely to object to the issuance of the PSD permit based on visibility impacts.
• If the single-source contribution to a change in extinction is <5.0% or if a cumulative analysis already exists or is required for some other reason, then the analysis should proceed to the next step and estimate its contribution to cumulative impacts.

3. Conduct a cumulative analysis and compare results with cumulative, distant/multi-source Decision Threshold.

• If cumulative change in extinction is <10%, for all modeled block 24-hour periods, and the new source contributes at least a 0.4% change in extinction to any of these periods, then the FLM will consider the magnitude, frequency, duration, and other factors to assess the impact, but is likely to object to the issuance of the permit.
• If cumulative modeling results indicate that the effects from the combined sources are expected to cause a change in extinction that is < 10%, for all modeled block 24-hour periods, then the FLM is not likely to object to the issuance of the permit.

This prescription is portrayed schematically in Figure V-1.

d. Summary

FLAG has provided guidance in the form of recommendations, specific prescriptions, and interpretation of results for assessing visibility impacts near Class I areas (although this guidance is generally applicable to Class II areas, as well). The guidance addresses assessments for sources proposed for locations near and at large distances from these areas. It also recommends impairment thresholds and identifies the conditions for which cumulative analyses of all increment-consuming sources would be necessary. The key components of the recommendations are highlighted below.

In general, FLAG recommends that an applicant:

• Consult with the appropriate regulatory agency and with the FLM for the affected Class I area(s) or other affected area for confirmation of preferred procedures and for the need for a cumulative analysis.
• Obtain FLM recommendation for the specified reference levels (estimate of natural conditions) and, if applicable, FLM recommended plume/observer geometries and model receptor locations.
• Apply the applicable EPA Guideline, steady-state models for regions within the Class I area that are affected by plumes or layers that are viewed against a background (generally within 50 km of the source).

Calculate hourly estimates of ΔE and plume contrast, with respect to natural conditions, and compare these estimates with the thresholds given in Section D.2.c.

• For regions of the Class I area where visibility impairment from the source would cause a general alteration of the appearance of the scene (generally 50 km or more away from the source or from the interaction of the emissions from multiple sources), apply a non-steady-state air quality model with chemical transformation capabilities (refer to IWAQM guidance documents), which yields ambient concentrations of visibility-impairing pollutants. At each Class I receptor:

Calculate the change in extinction due to the source being analyzed, compare these changes with the reference conditions, and compare these results with the thresholds given in Section D.2.c.

If necessary, calculate the cumulative change in extinction due to new source growth.

• Appendix 2.A
• Appendix 2.B
• Appendix 2.C