ALTERNATIVE .SHORT-TERM N02 STANDARDS:

        SECOND ROUND ANALYSES
ENERGY AWD ERIVIROMiytEfliTAL ANALYSIS, It^C.
1111 North 19th Street
Arlington, Virginia 22209
(703)528-1900

-------
       ALTERNATIVE SHORT-TERM N02 STANDARDS:

               SECOND ROUND ANALYSES
                   Submitted to:

   Office of Air Quality Planning and Standards
          Environmental Protection Agency
   Research Triangle Park, North Carolina  27711
                   Submitted by:

Dale L.  Keyes, James H. Wilson, Jr., Vivian M. Daub
      Energy and Environmental Analysis, Inc.
         1111 North 19th Street, 6th Floor
            Arlington, Virginia   22209
                  August 22, 1979

-------
                           TABLE OF CONTENTS
1.   Summary



1.1  Findings and Conclusions



1.2  Sensitivity Analysis and Caveats



1.3  Comparison with the Previous Analysis



1.3.1  Methodology and Impact Data



1.3.2  Results



2.   Background Investigations



2.1  Frequency Distribution of Hourly N02 Concentrations



2.1.1  Sites Investigated



2.1.2  Best Fit Functional Forms of the Distributions



2.1.3  Peak-to-Mean Ratios



2.1.4  Implications for the Regulatory Analysis



2.2  Spatial Variation in NO. Concentrations



2.2.1  Regional Scale Variation in NO. in Los Angeles



2.2.2  Microscale Gradients in NO- Downwind of a Roadway



2.2.3  Implications for the Regulatory Analysis



3.   Attainment Strategies for Existing Sources



3.1  Cost and Effectiveness of Point Source Controls



3.1.1  Analytical Procedures



3.1.1.1  Screening NEDS and Selecting Sources

-------
                     TABLE OF CONTENTS (CONTINUED)
3.1.1.2  Modeling Conditions



3.1.1.3  Emission Control Options



3.1.1.4  Least Cost Control Strategies



3.1.2  Estimated Control Costs for Alternative Standards



3.2  Cost and Effectiveness of Area Source Controls



3.2.1  Mobile Source Controls



3.2.1.1  Inspection and Maintenance



3.2.1.2  Transportation Control Measures



3.2.1.2.1  TCM's Used in the Analysis



3.2.1.2.2  Other TCM's



3.2.1.3  Alternative Emission Standards



3.2.1.3.1  Automotive Emission Standard of 0.4 Grams per Mile



3.2.1.3.2  Alternative Emission Standards for Trucks



3.2.2  Stationary Area Source Controls



3.2.3  Sensitivity Analyses



3.2.3.1  The Effect of Growth on Nonattainment



3.2.3.2  The Effect of Hydrocarbon Control on NO- Levels



3.2.3.3  The Effect of FMVCP Waivers for Diesels



3.2.4  Summary



3.3  Area and Point Sources Combined



4.   New Source Controls

-------
                     TABLE OF CONTENTS (CONTINUED)
4.1  Cost of New Source Performance Standards



4.1.1  Utility Boilers:  Coal-fired



4.1.2  Utility Boiler:  Oil and Gas-fired



4.1.3  Industrial Boilers:  Coal-fired



4.1.4  Industrial Boilers:  Oil- and Gas-fired



4.1.5  Stationary Gas Turbines



4.1.6  Reciprocating Internal Combustion Engines



4.1.7  Nitric Acid Plants



4.1.8  Summary



4.2  Federal Motor Vehicle Control Program Costs



4.2.1  Light-Duty Vehicles



4.2.1.1  Initial Cost of Emission Control Systems



4.2.1.2  Maintenance Costs



4.2.1.3  Differences in Fuel Economy



4.2.1.3  The Use of Unleaded Fuel



4.2.2  Light-Duty Trucks



4.2.2.1  Initial Cost of Emission Control Systems



4.2.2.2  Annual Costs of Light Truck Controls



4.2.3  Heavy-Duty Trucks



4.2.4  Summary



4.3  Implications for PSD and New Source Siting

-------
                     TABLE OF CONTENTS (CONTINUED)
3.1.1.2  Modeling Conditions



3.1.1.3  Emission Control Options



3.1.1.4  Least Cost Control Strategies



3.1.2  Estimated Control Costs for Alternative Standards



3.2  Cost and Effectiveness of Area Source Controls



3.2.1  Mobile Source Controls



3.2.1.1  Inspection and Maintenance



3.2.1.2  Transportation Control Measures



3.2.1.2.1  TCM's Used in the Analysis



3.2.1.2.2  Other TCM's



3.2.1.3  Alternative Emission Standards



3.2.1.3.1  Automotive Emission Standard of 0.4 Grams per-Mle



3.2.1.3.2  Alternative Emission Standards for Trucks



3.2.2  Stationary Area Source Controls



3.2.3  Sensitivity Analyses



3.2.3.1  The Effect of Growth on Nonattainment



3.2.3.2  The Effect of Hydrocarbon Control on NO- Levels



3.2.3.3  The Effect of FMVCP Waivers for Diesels



3.2.4  Summary



3.3  Area and Point Sources Combined



4.   New Source Controls

-------
                     TABLE OF CONTENTS (CONTINUED)
4.1  Cost of New Source Performance Standards



4.1.1  Utility Boilers:  Coal-fired



4.1.2  Utility Boiler:  Oil and Gas-fired



4.1.3  Industrial Boilers:  Coal-fired



4.1.4  Industrial Boilers:  Oil- and Gas-fired



4.1.5  Stationary Gas Turbines



4.1.6  Reciprocating Internal Combustion Engines



4.1.7  Nitric Acid Plants



4.1.8  Summary



4.2  Federal Motor Vehicle Control Program Costs



4.2.1  Light-Duty Vehicles



4.2.1.1  Initial Cost of Emission Control Systems



4.2.1.2  Maintenance Costs



4.2.1.3  Differences in Fuel Economy



4.2.1.3  The Use of Unleaded Fuel



4.2.2  Light-Duty Trucks



4.2.2.1  Initial Cost of Emission Control Systems



4.2.2.2  Annual Costs of Light Truck Controls



4.2.3  Heavy-Duty Trucks



4.2.4  Summary



4.3  Implications for PSD and New Source Siting

-------
                     TABLE OF CONTENTS (CONTINUED)
5.  Economic Impacts



5.1  Electric Utilities



5.1.1  Industry Profile



5.1.2  Recent Financial History



5.1.3  Impact of NO  Control Costs on Ability to Raise Capital



5.1.4  Product Price Increases



5.1.5  Conclusion



5.2  Iron and Steel Industry



5.2.1  Profile



5.2.1.1  Production



5.2.1.2  Demand



5.2.2  Recent Financial History



5.2.3  Impact of NO  Control Costs



5.2.4  Product Price Increases



5.2.5  Conclusion



5.3  Urban and Community Impact Analysis

-------
                              1.  SUMMARY
This report communicates the methodology and findings of a second-round

study of potential impacts from setting alternative one-hour NO. standards.

The previous study was reported in three separate volumes in December

1978 (Keyes, et al, 1978a,b,c).


1.1  FINDINGS AND CONCLUSIONS

Following is a brief summary of the new findings for 1984:
                              Alternative One-Hour
                              NO., Standards (ppm)
                   .10      .15     .20     .25     .35     .50

Number of AQCR' s   132      64      36      16       4       0
Unable to Attain
the Standard

Capital Cost of
Control (Billions
of 1979 Dollars)
Point Sources
Area Sources
Total
5.2
3.7
8.9
1.1
3.2
4.3
.2
2.8
3.0
.1
2.6
2.7
Neg.
1.7
1.7
Neg.
0
0
Annual Cost of
Control (Billions
of 1979 Dollars)
  Point Sources

  Area Sources

  Total
1.1
1.0
2.1
.2
.9
1.1
Neg.
.7
.7
Neg.
.6
.6
Neg.
.3
.3
Neg.
0
0
These results include control requirements for both point and area

sources; they are based on dispersion modeling of individual point

-------
sources in NEDS and proportional modeling of area sources on an AQCR-
wide basis in 150 AQCR's.

The costs shown above may well increase by 1990 if a more stringent
automobile NO  emission standard (from 1.0 to 0.4 grams/mile) were
instituted in 1985.  Under this scenario, an additional initial invest-
ment of about 6.7 billion and an annual charge of $1.8 billion should
be included for each standard up to 0.35 ppm.  This would bring between
zero and six additional AQCR's into attainment depending on the stan-
dard being attained.  On the other hand, some existing sources will
be retired by 1990 thus reducing the point source costs to some
extent.

With respect 'to the controls required on point sources, sufficient
emission removal capability is or will be available to bring the
air quality impact from individual sources below all but the most
stringent one-hour standard (0.10 ppm).  Total capital costs are
less than $250 million for all standards above 0.20 ppm; they reach
over $1.1 billion at a standard of 0.15 ppm and over $5 billion
at a standard of 0.10 ppm.  Industrial boilers, utility boilers,
industrial furnaces, and I.C. engines are the source types most
commonly needing control, in that order.  From an economic impact
perspective, utilities and the iron and steel industry shoulder the
major burden, though the costs are relatively modest at a 0.20 ppm
standard:

                                   Capital       Annual
                                   (Millions of Dollars)
Gas and Electric Utilities           110           18
Iron and Steel                        71           11

-------
Area sources are both more  costly to control and less effectively
controlled.  Initial costs  range between $1.7 billion (0.35 ppm
standard) and $3.7 billion  (0.10 ppm standard) in 1984.  Again,
these costs may increase  by approximately $6.7 billion in 1990 if
a 0.4 gram/mile NO  standard is mandated for autos starting in 1985.
                  A
Even with this level of expenditure, several AQCR's can not attain
the required air quality  levels under all but the least stringent
standard (0.50 ppm).

The bulk of area source control costs will be incident directly on
households, in terms of automotive inspection and maintenance fees,
and, perhaps, increased new housing costs.  Moreover, if a 0.4 gram/
mile automotive standard  were instituted substantial increases in new
car costs would be experienced.  In addition to these area source costs,
cost passthrough in the form of higher electricity prices may be
experienced, though this  should be negligible.

Of the variuos control options listed, the 0.4 grams of NO /mile
                                           ,    •           ^^
automotive emission standard appears to be the least cost-effective.
The costs as noted above  are between 6 and 7 billion dollars (initial
investment) with only a small improvement in AQCR attainment status
realized.  On the other extreme, point source controls are relatively
modest and lead to attainment for all but the most stringent standard
(0.10 ppm).

1.2  SENSITIVITY ANALYSES AND CAVEATS
Analyses were conducted to  determine how sensitive the area source
cost and attainment estimates were to various factors.  Specifically,
the impact of (a) low growth rates,  (b) hydrocarbon emission reduc-
tions, and  (c) waivers to the 1.0 grams/mile emission standard for

-------
diesel autos were considered.  The first two factors considered alone
produced significant reductions in the number of nonattainment AQCR's
under baseline conditions (i.e., no additional NO  controls).  For
example, assuming low growth (one percent annual growth for mobile
sources and zero percent for stationary area sources) or assuming that
hydrocarbon emissions would be reduced in all 150 AQCR's by 50 percent
(and thus produce a 15 percent reduction in peak one-hour N02 levels)
reduced the number of nonattainment areas from 41 to 28 for a 0.20 ppm
standard.  Thus, the estimates of nonattainment areas summarized above
are subject to downward revision if growth is low or extensive hydro-
carbon controls are realized.  Likewise, the control costs may be reduced
somewhat.  The diesel analysis, on the other hand, revealed no sig-
nificant impact from granting a waiver.  The number of cars, the degree
of allowable emission increase, and the number of years a waiver could
be granted are all too small to provide a discernible impact.

The same set of caveats enumerated in the previous report apply to
these findings.  The key ones are repeated here in summary form:

  •  The analytical approach used for point sources is subject   '
     to considerable error:
     -  interaction among plants and among point and area
        sources is treated in a very approximate way.
        interaction among sources within plants is assumed
        perfectly additive.
     -  flat terrain is assumed
     -  a simple and largely unverified technique is used
        to related ambient NO  to N02>
  •  The area source analysis is also approximate:
     -  peak one-hour N0_ levels are estimated from annual
        average values based on typical relationships at
        a few monitors.

-------
        all area sources are assumed to contribute equally
        per unit emission  level to air quality levels.


1.3  COMPARISON WITH THE PREVIOUS ANALYSIS

1.3.1  Methodology and Impact Data

The basic methodology remains unchanged.  However, new input data,

changed parameters, slightly improved algorithms, and new output

specifications have introduced potentially significant modifications.
The key changes are noted here:
  •  More and different standards were,assessed; these were
     expressed in ppm rather than yg/m  units.

  •  A new NEDS file (updated through February 1979) was
     used for the point source analysis.

  •  A more extensive screening procedure was used to
     extract usable data  from NEDS.

  •  Slightly improved NO-  air quality modeling procedures
     were applied in the  point source analysis.

  •  Updated unit control costs for point and area sources
     were incorporated.

  •  New AQCR air quality values were derived for the area
     source analysis based  on 1976 SAROAD data and on new
     estimates of peak-to-mean ratios.

  •  Additional sensitivity runs were conducted.

  •  1984 rather than 1982  was used as the near-term
     attainment deadline.
1.3.2  Results
The results of this  analysis  are  quite  similar for point sources and

difficult  to  compare for area sources.  The point source costs are

compared below:

-------
                              Standard (ppm)
                       .135     .150     .250     .266
Previous Study          1.5                       Neg.
 (Billions of 1976$)
This Study                       1.1       .1
 (Billions of 1978$)
However, the number and types of sources controlled in the two analyses
were not identical.  In general, fewer sources were controlled in this
analysis at least partly because a baseline level of control (LEA) was
assumed on most sources.  Thus, some sources would need no additional
control.  However, the distribution of controlled sources also differs.
The first analysis controlled more industrial furnaces, 1C engines and
gas turbines and fewer industrial boilers.

Some differences appear simply because the alternative NO- standards
were not identical between the two studies.  However, there would
still appear to be relatively fewer industrial non-boiler combustors
controlled in the second study.

The previously reported 1990 area source costs in billions of 1976
dollars were roughly as follows for a 0.133 ppm standard:

                                        Capital      Annual
0.4 g/mi Auto Standard                  4.2-5.4      0.8-1.1
I&M (Loaded Mode) Program                  *         0.5
Existing Stationary Source Controls      50            *
*Not Estimated

-------
The total estimated cost for all area source controls in this analysis
to meet a 0.15 ppm standard is approximately $10 billion (1979$) capital
and $2.7 billion annual in 1990.  However, the most costly option con-
sidered in the previous study  (replacing all existing stationary area
source boilers and furnaces) was judged too unrealistic in this analysis.
Instead, only burner replacement of commercial and industrial units as
well as the mandated use of low NO  burners in new units was considered.
                                  Jw
On the other hand, the costs of a 0.4 gram/mile auto standard are
identical if a high growth rate is assumed and if the inflation between
1976 and 1979 is taken into account.  The estimates for ISM programs are
more detailed and presumably more accurate in the second study.

The number of AQCR's expected  to meet the alternative NO. standards in
the two studies is roughly equivalent.  Again, exact comparisons are not
possible due to non-identical  standards and, except for 1990, non-
identical attainment years.

-------
                     2.  BACKGROUND INVESTIGATIONS
2.1  FREQUENCY DISTRIBUTIONS OF HOURLY NO., CONCENTRATIONS
Given the sparcity of data on hourly NO. concentrations in most AQCR's,
some means of estimating likely peak concentrations had to be developed
for purposes of this regulatory analysis.  Developing relationships
between peak and mean values for recorded data at continuous monitors
and applying these relationships to annual means recorded at the more
common 24-hour monitors is a convenient way to extrapolate the observed
short-term values.  Previous investigations have shown that peak (observed
maximum hourly value) to mean (observed annual arithmetic average)
ratios for approximately 120 continuous monitors are, on average (median
ratio) 6:1 (Trijonis, 1978).  However, the degree of variability in
peak:  mean values with time at any one site, and the possible consistent
variation of the ratio with geographic location (i.e., source influence)
were not investigated.  In order to provide additional insight and to
develop a more robust estimate of this ratio, 18 we11-characterized
continuous N0_ monitoring sites were selected for detailed investigation.
Several years of recorded hourly values at each of these sites were
obtained from EPA1s SAROAD data file and curves approximating the distri-
butions for each year were fitted to the data.  Estimates of predicted
maximum hourly concentration were then compared to predicted annual
means, ratios were computed, and the temporal and within site variations
in these ratios were examined for 44 station years of data.*
 *The predicted maximum is the "characteristic high" or the value
 exceeded on average one-hour per year.  In magnitude, it lies
 between the expected high and expected second high.  Predicted
 annual means are the expected means (i.e., the concentration with
 an expected probability of occurrence of 0.5).  The ozone NAAQS is
 specified in terms of a characteristic high daily value.  If a
 similar form is used for an hourly N0_ standard, our approach to
 estimating peak-to-mean ratios is fully justified.

-------
2.1.1  Sites Investigated
The 18 sites were chosen to represent a variety of source influences.
Although the majority are either mobile source dominated or influenced
by a mix of source types, at least two are dominated by emissions from
point sources.  Source influence was determined from the site description
in the EPA site directory (EPA, 1978) and through discussions with EPA
personnel.*  The sites, locations, years of record, and type of source
influence are shown in Table 2-1.  A second criterion was completeness
of the record.  At least 75 percent of the possible 8760 hourly readings
was considered the lowest acceptable level of coverage.  Furthermore,
although these tended to be sites with fairly reliable data records, an
EPA computer program to filter probable anomalies from each year's
record was applied to further improve the data sets.  (EPA, 1978b).

2.1.2  Best Fit Functional Forms of the Distributions
Once the data had been assembled and screened, the frequency distri-
butions of the upper 50 percentile values were fit to two alternative
functional forms:  lognormal and Weibull.  Two forms were tested since
the traditionally used two parameter lognormal function, though generally
adequate in reflecting the bulk of air quality data, frequently provides
a poor fit for values at the upper tail of the distribution (Curran and
Frank, 1975).  The Weibull function, on the other hand, is more adaptable
to "light tailed" distributions and has been shown to be a better
predictor of extreme ozone values  (Johnson, 1979).

In brief, the analytical procedure used here consisted of transforming
the two distributions into linear  forms which could then be evaluated
with standard regression techniques.   (Details can be found in Appendix
A.)  Using the 44 station-years of SAROAD data and all N02 values above
the 50 percentile level for each station-year record, the equations
*Primarily, Hal  Richter and Don Sennett, MDAD

-------
                                          TABLE  2-1
                          STATIONS USED  IN THE  STATISTICAL ANALYSIS
 Site Code
 f '
 IJ0600002G01

 p5880001I01
 i
 056800004101

 1 J6980004A05

 $58720001101

 U>0580002F01

 \ ;0080008F01
 i
 222160005F01

 i 4340003F01

 ?.^4760002F01
 I
 ia4880002F01

 :"4680050F01
 j
 050230001101
 i —
 0  0500002101

-^3900001101

 Crj4180001I01
 (,  4180002101
 054200001101
Location
Phoenix, AZ
Pittsburg, CA
San Diego, CA
San Jose, CA
Whittier, CA
Denver, CO
Ashland, KY
Springfield, MA
Port Huron, MI
Saginaw, MI
Southfield, MI
Welfare Island, NY
Anaheim, CA
Azusa, CA
Lennox, CA
Los Angeles, CA
Los Angeles, CA
Los Angeles, CA
Years of Record
1976
1975^-1977
1976-1977
1975-1977
1975-1977
1975-1976
1975-1977
1977
1975-1977
1975-1977
1976-1977
1975
1975-1976
1975-1977
1975-1977
1975-1977
1975-1977
1975-1977
Source Influence
Mobile
Mixed
Mobile
Mobile
Mobile
Mixed
Point
Mixed
Mixed
Mixed
Mobile
Point
Mobile
Mixed
Mixed
Mixed
Mobile
Mixed

-------
were specified and tested for goodness of fit.  The coefficients of
                2
determination (R ) for each of the site years varied from .965 to .999
for individual years, with no consistent difference between the two
functional forms.  Other statistics, however, were used to judge the
goodness of fit for the upper tails of each distribution.*  On this
basis, the lognormal distribution proved superior in 24 of the 44 site-
years .

2.1.3  Peak-to-Mean Ratios
Once the best fit function was selected, predicted peak and mean values
for each site were estimated and their ratio computed.  The results
across all 18 sites were as follows:

               Peak:Mean Ratio          Statistic
                    6.8                 Median (Md)
                    7.0                 Mean (M)
                    1.6                 Standard Deviation (SD)
                    23%                 Coefficient of Variation (SD/M)
                    5.9-8.1             50% Confidence Interval
                    3.7-10.3            95% Confidence Interval

The variation across sites indicated above is substantial, due pre-
sumably to variations in both meteorology and type of source influence.

To distinguish these effects, variations in peak-to-mean ratios over-
time at individual sites and variations among types of sites were
examined.  Though the time variation at individual sites was less
than the total variation in the data set, it was still considerable:
*The Durbin-Watson  statistic was used.   For details, see Appendix A.

-------
an average standard deviation of over 1.0 (or a coefficient of variation
of about 15 percent) for each site.  Thus, meteorology would appear to
play a significant role in determining peak-to-mean ratios.  On the
other hand, the variation between types of monitors was not as great as
anticipated:
               Point Source          Mixed Source          Mobile Source
               Dominated             Influenced            Dominated
Number of           2                   9                      7
  Sites
Average Peak:     9.6                 6.1                    7.7
  Mean Ratio
Standard          2.3                 1.2                    1.6
  Deviation
However, the limited number of point source dominated sites makes this
a considerably less than sufficient test.   Still, given this outcome,
we can only assume that no statistically significant difference can be
discerned between point and area source-dominated monitors.  Thus, the
average peak-to-mean values for all sites should be used in the area
source component of the regulatory analysis even though most monitors
reflect area source contributions.

2.1.4  Implications for the Regulatory Analysis
The results of this statistical analysis can be used to infer the relative
stringency of alternative hourly N0_ standards with respect to the
current annual NAAQS.  An average peak-to-mean ratio of 7.0 implies that
a new hourly standard of 0.35 ppm would be just as stringent as the 0.05
ppm annual standard for about half of all AQCR's, assuming that the 18
sites employed in this analysis are representative of all regions.  Expressed
alternatively, the probability is approximately 0.5 that an hourly

-------
standard of 0.35 ppm could be achieved in any one area if the current
standard is attained.  Likewise, attainment of an hourly standard of
0.125 ppm would assure attainment of the current standard in over 98
percent of all areas, while meeting a 0.25 or 0.50 ppm hourly standard
would assure attainment of the current annual standard in about 60 and
5 percent of all regions, respectively.

For purposes of this analysis, we will assume that the peak hourly NO.
concentration in each AQCR is 7.0 times the observed annual average in
the base year.  The ratio will be varied to test the sensitivity of the
output to this factor.

2.2  SPATIAL VARIATION IN NO,, CONCENTRATIONS
The preliminary regulatory analysis adopted as a major tenet that short-
term NO- air pollution from area sources was an AQCR-wide problem
(Keyes, et al.  1978).  This was then used to justify a linear rollback
approach in the control strategy analysis.  Limited tests of this
hypothesis were undertaken at both the regional and micro scales.

2.2.1  Regional Scale Variation in NO., in Los Angeles
Few AQCR's have a sufficiently dense monitoring network to support an
investigation of spatial variation in ambient NO- levels.  Los Angeles
is one exception.  Figures 2-1 and 2-2 are maps of air quality isopleths
drawn according to standard cartographic practices from ambient data at
20 continuous NO- monitors.  Figure 2-1 demonstrates that annual average
NO- values vary in a relatively uniform way from the city center outward,
with most of the AQCR recording violations of the current NAAQS.  However,
the second high hourly values in Figure 2-2 show considerably more
variation with geographic location.  Though high levels are recorded in
most parts of the AQCR, a steep gradient is apparent from the central

-------
                                                     FIGURE 2-1
                  1977 N02 CONCENTRATIONS1 IN THE LOS ANGELES AIR QUALITY CONTROL REGION
A • Aiuhcil..

8 Una....

C-6uib.il...
.in
.HI
. IM
0-Ciioiillo	(1

E-CotliMtn	S41
          MMITORIN6 UTE MO READIRGI

f • L« H»M	IM1          K • lot Anftltl Of	t«I

G • Lami.	la          L • I|n«oo4...'.-.	l»

H-LonfBuck	Ill          ••ReiikiH	(t

I lon»	Ill
                               1977 ARITHMETIC MEAN  (ug/mj)
                              > I kiurcj If Siltnn (MnlMilc Keliot
R-SnB«ni4l»	B

           It*
                                                                                                    p>»%lt otumlloat
                                                                                                   •La

-------
                                                FIGURE  2-2
                1977 NOz CONCENTRATIONS1 IN THE LOS ANGELES AIR QUALITY CONTROL REGION
                                                                                     I	I	«QCi t1
A •



l-Aiiu
0-Cuuiilto	Bi




(•CiiliHta	Ul
         •MUORMG »!£ AND REAOIWS





F • li Hibn	Ml         R. Lit tophi Ct|	Ill




6 • Itowt	IH         I • l|»nf	JU




N • l««| Buck	M         • • Murkill	Ot




I • In (UplH	»A ^«,      I • r»ufcm	Ill




J • In «U|tlti	IMS         0 • POMU:	]||
  	UI



  	IN



  	M




	»)



	in
1971 HOURLY SECOND HIGH (ug/rn1)
                                                                                        I •UMn4l|fattlBMCll«(M*lcM<4
                                                                                        •III

-------
city outward.  Moreover, nodes and corridors of high concentration are
evident east and southeast of the center.  This suggests that ambient
levels of N02 are not necessarily uniform throughout an AQCR.

2.2.2  Microscale Gradients in NO., Downwind of a Roadway
To probe potential "spatial variations in more detail, the results of a
recent study of a roadside air quality were reviewed CU-S. EPA. 1977).
During this study, continuous monitors measured hourly NO , NO-, NO, and
                                                         X    *»
ozone, along with wind speed and direction and traffic count from July
20-August 30, 1978 at the site of the Los Angeles Catalyst Study (LACS).
The LACS measured concentrations upwind and downwind of the San Diego
Freeway.  The San Diego Freeway is a major urban expressway with an
average daily traffic approaching 200,000 vehicles per day and should
lead to high downwind concentrations of NO-.  Figure 2-3 shows the
configuration of the highway and the monitors.  These monitors were
located in an open area that is unlikely to be impacted by NO  sources
                                                             Jv
other than the roadways shown.

Data on NO and N02 levels for four afternoon hours during one represen-
tative day are shown in Figure 2-4.  A significant decrease in NO- with
distance from the highway is apparent; at about 400 meters the con-
centrations approximate 20 percent of those at 5 meters.
This variation is substantial, especially given the derived nature of
NO-.  To further examine the cause for the degree of change, NO--to-NO
  £                                                            ^B
ratios were computed at these sites for two different levels of 0, for
those days which had the most complete data sets and during which the
wind direction was roughly perpendicular to the freeway.  The results
are shown in Table 2-2.

-------
                                                  FIGURE 2-3
PREVAILING
   WIND
            to
            oo

                  en
                  CQ
                                          LOS ANGELES CATALYST STUDY

                                               MONITORING SITES
                                                                                     SCALE  1" = 50m
•
6
•
9
•
10
•
11
t
12
                                      DISTANCE FROM NEAR EDGE OF FREEWAY

                                        RECEPTOR          DISTANCE  (M)
                                                                      3
                                                                      5
                                                                      6
                                                                      9
                                                                     10
                                                                     11
                                                                     12
                                                             30
                                                              8
                                                             30
                                                             121
                                                             195
                                                             286
                                                             385

-------

-------
                               TABLE 2-2


            NO-/NO  RATIOS FOR MONITORS NEAR THE LACS SITE*
              ft   A
                              Distance From Roadway

                        30        121        195        286        385
Ozone Level

Medium         .23      .35         .60        .86        .84         .66
(.06-.08 ppm)

High           .25      .47         .84        .91        .82         .70
(0.08-.21 ppm)
*These ratios were determined from measurements made during
 afternoon hours  (12 noon - 7 p.m.) when the Seabreeze effect
 was most likely  to occur.  The wind direction at this time
 was approximately perpendicular to the roadway.

-------
Since the ratios reach a peak and then decrease after several hundred
meters, it would appear that chemical dissociation of NO. as well as
atmospheric dilution contributes to this rapid decrease in ambient
levels.  However, the decrease in NO,: NO  ratios is somewhat unusual
                                    ^    A
given the levels of observed 0_.

These findings must be considered preliminary at this time.  The recorded
NO- levels are considerably below peak levels in Los Angeles, due perhaps
to the perpendicular wind field characteristic of the test site and the
low background levels at the LACS site.  Oblique, low speed winds are
expected to inhibit dilution and cause high concentrations over larger
areas.

2.2.3  Implications for the Regulatory Analysis
The above findings at both the regional and microscale levels suggest
that treating N0_ as a homogeneous problem for entire AQCR's may not be
wholly appropriate.  On the other hand, many of the area source control
strategies considered in Section 3 are difficult if not impossible to
implement on less than an AQCR-wide basis (.e.g., mobile source ISM.
programs).  Moreover, many area sources are ubiquitous in urban areas
(e.g., autos) making them likely contributors to high concentrations
wherever they occur.  Finally, major point sources may account for some
of the "nodes" of high N02 observed in Los Angeles.  Our independent
treatment and control of point sources may thus render the assumption of
uniform region-wide N0_ levels from area NO  sources more acceptable.

-------
                              REFERENCES
U.S. EPA.  1977.  Proceedings of the Los Angeles Catalyst Study
Symposium, ORD, Research Triangle Park, NC.

U.S. EPA.  1978a.  Directory of Air Quality Monitoring Sites Active
In 1976, OAQPS, Research Triangle Park, NC.

U.S. EPA.  1978b.  Screening Procedures for Ambient Air Quality Data,
OAQPS, Research Triangle Park, NC.

Johnson, T.  1979.  " A Comparison of the Two-Parameter Weibull and
Lognormal Distribution Fitted to Ambient Ozone Data" in Proceedings
of specialty Conference or Quality Assurance in Air Pollution Mea-
surement, New Orleans, LA.

Curran, T. et al.  1975.  "Assessing the Validity of the Lognormal
Model When Predicting Maximum Air Pollution Concentrations" Paper
No. 75.3 68th Annual Meeting of the Air Pollution Control Association,
Boston, MA.

-------
            3.  ATTAINMENT STRATEGIES FOR EXISTING SOURCES
As in the previous analysis, two distinct analytical approaches have
been adopted for assessing control strategies to obtain the various
hourly NO- standards (Keyes, et al, 1978a).  Individual point sources of
NO  emissions were modeled using a standard Gaussian dispersion model
  X
and time-dependent NO-to-N02 conversion curves.  Monitoring data and
proportional.modeling were used to select control strategies for area
sources of NO  on an AQCR-wide basis.
             Jk

The rationales for the separate approaches are briefly reviewed in
Sections 3.1 and 3.2.

3.1  COST AND EFFECTIVENESS OF POINT SOURCE CONTROLS
As noted previously, the placement of most continuous and 24-hour
monitors produces air quality data which reflect area rather  than point
source contributions.  For this reason, the assessment of point sources
was based on dispersion modeling of individual sources in NEDS without
reference to air quality readings at the nearest monitor.  However,
recorded air quality in each AQCR was used to estimate an NO. "back-
ground" level  (i.e., contributions from other sources at the modeled
receptor of maximum impact for each individual point source).

3.1.1  Analytical Procedures

3.1.1.1  Screening NEDS and Selecting Sources
Data quality problems associated with NEDS are widely acknowledged.
There exists, however, no alternative nationwide emissions data base.

-------
Our goal was to save  as many  of  the  source  records  as possible  (recog-
nizing that NEDS contains  less than  total coverage  for many  types of
sources) while eliminating clearly erroneous  data Items.   To accomplish
this, a series of threshold values and, where values fell  above or below
the indicated threshold, default values were  specified for each of the
key data items:  source size  (design capacity), hours of operation,
stack height and diameter, flue  gas  temperature, and gas flow rate.   In
some cases the threshold and  default values applied to large categories
of sources, and in  others, they  were SCO-specific.  A listing of the
threshold/default values used and the screening procedures in the analysis
appear in Appendix  C.
Following this  initial  screening, NO   sources  were extracted  from NEDS
based on a preliminary  assessment of  their  air quality  impact.   Emissions
were estimated  using  hourly design  rate or  boiler design capacity and an
emission factor (see  Appendix A), and maximum  ambient levels  of  NO   were
                                                                  X
estimated for each  source (i.e., combustor  or  process)  using  EPA's
dispersion model PTMAX.*  Plants were included in the extracted  file if
the sum of the  maximum  concentrations for each source within  the plant
plus a background concentration (from area  and other point  sources)  was
_>0.10 ppm NO .*  This  is a conservative approach since (a) maximum
impacts from individual sources are not totally additive, and (b) not all
NO^ is N0r

3.1.1.2  Modeling Conditions
The sources selected  in this manner were next  subjected to  additional
air quality modeling.   Using EPA's  dispersion  model PTMTP and EEA's NO-
to-NO. conversion curves (Keyes, et al, 1978a), maximum concentrations
of NO. from each source were estimated and  summed on a  plant-wide basis
to obtain maximum impacts from  each isolated plant. The meteorological
 *See  Keyes,  et al,  1978a for a description of the assumed meteorologic
  conditions  and background concentrations.

-------
conditions were a mixing height of 800 meters and wind speeds of either
1.0 or 5.0 meters/second, whichever produced higher plant-wide concen-
trations.*  To this sum was added an area and other point source back-
ground level, specified at either 1.0 or 1.5 times the highest recorded
annual average NO- value (1976 SAROAD data) in the AQCR, for a wind
speed of 1.0 and 5.0 meters/second, respectively.**

3.1.1.3  Emission Control Options
Essentially the same set of control options were used here as were
employed in the previous analysis.  However, two important changes were
made.  First, the fuel savings or penalty was added explicitly to the
annual cost for each option.  This had the effect of eliminating off-
stoichiometric combustion as a viable option for many sources since it
is more expensive (on an annual basis) but is less effective than other
available options.  Secondly, the one option with net cost savings  (low
excess air) was assumed to be in place on all sources and emissions were
adjusted accordingly.  This is a reasonable assumption if the use of LEA
is indeed economic.

The resulting set of control options are shown in Tables 3-1 to 3-10.
Included are the capital and 0 & M costs, control efficiencies and
energy savings/penalty.  Total annual costs are computed by annualizing
the capital cost at 16 percent (assuming a weighted cost of capital of
11.5 percent and a pay back period of 10 years) and adding this to  the
0  §  M  charges.

3.1.1.4  Least Cost Control Strategies
Once the degree of ambient air quality reduction was determined for each
plant, the sources within the plant were controlled in a cost-effective
manner.  That is, those combinations of sources and controls with the
 *The  same  conditions were  applied to  all  sources within a plant.
 **See Keyes,  et  al, 1978a  and  b  for a discussion of the derivation
   of  these coefficients.

-------
lowest control cost per air quality improvement  ($ per ug/m. ) were.
selected so that the total required improvements were achieved at least
cost to the plant.

3.1.2  Estimated Control  Costs for Alternative Standards
Table 3-11 shows the estimated nationwide costs  for controlling NO
                                                                  X
emissions at point sources to meet each of six one-hour NO. standards.
Costs to meet the two highest standards (0-40 and 0.50 ppm) are negligible.
Also shown are the energy penalties associated with these controls.

Figure 3-1 diagrams the rate of change in capital costs with level of
the standard.  A sharp increase in the rate occurs between 0.15 and 0.20
ppm.

The sources needing controls show a wide distribution geographically. At
the low end (0.10 ppm standard), 115 AQCR's have one or more plants
requiring control.  This  number shrinks to about 25 at 0.25 ppm and less
than 15 at 0.35 ppm.  Significantly, all sources can be controlled to
the required level except in the case of a 0.10 ppm standard.  Area and
minor source background levels are above 0.10 ppm in a few AQCR's thus
precluding attainment of  this standard.

Table 3-12 shows the types of source likely to need NO  controls above
                                                      ^
the assumed baseline control level Clow excess air) for two standards.
In both cases industrial  boilers account for the greatest fraction
(slightly under half) of  total sources needing control.  Utility boilers
are the next major class  followed by industrial  furnaces and internal
combustion engines.  Although utility boilers produce the majority of
point source NO  emissions, the tall stacks typical of power plants
provide a good measure of atmospheric dispersion before the NO. reaches

-------
ground level.  Tables 3-13 and 3-14 display the distribution of plants
and sources needing control (3-13) and control costs  (3-14) among
industrial groupings.  Electric and gas utilities and iron and steel
plants have the largest number of sources to control and incur the
highest costs.

3.2  COST AND EFFECTIVENESS OF AREA SOURCE CONTROLS
Based on the assumption that area sources are dispersed throughout urban
areas and high hourly NO- values are widespread, proportional modeling
was used to relate changes in monitored NO- concentrations to changes in
current NO  emission levels in those AQCR's which may experience  future
          3&
problems meeting short term standards.  The base year emissions data
were compiled from NEDS, and are assumed to be representative of  1976
emissions.  Base year (1976) air quality was defined as the characteristic
high hourly (peak) N02, and was estimated by multiplying the highest
recorded annual average NO- level in each AQCR by seven.*  This is based
on an average 7:1 peak-to-mean ratio obtained from a statistical  evaluation
of 44 station-years of data from continuous monitors  (See Section 2.1).
Background NO- was assumed to be zero in the analysis since natural and
transported NO- levels are typically very low.  Emissions were grown
and/or reduced to reflect area source growth rates (three percent for
mobile sources and one percent for stationary area sources) and the
schedule of increasingly stringent mobile source emission standards.
Using proportional modeling assumptions, future year air quality  was
estimated and compared with alternative NO- standards. Area source
controls were applied (most cost effective first) in AQCR's where they
were necessary to achieve the standard, and estimations of control
requirements and costs were made for 1934 and 1990.  Several sensitivity
analyses were then conducted for the key input variables.
 *See Section 2.1 for a definition of the statistical term "characteristic
  high".   In magnitude, it falls between the expected high and the expected
  second high.

-------
3.2.1  Mobile Source Controls
The Clean Air Act Amendments of  1977 mandate  the oxides of nitrogen
emission standards shown  in Table 3-15.  These emission standards are
detailed for five different vehicle categories.  The new vehicle emission
rates and the deterioration rates used in EPA's MOBILE 1 model and
assumed In this analysis  are listed along with the corresponding emission
standards.

Our baseline analysis  extends  the 1.0 gram/mile standard through 1990.
In order to reduce automotive  NO emissions even more than the reduction
                                 X
achieved through the Federal Motor Vehicle Control Program  (FMVCP), a
number of alternative  control  options were included in this analysis.
These include Inspection  and maintenance for  NO  control, transportation
control measures, and  a 0.4 gram/mile NO  emission standard.-  Alternative
                                        &
light and heavy duty truck standards are also evaluated.

3.2.1.1  Inspection and Maintenance
Since I&M programs currently monitor only CO  and HC emissions, it is
difficult to predict with any  certainty what  the exact reductions in NO
                                                                       &
will be from an I&M program.   All of the evidence which can be used to
estimate the potential reductions from a NO   inspection is preliminary
                                           X
and subject to change.  The source of the information used in this
analysis is a technical paper  written by the  Inspection and Maintenance
Staff of EPA's Office  of  Mobile  Source Air Pollution Control  (U.S. EPA,
1979).  Unlike HC and  CO, NO   emissions from  cars increase as vehicle
speed increases.  Consequently,  if the idle test presently used in I&M
programs for measuring CO and  HC were used to measure NO. test values
                                                        jC
would not correlate well  with  actual in-use NO  emissions.  Instead, an
I&M program designed to test for NO  would be more complex than any
existing I&M program.  The two systems currently viewed as practical for

-------
NO  controls are (1) an inspection of the exhaust gas recirculation
  2v
(EGR) system, and (2) a dynamometer test under varied driving conditions.
The EGR inspection would simply be added to an idle test regime.  Cars
with obviously malperforming EGR systems would be identified and main-
tenance would be required just on this group.  A dynamometer test would
be more costly and complicated to implement, but it would allow measure-
ments to be made at high speed load conditions under which the highest
NO  levels occur.
  X

The estimated cost and emission reduction effectiveness values given  in
the EPA technical paper for each of the two alternative NO  I&M programs
are shown in Table 3-16.  The credits shown are for 1975-77 model year
vehicles.  While it seems reasonable to assume that the loaded-mode test
credits for these model year vehicles are likely to also apply to 1978-
1990 model year cars  (because most, if not all, will be equipped with a
catalyst), it is less reasonable to assume similar reductions in pre-
1975 cars.  Therefore, the methodology used here conservatively assumes
that NO  emissions from pre-75 cars are unaffected by I&M when the
       X
dynamometer test is applied.  Because EGR appeared in cars starting with
the 1973 model year, emission reduction credits for an EGR inspection
are assumed to start with 1973 models.  Using the travel weighting
factors for light-duty vehicles in the mobile source emission factors
documentation (U.S. EPA, 1978), the aggregate travel fractions for pre-
1975 cars in 1984 and 1990 are seven percent and one percent, respectively.
Therefore, the automotive fleet percentage reductions in NO  from a
                                                           X
loaded mode test are calculated by multiplying the percentage reduction
in NO  (from Table 3-16) by the fraction of travel accounted for by 1975
     X
and later model year autos:*
*For the purpose of this analysis, all  I&M programs  are  assumed  to  be  in
 operation for at least a  full year before 1984.   These  I§M programs are
 assumed to affect all post-1974 cars for the  loaded test  and  all post-
 1972 cars for the EGR inspection.

-------
Projection
Tear
1984


1990


Loaded Mode Test
Failure Rate
(2)
10
20
30
10
20
30
Z Reduction in NO Travel
(From Table 3) Fraction
1975-77 Models of post
1974 cars
5.9
10.2
12.9
5.9
10.2
12.9
.93
.93
.93
.99
.99
.99
Z Reduction
in NO (Auto
fleet?
5.5
9.5
12.0
5.8
10.1
12.8
A similar calculation  is  made  to  estimate  the  auto  fleet  emission reduction
credit for NO  resulting  from  an  EGR  inspection.  The  travel  fractions for
             X
pre-1973 cars in  1984  and 1990 are  three percent  and zero percent,
respectively:
Projection
Year
1984
1990
EGR Inspection
Failure Rate
13
13
Z Reduction in
(From Table 3)
1975-77 Models
6
6
Travel
Fraction
of post
1972 cars
.97
1.00
Z Reduction
in NO (Au
fleet?
5.87 .
6.00
For AOjCR's with concentrations  projected  to  be  above  one of  the alternative
standards in  1984 and  1990,  the I&M emission reductions  for  the auto  fleet  ~
shown above are applied  to determine if these controls reduce emissions
enough to bring any AQCR's into compliance.   If the I&M  programs do not
bring all of  the AQCR's  into compliance with each  alternative NO. standard,
additional controls are  applied.  The impacts of an I&M  program are shown in
Table 3-17 for 1984 and  in Table 3-18 for 1990.
Tables 3-19 to 3-22  show  the capital  and  annual  cost  estimates  (1979  dollars)
for both the EGR inspections and  the  NO   loaded  mode  test  for 1984  and
1990.  These costs are  the  incremental costs  of  adding  these  programs to

-------
an existing HC and CO idle inspection program.  It is assumed that EGR
inspections can be performed without any additional investment in capital
equipment.

In order to estimate the capital costs of implementing a loaded mode
test, it was necessary to make assumptions about the cost of a dynamometer,
the number of vehicles one dynamometer can inspect in a year, and the
number of registered vehicles in a typical area needing a NO  I&M program.
                                                            3t
Estimates of these parameters were based on a conversation with Harold
Davis, EPA Office of Mobile Source Air Pollution Control.  The price of
the required dynamometer is estimated to be $5000.  The type of dyna-
mometer assumed for use in NO  testing has road load but no inertia
                             A
capability.  In other words, it can be used to test at constant high
speeds but cannot simulate the variable driving conditions in the Federal
Test Procedure.  Road load capability alone is consistent with the I&M
annual cost effectiveness estimates used in the analysis.  Dynamometers
with inertia capability would cost at least $10,000.

Each dynamometer was assumed to be able to make 25,000 inspections
annually.  Therefore, if an average AQCR has 500,000 registered vehicles,*
20 dynamometers will be required.  Using these assumptions, capital
costs for loaded mode tests in 1984 and 1990 were estimated by multiplying
the number of nonattainment AOjCR's identified in Tables 3-17 and 3-18
for the baseline cases, by the estimated $100,000 cost to purchase 20
dynamometers for each nonattainment AQCR.  These capital costs may be
somewhat understated because they assume that all areas have an existing
idle inspection program.  Although it is recognized that not all of the
AQCR's projected to have NO- concentrations greater than the alternative
standards will have idle inspection stations in operation, no attempt
was made to determine which areas would not have HC and CO I&M programs
operating by 1983.
*This is based on an average ownership rate of  .41 autos per person
 and an average of 1.2 million people per AQCR  for those AQCR's  in
 this analysis, (U.S. Bureau of the Census, 1974).

-------
Annuallzed costs of I&M were estimated on an AQCR-by-AQCR basis using
the cost effectiveness estimates shown in Table 3-16.  All AQCR's iden-
tified as being in nonattainment in the baseline cases in Tables 3-17
and 3-20 were assumed to require I&M programs for the purpose of esti-
mating costs.

3.2.1.2  Transportation Control Measures
3.2.1.2.1  TCM's Used In The Analysis
It is likely that many of  the areas that are projected to violate a one-
hour NO2 standard also violate either or both of the 0_ or CO standards.
Therefore, these areas may have instituted or plan to institute trans-
portation control measures (TCM's) to reduce CO or HC emissions.  These
TCM's may bring reductions in NO  emissions as well.  Therefore, NO
                                &      •                            X
credits from these TCM's are assumed to be accrued at no cost.  This is
a reasonable assumption to the extent that TCM's which reduce HC and CO
emissions also reduce NO .  In general, this is the case for TCM's which
                        3C
decrease VMT.  However, no attempt was made to identify the TCM's pro-
posed to CO and HC control in those AQCR's estimated to need additional
mobile source NO  control  in this analysis.
                *"                                         "         r

Present guidance on NO  emissions from TCM's deals only with the following
programs: (U.S. EPA. 1978b)

  •  Bus lanes
  •  Carpool/Vanpool
  •  Transit
     — Fare reduction
     — Service improvements

-------
The effectiveness of these TCM's was analyzed by a combination of observed
and model-estimated impacts.  Figure 3-2 shows estimated NO  emission
reductions for nine TCM scenarios in a large urban area,  (U.S. EPA,
1978).  Case 19 was used in our area source analysis to represent a
typical TCM affecting NO .  Case 19 includes a carpool/vanpool program,
reserved lanes for buses and carpools, ramp metering, and bus-by-pass
ramps.  If these measures were implemented in a large urban area, it is
estimated that regional NO  emissions would be reduced by 1.6 percent.
The impact of TCM's on attainment of alternative one-hour NO. standards
is shown in Table 3-17 for 1984 and in Table 3-18 for 1990.

3.2.1.2.2  Other TCM's
Other transportation control measures which have been analyzed by Alan
M. Voorhees in draft working papers include  (Voorhees, et al, 1979):

  •  Traffic Signal Improvements
  •  Work Rescheduling
  •  Traffic Engineering Improvements
  •  Parking Management
  •  Truck Restrictions and Enhancements

Most of these measures are designed to reduce congestion  and increase
vehicle speed, which is appropriate for reducing CO emissions but
unlikely to have any positive benefit for NO  control.  A more detailed
assessment of the impact of these TCM's on reducing NO  levels is given
in Appendix B.

-------
3.2.1.3  Alternative Emission Standards
3.2.1.3.1  Automotive Emission  Standard of 0.4 Grams Per Mile
The Clean Air Act Amendments of 1977 state that EPA shall study "the
public health implications of attaining an emission standard on oxides
of nitrogen from light-duty vehicles of 0.4 grams per vehicle mile."  We
have evaluated the effect of a  mandated 0.4 grams per mile emission
standard on attainment  of alternative one-hour N02 standards,  tfe assumed
that the 0.4 gram/mile  standard will start with the 1985 model year and
simulated its effect by re-estimating composite light-duty vehicle
emission factors using  MOBILE 1.  An initial emission rate of 0.12
grams/mile and a deterioration  rate of 0.22 grams/mile were assumed for
this standard in accordance with MOBILE 1 assumptions.
                                                            *
tfcile the ability of cars to attain the 0.4 NO  standard has been ques-
                                              3C
tioned, certification data from California for 1979 model year vehicles
shows that five different manufacturers can meet the 1983 California
emission standards (.41 HC, 7.0 CO, 0.4 NO ).  Most of these cars use
three-way catalysts and fuel injection.

In the Three-Agency Study,  (U.S. EPA, DOT. 1977), EPA estimated that
achievement of the (0.41/3.4/0.4) emission standards is possible by
adding electronic fuel  injection to a three-way catalyst system.  This
can be accomplished with little or no fuel economy penalty, and at an
average increase in vehicle sticker price of about $330 (1977 dollars)
relative to today's cars (1.5/13/2.0).  The capital cost increase per
vehicle is $90 when compared with the .41/3.4/1.0 standards and inflated
to 1979 dollars.  The total capital cost of a 0.4 gram/mile NO  standard
                                                              X
in 1990 was estimated by multiplying the $90 incremental initial cost by
the expected new car sales for  the years 1985-1990.  The total capital
cost estimated in this  manner is $6.7 billion.

-------
Because there is no assumed difference in fuel economy due to the 0.4
gram/mile NO  standard, the only annual costs are the annualized capital
            X
charges and the increase in maintenance cost.  A capital recovery factor
of 18 percent was assumed.*  The Three-Agency Study estimates a $40
(1977 dollars) increase in lifetime maintenance cost due to the 0.4
gram/mile standard.  When annualized and inflated to 1979 dollars
(assuming an average car life of 10 years) the annual per car maintenance
cost is assumed to increase by $9.  This additional operating and
maintenance cost, when multiplied by the new car sales from 1985-1990
and the scrappage rate for each model year yields a total annualized
capital cost of $0.67 billion,  When added to the estimated annualized
capital charge of $1.2 billion, the total annual cost of a 0.4 gram/mile
NO  emission standard is estimated to be $1.9 billion  (1979 dollars).
  &

3.2.1.3.2  Alternative Emission Standards For Trucks
At the request of EPA's Office of Mobile Source Air Pollution Control
(OMSAPC), we have also estimated the air quality impacts of varying
emission standards for trucks.  The results of these projections can
be compared with the baseline projection for 1990 to determine the impact
of alternative truck standards.  The assumed standards for each case are
detailed below:
Vehicle                  Baseline       Low-truck      High-truck
Classes                    Case         Standards      Standards
Light-duty Trucks
    < 6000 Ibs.           1.4 gm/mi.     1.2 gm/mi       1.7  gm/mi.
6000-8500 Ibs.           1.4            1.2             1.7
Heavy Gas Trucks         5.35           3.0             6.0
Heavy Diesel Trucks      5.35           3.0             6.0
*This is based on an average automobile  life of  10 years  and 100
 percent debt financing at a rate of 12.5 percent.

-------
All of the above standards are  assumed  to start with  the  1985 model
year.

There is no published  guidance  on how to adjust the new vehicle  emission
and deterioration rates  to account  for  changing truck standards.  There-
fore, the new vehicle  emission  and  deterioration  rates for  alternative
standards were estimated by multiplying the new vehicle emission and
deterioration rates  for  the baseline standards by the ratio of the
alternative standard to  the baseline.   The emission rates used in the
alternative truck standard analysis are listed in Table 3-23.

Table 3-18 reveals that  low truck standards have  a slightly greater
impact on the number of  NO. nonattainment areas than  the  previously
analyzed automotive control strategies.  Comparing the emission  factors
in Table 3-23 with those in Table 3-15  shows that the major difference
is in the heavy truck  emission  rates.   In order to separate out  the
effect of changing the light  trucks vs. the heavy truck standards, each
case was analyzed separately.   Table 3-24 shows these results.   The
baseline assumptions were used  for  all  vehicle categories except the
specific ones listed.

As indicated, changing the light truck  standards  from 1.2 grams/mile to
1.7 grams/mile is predicted to  have no  impact on  the  number of non-
attainment areas in  1990.  Changing only the heavy truck  standard has
almost the same impact as changing  both the light and the heavy  truck
standards.

3.2.2  Stationary Area Source Controls
In the area source analysis,  both the emission reduction  and the cost of
applying controls to existing and newly constructed stationary area

-------
sources were modeled.  Twelve different stationary area source categories
were analyzed.  The four major categories of stationary area sources
were residential, commercial-institutional, and industrial fuel combustion
and solid waste.  Further categorization was based on types of fuel
burned.

Table 3-25 shows the assumed emission reduction effectiveness, and
capital and annual costs associated with control of specific source
types.               :

Stationary area source emissions were assumed to grow at a compounded
annual rate of one percent.  For AQCR's predicted to not meet the stan-
dards without controls on existing stationary area source, controls were
applied (most cost-effective first) until either the standard was met or
all of the potential controls were applied.  Controls on new stationary
area sources  (i.e., low NO  burners on combustors) were assumed necessary
                          A
in all AQCR's for each standard equal to or lower than 0.35 ppm.  Tables
3-19 to 3-23 detail the capital and annual cost estimates for 1984 and
1990 for alternative levels of the one-hour N02 standard.

While it would be possible to achieve greater reductions in stationary
area source emissions by assuming that all existing sources are replaced
by new units, the cost of such a measure would be prohibitive and is not
analyzed here.  Preliminary estimates included in the previous report
placed the costs at several tens of billions of dollars and indicated
that few AQCR's moved into attainment (Keyes, et al, 1978).

-------
3.2.3  Sensitivity Analyses
3.2.3.1  The Effect of Growth on Nonattainment
In order to test the effect of changing the growth assumptions for
mobile and stationary sources on the number of areas attaining the
standard, the baseline case was run assuming one percent mobile source
and zero percent stationary area source growth.  This contrasts with the
prior assumptions of three percent mobile source growth and one percent
stationary area source growth.  The number of AQCR's not attaining
alternative levels of the one-hour NO- standard for the low growth case
are shown below,  values for high growth are included for comparison
purposes.

               NUMBER OF AQCR'S NOT MEETING ALTERNATIVE
                   LEVELS OF THE 1-HOUR N02 STANDARD
Year

1984
1990
Standards (ppm)
ilO
Low
Growth
126
117
«
High
Growth
135
135
.15
Low
Growth
57
44

High
Growth
81
75
.20
Low
Growth
28
16

High
Growth
41
40
.25
Low
Growth
15
8

High
Growth
21
19
.35
Low
Growth
3
1

High
Gro
6
6
The largest impact appears in 1990.  Compared to the baseline, the lower
growth rate can reduce the number of nonattainment AQCR's by as much as
60 percent in 1990.  The number of nonattainment AQCR's in 1984 is
reduced as much as 30 percent with the lower growth rates.

3.2.3.2  The Effect of Hydrocarbon Control on NO., Levels
Although the atmospheric chemistry of NO  and HC is complex and poorly
                                        A
understood, recent empirical evidence suggests that hydrocarbon control

-------
                       0
yields slight to moderate benefits in reducing yearly maximum one-hour
NO- concentrations (Trijonis, 1978).  It is estimated that reducing
hydrocarbons by 50 percent should decrease yearly peak NO- by about 10-
20 percent.  For this analysis, it was assumed that hydrocarbons may be
reduced by 50 percent from 1976 levels by both 1984 and 1990, and that
this HC reduction will reduce NO- levels by 15 percent.  The effects of
a 15 percent reduction in NO- due to HC controls on the baseline case
are shown below:

               NUMBER OF AQCR'S NOT MEETING ALTERNATIVE
                   LEVELS OF THE 1-HOUR N02 STANDARD

Year                               Standards  (ppm)

          .10             .15             .20              .25            .35
     High HC Base HC High HC Base HC High HC Base HC High HC Base HC High HC Base HC
     Control Control Control Control Control Control Control Control Control Control
1984
1990
125
123
135
135
57
52
81
75
28
25
41
40
15
12
21
19
3
2
6
6
As can be seen from the values in the table above, the effect of HC control
on NO- concentrations in 1984 is essentially the same as  that of assuming
the lower growth rate and somewhat less pronounced than the  lower growth
rate in 1990.  This is due to the fact that the growth rate  is  compounded
while the reductions in HC emissions are assumed constant  over  time.

3.2.3.3  The Effect of FMVCP Waivers for Diesels
The 1977 amendments to the Clean Air Act allow for the granting of waivers
to the Federal Motor Vehicle Control Program  (FMVCP) emission standards
for diesel powered automobiles.  The waiver can raise the  FMVCP standard
from 1.0 up to 1.5 grams/mile for no more than four model  years (1981-84).
A preliminary assessment of the impact of such waivers indicated that  the

-------
attainment status of  all  AQCR's  In the  area  source  analysis  would  remain
the same with or without  the waivers   (Keyes,  et  al.  1978).  This  result
is not surprising given the small  fraction of  the total vehicle  fleet
accounted for by diesel automobiles between  1981  and  1984, and the
relatively small increase in emissions  per diesel (up to  0.5 grams/mile).
On an AQCR-wide basis,  the net change in  emissions  is not substantial
enough to affect region-wide air quality.  On  a smaller geographic
scale, however, ambient NO2 levels may  be elevated.   This is especially
true near roadways.
To evaluate the potential  microscale  impact  of  granting  a NO  waiver  to
diesels, a dispersion  modeling  investigation was  undertaken using  the
HIW4.Y model applied  to a major  urban  highway — the  San  Diego Freeway in
Los Angeles.  As noted in  Section  2.2.2,  a site on the San Diego Freeway
was the subject of a recent air quality field investigation  (U.S.  EPA,
1977).  As such, it  is well characterized with  respect to vehicle  use
and downwind NO. levels.   This  provided the  basis to estimate emissions
from a large line source and to check the reasonableness of the modeling
result.

The initial conditions and modeling assumptions are  detailed in Table 3-
26.  The meteorological and traffic assumptions were chosen to simulate
a "worst-case" meteorological situation and  peak  hour traffic.  They  are
conservative, leading  to overestimates of ambient concentrations.   As is
well documented in the literature, HIWAY  overpredicts for stable atmos-
pheric conditions, low wind speeds, and wind directions  nearly parallel
to the road  (Chock,  1977).   Moreover, it  is  unlikely that the meteoro-
logical conditions specified would remain unchanged  for  an entire  hour,
especially wind direction.   Therefore, the absolute  differences in

-------
predicted concentration between the NO  waiver  and  the  no  waiver cases
                                      X
also are very conservative  (high) estimates.  For comparative  purposes,
more reasonable conditions are postulated  (a perpendicular wind field)
and analyzed.  Specific background concentrations were  not specified  for
two reasons — natural levels of these pollutants are small in comparison
to the levels estimated here and the emphasis of this analysis is on  the
incremental impact of granting a FMVCP NO  waiver,  which is independent
of background concentration assumptions.

Table 3-27 depicts emission factors and VMT mixes from  which total
emissions for the San Diego Freeway were computed.   Basically, the
composition of the vehicle stock reflects  an assumed market penetration
for light-duty diesels of 5.6 percent in 1981 and 10.0  percent in 1989.
The 1984 figure corresponds with the highest estimate given in testimony
at EPA's March 1978 public hearing on the  proposed  rule to regulate
particulate emissions from diesel light-duty passenger  vehicles.
Three scenarios are depicted:  Case  1 - a NO  waiver is requested and
                               ^^^^™""*^^™"™       X
granted to all diesel manufacturers; Case 2 - a waiver is refused but
manufacturers are able to build a diesel which  meets the 1.0 gram/mile
NO  standard; and Case 3 - a waiver  is refused, no diesels are marketed,
  X               ^•••^^^^^••^
and all new sales are gasoline vehicles.
N0? concentrations were computed  from  the NO   concentrations estimated
  £                                         X
using the HIWAY model by applying a NO,/NO  conversion factor which
                                       £*   X
varies by distance from the roadway.   These conversion factors were
determined from N09 and NO  monitoring data collected at receptors
                  &       X
aligned with the wind field during a special study at this site from
July to August 1978.  The conversion factors for  the receptor locations
in the LA Catalyst Study were shown previously (Section 2.2.2, Table 2-2),

-------
These factors have been adjusted  to  reflect  current  evidence that more
NO  Is emitted as NO.  for diesels (about  20  percent)  than for gasoline
  X                  £
vehicles  (about  10 percent)  under high speed ,  steady-state cruise condi-
tions:
                                    NO, /NO,.  Ratios
                                      £»    X
                               Distance  Downwind of  the Roadway (m)
                                    (wind speed  • 1m/ sec.)
Distance Downwind        ^      10_     25_    50_     100     250     500
of the Roadway  (m)
(wind speed • 1 m/sec)
Case
Case
Case
1
2
3
•
•
•
2523
2517
2500
•
•
•
3523
3517
3500
.5023
.5017
.5000
.6523
.6517
.6500
.8523
.8517
.8500
•
•
•
8523
8517
8500
.7023
.7017
.7000
Table 3-29 shows the  estimated  impact  in 1984  of  each of the three
cases analyzed.  The  validity of  the absolute  estimates of NO. levels
can be examined by  comparing them with observed air quality levels at
the original monitoring study site and with ambient measurements at
other similar sites across  the  country.   The peak one-hour NO,, con-
centration observed downwind of this roadway during a special three
month (summer) study  was approximately 0.22 ppm  (U.S. EPA. 1977).   This
concentration was observed  with the wind direction roughly perpendicular
to the roadway.  Therefore, the predicted 1984 NOi concentrations  shown
in Table 3-29 for a 90° wind are within a factor  of two or three of the
observed peak.  Using HIW&Y with  wind  speed set equal to two meters per
second  (the average speed for the hour when the high NO^ levels were
actually observed,  (U.S. EPA, 1977), the predicted levels are 0.22-0.28
ppm.  This is excellent correspondence between estimated and observed
levels.  However, it  was felt that the more stringent conditions could
exist with a similar  roadway configuration at  another location.

-------
An estimate of the differences among the various cases also can be made
using the data from the LA. Catalyst Study.  Concentrations caused by any
air pollution source are directly proportional to the emission strength
if all other variables are held constant.  Based on the vehicle mix
observed during the period of the catalyst study, the approximate aver-
age emission rate was 0.012183 grains/meter/second.  For the conditions
assumed for this analysis, the average emission rate for Case 1 was
0.010034 grams/meter/second, for Case 2, 0.00998 grams/meter/second, and
for Case 3, 0.009995 grams/meter/second.  The corresponding difference
in NO2 concentrations between Case 1 (waiver) and Case 2  (no waiver), or
Case 1 and Case 3 (no diesels), at the monitor recording the highest
levels is 0.001 ppm.  These impacts are quite similar to the differences
shown in Table 3-29 for the perpendicular wind case.

The magnitude of the N02 concentrations shown in Table 3-29 for the 5°
wind appear to be much too high.  These levels probably reflect the
tendency of the HIU&Y model to overpredict.  Based on data from the LA
Catalyst Study, they appear to be high by a factor of 3 to 5.  Accordingly,
the estimates of differences in concentration also are high by about the
same amount.  Nevertheless, the more important feature of the results is
the relative size of the impacts (differences among cases) compared with
the predicted absolute levels (<1.0 percent change).

The most likely estimate of the maximum difference between the three
cases is 0.005 ppm or less.  In other words, granting a NO  waiver  for
                                                          A
all diesels for all four years is likely to raise peak one-hour N0«
levels near a high volume highway by no more than 0.005 ppm.  Furthermore,
this maximum differential will diminish after 1984 as all new autos must
meet the .1.0 grams/mile standard.

-------
3.2.4  Summary
Figure 3-3 shows the rate of change of 1984 area source control costs
across the alternative one-hour NO. standards.  Unlike the costs for
point source control, the costs curves are approximately linear.  This
is a result of the fact that, even at a 0.35 ppm standard, not all
AQCR's can be brought into attainment.  Thus, the same degree of application
is required for control strategies such as the 0.4 gram/mile auto
standard and mandatory low NO  burners for new stationary area combustors.
                             A
Some degree of additional control on existing stationary area and mobile
sources is required for a few AQCR's to meet more stringent standards,
but for many, the costs remain constant while their attainment status
changes.

Although all of the control options are necessary in order to bring as
many AQCR's into attainment as possible, clearly not all are equally
cost-effective.  Based on the annual cost per AQCR brought into attain-
ment of a 0.20 ppm standard in 1990, the following ranking is obtained:

                                     Increase in         ..    ,
                    Annual Cost      Attainment        $ x 10 /Attainment
Strategy          C1979 $ x 10 )   (No. of AQCR's)           AQCR	

ISM-Loaded Test         150              5                    30
New Stationary          280              3                    93
 Area
Source Controls
 I&M-EGR Test           100              1                   100
Existing Stationary     300              1                   300
 Area Source Controls
0.4 g/mi Auto          1860              3                   620

-------
3.3  AREA AND POINT SOURCES COMBINED
Table 3-29 summarizes the AQCR attainment status and the control costs
for three representative standards, taking into account both point and
area source control strategies.  The 1984 results for area sources are
used here since most existing point sources will still be in operation
in 1984.  Thus, the costs for.the two major categories of sources can
justifiably be summed.

In no case (and, indeed, for no standard up to 0.35 ppm) can all AQCR's
be brought into attainment.  However, for a 0.25 ppm standard, over
half of half of the AQCR's needing additional NO  control can be brought
                                                X
into attainment with available control technology.  The total capital
cost is over $2.6 million dollars.  The AQCR attainment fraction falls
to about 40 percent for a standard of 0.15 ppm, and the initial invest-
ment cost raises to $4.3 million.

-------
                                              TABLE 3-1
                         COST AND EFFECTIVENESS OF NO  CONTROLS FOR UTILITY
                            BOILERS - COAL-FIRED (PULVERIZED AND CYCLONE)
                                                  Differential Control Costs
Control .
o /
Techniques '
LEA
Retrofit: Low
NO Burner
X
Control
Potential
11%
40%
Earliest
Year
Available
Present
1980
Initial Investment
(100 $/10° Btu/hr.)
0.75
2.50
Annual Cost
(f/10 Btu)
-.57 ,
0.76
Effect on
Fuel
Consumption
0.5% Decrease
0
Retrofit: Dry
 SCR (only NO )
               90%
1984
75.00
14.2
3.0% Increase
a/
b/
LEA = Low Excess Air
SCR - Selective Catalytic Reduction

Annual Cost = Operating and maintenance costs (includes fuel costs)
If/106 Btu annual heat input = 0.54 $/KW § 5400 hours of operation and 6 104 Btu = 1 kWh.
                             = 0.1 mil/KW hr.
Nominal heating value of coal= 28 KJ/KG (12,000 Btu/lb.)
SOURCE:  Constructed from data in Evans et al,  1978.

-------
                                              TABLE 3-2
                         COST AND EFFECTIVENESS OF NO  CONTROLS FOR UTILITY
                                    BOILERS - COAL-FIRED (STOKERS)
                                                  Differential Control Costs
Control ,
Techniques
LEA
LEA and OSC
Retrofit: Dry
Control
Potential
11%
22%
90%
Earliest
Year
Available
Present
Present
1984
Initial Investment
(100 $/10 Btu/hr.)
0.75
1.60
75.0
Annual Cost ,
(*/10b Btu)
-.57
0.26
14.2
Effect on
Fuel
Consumption
0.5% Decrease
0.5% Increase
3.0% Increase
 SCR (only NO )
a/
b/
LEA = Low Excess Air
OSC = Off-Stoichiometric Combustion
SCR = Selective Catalytic Reduction

Annual Cost = Operating and maintenance costs (includes fuel costs)
       Btu annual heat input  =  0.54 $/KW @ 5400 hours of operation and @ 10  Btu = 1 kWh.
                                  =  0.1 mil/KW hr.
    Nominal heating value of coal =  28 KJ/KG (12,000 Btu/lb.)
SOURCE:  Constructed from data in Evans et al, 1978.

-------
                                              TABLE 3-3
                         COST AND EFFECTIVENESS OF NO  CONTROLS FOR UTILITY
                                     BOILERS - GAS AND OIL FIRED
                                                  Differential Control Costs
Control .
Techniques8'
LEA
LEA and OSC
LEA and OSC
and FGR
LEA and OSC
and NH3
Injection
Retrofit: Dry
SCR
a/ FGR = Flue
Control
Potential
11%
40%
59%

70%

Earliest .,
Year Initial Investment Annual Cost
Available (100 $/10 Btu/hr.) U/10 Btu)
Present 0.38 -1.11
Present 1.00 2. OS
Present 11.3 2.15

1981 10.4 16.9

90% 1982 75.0 16.8
Gas Recirculation
SCR = Selective Catalv
'tic Reduction
Effect on
Fuel
Consumption
0.5% Decrease
1.0% Increase
1.0% Increase

1.0% Increase

3% Increase

 '
           Btu annual heat input = 0.54 $/KW for 5400 hours of operation
                                 =0.1 mil/kWh
104 Btu
1 KW hr.
SOURCE:  Constructed from data in Evans et al,  1978.

-------
                                              TABLE 3-4
                         COST AND EFFECTIVENESS OF NO  CONTROLS FOR INDUSTRIAL


                               BOILERS - COAL-FIRED (PULVERIZED-LARGE)
                                                  Differential Control Costs

Control
Techniques
LEA
Retrofit: Low
NO Burner
X

Control
Potential
10%
40%

Earliest
Year
Available
Present
1984


Initial Investment
(100 $/10 Btu/hr.)
0.9
2.5


Annual Cost ,
U/10 Btu)a/
-1.07
0.26

Effect on
Fuel
Consumption
1% Decrease
0

Retrofit: Dry

 SCR (NO  Only)
        A
90%
1984
63.0
13.9
3% Increase
a'   1{/10  Btu annual heat input = 0.17 $/KG/hr. steam @ 1400 Btu/lb. of steam and

    5400 hours of operation.
SOURCE:  Constructed from data in Evans et al, 1978.

-------
                                              TABLE 3-4
                         COST AND EFFECTIVENESS OP NO  CONTROLS FOR INDUSTRIAL

                               BOILERS - COAL-FIRED (PULVERIZED-LARGE)
                                                  Differential Control Costs
Control
Techniques
LEA
Retrofit: Low
NO Burner
Control
Potential
10%
40%
Earliest
Year
Available
Present
1984
Initial Investment
(100 $/10 Btu/hr.)
0.9
2.5
Annual Cost .
U/10° Btu)a/
-1.07
0.26
Effect on
Fuel
Consumption
1% Decrease
0
Retrofit: Dry    90%
 SCR (N0x Only)
1984
63.0
13.9
3% Increase
a'  If/10  Btu annual heat input = 0.17 $/KG/hr.  steam 8 1400 Btu/lb.  of steam and
    5400 hours of operation.
SOURCE:  Constructed from data in Evans et al, 1978.

-------
                                              TABLE 3-5
                         COST AND EFFECTIVENESS OF NO  CONTROLS FOR INDUSTRIAL

                               BOILERS - COAL-FIRED (PULVERIZED-SMALL)
                                                  Differential Control Costs
Control
Techniques
LEA
Retrofit: Low
NO Burner
Control
Potential
10%
50%
Earliest
Year
Available
Present
1984
Initial Investment
(100 $/10 Btu/hr.)
0.9
3.8
Annual Cost . ,
(*/10b Btu)a/
-1.07
0.27
Effect on
Fuel
Consumption
1% Decrease
0
Retrofit: Dry
 SCR (N0x Only)
90%
1984
63.0
13.9
3% Increase
a'   l
-------
                                              TABLE 3-6
                         COST AND EFFECTIVENESS OP NO  CONTROLS FOR INDUSTRIAL

                                   BOILERS - COAL-FIRED (STOKERS)
                                                  Differential  Control  Costs

Control
Techniques
LEA
LEA and OSC
Retrofit: Dry

Control
Potential
10%
20%
90%
Earliest
Year
Available
Present
Present
1984

Initial Investment
(100 $/10 Btu/hr.)
0.9
2.3
63.0

Annual Cost ,
(t/10° Btu)a/
-1.07
1.32
13.9
Effect on
Fuel
Consumption
1% Decrease
1% Increase
3% Increase
 SCR (N0x Only)
a'  If/10  Btu annual heat input = 0.17 $/KG/hr.  steam 6 1400 Btu/lb.  of steam and
    5400 hours of operation.
SOURCE:  Constructed from data in Evans et al,  1978.

-------
                                              TABLE 3-7
                         COST AND EFFECTIVENESS OF NO  CONTROLS FOR INDUSTRIAL
                                     BOILERS - GAS AND OIL-FIRED
                                                  Differential Control Costs

Control
Techniques
LEA
Retrofit: Low

Control
Potential
10%
50%
Earliest
Year
Available
Present
1981

Initial Investment
(100 $/10 Btu/hr.)
0.9
3.8

Annual Cost .
U/10 Btu)a/
-1.94
0.27
Effect on
Fuel
Consumption
1% Decrease
0
 NO  Burner

Flue Gas Treat-  90%
ment Retrofit:
Dry SCR (NO  only)
           A
1984
63.0
16.3
3% Increase
a/
    Annual Cost = Operating and maintenance costs (includes fuel costs)
    If/10  Btu annual heat input =0.1 mil/kW hour.
SOURCE:  Constructed from data in Evans et ai, 1978.

-------
                                              TABLE 3-8

               COST AND EFFECTIVENESS OF NO  CONTROLS FOR GAS TURBINES AND 1C ENGINES
                                                  Differential  Control Costs
Control
Techniques
Gas Turbines
Water
Injection
Control
Potential
60%
Earliest
Year
Available
Present
Initial Investment
(100 $/10 Btu/hr.)
15.7
Annual Cost .
U/105 Btu)a/
7.4
Effect on
Fuel
Consumption
2% Increase
1C Engines
  Fine Tuning
  and Changing
  A/F
30%
Present
0
33.3
10% Increase
  Retrofit Dry    90%
  SCR
            1984
                 38.0
                      13.2
                      1% Increase
a/
    Annual Cost = Operating and Maintenance Cost (Includes Fuel  Costs)

    M/106 Btu = 0.1 mil/kWh 8 104 Btu/1 kWh.
SOURCE:  Constructed from data in Evans et al,  1978

-------
                                              TABLE 3-9

               COST AND EFFECTIVENESS OF NO  CONTROLS FOR INDUSTRIAL PROCESS FURNACES
                                                  Differential Control Costs
Control
Techniques
LEA
Advanced De
Control
Potential
25%
sign 50%
Earliest
Year
Available
Present
1981
Initial Investment ,
(100 $/10b Btu/hr.)a/
1.30
3.8
Annual Cost, ,
U/105 Btu)b/
-1.5
0.27
Effect on
Fuel
Consumption
1% Decrease
0
 Burner New
 Equipment/
 Retrofit

Retrofit Dry
 SCR (NO  Only)
        Jv
90%
1985
63.0
9.7
2% Increase
a/
    Annual Cost = Operating and Maintenance Costs  (Includes Fuel Costs)
    1 Kwh =10  Btu/hr.

b'  If/106 Btu annual heat input =0.1 mil/kWh.
SOURCE:  Constructed from data in Evans et al, 1978.

-------
                                              TABLE 3-10




                COST AND EFFECTIVENESS OF M>x CONTROLS FOR NITRIC ACID MANUFACTURING
                                                  Differential Control Costs
Control
Techniques
Chilled
Absorption
Control
Potential
90%
Earliest
Year
Available
Present
Initial Investment
10 $/Tpn/Day
2.5
Annual Cost
$/Ton
2.0
Effect on
Fuel
Consumption
0
a/
b/
Based on a model plant of 300 tons/day capacity operated for 8,000 hours per year.





Annual cost = operating and maintenance costs (includes fuel costs)
SOURCE:  Constructed from data in Evans et al,  1978.

-------
                                             TABLE 3-11

              ESTIMATED COST AND ENERGY IMPACTS FOR ALTERNATIVE ONE-HOUR N02 STANDARDS


                                                                 N00 Levels (ppm)a/
.10
5180
.15
1110
.20
242
.25
79
.30
38
.35
25
1.  Capital Cost,
    (1979 $ x 10 )

2.  Annual Cost ,                             1140       194       44        15         8         5
    (1979 $ x 10 )

3.  Additional Barrels of Oilb/               2.5xl07   2.2xl06   6.1xl05   2.1xl05   8.4xl04    6.2xl04
     Used Per Year
    LEA costs are not included.  The additional capital cost due to LEA controls is $240 x 10 .
    The annual fuel savings due to LEA controls is estimated to be 4.0 x 11  barrels/year.

    Assumes that the heat content of crude petroleum is 5.8 x 10  Btu/barrel.

-------
                                   TABLE 3-12
      TYPES OF SOURCES NEEDING ADDITIONAL CONTROL TO MEET SPECIFIED LEVELS


Source Category                                   NO- Levels (ppm)
                                              .15                  .20
1.  Utility Boilers - Coal                     65                   21
2.  Utility Boilers - Oil and Gas          .   455                  184
3.  Industrial Boilers - Coal                 116                   60
4.  Industrial Boilers - Oil and Gas          764                  427
5.  Gas Turbines                               35                    7
6.  1C Engines                                179                  135
7.  Industrial Process Furnaces               365                  164
8.  Nitric Acid Plants                          2                    1
9.  Municipal and Industrial Incinerators       0                    0

                                              1981                 999

-------
                                                                TABLE  3-13
                                           TYPES OF  INDUSTRIES  NEEDING ADDITIONAL CONTROL TO MEET
          SIC
131 - Oil and Gas Extraction



 28 - Chemicals




 29 - Petroleum Refining




331 - Steel



333 - Metal Melting




 49 - Utilities




      Misc.
                                                      SPECIFIED ONE-HOUR N02 STANDARDS
NO, Levels (ppa)

Nos. of
Plants
5
24
25
20
3
67
46
190
JIS
Nos. of
Sources
214
152
3S1
4 SO
11
555
248
1981
.20
Nos. of
Plants
4
9
17
7
1
33
19
90

Nos. of
Sources
153
S3
184
261
8
214
126
999
.25
Nos. of
Plants
2
6
9
3
1
IS
11
47

Nos. of
Sources
109
27
92
143
6
78
89
S44

-------
                                                                 TABLE 3-14
                                       ESTIMATED INDUSTRY CONTROL COSTS FOR ALTERNATIVE ONE-IOUR
                                                          STANDARDS (1979 f x 106)
          SIC
131 - Oil and Gas Extraction
 28 - Cliealcal
 29 - Petroleua Refining
331 - Steel
333 - Metal Melting
 49 - Electric and Gas Utilities
      Misc.
NO, Levels (PPM)

Capital
IS
21
63
253
10
741
27
1110
.15
Annual
6
4
IS
37
2
119
11
194

Capital
S
9
24
71
7
110
16
242
.20
Annual
2
2
S
11
1
18
S
44

Capital
1
6
11
29
1
28
3
79
.25
Annual
NIJG.
1
3
5
NEC.
4
2
IS

-------
                                   TABLE 3-15
       CURRENT OXIDES OF NITROGEN EMISSION STANDARDS AND EMISSION FACTORS
Vehicle
Category
Autos

Light Trucks
0-6000 Ibs.

Light Trucks
6000-8500 Ibs
Emission
Standard
(gm/mi)
2.0
1.0
3.1
2.3
1.4
2.3
. 1.4
Model
Year
1977-80
1981+
1975-78
1979-84
1985+
1979-84
1985+
New Vehicle ,
Emission Ratea'
(gm/mi)
1.5
0.29
2.45
1.73
.41
1.73
.41
Deterioration
Rate*/
(gm/mi)
0.16
.22
0.00
.11
.22
.11
.22
Heavy Trucks   5.35
Gasoline
1985+
3.99
.34
Heavy Trucks   5.35
Diesel
1985+
5.35
.00
a/
    These are the rates assumed in EPA's mobile source emission model
    (MOBILE 1)

-------
                              TABLE 3-16
          ESTIMATED COST AND EMISSION REDUCTION EFFECTIVENESS
           FOR MOBILE SOURCE ISM PROGRAMS N0x INSPECTION AND
                    LOADED MODE TEST 1975-77 MODELS

                                   Loaded Short Test Failure Rate*
                                   10%         20%           30%
Percent Reduction in N0x           5.9         10.2          12.9
Cost Per Inspected Vehicle($)      2.20         2.85          3.60
Annual Cost-effectiveness,         1270         994           990
  ($/ton removed)
*A short test is a dynamometer test at constant vehicle speed.

                   EGR INSPECTION, 1975-77 MODELS

Percent vehicles failing inspection               13
Percent NO  reduction                              6
Annual Cost of inspection and repair ($)        2.00
Annual Cost-effectiveness                       1210
  ($/ton removed)
Source:  (U.S. EPA, 1979)

-------
                              TABLE 3-17
            THE NUMBER OF NONATTAINMENT AQCR'S IN 1984 FOR

            ALTERNATIVE ONE-HOUR N02 STANDARDS AND VARIOUS

                    AREA SOURCE CONTROL STRATEGIES
Strategy
                    Standard (ppm)

Baseline
im
A. Loaded Test
B. EGR Inspec-
tion
.10
135

135
135
.15
81

78
79
.20
41

39
40
.25
21

19
21
.35
6

6
6
.50
0

0
0
ISM (Loaded
  Test) plus
  TCM's
134
77
38
19
ISM (Loaded
  Test) plus
  TCM's and
  Controls
  on Existing
  Stationary
  Area Sources
133
69
36
16
ISM (Loaded
  Test) plus
  TCM's and
  Controls on
  New and
  Existing
  Stationary
  Area Sources
132
64
36
16

-------
                                   TABLE 3-18
                 THE NUMBER OF NONATTAINMENT AQCR'S IN 1990 FOR

                 ALTERNATIVE ONE-HOUR N02 STANDARDS AND VARIOUS

                         AREA SOURCE CONTROL STRATEGIES
Strategy
Baseline

ISM

  A.  Loaded Test

  B.  EGR Inspection

ISM (Loaded Test)
  Plus TCM's

ISM Plus TCM's
  and a 0.4 g/mile
  Auto Emission
  Standard

ISM Plus TCM's
  Plus a 0.4 g/mile
  Auto Emission
  Standard Plus
  Controls on
  Exited Stationary
  Area Sources

ISM Plus TCM's,
  0.4 g/mile
  Emission Stan-
  dard, and
  Controls on
  New and
  Existing
  Stationary
  Area Sources

Low Truck Emission
  Standards

High Truck Emission
  Standards
              Standard
.10


135



134

134

134


134




129
127
134


135
.15


 75



 68

 71

 68


 62




 59
           55
           68
           79
.20


 40



 35

 39

 35


 32




 31
           28
           35
           40
                            .25


                             19



                             16

                             17

                             16


                             16
                             15-
         14
         16
         19
.35


  6



  5

  5

  5


  5
.50


  0
          0


          0

-------
                                   TABLE 3-19

                    CAPITAL COSTS OF AREA SOURCE NO  CONTROLS
                                                   A
                     FOR ALTERNATIVE ONE-HOUR N02 STANDARDS

                             (1979 $ x 106) IN 1984


Strategy                 	Standard (ppm)	

                         .10          .15          .20           .25           .35
ISM

     A.  Loaded Test      14            8            4            2            1

     B.  EGR Inspection    0            0            0            0            0
ISM (Loaded Mode)       2630         2180         1790         1520          620
Plus Controls for
Existing Sta-
tionary Area
Sources
    (Loaded Mode)       3660         3210         2820         2550         1650
Plus Controls for
New and Existing
Stationary Area
Sources

-------
                                   TABLE 3-20
                  ANNUALIZED COSTS OF AREA SOURCE NO  CONTROLS

                     FOR ALTERNATIVE ONE-HOUR N02 STANDARDS

                             (1979 $ x 106) IN 1984


Strategy                	Standard (ppm)	

                         .10          .15          .20          .25          .35


I§M
  A.  Loaded Test        260          200          150          100           40

  B.  EGR Inspection     160          120           90           60           20
ISM (Loaded Test)        820          670          540          400          160
  Plus TCM's
  and Controls
  on Existing
  Stationary
  Sources


ISM  (Loaded Test)      1010          860          730          580          324
  Plus TCM's
  and Controls
  on New and
  Existing
  Stationary
  Area Sources

-------
                                   TABLE 3-21
                    CAPITAL COSTS OF AREA SOURCE NO  CONTROLS

                     FOR ALTERNATIVE ONE-HOUR N02 STANDARDS

                             (1979 $ x 106) IN 1990
Strategy
                                        Standard (ppm)
ISM
     A.  Loaded Test

     B.  EGR Inspection
                         .10
                          14

                           0
                                      .15
    8

    0
               .20
  4

  0
.25



  2

  0
                          .35
    1

    0
ISM (Loaded Mode)
Plus 0.4 g/mile
Auto Emission
Standard

ISM (loaded Mode)
Plus 0.4 g/mile
Emission Standard
Controls and
Existing Stationary
Area Sources
    (Loaded Mode)
Plus 0.4 g/mile
Emissions Standard
and Controls on
New and Existing
Stationary Area
Sources
                        6720
 6720
6710
                        8920
 8390
8040
                       10610
10080
9730
6710
6710
7440
6870
9130
8560

-------
                                   TABLE 3-22
                  ANNUALIZED COSTS OF AREA SOURCE NO  CONTROLS

                     FOR ALTERNATIVE ONE-HOUR N02 STANDARDS

                             C1979 $ x 106) IN 1990
Strategy                    	Standard (ppm)	

                              .10       .15       .20       .25       .35
ISM

  A.  Loaded Test             260       190       150        90        40

  B.  EGR Inspection          160       120       100        60        20

ISM (Loaded Mode)            2100      2030      2000      1950      1900
  Plus a 0.4 g/mile
  Auto Emission
  Standard
ISM (Loaded Mode)            2560      2390      2300      2110      1980
  Plus a 0.4 g/mile
  Emission Standard
  and Controls on
  Existing Stationary
  Area Sources

ISM (Loaded Mode)            2860      2690      2580      2400      2240
  Plus a 0.4 g/mile
  Emission Standard
  and Controls on
  New and Existing
  Stationary Area
  Sources

-------
               TABLE 3-23
 NEW VEHICLE EMISSION AND DETERIORATION
RATES FOR ALTERNATIVE EMISSION STANDARDS
Emission
Vehicle Standard Model
Category ^gm/mi) Year
Autos
Light Trucks,
0-6000 Ibs.
Light Trucks,
6000-8500 Ibs.
Heavy Trucks,
Gasoline
Heavy Trucks,
Diesel
0.4 1985+ :
1.2 1985+
1.7 1985+
1.2 1985+
1.7
3.0 1985+
6.0
3.0 1985+
6.0
New Vehicle
Emission Rate
Cgm/mi)
0.12
.35
.50
.35
.50
2.24
4.47
3.00
6.00
Deterioratioi
Rate
Cgm/mi)
0.22
.19
.27
.19
.27
.19
.38
.00
.00

-------
                              TABLE 3-24
              NUMBER OF NONATTAINMENT AQCR'S IN 1990 FOR

                        DNE-HOUR N02 STANDARDS

                         TRUCK CONTROL OPTIONS
ALTERNATIVE ONE-HOUR N02 STANDARDS AND VARIOUS
Cases               	Standard  (ppm)	

                        .10        .15        .20        .25        .35
Light Truck Std.        135        75        40        19
  3 1.2 gin/mi.
Light Truck Std.        135        75        40        19
  » 1.7 an/mi.
Heavy Truck Std.        134        68        36         16
  =3.0 gm/mi.
Heavy Truck Std.        135         79        40         19
  =6.0 gm/mi.

-------
                                 TABLE 3-25
                       STATIONARY AREA SOURCE CONTROLS
                                           Source Categories
Residential Comm./Inst.
Oil/Gas Coal
Controls on New Sources
Emission Reduc- 50 24
tion (%)
Capital Cost 1500 2430
($ per ton
Reduced)
Annual Cost 270 495
($ per ton
Reduced)
Controls on Existing Sources
Emission Reduc- 0 20
tion (%)
Capital Cost 0 3600
Comm./Inst. Industrial Industria!
Oil/Gas Coal Oil/Gas

50 24 50

12000 900 4440
2390 175 890


40 20 40

18460 1350 8550
  ($ per ton
  Reduced)

Annual Cost
  ($ per ton
  Reduced)
740
4120
290
1530

-------
                              TABLE 3-26

           ROADWAY CHARACTERISTICS AND MODELING ASSUMPTIONS

                      USED  IN THE HIWAY MODELING
Roadway Characteristics

1.  Line source segment is five kilometers long.

2.  Emission height » 0 m.*                     :

3.  The eight-lane roadway has peak hour traffic and free-flow conditions:

     Northbound » 7000 vehicles/hour

     Southbound = 5500 vehicles/hour

     Average vehicle speed "55 mph

     Vehicle mix;  See Section 3

4.  Highway width * 32 meters

5.  Median width =1.0 meters


Modeling Assumptions

1.  Wind speed =1.0 m/sec.

2.  Mixing height does not limit vertical dispersion in the vicinty
    of the roadway.

3.  Neutral atmospheric stability (Pasquill-Gifford Class D) is assumed.

4.  Angle of wind with the roadway:

    a.  90°

    b.  5°
  Higher emission heights were assumed to simulate the turbulent mixing
  zone on the highway, but the effect on the estimates of ambient levels
  was negligible.

-------
                                   TABLE 3-27
            VEHICLE EMISSION FACTORS AND VEHICLE MILES TRAVELED (VMT)
                        SHARES USE IN THE HIWAY MODELING
                               (55 mph, 75°F, 1984)
Light-Duty Passenger
Gasoline
Diesel
                                   ALL WAIVERS APPROVED (CASE 1)
Light-Duty Truck
(Less Than 6000 Ib.)
Gasoline    Diesel
Light-Duty Truck
(6000-8500 Ib.)
Gasoline  Diesel
            Heavy-Duty Truck
                Or Bus	
            Gasoline  Diesel
Emission Factor      2.544
 (g/mi)
VMT Share           60.87
               1.517
               4.01
              3.805
             13.05
             2.007
             0.88
 4.817
10.97
2.000
0.84
13.456
 2.59
35.070
 4.25

-------
                 TABLE 3-27  (CONTINUED)

VEHICLE EMISSION FACTORS AND VEHICLE MILES TRAVELED  (VMT)
            SHARES USE IN THE IIIWAY MODELING
                   (55 mph, 75°F, 1984)
NO WAIVERS APPROVED, DIESEL AUTOS MEET
Emission Factor
(g/mi)
VMT Share
Light-Duty Passenger
Gasoline Diesel
2.544 1.099
60.87 4.01
Light-Duty Truck
(Less Than 6000 Ib.)
Gasoline Diesel
3.805 2.007
13.05 0.88
1.0 GRAMS PER MILE (CASE 2)
Light -Duty Truck
(6000-8500 Ib.)
Gasoline Diesel
4.817 2.000
10.97 0.84
Heavy- Duty Truck
Or Bus
Gasoline Diesel
13.456 35.070
2.59 4.25

-------
                             TABLE 3-27 (CONTINUED)


            VEHICLE EMISSION FACTORS AND VEHICLE MILES TRAVELED (VMT)

                        SHARES USE IN THE IIIWAY MODELING

                               (55 mph, 75°F, 1984)
          NO WAIVERS APPROVED, DIESELS REPLACED BY GASOLINE AUTOS (CASE 3)
Emission Factor
 (g/mi)

VMT Share
Light-Duty Passenger
Gasoline      Diesel
            Light-Duty Truck
            (Less Than 6000 Ib.)
                       Light-Duty Truck
                       (6000-8500 Ib.)
            Gasoline    Diesel     Gasoline  Diesel
 2.477
64.26
1.600
0.61
 3.805
13.05
2.007
0.88
 4.817     2.000
                                 Heavy-Duty Truck
                                     Or Bus	
                                 Gasoline  Diesel
            13.456    35.070
10.97
0.84
2.59
4.25

-------
     3-1
.10
.10

-------
                                  FIGURE 3-2
             ESTIMATED IMPACTS FOR NINE REGIONAL SCENARIOS IN A LARGE URBAN AREA:

                         REGIONAL NITROGEN OXIDES EMISSIONS
       MX
      •ESIIMAUO AlSOt UIE RfCIONAl CIIANCf IHQI)  [MISSIONS »OH
      fROIOIYM URIAH BtCIOH Of ArfROIIMAICIVViOO.OM- 1.UI.OOI
      stUA roruiAiion AND AH AVIRAGI IA« WUXOAY HO IIICHWA*
      EMISSIONS or lit Mr IVY>.
      Esufliitat usumi uointainiplad Utliic Haw aindiuaas.
Reference:   Air Quality  Impacts  of Transit Improvement, Preferential
              Lane,  and Carpool/Vanpool  Programs, U.S.  EPA,  OTLUP,
              Washington,  D.C., March 1978.

-------
%f    \
           •ID
                                CAPITAL
                                   • It
2 -I-
                             7.1
                                                             .55    Pff.
                                                             r/v;

-------
                                    TABLE 3-28
                     PREDICTED PEAK HOUR N02 CONCENTRATIONS IN

                                    1984 (ppm)
                                Oblique Wind (5°)*
Receptor
Legation (m)**
,•
(- 4
" 10
25
-50
LOO
250
>00
Case 1
With NO
Waiver x
1.273
1.749
2.389
2.870
3.220
2.243
1.255
Case 2
No Waiver,
With Diesels
1.263
1.738
2.376
2.856
3.205
2.232
1.249
Case 2-
Case 1
-.010
-.011
-.013
-.014
-.015
-.011
-.006
Case 3
No Waiver,
No Diesels
1.250
1.725
2.363
2.846
3.197
2.226
1.245
Case 3'
Case 1
.023
.024
.025
.024
.023
.017
.010
                             Perpendicular Wind (90 )
RL^eptor
Location (m)**
../-
) 4
10
"25
50
loo
250
•>oo
f
}
Case 1
With NO
Waiver x
.191
.286
.423
.503
.513
.297
.151
Case 2
No Waiver
With Diesels
.190
.284
.421
.501
.511
.295
.150
Case 2-
Case 1
-.001
-.002
-.002
-.002
-.002
-.002
-.001
Case 3
No Waiver,
No Diesels
.188
.282
.419
.499
.509
.294
.149
Case 3-
Case 1
.003
.004
.004
.004
.004
.003
.002

Results are unrealistically high; see text for a discussion.
Distance is measured from the edge of the roadway in the
direction of the wind field.

-------
                              TABLE 3-29

        ESTIMATED NUMBER OF AQCR'S NEEDING CONTROLS AND CONTROL

       COSTS FOR POINT AND AREA SOURCES TOGETHER FOR ALTERNATIVE

                    ONE-HOUR N02 STANDARDS IN 1984


                              .	Standard (ppnp	
                           .15           .20            .25

Number of AQCR's Not       106            61             37
 Attaining the Stan-
 dard Without Additional
 NO  Controls
Number of AQCR's Not        64            36             16
 Attaining the Stan-
 dard with all
 Additional NO
 Controls     x
Capital Cost of           4320          3070           2631
 Controls    ,
 (1979 $ x 10°)
Annual Cost of            1074           794            608
 Controls    ,
 (1979 $ x 10°)

-------
                              REFERENCES
Chock, D.  January 1977.  "General Motors Sulfate Dispersion Experiment,
Assessment of the EPA HIWAY Model", JAPCA. 27,

Evans, M., and Castaldini, c.  March 1978.  Summary of Combustion
Modification NO  Controls - Emission Levels, Costs, and Fuel Impacts,
Acurex, Mountainview, CA.

Keyes, D., et al.  December 1978a.  Short-Term NO., Standards, Volume II;
Estimated Cost of Meeting Alternative Standards, EEA, Arlington, VA

Keyes, D., et al December 1978b.  Short-Term NO,, Standards, Volume III:
An Investigation of Short-Term NO., Concentrations in Chicago, EEA,
Arlington, VA.

Trijonis, J. February 1978.  Empirical Relationships Between Atmospheric
Nitrogen Dioxide and its Precursors, (EPA-600/3-78-018), U.S. EPA, ORD-
ESRL, Research Triangle Park, NC.

U.S. Bureau of the Census.  1974.  Statistical Abstract of the
United States; 1974.  (95th Edition) Washington, D.C.

U.S. EPA.  1977.  Proceedings of the Los Angeles Catalyst Study Symposium,
ORD, Research Triangle Park, NC.

U.S. EPA.  April 1979.  "Questions and Answers Concerning the Technical
Details of Inspection and Maintenance", OMSAPC, Ann Arbor Michigan.

U.S. EPA.  March 1978a.  Mobile Source Emission Factors, (EPA-400/ 9-78-
005), OAWM, Washington, D.C.

U.S. EPA.  March 1978b.  Air Quality Impacts of Transit Improvement,
Preferential Lane, and Carpool/Vanpool Programs, OTLUP, Washington D.C.,
(NTIS No. PB-282-346).

U.S. EPA, U.S. FEA, U.S. DOT.  May 19, 1977.  An Analysis of Alternative
Motor Vehicle Emission Standards, Washington, D.C.

Voorhees, A. and Associates. April 1979.  Transportation System
Management-Institutional and Planning Research-Working Papers.

-------
                        4.  NEW SOURCE CONTROLS
The costs and impacts of meeting alternative ambient standards of NO-
are the primary focus of the regulatory impact analysis.  However,
another perspective on the magnitude of these costs can be gained by
comparing them to the costs of meeting the emission standards for new
sources — the New Source Performance Standards (NSPS's) for stationary
sources and the Federal Motor Vehicle Control Program (FMVCP).  This
section will focus on these costs. In addition, preliminary judgments
will be made regarding possible PSD increments for a short-term NO-
standard and the difficulty new sources may have in meeting them.
                *
4.1  COST OF MEETING NEW SOURCE PERFORMANCE STANDARDS
Total costs of new source control to meet the NSPS's for NO  were
estimated for the following major categories of stationary sources:

  •  Utility Boilers:  Coal-fired
  •  Utility Boilers:  Oil- and gas-fired
  •  Industrial Boilers:  Coal-fired
  •  Industrial Boilers:  Oil- and gas-fired
  •  Gas Turbines
  •  Reciprocating 1C Engines
  •  Nitric Acid Plants

The capital costs are summed for the period beginning with the promul-
gation of the appropriate NSPS  (see Table 4-1) and ending in 1990.
Annual costs are for 1990 and include an annualized capital charge and
operating and maintenance costs.  All costs are in 1979 dollars.

-------
                               TABLE 4-1

                NEW SOURCE PERFORMANCE STANDARDS FOR NO
     SOURCE CATEGORY
   NSPS
DATE OF
PROPOSAL
1.  Utility Boilers:  Coal-fired   0.70 lb/10  Btu

                                   0.60 lb/106 Btu
2.  Utility Boilers:  Oil-fired

3.  Utility Boilers:  Gas-fired

4.  Nitric Acid Plants

                *
5.  Gas Turbines
0.30 lb/10  Btu

0.20 lb/106 Btu

3 Ib/ton of 100%

  acid produced
75 ppm NO
   1971

   1978

   1971

   1971

   1971


   1977
  The proposed standards would apply to all new, modified and
  reconstructed stationary gas turbines with greater than 1000 hp
  heat input.  Gas turbines with a heat input at peak load from
  1000-10,000 hp would be exempt from the NO  emission limit for
  5 years from the date of proposal.  Emergency — standby gas
  turbines would be exempt permanently from the NO  emission limit.

-------
4.1.1  Utility  Boilers;   Coal-fired
Based on the uncontrolled emission factor  in AP-42,  (U.S. EPA.  1978a)
new coal-fired  utility boilers would have  to institute control  procedures
to meet the 0.7 Ib./lO  Btu NSPS proposed  in 1971.   However, the  lag
time between NSPS proposal date and boiler start-up  date has averaged
about five years for utilities, so that boilers constructed between 1971
and 1975-76 were not affected by this regulation*.   Furthermore,  boiler
fire-box designs appearing after 1975 are  reported to have reduced
uncontrolled N0x levels  to at least 0.7 Ib./lO  Btu, and at no  additional
cost* (U.S. EPA. 1978b).   Thus, the cost of this NSPS is assumed  to be
zero.

Similarly, recent studies have indicated that the current boiler  design
employed by each of  the  four major manufacturers of  coal-fired  utility
boilers should  be sufficient to comply with the proposed 0.6 lb./106 Btu
emission limits (U.S.  EPA.  1978b).  Where  these designs prove to  be less
than adequate,  minor modifications to current designs should alleviate
the problem.  Thus,  coal-fired utility boilers should experience  no
increase in capital  or 0§M costs as a result of the  NSPS for NO .
                                                                A

4.1.2  Utility  Boiler:   Oil  and Gas-Fired
Based on the AP-42 emission factors for oil and gas  utility boilers,        ''
units installed since 1971 should have been able to  meet NSPS levels
without additional control (U.S. EPA. 1978a).  Where control has  been
found necessary for  some residual fuel oil boilers,  combustion  modification
(low excess air) at  negligible capital and no annual expense has  undoubtedly
been the preferred option (Evans and Castaldini. 1976).

4.1.3  Industrial Boilers:   Coal-Fired
There currently are  no NSPS's for coal-fired industrial boilers.  The
average SIP requirement  is 0.7 lb/10  Btu  (August 1978)  (Broz,  et al. 1978).
*Personal communication with John Copeland, U.S. EPA, OAQPS-ESED, Durham
 NC, February 1979.

-------
Based on the latest AP-42 emission factors, (U.S. EPA. 1978a) it appears
that new industrial boilers can meet this level without control.

However, an NSPS for coal-fired industrial boilers is now under review.
The present schedule calls for proposal of the standard in October 1980.
Assuming at least one-year time lag for the construction of a new indus-
trial boiler, the costs calculated here are for 1982-1990.  Because the
level of the NSPS for coal-fired industrial boilers is uncertain, the
following alternative control levels are evaluated:

     Recommended NO  Control Levels for Industrial Boilers

                                    Control Level* (Ib. N0x**/106 Btu)
                               Moderate        Intermediate        Stringent

Units > 29 MW Heat Input          0.7              0.6                0.5
Units < 29 MW Heat Input          0.5              0.4                0.3
 * Moderate control represents that level which is achievable applying
   currently available technology.  The stringent control option is
   technology-forcing and encompasses technology expected to be commer-
   cially available within five years.
** Measured as N0_

Based on recent simulations using EEA's Natural Gas Pricing Model, new
coal-fired industrial boiler capacity between 1982 and 1990 is estimated
to be 3.45 X 10   Btu/hr.  This assumes that 60 percent of the new
capacity will be coal-fired.

Other assumptions used in this cost analysis are as follows:

-------
  •  Average annual operating rate is 5500 hours
  •  10  Btu/hr heat input * 1250 Ibs/hr steam output
  •  All coal-fired industrial boilers will be stokers
  •  Based on projections of the size distribution of future
     industrial boilers, one-third are assumed to be less
     than 29 MW heat input and two-thirds are assumed to be
     greater than 29 MW heat input.
  •  Differential control costs are taken from Evans and
     Cast.aldini, 1976) and assume a capital recovery rate
     of 16 percent per year
  •  All boilers purchased in this time period are assumed
     to be operating in 1990

The results of the industrial boiler cost analysis are shown in Table
4-2.

4.1.4  Industrial Boilers:  Oil- and Gas-Fired
As with coal-fired boilers, there currently are no NSPS's for oil-and
gas-fired industrial boilers.  The average SIP requirements for these
boilers are 0.3 Ib of NO^IO6 Btu for oil-fired and 0.2 Ib of N0x/106
Btu for gas-fired units (Broz, et al. 1978).  More stringent standards
for these categories may be promulgated at the same time as the coal-
fired industrial boiler NSPS's.  Therefore, a range of alternative
NSPS's is evaluated in this cost analysis.

Assuming that 40 percent of the new industrial boiler capacity from
1982-1990 will be fired by oil or gas, new boiler capacity for these
units is projected to be 2.3 X 10   Btu/hr between 1982 and 1990.  The
assumptions used are essentially the same as those used for the coal-
fired analysis.  Differential control costs are taken from Evans, Castaldini.
1976.  Because the costs in this reference were given separately for
watertube and firetube boilers, two cases are shown in Table 4-3.

-------
                               TABLE 4-2

            COAL-FIRED INDUSTRIAL BOILER CONTROL COSTS FOR

               ALTERNATIVE LEVELS OF NEW SOURCE CONTROL
                      (1982-1990) (1979 DOLLARS)
                                              Initial       Annual
                                              Investment    Cost in 1990
                                              (10  $)       (106 $)
Technology 1:  Moderate Control

  0.7 Ib of NOX/106 Btu for Units >29MW           0            0

  0.5 Ib of NOX/106 Btu for units <29MW           0            0
                                                  0            0
Technology 2:  Intermediate Control
  0.6 Ib of NOX/106 Btu for units >29MW         24.9          5.5

  0.4 Ib of NOX/106 Btu for units <29MW         	0_           0

                                                24.9          5.5

Technology 3:  Stringent Control                          :

  0.5 Ib of NOX/106 Btu for units >29MW         73.2         14.6

  0.3 Ib of NOX/106 Btu for units <29MW         35.4          7.1

                                               108.6         21.7

Technology 1:  No controls necessary.

Technology 2:  Low excess air for units >29MW.  No control for units <29MW.

Technology 3:  Low excess air and off-stoichiometric combustion for all
               units.

-------
                                         TABLE 4-3
              OIL- AND GAS-FIRED INDUSTRIAL BOILER CONTROL COSTS FOR ALTERNATIVE
                         LEVELS OP NEW SOURCE CONTROL (1982-1990)
                                      (1979 DOLLARS)
CASE I - Assume all oil- and gas-fired boilers are watertube boilers.
Alternative NSPS's
(Ib. of NO../ 10 Btu)







CASE

	 x
0.3
0.25

0.20

0.15
0.10
II - Assume all

Alternative NSPS's
(Ib. of NO../ 10 Btu)






0.3
0.25
0.20

0.15
0.10
Controls
Needed
No control
Initial Investment
' do6 $)
0
Annual Cost
in 1990
do6 $)
0
Low excess air and off-stoichio-
metric combustion
Advanced staged combustion and
LEA
Low NO burners and LEA
Low NO burner and advanced
firebox design
54.5

90.8
120.6
145-255
11.0

18.1
23.6
30-52
oil- and gas-fired boilers are firetube boilers.

Controls
Needed
No control .••'
Low excess air (LEA)
LEA and flue gas recirculation
(FGR)
Low NO burners and FGR
X
Low NO burners, FGR, and

Initial Investment
(106 $)
0
72.7

253
398

Annual Cost
in 1990
do6 $)
0
14.9

51.1
79.2

                              advanced firebox design
545
109.0

-------
4.1.5  Stationary Gas Turbines
The total nationwide cost of a 75 ppm NO  emission standard for stationary
                                        A
gas turbines is calculated on the basis of 00 expected sales from the
time the proposed standard takes effect on different sized gas turbines
through 1990 and (2) the costs for meeting the standard estimated in
U.S. EPA. 1976.  The sales projections used here are extrapolated from
Table 7.27 in this reference.  Because the timing of the NSPS differs by
the size of the gas turbine Csee Table 4-1), the average unit size was
determined for each application listed in Table 7.27 in U.S. EPA. 1976.
Based on information gathered from this reference, it was assumed that
stationary gas turbines used by electric utilities have an average unit
size of 80 MW (100,000 hp).   Therefore, all stationary gas turbines
built after October 3, 1977 (NSPS proposal date) for use by electric
utilities will have to comply with the 75 ppm NO  standard.  Because
                                                A
stationary gas turbines greater than 20,000 hp heat input are produced
on a custom order basis only, the lead time for delivery is approximately
2 years.  Therefore, it was assumed that costs of the NSPS would not be
incurred until 1979 for turbines of this size.  This is a much longer
lead time than that required for the smaller mass-produced models.

For the other applications of stationary gas turbines listed in Table
4-4, the average unit size is assumed to be between 1000-10,000 hp.
Therefore, they are exempt from the proposed NSPS for five years from
the date of proposal.  Therefore, costs of the NSPS for 1000-10,000 hp
gas turbines are assumed to be incurred starting in 1983.  Emergency-
standby stationary gas turbines were exempted from standards of performance
limiting NO  emissions, so this application is not listed in Table 4-4.
           A

The total capital and annualized cost estimates shown in Table 4-4
were made using the cost estimates in Table 7.26 of U.S. EPA. 1979 and

-------
                                                                  TABLE  4-4
                                         STATIONARY GAS TURBINES - TOTAL CAPITAL AND ANNUALIZED COST
                                                               (1979  DOLLARS)
                      Year Costs
                                           Total  Annual  Cost
of NSPS Incremental Sales Capital Total Capital , Assuaed Ave. Annual lied in 1990
Begin to be High Low Costs Cost Thru 1990 (10 1) Annual Usage Cost per Kwh (10 $)
Applications Incurred (MM) ($/Kt«h) High Low (hr/year) (pHUs/Kwh) High Low
1.
2.
S.


4.
Utilities 1979 . 47,830 . 29,300 3. 25 155.4 ' 95.2 500 1.9
Oil and Gas 1983 7,970 1,880 6.03 48.1 > 11.3 8000 0.8
Industry • . •
Private Industry
Electric Power
Ceneration
a. oil and gas 1983 600 55 6.02 4.0 0.3 8000 1.5
b. other 1983 1380 6.02 8.3 2000 1.3
industry
Other 1983 835 6.02 • 5.0 2000 1.3
45.4 27.9
48.2 11.3

7.9 0.6
3.5
2.1
                                                                       220.8
120.1
107.1
45.4
Source:  U.S.  CPA.  1976

-------
the projection of new sales.   Annual!zed costs are a function of assumed
hours of operation.  These are shown in Table 4-4.

4.1.6  Reciprocating Internal Combustion Engines
The most recent estimates of the soon-to-be-proposed NSPS's for 1C
engines are as follows:

          Fuel Type                          NSPS (ppm NO )
            Gas                                   700
            Diesel                                600
            Dual Fuel                             600

These standards should be proposed in either March or April of 1979.
They will take effect 30 months after they are proposed.  For the purposes
of this analysis, it is assumed that the NSPS's for 1C engines take
effect at the beginning of 1982.  Therefore, costs of NO  control for 1C
                                                        A
engines are summed from 1982-1990.

Based on comparisons of uncontrolled emission rates for new 1C engines
with the estimated standards, the following percentage reductions
necessary to meet the NSPS's are estimated:

     Fuel Type                     Percentage Emission Reduction Needed
                                       to Meet the Estimated NSPS's

     Gas                                          60%
     Dual fuel                                    25%
     Diesel                                       45%

The cost information as it appears in Reference 7 is shown below:

-------
             Annual Costs of Alternative Standards in the
                Fifth Year After  Implementation  CIO $)
                Application         Alternative NO  Reductions
                                    20%       40%          60%
Gas Production and
Transmission
Electric Generation
Other Application
All Applications

6.5
8.7
2.9
17.6

15.0
9.8
3.7
28.5

19.6
21.9
7.8
49.3
Because the cost information for alternative NO  reductions was broken
                                               A
down by application rather than by fuel, a conservative assumption is
made that a 60 percent reduction in NO  is needed for all applications.
                                      A
Based on this assumption, and the assumption that the constant annual 1C
engine sales estimated in Acurex. 1978 will continue thru 1990, the.
following costs are estimated for the period 1982-1990:

                      NO  Control Costs for 1C Engines

               Capital Cost                       Annual Cost
                 (106 $)                          in 1990 (106 $)
                    16                                   80

 The capital costs  for the 1982-1990 period are  estimated  by multiplying
the $2 million per  year capital  cost figure given  in Reference 7  by
eight.  The annual  cost in 1990  is  estimated  by  multiplying by 1.6 the

-------
$50 million fifth year cost associated with a 60 percent NO  reduction
(i.e., the annual cost during the fifth year after implementation of the
NSPS) reported in this reference.

4.1.7  Nitric Acid Plants
Capital and operating costs for control of NO  in new nitric acid
                                             Jt
plants are taken from U.S. EPA. 1978c.  These costs (representative of a
270 Mg/day plant) are used along with growth estimates provided by MITRE
Corporation to calculate the total cost of attaining an NSPS of 3 Ibs.
of NO /ton of 100 percent acid produced.  MITRE estimates that an average
     A
of four nitric acid plants of 420 Mg/day capacity will be built each
year*. Actual growth in nitric acid plants from 1971-1978 has been
somewhat less than four plants per year.  However, the four new plants
per year assumption is used in this analysis.

Costs for six different NO  abatement systems for nitric acid plants are
                          Jv
shown in Table 6-4 of U.S. EPA. 1978c.  In order to determine a high and
a low cost estimate for new source controls, costs were developed for
both a molecular sieve control process and a chilled absorption process.
Chilled absorption is used primarily for retrofit of existing plants,
however, and cost estimates for this control technique may be unrealistically
low.  The catalytic reduction process has higher costs than the molecular
sieve process and was used to establish the NSPS for nitric acid plants
originally.  However, since that time, fuel costs have risen to the
point where catalytic abatement is not economically attractive for new
nitric acid plants.  Therefore, the molecular sieve process is analyzed
here to generate a high cost estimate.
*Personal communication with Linda Duncan, MITRE Corporation, McLean,
 Virginia, December 1978.

-------
The estimates of total cost of all nitric acid plants built from 1971-
1990 attaining an NSPS of 3 Ibs of NO /ton are shown below:
                                     Ai

     Control                                           Annual
    Technique            Capital Cost CIO6 $1      Total Cost CIO6 $)

   Molecular Sieve                13                      40.8
   Chilled Absorption              6                      15.9
   (CDL/Vitok Process)

4.1.8  Summary
Tables 5 and 6 summarize the cumulative capital costs and annual costs
for stationary source NOX control to meet the new source performance
standard levels.  The moderate control scenario assumes that the current
NSPS levels do not change between now and 1990.  The intermediate and
stringent control levels show how costs would increase if stricter
standards for industrial boilers were promulgated in 1981.  The recommended
NO  control levels for industrial boilers were detailed earlier in
  A
Section 4.1.3.  The ranges in control costs indicate the uncertainty in
both the number of new sources that will be constructed between now -and
1990 and the costs of control techniques.

-------
                               TABLE 4-5
       CUMULATIVE CAPITAL COSTS FOR STATIONARY SOURCE NO  CONTROL

                TO MEET THE NSPS LEVELS Q.979 $ x 106)
Source Categories
                                        Moderate
        Control Levels*
           Intermediate
             Stringent
Utility Boilers

Industrial Boilers

   Coal-fired

   Oil and gas-fired

     Case I

     Case II

 Stationary Gas Turbines

     High and (Low) Sales

Internal Combustion Engines

Nitric Acid Plants
  0

  0
                                          156-256
 25



 41

253
              272-550
  For source categories with existing NSPS's, one cost number is
  listed under all three control levels.
  109



145-255

  545
220 (120)   220 (120)      220 (120)

  30          ..  30    . ,       30

   6             6            13
              417-917

-------
                               TABLE 4-6
            ANNUAL COSTS FOR STATIONARY SOURCE NOX CONTROL

                    TO MEET THE NSPS LEVELS (106 $)
Source Categories
           Control Levels*

Moderate    Intermediate    Stringent
Utility Boilers

Industrial Boilers

     Coal-fired

     Oil and gas-fired

          Case I

          Case II

Stationary Gas Turbines

     High and (Low) Sales

Internal Combustion Engines

Nitric Acid Plants
    0

    0
18

51
 22



30-52

109
107 (45)
ISO
16
102 (45)
150
16
107 (45)
ISO
41
                                          211-278 .
              235-330
            288-429
  For source categories with existing NSPS's, one cost number is
  listed under all three control levels.

-------
                              REFERENCES
Acurex Corporation. March 1978.  Standards Support Environmental Impact
Statement for Reciprocating Internal Combustion Engines  (Draft),
Mountain View, CA.

Acurex Corporation.  August 29, 1978.  Memorandum from Larry Broz et
al, to the Industrial Boiler Mailing List on "ITAR Average SIP
requirements and Recommendations for Moderate, Intermediate and
Stringent Control Levels,"  Raleigh, NC.

Duncan, L. (MITRE Corporation) December 1978.  Personal communication.

Evans, R. et al.  March 1976.  Summary of Combustion Modification NO
Controls - Emission Levels, Costs, and Fuel Impacts, Acurex Corporation.

U.S. EPA.  1976.  Standards Support Environmental Impact Statement for
Stationary Gas Turbines, Chapters 3 and 7, Research Triangle Park,
NC.

U.S. EPA.  May 1978a.  Compilation of Air Pollutant Emission Factors,
Third Edition, (AP-42), Research Triangle Park, NC 27711.

U.S. EPA, OAQPS.  July 1978b.  Electric Utility Steam Generating Units -
Background Information for Proposed NO  Emission Standards, (EPA-450/2-
OOSa), Research Triangle Park, NC27711.

U.S. EPA, OAQPS.  January 1978c.  Control Techniques for Nitrogen Oxides
Emissions from Stationary Sources, Second Edition, (EPA-450/1-78-001),
Research Triangle Park, NC.

-------
4.2  FEDERAL MOTOR VEHICLE CONTROL PROGRAM COSTS
The costs of the FMVCP are estimated for the following vehicle categories:

  •  Light-duty vehicles
  •  Light-duty trucks
  •  Heavy-duty vehicles

The costs of NO  control for motorcycles and aircraft are negligible
               ^t
and are hot included here.  All costs are expressed in 1979 dollars.
Cost estimates for the most stringent currently mandated future emission
standards, for the: three vehicle categories were taken directly from the
regulatory analysis for ozone  (U.S. EPA. 1979).  For current in-use
technologies, other available  evidence on the cost of controls was used.

4.2.1  Light-Duty Vehicles
4.2.1.1  Initial Cost of Emission Control Systems
There were no standards for control of oxides of nitrogen from auto-
mobiles until 1973.  Starting  with the 1973 model year, the NO  stan-
                                                              Jv
dard was 3.0 grams/mile.  The  automotive NO  standard decreased to 2.0
                                           A
grams/mile in 1977, and is scheduled to drop to 1.0.grams/mile in 1981.
The technology used to meet the 2.0 and 3.0 grams/mile standards for
NO  was exhaust gas recirculation (EGR).  Because-this control technology
  A
is specific to NO , the entire cost of EGR can be attributed directly
to NO .  Likewise, the cost of all other controls prior to 1981 can
be attributed to CO and HC.  According to data prepared by EPA for
the ozone cost analysis,  (U.S. EPA. 1979) the cost of the EGR components
designed to meet a 2.0 gram/mile standard is $11 (1978 dollars), or
about $.12 in 1979 dollars.  Therefore, in this analysis, we have
assumed that $12 is representative of the cost of EGR needed to meet a

-------
2.0 gram/mile standard.  Because the EGR value needed to meet a 3.0
gram/mile standard is not as complex, its cost is estimated at $10.

More sophisticated technology is needed in order to meet the 1.0 gram/mile
N0x emission standard mandated to start in 1981.  This technology is a
three-way plus oxidation catalyst system.  Because the three-way catalyst
needed to meet the 1.0 gram/mile NO  standard is also needed to meet the
                                   A
0.41 gram/mile HC and 3.4 gram/mile CO standards, the initial cost of
installing this three-way catalyst must be apportioned between the three
pollutants.  Therefore, the $285 cost of a three-way catalyst (in 1978
dollars) identified in the ozone cost analysis (U.S. EPA. 1979) is
apportioned to NO  by dividing by three and inflating to 1979 dollars,
producing a cost is $105 per car.*

The capital costs of automotive NO  control in 1984 and 1990 are estimated
                                  A
by multiplying the projected or actual new car sales for each model year
back to 1973 by the increased cost per car due to NO  controls.  These
                                                    A
products are then summed to determine the total capital cost of NO
control.  The cummulative capital cost is estimated to be at $5.7 billion
by 1984 and $13.5 billion in 1990.  The calculation procedure used for
estimating the 1990 capital costs is shown in Table 4-7.  The 1984
capital costs were estimated by summing the yearly capital costs in
Table 4-7 from 1973-1984.  The annualized capital costs for 1984 and
1990 were calculated by multiplying the capital cost attributed to each
model year by the scrappage rate  (determined by vehicle age) and a
capital recovery factor of 18 percent (assuming, that the discount rate
for households is 12.5 percent, and that cars have an average life of
ten years.)  The 1984 annualized capital cost for automotive NO  control
                                                               A
is estimated to be $1.0 billion, while the 1990 annual cost is estimated
to be $2.1 billion.  The calculation of the 1990 costs is shown in
Table 4-8.
*An equal opportioning of the costs is recommended by the EPA's Office of
 Mobile Source Pollution Control.

-------
                                    TABLE 4-7






        THE CAPITAL COST INCREASES FOR AUTOS DUE TO N0x CONTROLS  IN  1990




                              CCOSTS IN 1979 DOLLARS)
Model
Year
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
1977
1976
1975
1974
1973
New Car ..Sales*
x 106
12.5
12.6
12.5
12.4
12.3
1.2. -3
12.0
11.5
11.0
11.1
11.5
10.9
11.3
11.2
9.85
8.2
8.79
11.48
Initial Cost Increase
Per Car
$105
105
105
105
105
105
105
105
105
105
12
12
12
12
10
10
10
10
Total Capital Cost
x 10*
$1.31
1.32
1.31
1.30
1.29
1.29
1.26
1.21
1.16
1.17
0.14
0.13
0.14
0.13
0.10
0.08
0.09
0.11
                                                                      $13.54
*  McNutt, et al. 1979

-------
                               TABLE 4-8



     ANNUALIZED CAPITAL CHARGES FOR NO  CONTROL IN 1990 FOR AUTOS
                                      A
Model
Year
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
1977
1976
1975
1974
1973
Total Capital Cost
x 10y
(From Table 4-7)
$1.31
1.32 "
1.31
1.30
1.29
1.29
1.26
1.21
1.16
1.17
0.14
0.13
0.14
0.13
0.10
0.08
0.09
0.11
Scrappage*
Rate
1.00
1.00
1.00
0.99
.99
.97
.91
.82
.71
.59
.47
.37
.27
.20
.16
.10
.06
.00
Capital
Recovery
Factor
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18 ...
.18
.18
.18
.18
.18
Annual i zed
Capital Charge
x 10
$.24
.24 :
.24
.23
.23
.23
.21
.18
.15
.12
.01
.01
• P1
0
0
0
0
0
                                                             $ 2.1
*  Source:   McNutt, et al. 1979

-------
4.2.1.2  Maintenance Costs
The assumed yearly operating and maintenance savings or costs are taken
directly from U.S. EPA, 1979.  For the cars equipped with three-way
catalysts, it is assumed that a $5 per year savings in maintenance cost
will be accumulated by each vehicle on the road.  For pre-1981 autos, it
is assumed that there is no change in operating and maintenance costs
attributable to NO  control.  In other words, it is assumed that the EGR
                  A
value does not need replacement.  The annual operating and maintenance
savings due to NO  controls for 1984 and 1990 are $0.07 and $0.54
billion, respectively.  The O&M cost calculation for 1990 is shown in
Table 4-7.  Costs for 1984 were calculated in the same manner.
4.2.1.3  Differences in Fuel Economy
The EPA technical staff estimated the composite fuel economy of catalyst-
equipped vehicles to be at least 7 percent better than that of uncon-
trolled vehicles  (U.S. EPA. 1979).  This value represents only a minimal
fuel savings.  For 1973 and 1974 model year vehicles, there is a slight
fuel economy penalty associated with an EGR system, estimated to be 5
percent (Murrell. 1979).

The dollars saved via fuel economy differences between controlled and
uncontrolled cars are calculated for 1990 as shown in Table 4-10.  The
average annual vehicle miles traveled (VMT) per car by vehicle age and
the composite fuel economy for each model year are used to estimate the
average number of gallons of fuel consumed per car by model year.  Then,
assuming that gasoline costs $1.00 per gallon  0-979 dollars), an annual
fuel cost per car is calculated.  This is then multiplied by the expected
new car sales, scrappage rate, and fuel economy benefit for each model
year to estimate  the dollars saved due to fuel economy differences.  The
$3.89 billion cost saving shown in Table 4-10  is representative of the
entire FMVCP, and must be divided by three to  determine the portion

-------
                               TABLE 4-9
             ANNUAL OPERATING AND MAINTENANCE COST SAVINGS
                       DUE  TO THE FMVCP IN 1990
Model
Year
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
1977
1976
1975
1974
1973
Yearly 0§M
Savings
$5
5
5
5
5
5
5
5
5
5
0
0
0
0
0
0
0
0
New Car-Sales*
x 106
12.5
12.6
12.5
12.4
12.3
12.3
12.0
11.5
11.0
11.1
11.5
10.9
11.3
11.2
9.85
8.2
8.79
11.48
Scrappage
Rate
1.00
1.00
1.00
.99
.99
.97
.91
.82
.71
.59
.47
.37
.27
.20
.16
.10
.06
.00
Total Annual,
Savings x 10
$62.5
63.
62.5
61.9
60.9
59.7
54.6
47.2
39.1
32.7
0
0
0 '
0
0
0
0
0
                                                         $ 543.6
* McNutt, et al. 1979

-------
                                        TABLE 4-10




                 ANNUAL FUEL ECONOMY DIFFERENCES DUB TO THE FMVCP IN 1990
Model
Year
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
1977
1976
1975
1974
1973

VMT per*
year per Car
15,900
15,000
14,000
13,100
12,200
11,300
10,300
9,400
8,500
7,600
6,700
6,600
6,200
5,900
5,500
5,100
5,000
4,700

Fuel**
Consump-
tion (mpg)
27.5
27.5
27.5
27.5
27.5
27.5
26.0
24.5
23.0
21.5
20.0
19.0
18.0
17.0
16.0
15.0
13.0
13.0

Gallons of
Fuel Used
578
545
509
476
444
411
396
384
370
353
335
347
344
347
344
340
385
362

New Car**
Sales
x 106
12.5
12.6
12.5
12.4
12.3
12.3
12.0
11.5
11.0
11.1
11.5
10.9
11.3
11.2
9.85
8.2
8.79
11.48

Scrappage**
Rate
l.QO
1.00
1.00
0.99
.99
.97
.91
.82
.71
.59
.47
.37
.27
.20
.16
.10
.06
.00

Fuel
Economy
Benefit
.07
.07
.07
.07
.07
.07
.07
.07
.07
.07
.07
.07
.07
.07
.07
.07
- .05
- .05

Dollars
Saved
X 10
0.51
0.48
0.45
0.41
0.38
0.34
0.30
0.25
0.20
0.16
0.13
0.10
0.07
0.05
0.04
0.02
-0.01
0.00
$ 389
*   Source:  U.S.  EPA,  1978
**  Source:  U.S.  EPA,  1979
    The price of gasoline is assumed to be |1.00/gallon (1979 dollars)

-------
attributable to NO .   The annual cost savings of the FMVCP in 1984 are
                  A,
estimated to be $3.85 billion.   Therefore,  $1.29 billion of these 1984
cost savings can be attributed to NO .
                                    A

4.2.1.3  The Use of Unleaded Fuel
Assuming that all cars manufactured after 1974 must use unleaded gasoline
and that the price differential between leaded and unleaded gasoline is
4 cents per gallon, the annual costs of the FMVCP in 1984 and 1990 are
estimated to be $2.3 and $2.2 billion,  respectively, with the NO  portion
set at one third of these values ($.77  and $.73 respectively).  These
costs are calculated using the same data shown in Table 4-10.

4.2.2  Light-Duty Trucks

4.2.2.1  Initial Cost of Emission Control Systems
Initial costs of emission control systems for light-duty trucks were
assumed to be the same as for autos.  Sales of light-duty trucks are
lower and the standard schedule is somewhat different when compared to
autos, so the capital cost for NO  control of light-duty trues is lower
than that of autos.  The calculation of the 1990 capital costs for
light-duty trucks is shown in Table 4-11.  The 1984 and 1990 capital
costs are estimated at $0.45 and $3.39  billion, respectively.

4.2.2.2  Annual Costs of Light Truck Controls
To be conservative, no maintenance savings or fuel economy savings are
credited to light-duty trucks.  Therefore, the only annual costs cal-
culated for light trucks are the annualized capital charges and the
added cost from the use of unleaded gas.  Both of these costs are estimated
using the same methodologies used for autos.  Fuel economies sales

-------
                   TABLE 4-11



THE CAPITAL COST INCREASES FOR LIGHT-DUTY TRUCKS


           DUE TO NO  CONTROLS IN 1990
                    A
Model
Year
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
1977
1976
1975
1974
1973
New Car-Sales
x 106
4.8
4.76
4.69
4.61
4.54
4.5
4 '.31
4.03
3.75
3.83
3.88
3.53
3.78
3.5
2.9
2.09
2. 25
2.56
Initial Cost Increase
Per Car
$105
105
105
105
105
105
12
12
12
12
12
12
10
10
10
10
10
10
Total Capital Cost
x 10y
$0.50
.50
.49
.48
.48
.47
.05
.05
.05
.05
.05
.04
.04'
.04
.03
.02
.02
.03
                                                  $3.39

-------
projections, and scrappage rates for light trucks were taken from McNutt,
et al. 1979.  The 1990 annualized capital charges and the costs from
using unleaded gas are detailed in Tables 4-12 and 4-13, respectively.
The total annual costs of light truck controls in 1984 are estimated to
be $1.03 billion, while the 1990 costs are estimated at $1.62 billion.

4.2.3  Heavy-Duty Trucks
The estimated cost of NO  emission controls for heavy-duty gasoline and
diesel trucks is taken directly from U.S. EPA. 1979, inflated by
10 percent to convert them to 1979 dollars.  Therefore, the total cost
for NO  control for heavy-duty trucks, which includes hardware, maintenance,
      A
fuel consumption, and the added cost of unleaded gasoline, is estimated
to be $0.6 billion in both 1984 and 1990.

                         Total Capital Cost            Total Annual
                              Of FMVCP.                Cost of FMVCP
Year                        (1979 x 10 )               ($1979 x 10 )
1984
1990
6.8
17.5
1.4
2.9 ....
4.2.4  Summary
Tables 4-13 and 4-14 summarize the total annual costs of the Federal
Motor Vehicle Control Program which can be attributed to NO  for 1984
and 1990.  Again, where the costs or benefits of NO  vs. CO or HC
control could not be differentiated, the total cost of the FMVCP was
divided by three to estimate the portion attributable to NO .

-------
                    TABLE 4-12
ANNUALIZED CAPITAL CHARGES FOR NO  CONTROL IN 1990
                 FOR LIGHT TRUCKS
Model
Year
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
1977
1976
1975
1974
1973
Capital, Cost
x 10°
504
500
493
,484
477
473
52
48
45
46
47
42
38
35
29
21
23
26
Scrappage
Rate
1.00
1.00
0.99
.97
.95
.93
.90
.86
.83
.77
.71
.65
.57
.50
.44
.38
.32
.26
Capital Recovery
Factor
.18 :
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
.18
Annualized Capita:
Charge x 10
$ 91
90
88
85
82
79
8
7
7
6
6
5
4 '
3
2
1
1
1
                                                  $566

-------
                    TABLE 4-13
INCREASE IN ANNUAL COST OF LIGHT TRUCKS DUE TO THE
            ADDED COST OF UNLEADED GAS
i )del
Year
i
f 990
-.J89
-1-988
387
1986
'
985
1984
983
1982
"981
bso
'-979
-.978
,.1 977
u?76
1975
974
1973
VMT per
Year
15,900
15,000
14,000
13,100
12,200
- 11,300
10,300
9,400
8,500
7.600
6,700
6,600
6,200
5,900
5,500
5,100
5,000
4,700
Fuel
Economy
(mpg)
22.8
22.8
22.8
22.8
22.8
22.1
22.2
21.6
20.0
18.7
17.5
17.3
17.0
16.5
15.5
12.9
10.9
11.2
Gallon of Fuel
Used Per Truck
697
658
614
575
535
511
464
435
425
406
383
382
365
358
355
395
459
420
Sales
x 106
4.8
4.76
4.69
4.61
4.54
4.5
4.31
4.03
3.75
3.83
3.88
3.53
3.78
3.5
2.9
2.09
2.25
2.56
Scrappage
Rate
1.00
1.00
0.99
.97
.95
.93
.90
.86
.83
.77
.71
.65..
.57
.50
.44
.38
.32
.26
4tf/gal Total Annu£
Penalty Cost Incres
x 105
$.04 $134
125
114
103
92
86
72
60
53
48
42
35
32
25
18
13
0
0
                                                                $1052

-------
                              TABLE 4-14
    TOTAL ANNUAL COST OF THE FEDERAL MOTOR VEHICLE CONTROL PROGRAM
                     IN 1984 (1979 DOLLARS x 109)
                                                       Total Cost of
Passenger Cars                                         NO  Control
                                                         x

Hardware                                                   1.0
Fuel Consumption                                          (1.3)*
Added Cost of Unleaded Gasoline                            0.8
Operating and Maintenance                                 fo.p*
                                                           0.4
Light-Duty Trucks

Hardware                                                   0.1
Fuel Consumption                                             0
Added Cost of Unleaded Gasoline                            0.3
Operating and Maintenance                                    0
                                                           0.4
Heavy-Duty Trucks
Hardware, Maintenance, Fuel Consumption,                   0.6
and the Added Cost of Unleaded Gasoline
TOTAL                                                      1.4
*  Negative Costs

-------
                         TABLE 4-15
TOTAL ANNUAL COST OF THE FEDERAL MOTOR VEHICLE CONTROL PROGRAM IN 1990
                         C1979; DOLLARS x 109)
                                                       Total Cost
Passenger Cars -                                        of NO  Control
Hardware                                                  $ 2.1
Fuel Consumption                                           (1-3)*
Added Cost of Unleaded Gasoline                             0.7
Operating and Maintenance                                  (0.2)*
                                                            1.3

Light-Duty Trucks

Hardware                                                    0.6
Fuel Consumption                                              0
Added Cost of Unleaded Gasoline                             0.4
Operating and Maintenance                                     0
                                                            1.0

Heavy-Duty Trucks
Hardware, Maintenance, Fuel Consumption,                    0.6
and the Added Cost of Unleaded Gasoline
TOTAL                                                       2.9
*  Negative Costs

-------
4.3  IMPLICATIONS FOR PSD AND NEW SOURCE SITING
Although the major subject of this analysis is the impact of setting a
short-term NAAQS for N02, whatever standard (if any) is eventually
promulgated by EPA may have implications for NO, PSD increments and thus
for new source siting and emission control.  According to the Clean Air
Act Amendments of 1977, EPA must promulgate regulations to prevent
significant deterioration of air quality no more than two years after a
new NAAQS has been established.*

Though Congress explicitly left open the use of PSD measures other than
setting maximum increments of ambient concentrations, the likely effect
of whatever measures are adopted will be to reduce the maximum allowable
ambient contribution of new sources below the short-term NAAQS.

In order to identify in a preliminary way at what PSD increment level
new sources would need to install emission control beyond that necessary
to meet NSPS, the results of previous dispersion modeling studies of
typical new power plants were examined and interpreted.**  The specific
plant modeled was a 1,000 MW unit with a single coal-fired boiler;-the
dispersion model employed was CRSTER, and meteorologic conditions similar
to Knoxville and Phoenix were used.  Two emission scenarios were modeled:

  •  The use of low sulfur coal alone to meet the old NSPS for
     coal-fired utility boilers (1.2 ibs. S02 per 10  Btu) —
     Case 1, and
  •  The use of both a low sulfur coal and flue gas dgsulfuri-
     zation  (FGD) system to meet the 0.12 Ibs. per 10 . Btu emission
     limit—Case 2

The emission and design characteristics of the plant are shown in Table 4-16.
 *PL 95-95, S.166
**The modeling was performed by EEA in late 1975 for both FEA and EPA.

-------
While the actual modeling was done for S02, it is possible to use the
results for estimating N02 levels by assuming that the ambient concen-
trations of NO  compared to SO- is the same ratio as the NO  to SO-
              «               »                            JC      ^
emission rates.  The level of N02 is then obtained by applying an NO to
NO- conversion rate.  (For purposes of interpreting the modeling results,
ambient NO- is assumed to be 90 percent of total NO .  Since the shortest
sampling period assessed for SO- was three hours, one-hour levels had to
be obtained by increasing the three-hour concentrations by 25 percent as
suggested by Turner (Turner, 1970).

The results for terrain at or below the stack heights appear in Table 4-
17, where Table 4-18 shows the results for plumes impacting on surrounding
terrain.  In both cases, NO  emissions corresponding to the proposed
                                                     6
NSPS for coal-fired utility boilers (0.60 Ibs. per 10  Btu) are assumed.
Two conclusions can be drawn:

  •  The use of a FGD significantly increases the concen-
     tration of NO  due to lower flue gas temperatures
     and thus lower plume rise, and
  •  The presence of complex terrain greatly increases
     ambient NO  levels
               A

It would appear that complex terrain locations for future power plants
may be greatly inhibited by PSD increments for NO-.  Even if NO- is
only 50 to 75 percent of total ambient NO  in Table 4-17, significantly
                                         Jv
high levels would be recorded on valley or mountain walls even 25 miles
away.  Of the 74 new or proposed plants assessed in the previous study,
12 were in locations with surrounding terrain at least 500 feet higher
than the plant.  If the proposed good engineering practice stack height
limits are applied to new sources, effectively limiting power plant
stacks to 250 feet, then the terrain threshold would be 250 feet above
the plant and many more sites would be affected.

-------
A preliminary bottom-line conclusion can be drawn.  Table 4-17 indicates
that an expressed or implied PSO increment for NO. of less than 0.18 ppm
(assuming 90 percent of NO  is NO- at the point of mart mm impact) may
                          A      w
impose additional control requirements on individual new 1000 MW or
larger power plants with 250 foot stacks and FGD systems, and located
in flat terrain.  For plants locating in complex terrain, PSD increments
of 0.22 ppm or less (again assuming 90 percent of NO  is NO.) would
require additional NO  control unless the impacted terrain were more
                     A
than 15 km. away; locations less than 10 km. would likely be prohibited
due to considerations of both the SO- and likely NO- PSD increments.
Locations near complex terrain features could only be allowed if the GEP
stack limit regulations were relaxed.  These upper bound PSD increments
for NO- imply a considerably higher one-hour NAAQS.  Conversely, if the
NAAQS is set at .18-.22 ppm, the PSD increment will probably be con-
siderably lower and the number of new sources requiring additional NO
                                                                     A
controls and/or site reconsiderations substantially higher.

-------
                              TABLE 4-16

              CHARACTERISTICS OF THE MODELED POWER PLANT
Parameters
Capacity (MW)
Exit Gas Temperature (°F)
Heat Rate (Btu/kWh)
Heat Content of Coal :
(Btu/lb)
Fuel Rate (tons/hr)
Lb - air/lb - fuel
S02 Emissions (Ib/hr)
NO Emissions (Ib/hr)
A
Stack Height (ft)
Stack Diameter (ft)
Exit Vel. (ft/sec)
Flue Gas Rate (ft3/sec)
Excess Air,
Case la/
1,000
300
9,250
11,000
421
10
11,080
5,557
500
24
95
44,600
20%
Case 2a/
1,039
170
9,250
11,000
437
10
1,148
5,768
500
24
95
42,450
20%
a/
    The Case 1 plant meets New Source Performance Standards by burning
    low-sulfur content (0.7%) coal.  The Case 2 plant also employs 90%
    removal efficiency flue gas desulfurization, assumed to be fully
    reliable.  An extra 39MW capacity is allowed for scrubber require-
    ments .

-------
                                        TABLE 4-17




                 ESTIMATE!) CONCENTRATIONS FROM A NEW 1,000 MW POWER PLANT



                          IN TERRAIN AT OR BELOW THE STACK HEIGHT






                                                  Ambient Concentrations (ppm)
Height of Stack
Above Terrain (ft)
0
250
500
750

3 hr SO.,
0 18
0,16
0,15
0 14
Case la/
3 hr NO b/
0.09
0.09
0.08
0.07

1 hr NO C/
0.11
0.11
0.10
0.09

3ll» Of) "
nr on /i
0.06
0.03
0.03
0.02
Case 2a/
' 3 hr NO b/
0.32
0.16
0.13
0.11

1 hr
0.40
0.20
0.17
0.13

N0,c/




a/                      •                           6
    Case 1 assumes meeting current NSPS (1.2 lbs/10  Btu) with low sulfur coal.   Case 2 assumes using low

    sulfur coal plus a 90 percent efficient PGD system.




 '  NO  emission level assumed to be 0.6 lbs/10  Btu (proposed NSPS).



c/
    One-hour levels obtained by increasing three hour levels by 25 percent.

-------
                                        TABLE 4-18



                 ESTIMATED CONCENTRATIONS FROM A NEW 1,000 MW POWER PLANT



                              WHERE TERRAIN IMPACTIONS OCCUR







Distance to Terrain  ,                     	Ambient Concentration (ppm)
Impaction Case (km)
2.5
5.0
10.0
15.0
20.0
25.0
3 hr S02b/
6.04
2.17
0.74
0.39
0.29
0.21
3 hr NO C/
3.02
1.09
0.37
. 0.20
0.14
0.10
1 hr N0xd/
3.78
1.36
0.47
0.25
0.18
0.13
a/
    Terrain at least 500 feet above physical stack.





 '  Based on SO- emission rate of 1.2 lbs/10  Btu and no FGD system.





c'  Based on NO  emission rate of 0.6 lbs/10  Btu.
               A.




    One-hour levels obtained by increasing three-hour levels by 25 percent.

-------
                              REFERENCES
McNutt, B., et al.  February-March 1979.  "Factors  Influencing Auto-
motive Fuel Demand" SAE paper #790226.

Murrell, J.  February-March 1979.  "Light Duty Automotive Fuel Economy
	Trends Through 1979"  SAE paper #740223.

U.S. EPA, OQWM.  March 1978.  Mobile Source Emission Factors, Washington,
D.C.,  (EPA-400/9-78-005)

U.S. EPA, OAQPS-SASD.  February  1979.  Cost and Economic Impact Assessment
for Alternative Levels of  the National Ambient Air  Quality  Standards  for
Ozone, Research Triangle Park, NC, CEPA-450/5-79-002;).

-------
                         5.  ECONOMIC IMPACTS
Whether the costs to meet or approach the ambient NO- standard are
significant depends on the level of the standard and the distribution
of costs in the public and private sectors.  The following sections
trace out some of the estimated control costs.

5.1  ELECTRIC UTILITIES
5.1.1  Industry Profile
The electric utility industry is comprised of two main sectors.  Private
or investor-owned firms account for approximately 78 percent of total
U.S. generating capacity and production.  Publicly-owned firms (including
those owned by municipalities and Federal and State agencies) account
for 20 percent of U.S. capacity and production.  Cooperatives, which are
frequently included among the publicly-owned firms, account for the
remaining two percent.  Publicly-owned utilities have lower financing costs
than private firms and tend to have a large percentage of hydroelectric
generation.  As a result, publicly-owned electric utilities comprise
only approximately 15 percent of the industry's total operating revenues,
while private firms account for about 85 percent, (Temple, Barker, and
Sloane, Inc., May 1976).

Most of the publicly-owned and cooperative electrical systems are small-
scale operations involved solely in distribution.  There are a few
exceptions to this generality, such as Bonneville Power Authority,
Tennessee Valley Authority, and County Public Utility Districts (Washington
State), which generate vast quantities of electricity and are large
suppliers to privately-owned systems.  There are approximately 2,200
publicly-owned and 1,000 cooperative supply systems with a combined
generating capacity of 121 million kilowatts in 1977.  Investor-owned

-------
electric utilities, on the other hand, are much fewer in number (approxi-
mately 500 systems), yet they account for about 78 percent of total
capacity (436 million kilowatts in 1977).  Total 1977 capacity was
therefore 557 million kilowatts, an increase of 4.9 percent over the
previous year,  (Temple, Barker, and Sloane, Inc., May 1976 and Bureau of
the Census 1978).  These patterns are displayed in Table 5-1.

Electricity is produced by several methods:  conventional steam, (i.e.,
fossil fuel-fired boilers), nuclear, hydroelectric, gas turbine, and
internal combustion engine units.  As shown in Tables 5-2 and 5-3,
nearly 82 percent of all electricity is produced by conventional steam
and hydroelectric plants combined.  The proportion of generating capacity
furnished by the various types of production plants differs for public
and private firms.  The primary distinction is in the number of hydroelectric
and steam plants; five percent of private capacity is furnished by
hydroelectric plants and 76 percent by conventional steam plants,  whereas
37 percent of public firm capacity is hydroelectric and 46 percent is
conventional steam, (Bureau of the Census 1978).

Public and private firms are essentially the same with respect to
customer consumption patterns.  In 1977, the residential sector consumed
33 percent, the commercial sector 24 percent, and the industrial sector
39 percent of total U.S. ultimate sales.  The remaining 4 percent of
sales were consumed by other types of ultimate customers.  U.S. electric
utility net generation in 1977 was 2.124 billion killowatt-hours,  an
increase of 4.3 percent over 1976, (Bureau of the Census 1978).  (See
Table 5-4.)

Privately-owned utilities had electric operating revenues of approximately
$50.6 billion in 1976.  Major municipals* had operating revenues of $4.6
billion and Federal projects had revenues of more than $2.6 billion in
     Major municipals are those municipally-owned electric utilities with
     annual operating revenues of $5,000,000 or more.  The indicated total
     operating revenues  for major municipals are only 72 percent of total
     municipal operating revenues for the U.S., the Virgin Islands, the
     Canal Zone, Guam, Puerto Rico, and Samoa.

-------
1976.  In light of these large revenues, however, it is important to
note that the industry is characterized by very high plant investment
per dollar of revenue.  For example, total capital expenditures for
construction and plant (including land) amounted to more than $17.3
billion in 1976 for privately-owned utilities.   Total investor-owned
electric utility assets (for electric plant and nuclear fuel only) were
nearly $178 billion in 1976.  The comparable figures for major municipals
and Federal projects were $19 billion and $17 billion, respectively,
(U.S. Department of Energy, December 1978).

5.1.2  Recent Financial History
The period from 1960 to 1965 was one of general prosperity for the
electric utility industry.  The costs of generating electricity decreased
due to technical changes in the design and operating performance of
generating equipment, which permitted more efficient conversion of fuels
to electrical energy.  Costs for transmission and distribution were
also on a downward trend so that the industry's total costs per kilowatt-
hour decreased significantly from 1960 to 1965.  This decrease in overall
costs led to a decline in the price charged to customers.  The resulting
1.2 percent average yearly decrease in revenues per kilowatt-hour from
1960 to 1965 was more than offset, however, by an average yearly increase
of 2.2 percent in the number of customers and 4.6 percent in the usage
per customer.  The net result was an increase in total revenues of 5.8
percent per year from 1960 to 1965.  Overall, problems within the industry
were very predictable and manageable, and uncertainty was at a minimum
in the early 1960's, (Temple, Barker, and Sloane, Inc., May 1976).

Since 1965, however, a series of events changed the picture to one of
uncertainty and adversity.  A dramatic increase in interest rates
created a credit crunch and increased the cost of loans.  Inflation
caused increases in costs which could no longer be offset by productivity

-------
gains.  The effect of inflation was felt particularly strongly in the
capital goods industries; the cost of building new electric generating
plants rose rapidly.  Equipment shortages and environmental protection
hearings led to delays in new plant construction.  Such delays meant a
further increase in costs due to longer financing requirements as well
as higher interim operating costs from the slowed retirement of older,
less efficient equipment.

Fuel cost increases also took their toll with the 1967 Suez Canal
closure and the increases in posted prices and taxes by the oil ex-
porting -countries.  The cost of coal rose as the competitive pressure
exerted by oil was lifted and as investments for improving safety and
health were required.  In addition, the Federal Trade Commission began
to permit interstate natural gas price increases, further releasing
competitive pressure on coal prices.

As a result of these events, the financial condition of the electric
utility industry changed.  By 1970 the industry faced growth without
prosperity.  The overall rate of growth in demand had increased, but so
did the degree of uncertainty.  With the long lead time for nuclear
units, even small differences in demand growth rates meant large dif-
ferences in projected capacity requirements by the time the nuclear
units would come on line.  The long construction periods combined with
the rapidly escalating construction cost of nuclear plants tied up large
sums of capital in nonproductive capacity.  Moreover, fossil fuel units
had reached a size where economies of scale had decreased or had been
eliminated, further raising the average cost per kilowatt of capacity.
Operating cost increases due to fuel cost surges and general infla-
tionary pressures could no longer be offset with increases in effi-
ciency.  Rising interest rates on new debt issues led to additional
increases in the industry's volume of outstanding debt.

-------
To add to these problems, conservation efforts in response to the 1973
oil embargo caused demand to plateau in 1974 and 1975.  Capacity additions
for this period had already been determined by construction programs.
This situation has changed somewhat in recent years, but consumption of
electricity is still below previous projections.

The economic impacts of alternative short-term NO- standards must be set
against this background of financial concern on the part of both public
and private utility companies.

5.1.3  Impact of NO  Control Costs on Ability to Raise Capital
Despite financial problems faced by the industry, the impact of costs
to meet alternative one-hour standards does not appear to be severe.
The total capital cost accruing to utilities indicates that the estimated
control costs for meeting even a 0.15 ppm standard are modest.  The
approximately $750 million capital cost will be spread over approximately
67 plants, making the per plant cost about $11 million on the average.
To put this cost in perspective, the estimated capital expenditure for
NO  controls on existing utility sources is less than 5 percent of the
  Jt                                                     • -         ,
estimated capital expenditures between 1976 and 1990 to meet the proposed
NSPS for S02, and about 0.1 percent of the total estimated capital
expenditures by utilities for the same period.

Financing the expenditures for a short-term NO. standard of 0.15 ppm
should not pose significant difficulties for the public utilities.
Many publicly-owned firms depend either directly on governmental funding
or indirectly on governmental guarantees, and so in most instances they
are relatively assured of access to adequate funds.  Therefore, the
financing of public firms would perhaps best be dealt with as part of
the financing of the government sector.

-------
Investor-owned  firms,  on the  other hand,  are  faced with  competition  for
funds in  the  capital market.   In  addition,  they  tend  to  be  larger than
publicy-owned firms and so  have higher  emission  levels and  require
larger total  capital expenditures for pollution  control.  Key  to the
industry's  financial strength and,  thus,  its  ability  to  raise  sufficient
capital through external financing,  is  the  allowed rate  of  return on
common equity.   If returns  are set  by state regulatory commissions at a
level commensurate with the rates of return required  in  the capital
markets,  the  industry's status as a regulated monopoly offering safe and
predictable returns should  enable it to compete  for the  necessary funds.

While the capital  markets may be  large  enough to meet the total needs of
'the electric  utility industry, some individual firms  may have  difficulty
in gaining  access  to sufficient funds to  meet even baseline capital  needs.
Once again, however, it is  within the power of the regulatory  systems to
permit changes  that would enable  all companies to meet both baseline and
pollution control  financing needs.

S.I.4  Product  Price Increases
The effect on the  average monthly  electric bill was estimated on the
basis of a previous  study done  in  1976 by Temple, Barker, and Sloane
CTBS) (Temple, Barker,  and  Sloane,  Inc., May  1976).  The TBS study
estimated that an  average residential monthly bill of $55.88 (.$1979)
in 1985 would increase  by $3.69 as a result of utility expenditures for
federal ambient  air  quality regulations  (sulfur dioxide and total suspended
particulates) and  water effluent guidelines combined.  Comparing the
total cost to meet these regulations with those estimated to meet a
0.15 ppm NO- NAAQS (about $200  million), the  average monthly residential
bill in 1985 would increase by  another $0.08.  This estimate was obtained
by first annualizing the TBS  capital, operating, and maintenance costs
for S0_, TSP, and  water pollution  control equipment.  The NO  annualized
      2                                                     X

-------
control cost was expressed as a percentage of the TBS annualized control
cost (2.3 percent),  and then multiplied by the TBS estimated increase in
an average monthly residential bill ($3.69)*.  Bills of some customers
in some regions would obviously be greater than this figure since the
nationwide control costs are not spread over all utilities, or more
corectly, over the same number of plants requiring SO- controls in
the TBS study.  However, the estimated average cost is so low that
a substantial increase in any region is highly unlikely.

5.1.5  Conclusion
On the basis of the above discussion, we can conclude with some confidence
that the economic impacts of meeting even a stringent standard, 0.15 ppm
are unlikely to be large.
*A11 dollar figures are given in 1979 dollars.

-------
                                   TABLE 5-1


ELECTRIC ENERGY  PRODUCTION  AND INSTALLED GENERATING CAPACITY

BY CLASS OF OWNERSHIP AND TYPE OF PRIME MOVER:   1960  to  1977
     I Production lor ote*d
                          ; other data aa of December 31. 1980 excludes Alaska and Hawaii. Seeelso HUtrttml
                    SulttUa, CbtMfai Tlma to 1910, antes 3 32-52. S 78-82. and S B6-94I
rriM
nopacnoK
Toul 	 	 	 bfl. kWh ..
Production, k Wh per It W o< capacity.
Industrial plants '........ ......bll. kWh.
Electric utUIUta • 	 	 bit kWh.
Privately owned- 	 ....blLkWh
Pfnant nf total utilitin. ... . _
PuMlety owned 	 Ml. k Wh.
Municipal 	 bll. kWh.
Federal......... 	 ._ bll. kWh.
Cooperatives «nd otter 	 bll. k Wb.
Soareeol mercy (percent):
CM) ',,...... i.. .... 	 	 	 u
Nuclear .. .. .
OO „ ... .. ...„. .. .
(las .... .— ............. ._
INSTAtUD CAPAOTT
TouL... ._...... 	 ...roll. kW..
Indintrtal plant* '_.._ 	 _mn. kW..
Electric utilities • 	 mil. k W..
Privately owned 	 mil. kW..
Percent of total utilities...... 	 	
PuMWy owned ml) kw
Municipal 	 _ 	 ..'.rail. kW..
FwtoraL 	 mil. kw..
ConpvrattTia end atheT_. ..mil. kW..
TO* or num uovu
Bteetrte utilities: •
Namber of plants, total > — -„ 	
Hydro .................... 	 .



Internal combustion...................
Pmductiou...-.......---»....blL k WB..
Hydra 	 	 	 bll. kWh..
" Steam conventional.. ......bll. kWh._
Oat turbine... ........... WLkWh.
Steam nwrtoar .,..., bll. k»h.
Internal eombotUon 	 blL k Wh..
[mtalltd capacity ....... mil. kW
Ilydro_.'. 	 '. 	 	 mil. kW..
Steam eonTentional.. .m..mll. kWu
Oas turbine 	 ...._.mll. kW..
Steam nuclear... 	 ......mil. kW..
Isttrnsi ccn-.husllss_-— —rsii. k W..
Z \jtea than .S million. ' Chance (rom
> Plants 0(100 kilowatt! and o»«r, Indud
' Bach prime mover type la combination
I«M
to
8.0
i,SK
88
ra
STB
78.8
179
37
112
28
S3.8
8.1
21.0
19.3
188
7.3
18
168
128
78.1
40
11
22
3.<3S
1.331
jl.OU
a
1.044
7S3
148
} 802
4
188
32
} 133
(»i
tfH
1.158
a.e
4.M8
102
1.03S
8DB
78.7
248
SO
148
31
54.5
8.1
21.0
18.4
259
8.8
18
238
178
75.2
a
15
32
11
3,190
1,231
1.089
9
991
1.095
194
882
4
.5
218
44
188
1
•
itn
1.838
7.2
4.438
104
1.832
1.183
77.2
349
71
188
91
1 48.1
\ 1.4
11.9
24.3
18.2
380
7.2
19
342
283
78.9
79
21
39
19
3.933
1.188
/ 998
( 323
18
1,008
1,492
249
/1.20I
1
340
IS
/ 280
1 ,J
im
1.899
6.3
4.448
109
1.790
1.399
77.7
391
79
207
105
44.3
3.1
11.6
21.5
15.8
417
7.1
19
398
314
78.9
84
23
40
21
3,614
1.173
998
413
28
1.001
1,750
273
1,387
29
84
398
M
294
• • 28
19
i
itn
1,985
5.9
4.282
105
1.880
1.432
78.1
408
81
212
116
45.3
4.}
16.8
18.3
14.8
481
10.6
19
442
348
78.7
94
25
45
24
3.681
1.181
992
480
30
1,038
1.881
272
1.448
' 30
83
7
442
62
321
33
21
1
1174
1,968
.1
3.980
101
1.867
1.142
77.2
429
79
220
126
44.6
6.1
18.0
17.1
16.3
497
7.8
IB
478
377
78.9
101
27
46
28
3.891
1.199
994
494
38
1.006
1.887
301
1.414
32
114
6
479
64
338
40
32
K
1175
2.003
1.8
3.801
89
1,918
1.487
77.9
411
82
221
128
44.6
9.0
16.1
19.6
11.8
827
6.0
19
908
399
78.5
109
29
SO
30
3.674
1.156
984
811
49
978
1.918
300
1.417
22
173
8
908
66
393
44
40
•
1978
2.124
6.0
3.883
87
2.037
1.982
77.7
458
78
236
141
46.1
9.1
15.7
14.6
14.1
150
4.6
IB
531
411
78.3
116
31
57
33.
3.682
1.149
978
820
48
969
'•S
1.534
24
IB1
8
831
68
368
47
43
5
itn
2.211
4.1
3.839
' 87
3.124
1.684
79.0
440
83
314
143
48.6
11.8
16.9
14.4
10.6
578
3.1
19
987
438
78.3
121
32
53
38
3.622
1.145
951
524
49
983
2,124
220
1.619
29
281
1
817
68
386
48
90
5
prior year shown; tor 1980, chance from 1963.
nf stationary powerplanU o( railroads. ' For public us*.
t waste end ceothermai sources.
plants counted separately.
      Source: 1960-1970. U.S. Federal Power Commbstoa, StaMe Power StefMla. end presVrelftases: thereafter.
     U.3. Energy Information Administration, Sir* Vcar Smuicrt of Power Prod act !•« 0114 Qntrattng CapmtUt Data.

-------
                        TABLE 5-2
PRIVATELY OWNED ELECTRIC UTILITY GENERATING PLANTS-NUMBER
 AND CAPACITY, BY TYPE AND SIZE OF PLANT:  1970 to 1977
                (CAPACITY AS OF DEC. 31)
TTrt AND SIZE
Of PtAWT
Toul 	
Steam conventional..
Under 100,000 kW..
100.0UO-500,OOOkVr.
Over 500,000 kW 	
Bleara. nuclear 	
Over 500,000 kW —
Hydro 	
Over 25.000 kW 	
Oas turbine 	 .„
Over 73,000 kW 	
tntenml com bastion.
Under 8.000 kW 	
Orer 5.000 kW 	
HOMBd
1170
1.937
672
220
299
153
U
r
744
185
285
128
331
238
83
an
1.101
684
IBS
274
IBS
27
20
731
189
380
226
119
218
103
1174
1.1 JT
671
182
276
213
31
24
734
170
387
251
314
208
106
1976
1.123
658
184
267
227
35
29
730
170
397
281
303
199
104
1976
1.10
859
159
257
243
37
31
732
17f
394
267
301
189
112
im
1.102
841
148
244
249
41
38
730
171
397
272
293
184
109
CAPACITY (mil. kilowatts)
1970
262.8
222.4
8.2
77.1
137.1
5.7
4.5
18.9
15.8
14.4
12.1.
1.3
.6
.9
1173
148.5
276.2
7.5
71.5
197.3
19.7
18.8
21.8
18.8
29.3
27.1
1.4
i.'o
I»74
377.3
290.1
7.2
72.4
210.5
28.2
26.9
22.8
19.6
34.8
32.5
1.6
.4
1.1
me
398.9
303.8
6.8
70.0
227.1
33.6
32:3
23.1
19.9
37.0
34.7
1.6
.4
l.l
1976
411.4
316.3
6.6
87.3
242.4
35.4
34.1
23.1
20.0
38.8
36.7
1.7
. .4
1.3
1977
416.1
330.1
6.0
63.8
280.3
41.7
40.8
23.1
20.0
39.6
37.4
1.6
.4
i.a

-------
                          TABLE 5-3
  PUBLICLY OWNED ELECTRIC UTILITY PLANTS-NUMBER AND
 INSTALLED GENERATING  CAPACITY,  BY TYPE AND SIZE OF
                   PLANT:  1970 to  1977
                 (CAPACITY AS  OF DEC. 31)
v
rm AMD aos
or rtAttr

TMal. .„....__.
Slcnm. coawnttotuJ....
Under 1 00.000 k\V 	
100,000- XXUWO k W 	
Ow 300,000 k\V___
gtfUH pucfoa* a_.
Over 5,000 kW 	
mm»«m
-

tm
1.496
328
254
U
14
t
1
434
164
«
a
877
208

1973
1.886
us
244
65
10
1
1
430
178
100
36
TIB
282

1974
1.154
133
234
«
24
T
3
CM
17»
HIT
41
tat
257

itn
t.Wl
326
234
67
25
10
4
426
IS4
114
46
6T8
259

1ST6
i.m
117
219
n
77
n
6
417
184
126
54
668
287

1OT
U5»
310
208
70
32
8
8
418
184
127
SB
660
266
CAPACITT (mil. kttowttta) .


IfTO
78.8
37.7
6.B
13.3
18.6
,

38.
38.
1.

3.0
2.0

1173
M.O
44.
7.
14.
23.
1.

40.2
3B.1
4.1
3.1
3.6
2.7

1174
106.1
47.6
7.0
14.1
26.5
3.4
2.6
40.9
38.7
4.8
3.8
3.6
2.7

1671
I6B.3
4B.6
6.B
14.8
27.9
6.3
4.6
43.8
41.7
7.1
6.0
3.6
2.8

1178
111.6
62.1
6.6
U.B
2B.6
7.8
6.6
44.6
43.6
7.7
6.6
3.6
L8

ttn
!».»
56.8
6.6
U.B
33.0
8.2
7.7
46.2
44.0
8.3
7.2
3.7
2.9
 Boaraotuhlel 102! md 1022: IBTfl, U.S. Federal Power Commlsiion. Electric Pwrtr Statlttta, uuituit, find press
ratetm; tbcnmtter, U.S. Eoort; latonnsttoo AdmtabuMton. Slx-Vmr Samm«r »/ Pomtr Pralmatfm and On-

-------
                               TABLE  5-4
ELECTRIC  LIGHT AND  POWER INDUSTRY-ENERGY GENERATED, SALES,
             REVENUE,  AND  CUSTOMERS:   1950  to  1977
         (PRIOR TO  1960,  EXCLUDES ALASKA AND  HAWAII)
CLASS
Energy generated i 	 	 bll. kWh..
Saiea la animate coatamere «. 	 Ml. kWh..
Residential or domestic 	 bll. kWh..
Percent of total.............. 	 ....
Comra.. small light ami power — bll. kWh..
Indus., large light and power 	 bll. kWh..
ttereaae fram ultimate ctuumare '.bll. del..
Residential or domestic ..... 	 .bit. dot..
Percent o( total ........... . . ...
Contra., small light and power — bll. dol..
Indus., large light and power 	 bll. dpi..
Ultimate coawmara. Dec. 11 ' 	 mllltaaa..
Ckunm., small light and power. ..millions..
Indus., large light and power 	 millions..
AT(. kWh «*ed aer natamer. 	 1.000..
Residential 	 . 	 	 1,000..
Commercial, small light and power..!, 000..
Artf. annval bill par t«ala«iit> 	 ..dal.*
Residential 	 dol..
Commercial, small light and power 	 dol..
HesldenttaJ 	 cent!..
Comm.. small light and power.. ....cents..
Indus., targe light and power 	 .cents..
ItW
129
281
70
25.0
62
142
I.I
2.0
31.7
1.
1.
45.
38.
8.
(.4
1.8
9.3
US
93
24S
1.81
2.88
2.83
1.02
IMS
647
481
12S
28.1
78
2S8
8.0
3 3
41 4
l.fl
2.4
S2.I
45.8
8.2
.4
«.3
2.8
12.7
155
73
316
l.(7
2.05
2.60
.04
ISM
753
(83
118
28.7
115
345
II. S
4.t
42.2
2.8
3.3
58.1
51.4
8:8
.5
11.7
3 0
17.0
118
95
418
i.n
2.47
2.48
.97
1*85
1 055
tS!
281
29.6
202
433
U.J
8 3
41 8
4.3
3.9
65.6
57.8
7.4
.3
u.r
4.9
27.4
234
111
584
1.5$
2.25
2.13
.90
1170
1 532
1.3S1
448
32.2
313
573
n.i
9 4
42 7
6.3
5.4
71.5
A4.0
7.9
.4
11.4
7 1
40.0
208
148
803
l.SS
2.10
2.01
.95
1*71
1 860
1.701
554
32.6
397
687
31.7
13 2
41 6
9.1
8.1
78.5
69.4
8.4
.4
21.0
8.1
47.6
408
192
1.094
1.86
2.38
2.30
1.17
1*74
1 887
1.701
655
32.6
393
689
3*.l
15.7
40.1
11.2
10.7
80.1
71.0
8.5
.4
21.4
7.9
46.6
493
224
1.327
2.30
2.83
2.85
1.55
1*75
1 818
1.713
586
33.8
418
682
46. 1
18 8
40
13.
12.
81.
72.
8.
21.4
8.2
49.0
578
262
1,582
2.76
3.21
3.23
1.92
1*76
2 037
1.850
613
33.2
441
725
63.1
21 1
39 6
15.2
15.0
83.1
74.2
8.8
.4
22.4
8 4
60.7
646
288
1.755
2.89
3.46
3.46
2.07
1177
2 124
1.S5I
652
33.4
469
757
62.6
24 7
39.4
18.0
17.6
85.6
76.9
8.9
.4
21.1
8.7
52.9
740
329
2,031
3.21
3.78
3.84
2.33
  1 Source: 1950-1976, U.S. Federal Power Commission; thereafter, U.S. Energy In format Ion Administration,
 Six- Year .Sumnurr of Power Production aiuf OtnaaUnt Capacity Data.
  ' Includes other types not shown separately.
  Source: Eicapt as noted, Edison Electric Institute, New York, N.Y., StatMieat Ytar Boo*;

-------
5.2  IRON AND STEEL INDUSTRY
5.2.1  Profile
5.2.1.1  Production
The U.S. iron and steel industry is composed of approximately 180 com-
panies which operate about 400 plants and employ about 450,000* workers,
(American Iron and Steel  Institute, AISI, 1978).  Many iron and steel
companies own coal and iron ore mining facilities in addition to their
manufacturing operations.  Most iron and steel products are manufactured
at integrated plants which convert iron ore and coking coal into steel
through a series of processes, including the production of coke, pig
iron, raw steel, and steel products.  Semi-integrated steel mills use
scrap steel and pig iron  as inputs, and non-integrated mills work
semifinished steel shapes into finished products.  These latter mills
are much smaller and more numerous than integrated mills.

In 1977, domestic iron and steel plants produced 125 million net tons**
of raw steel.  This represents 17 percent of world production of 743
million net tons.  U.S. production as a percent of world production
has declined over the past decade; in 1968, U.S. production comprised
23 percent of the world total for raw steel (Standard and Poors, 1978).

Steel is produced in 36 states, but production is heavily concentrated
in the Great Lakes region.  Pennsylvania is the leading state producer. .
*This figure includes only those employees involved in the production and
 sale of iron and steel products, and does not include those involved in
 mining and quarrying operations, transportation, warehousing, fabrication,
 and other non-steel production activities.
**A net ton is equal to 2000 pounds.

-------
contributing 21 percent to total domestic raw steel production.  Indiana
and Ohio each contribute 17 percent to domestic production, and Illinois
and Michigan produce nine and eight percent, respectively.  These five
states alone account for 72 percent of domestic production, (AISI, 1978).

This concentration of production capacity is not expected to change in
the near future.  The present system developed because of proximity to
major supplies of coal and iron ore and the availability of relatively in-
expensive transportation.  These considerations still bear enough importance
that most expansion projects under construction or in the planning stage
are additions to existing facilities.

Over the past decade, the U.S. steel industry has undergone a major change
in its production capacity.  Until 1967, a majority of production was
attributable to open-hearth furnaces.  Since then, the use of open-hearth
furnaces has declined to the point where these furnaces primarily are used
only during periods of peak demand.  By 1977, open-hearth production had
decreased to 16 percent, whereas basic-oxygen furnace production had
soared to 62 percent and electric furnace production had risen to 22
percent (Standard and.Poors, 1978).  This change is due, at least' in
part, to the higher requirement of fuel and labor for open-hearth as
opposed to basic-oxygen furnace production.  The electric furnace has
the advantage of improved quality control in addition to a reduction in
raw material requirements.

Capability utilization* has been improving over the past two years.  With
increasing demand, capability utilization was predicted to reach approxi-


*"Capability utilization" is a measure of capacity utilization taking
 into account, availability of raw materials, fuels, supplies, and
 current environmental and safety requirements.

-------
mately 83 percent in 1978 and to remain above 80 percent in 1979.  These
are the highest levels since 1974.  Capability utilization dropped from
a high in 1974 to 81 percent in 1976 and 78 percent in 1977 (AISI, 1978).

5.2.1.2  Demand
Steel industry shipments in 1977 increased 1.9 percent from 1976 shipments.
This increase was due primarily to a 38 percent increase in deliveries to
the oil and gas industry, raising their fraction af total demand to four
percent.  The automotive market did not expand nearly as much, reflecting
the impact of weight reductions in vehicles, but it did remain the
leading market in 1977 receiving 24 percent of total shipments.  Other
important markets include steel service centers (17 percent of 1977
industry shipments), the industrial and electrical machinery and equipment
market (nine percent), and the construction market (13 percent).  The
first two of these markets, machinery and steel service centers, each
absorbed about five percent more steel in 1977 than in 1976 while the
construction market remained essentially the same as in 1976.  In general,
shipments to consumer-oriented markets have been strong during the 1975-
1977 period, but demand in the capital goods and construction sectors
has been sluggish   (Standard and Poors, 1978).

5.2.2  Recent Financial History
While the position of the steel industry has been improving over the
past couple years, there are still some problems which must be resolved
for the industry's  long-term health.  The primary problem facing the
steel industry is an inadequate return on investment.  Even with improved
profits in 1978, earnings were still far below the level required for
funding much needed modernization and expansion programs, paying dividends,
and maintaining a firm financial position.  Over the past 20 years, the
industry's return on equity has exceeded the average for all manufacturers
only one time, the peak earnings year of 1974.

-------
In addition to these problems, costly environmental control programs
have been forcing the individual companies to allocate considerable
amounts of their already limited resources to pollution control equipment.
In 1977, iron and steel industry capital expenditures for air and water
pollution control equipment for domestic operations (including steel
company owned mines, quarries and fabricating plants) were $535 million.
Total capital expenditures for all purposes made by steel companies in
both steel and nonsteel operations (domestic and foreign) were $2,858
million* in 1977, (AISI, 1978).

Pollution control therefore, accounted for approximately 19 percent of
total capital expenditures of steel companies.  These expenditures produce
no return and often raise the operating costs of a plant.  Pollution con-
trol expenditures will remain substantial over the next few years.

The factors discussed above have had an adverse effect on the industry's
financial condition.  The steel industry has been forced to rely increas-
ingly on debt financing rather than internally-generated funds over the
past decade.  The debt/equity ratio increased from 0.36 at year-end 1968
to 0.45 in 1977.  The debt/equity ratio was at a minimum of 0.29 'in 1974
which was a peak year for the industry (Standard and Poors, 1978).

Sharp increases in labor costs have also posed a problem for the industry.
Labor costs have risen at an annual rate of about 10 percent over the
past several years, and will continue to do so due to the provisions in
the present steelworkers contract.  Steel industry labor costs have been
increasing more rapidly than average manufacturing wages.  Average
hourly earnings of steelworkers exceeded the average wage paid to all
manufacturing employees by approximately 64 percent in 1977 and 39
*Represents companies who produced 89 percent of the reported raw  steel
 production in 1977.

-------
percent in 1972  (Standard and Poors, 1978).  Rising labor costs have had
a significant effect on profitability, and since these costs cannot be
offset by increases in productivity, they must be recovered through
higher prices.  Employment costs for the industry amounted to $14.419
billion in 1977, or 36 percent of total costs of $39.764 billion  (AISI,
1978).

An additional problem facing the steel industry is the fact that foreign
steel companies have captured a significant share of the U.S. steel
market in recent years.  Imports as a percent of apparent steel supply
were 17.& percent in 1977, up from a low of 12.4 percent in 1973  (AISI,
1978).  This created a situation in which domestic producers had to
offer discounts to retain a competitive position.  By 1977, import
pressure was so severe that several large steelmaking operations were
forced to close.

With substantial job losses, plant closings, and the increasing deficit
in the U.S. balance of trade, the federal government came to the industry's
aid.  The program included tax relief, loan guarantees for weaker
companies, assistance .to communities affected by plant closings, and the
establishment of a trigger-pricing mechanism.  Trigger-pricing is a
system which is designed to prevent sales of foreign steel at prices
below "fair value".*  It appears that this system is beginning to slow
the influx of foreign steel.  Imports as a percent of apparent supply
were expected to be down to about 15.6 percent in 1978 and 14.4 percent
in 1979 (Standard and Poor, 1978).
*"Fair value" is defined as the price at which the same product is sold
 in the home market of the exporter provided that the price is above the
 cost of production in that country.

-------
The mix of imports by country has also been undergoing some change.
Japan, which accounted for 56 percent of U.S. imports in 1976, has
slipped to a 40 percent share because of restricted shipments, while
the European countries, South Korea and Mexico have increased their
shares.  Canada makes up 10 percent of the import market, and the European
countries combined, account for 40 percent CStandard and Poors, 1978).
Japanese and European steel companies are presently operating well below
capacity, and several developing countries are bringing on new capacity
beyond their internal needs.

Therefore, while the position of the iron and steel industry has improved
over the past few years, there are still some fundamental problems which
need to be resolved before the industry will become truly viable.

5.2.3  Impact of NO  Control Costs
Using the 1977 reported industry-wide costs for pollution control as a
guide, the impact of NO  controls on existing sources would appear to
                       J\
be modest.  Assuming that annual pollution control spending would
remain at the 1977 (about $500 million) level through 1984, the-industry-
wide investment would total about $2 billion for the four year period.
This is a low-side estimate in comparison with a 4.1 billion  (1979 dollars)
estimate for controls on existing plants for the period 1977-83
(Little, A. 1975).  The approximately 250 million dollars estimated
in this study for a 0.15 ppm is thus about 10 percent of the total
pollution control costs for the entire industry.  However, on a per
plant basis, the estimated NO  costs are approximately $13 million.
                             A
This is only slightly less than the average investment costs for all
pollution control equipment for all 130 major steel-making plants.

-------
5.2.4  Product Price  Increases
Based on other studies of the impact on steel costs of annual pollution
control expenses, the estimated $37 million annual costs for NO  controls
                                                               Jv
are insignificant.  The Arthur D. Little study predicted a steel making
cost increase of $25-30 per ton from pollution control investment of
$15-17 billion, or,, as a first approximation, an annual cost of about
$3 billion  (Little, A. , 1975).  The annual NO  control expenditure would
thus cause a per ton  increase in steel of less than $.50.

5.2.5  Conclusion
The capital costs of  additional NO  controls may be significant based on
                                  Jv
a simple analysis of  average pollution control costs per plant.  A more
detailed analysis of  model plants would have to be undertaken in order
to determine the impact on capital raising ability for the typical
company.  On an industry-wide basis, the estimated NO  control costs
are insignificant.

5.3  Urban and Community Impact Analysis
One of the main purposes of the Urban and Community Impact Analysis
(UCIA) is to identify those cities or types of cities which are more
likely to suffer an economic burden due to added costs of control
needed to achieve a one-hour NO- standard.  Specifically, the UCIA is
supposed to examine potential changes in employment, population, and
income due to a new NO- NAAQs.  Several studies have identified the
Nation's cities facing the most hardship.  One recent report by the
Urban Institute examined the Nation's 153 largest cities on the basis
of three distress indicators:
  •  Population decline between 1970 and 1976 of two percent or
     more.
  •  Per capita incomes less than the all-city average in 1970.

-------
     Unemployment rates greater than the all-city average in
     1976.
These cities were than grouped into four categories (0-3), with category
0 being the cities not exceeding any of the distress indicators, and
category 3 being the cities exceeding all of the distress indicators.
Table 5-5 below shows the number of AQCR's with cities in each of the
four city types which will need to apply area source controls in order
to meet alternative levels of the one-hour NO- standard.  If the per-
centage of areas in each category exceeding a standard is used as an
indicator, there does not appear to be a significant difference in the
type of city impacted by a one-hour standard.  However, the table indi-
cates that the most distressed cities tend to have the most severe NO-
problems.  Therefore, more controls would be needed in those-areas.

                               TABLE 5-5
            NUMBER OF AREAS EXCEEDING ALTERNATIVE ONE-HOUR
                N02 STANDARDS CATEGORIZED BY CITY TYPE

          Total No. of     	    One-Hour NO,, Standards (ppm)
City      Areas in This
Types       Category

  0            8
  1           29
  2           23
  3           18_
              78
.15
6
21
14
11
54
.20 -
3
12
10
il
38
.25
2
6
4
9
21
.35
0
1
2
3_
6

-------
A second purpose of the UCIA is to examine the impact of NO  control
                                                           Jt
costs on households and communities.  In other words, it is important

to identify potential control costs that might create hardships for

certain sectors of a community.  NO  controls could directly affect
                                   A
the following consumer costs:
     Utility bills could increase when costs of NO  controls
     (low NO  burners or selective catalytic reduction) are
     passed on to consumers.

     For areas with NO  related inspection and maintenance
     programs, direct costs will be incurred for inspections
     and inspection related repair.

     If a 0.4 gram/mile automotive emission standard for
     NO  is implemented, costs will be incurred nationwide
     for all new car buyers from 1985 (if that is when the
     standard is implemented) on.

     The NO- area source analysis showed that for some of
     the most stringent one-hour standards, new low NO
     furnaces for space heating may be needed to bring
     all AQCR's into attainment.  The cost of replacing
     these furnaces would be substantial and would be
     direct costs to households.

-------
                              REFERENCES
American Iron and Steel Institute. 1978.  Annual Statistical Report
1977.  Washington, D.C.

Little, A.  May 1975.  Steel and the Environment:  A Cost Impact Analysis,
Cambridge. MA.

Standard and Poors.  September 14, 1978.  Industry Surveys, Steel-Coal:
Basic Analysis.

Temple, Barker, and Sloane, Inc., May 1976.  Economic and Financial Impacts
of Federal Air and Water Pollution Controls on the Electric Utility Industry,
EPA-230/3-76-013, prepared for Office of Planning and Evaluation, U.S.
Environmental Protection Agency, Washington, D.C.

U.S. Bureau of the Census, 1978.  Department of Commerce, Statistical Abstract
of the United States, Table numbers 1018, 1021, 1022, and 1025.

U.S. Department of Energy, December 1978.  Statistics of Privately Owned
Electric Utilities in the United States 1976, DOE/EIA-0044, Tables 1, 2,
and 3, and Statistics of Publicly Owned Electric Utilities in the United
States 1976, DOE/EIA-0146, Tables 1M, 2M, 1FP, and 2FP.

-------
             APPENDIX A
DETAILS OF THE FREQUENCY DISTRIBUTION
        ANALYSIS OF N02 DATA
   TO BE INCLUDED AT A LATER DATE

-------
                              APPENDIX B
              TRANSPORTATION CONTROL MEASURES OTHER THAN
                      THOSE USED IN THE ANALYSIS
Traffic Signal Improvements
This measure would produce no significant change in NO  emissions.  If
                                                      A
vehicle speeds increase, NO  emissions may increase.  Analyses for
hypothetical cities by Voorhees showed that NO  emissions would not
change by more than 1 percent due to traffic signal improvements.

Work Rescheduling
Variable work hours can reduce CO emissions by increasing average speed
during the peak hour.  However, increases in average speed may increase
NO .  Compressed work week programs are aimed at reducing average daily
  1\
work trips and vehicle miles traveled during peak periods.  Four day
work schedules would deter the formation of carpools and could dissolve
some existing ones.  Therefore, the reduction in average daily work trip
VMT would be less than the reduction in average daily work trips.  On
balance, overall changes in VMT resulting from four-day work week schedules
would be negligible.

Traffic Engineering Improvements
Right-turn-on-red provisions could lead to NO  reductions of 1.7 percent
                                             A
(0.1 grams/mile).  This reduction was estimated over an urban grid net-
work for all vehicles (including buses).

Curb parking restrictions help speed the flow of traffic.  Roadways with
curb parking operate at two-thirds the capacity of similar facilities
without parking.  The results from a Washington, D.C. study were used as

-------
a basis for estimating fuel consumption and vehicle emission changes
resulting from curb parking prohibitions along an arterial.  NO  emissions
                                                               A
were estimated to increase by 7 percent.

Turn prohibitions are likewise designed to keep traffic flowing freely at
a constant speed.  Localized increases in NO  emissions resulting from
turn prohibitions are 30 percent for peak hour conditions and 4 percent
for off-peak hours.

Two-way left turn lanes reduce travel time and increase average speed.
Based on a model intersection, two-way left turn lanes would produce
approximately a 2 percent increase in NO .
                                        A

Bus stop relocation from the near-side to the far-side of a signaled-
intersection can impact fuel consumption and auto emissions.  The
following impacts were estimated in the Voorhees study:

               A.M. Peak           Midday       P.M. Peak
Fuel Con-
sumption          +7%               +9%           -6%
N0v               -2                -2            +2
  A

Geometric improvements, such as widening intersection approaches, serve
to improve the flow of traffic, but they would not reduce N0v levels.

One way streets can increase the roadway capacity from 0-29 percent when
compared with a pair of 2-way streets, depending on widths and parking
conditions.  Changing from two-way to one-way streets will reduce travel
times, but vehicle miles traveled will increase.

-------
Reversible lanes are used to take advantage of underutilized lanes in
the light volume direction by assigning them to serve the peak direction
of flow.  The impacts of reversible lanes can vary widely and include:

  o  Capacity increases up to 50 percent.
  o  Traffic volume increases from 4-41 percent
  o  Travel times decrease up to 20 percent.
  o  Vehicle speeds increase up to 23 percent.

Therefore, reversible lanes are more likely to increase than decrease NO .
                                                                        A

Freeway surveillance and control involves rapidly detecting and solving
congestion problems as well as controlling flow via ramp metering and
other congestion reducing measures.  This TCM serves to increase vehicle
speed.  Therefore, it would not reduce NO  levels.
                                         A

Parking Management
The conclusion of the Voorhees analysis regarding parking management controls
was that actions directed at increasing the cost of commercial parking for
CBD commuters are relatively ineffective.  Work travel is relatively
insensitive to parking cost increases.  The evidence from a number of
studies suggests that actions designed to discourage auto use through
price are less effective than actual constraints on the number of
spaces.

However, parking management actions can contribute to overall Transportation
System Management (TSM) programs.  Parking management often combines
well with other measures to produce impacts beyond what could be achieved
with individual measures.

-------
Truck Restrictions and Enhancements
These involve changing regulations to provide curbside loading zones for
pick-up and delivery activities and optimize the movement of through
traffic.  The major effect of reducing truck-auto interference is to
reduce the delay encountered by through vehicles.  This will increase
the average speed along the affected roadways, but again, will not
reduce NO  emissions.

-------
                                  *-r  DiNT~^jRcr  nALYi  i s—;NIN:iocr~~us
(DEC 77)        "             	 05/360  FORTRAN H EXTENDED	——   DATE"7972Z5/16*5.9 V40PAGE

IONSI  ALCt

-ECTI  NAME(MAIN)  OPTIMIZE(2) LINECOUNT(60) SIZE(MAX)  AUTODBL(NONE)-__._. 	
      SOURCE EBCDIC NOLlSf NODECK OHJECT NOMAP NOFOHMAT  NOGOSTMT  NOXREF ALC NOANSF TERM FLAG(I)
   C  SELECT SIGNIFICANT NOX SOURCES FROM NEDS    MT~ADES"™~~ 	  	

         DIMENSION AK200) tA2(200) . A3 (200) »A4 (200) • ISCC (200) » AQCR (200) *
       . & JSICUOO)»HE1GHT(200)»
        &[> J AM (200)»TEMP (200) ,FLOW (200) iDES (200)   '	~~   " 	~ '  	
         DIMENSION OHATE(200),XMAX(200)tSOX(200)fXNOX(200)tCON(200)
i         DIMENSION IFLAG(200)»K1(300)»E(300»2)iIHOURS(200).HF(200)»MB(HOO)
         INFEOEH AUTAB(200)iAQCR
         UtAL MAX(200)
         DIMENSION ICT (4) t AQ(200) »ICOUT (300) » INJOUO) tQ(3oO)
         DATA Kl/300«0/*E/600»0.y/iICOUT/300»0/
         DATA ICT/4»0/ . IAU/0/ t K1X/0/ »10UT/0/  »IMB/0/
         DATA iPL/0/»ix/o/tJ/o/tND/o/tNx/o/»NO/o/  fJQ/o/~
         IFNsQ
         AU(200)  = 13.  /1880.
         1STOP=0
   DO  b  I=Jt200
      D(4»«*ENDs7)  A
   A()INE
         IF ( I SCC (J)^LE. 1999.9999 LOO Tp_500
   c  .....             .......... •
   C   TEST  FOR  NON BOILERS
   C
                   EQ.O«0) MAX (jr=ORATE(J)/FLOAT(IHOURS(J) )

   C   SCREEN FOR TOO LARGE VALUES
  .c... ...... ............................ : .............................. ____ ............. . ............ ,„...... ................... ;__:.. ...... ^:_ ........

-------
OEC 77)
              MAIN
                        OS/360  FORTRAN"if EXTENDED
                                                            DATE~797225V 16.59440
                                                                                              PAGE
  C
  C
  C
        IF(U /FLOAT UHOURS (J))
        1F(MAX(J).GT.O<1NE>) GO TO 06
        IF(MAXU).EQ.O.O) GO TO 07
        UO TO 6UO            	
     50 TIMES TF.SI

    430 OMH=MAX(J» • FLOAT (IHQUHS(J))
        00=OMAfE(J) * 50.
        IF(DMH.GT.OG) GO TO 420
        GO TO 600
 TEST FOR BOILERS
    500
     F(
DE
OE
      (J
       J
           IIESU
           OflNE
             .EU.0.0 .AND. MAX(J).£0*0.0) GO  TO  07
              .EU.0.0
              W^BlMWWV**,
              .EO.0.0)
              .EU.0.0)
                     • ANO. UHAIC
                      OES(J)=MAX
                   .   GO TO 530
             .LE.O(INE)) GO TO 530
     ES(J)=MAX7J) ..« EUNE.2)   	
     : (OES(J).GT.QllNE)) GO TO 06
    GO TO 600
530 OH=DES]J) « FLOAT(
          UEi
S9
                                                   JO 87
    pOsORATE
-------
(DEC 77J
             MAIN
OS/36UFORTRAN"H EXTENDED"^"
DATE~7 97Z25/
PAGE
   PM s PM

   IFlPM.GE.O.l)
   PM=0.0
   UO 51 K=ltJJ
   CALL PTMAX(5.»HE1GHT(K) .FLOW 
A3d)=A3 J)
ISCCM = SCC
AOCR( »=AUCR
15IC( )=IS1C
HEIGHTtD«HEl
W\ l:?i»
FLOW( )=FLOW
UESd =l)ES(J)
	 ORATE (1)=ORA1
1 GO TO 90 	 : 	 	 	 	 - 	 	
 v •
GHT(J) ' • ' • ,•• ' ' .'• : ' : : '..--.'.• ' ' '•••/•:•'••• .-•.--
JJ .••.•:••:••• •-. '••• . • • . • • • • •"••f;- .-. . ' ••
J
J)
E(JJ ...... ' _
MAXd)=MAX(J)
SOXdi=SOX(J) \ :
XNOX(l)sXNOX(J»
1NJ(1)3INJ(J) , -

PMsO.O
J=l
1PL=IPL » 1
. ....„ GO TO 230 ...
(IS NO=NU»I
GO TO 20
86 NX=NX*1
.....: 	 GO TO 20
'
•; s •

-------
 LEVEL 2.2.1 IDEC 771         :  MAIN          OS/360  FORTRAN H EXTENDED	DATE  79.225/16.59.40

   |SN 0 52        87 NQsNO*!                         .
   ISN 0 53           GO TO 20      '
   1SN 0 5*       997 MHIIE(0,6001) 10UT,Ai<50)*A2<50)*A3(50>tISCC(50>
   ISN 0 55      6001 FOUMAld UT 200 CASES* OUT a i«18, iXtZAJt A«* IU)
   ISN 0 56           GO IQ 499
   J5N 0 57        99 WHI]EI(>f6002> (ICT (J> . J= 1, 4 >             _               __
   ISN 0 5B           WIU1E]6.«I  NOiNXtNUiJO                                "
    SN 0 59      6002 FOHMAT(//t'4lfU                                   •
    SN 0 60           MIUTE(6t«!  IOUT flPL *IXtIHII
    SN 0 61           1)0 998 K«1»K|X
    SN 0 6§       99B W«llE(6V«l
    SN 0 63       999 STOP
    SN 0 64           END

•OPTIONS IN EFFECT«NAME(MAINr OPTIMIZE12I LINECOUNTI60r SIZE(MAX)

"OPTIONS IN EFFECT'SOUHCE EBCDIC NOLIST NODECK OBJECT NOMAP NOFORNAT NOOOSTMT NOXREF AlC  NOANSF  TERM  FLAQI1)

«STATISriCS«     SOURCE STATEMENTS a     1631 PROGRAM SIZE *    29668»' SUBPROGRAM"NAME e  MAIN

•STATISTICS*  NO  DIAGNOSTICS GENERATED

«••••* END OF COMPILATION »••«•»	TDK BYTES"OF CORE'NOT'USED 	~

-------
 sec
            EMISS. FACTOR   HEAT CONTENT SCREEN UPPER BOUND
10100101
10100102
10100103
101001U4
10100105
10100106
1010019V
1010U201
10100202
10100203
10
10
10
10
lu
   .00204
   00205
   00206
   00207
  _00208
10100209
10100210
10100211
10100212
10100299
10100301
10100302
10100303
10100304
10100305
10100306
10100307
10100308
1010030V
10100310
10100311
10100312
10100313
10100314
10100315
10100401
10100402
10100*03
10100501
10100502
10100503
10100601
10100602
10100603
10100701
10100702
10100703
10100801
10100901
U.690
0.400
U.690
0.4UO
0.690
0.330
0.690
1.360
U.620
1.300
0.680
0.680
1.360
0.820
0.6bO
0.680
0.270
0.270
0.820
0.820
0.810
0.810
0.880
0.810
0.810
0.810
O.ttlO
0.810
0.810
0.810
0.810
0.810
U.810
0.010
0.810
0.710
0.710
0.710
0.7 00
0.700
0.700
0.590
0.230
0.120
U.590
0.230
0.120
0.770
0.580
  26.000
  26.000
  26.000
  26.000
  26.000
  26.000
  26.UOO
  22.000
  22.000
  22.000
  22.000
  22.000
  22.000
  22.000
  22.000
  22.000
  22.000
  22.000
  22.000
  22.000
  16.000
  16.000
  16.000
  16.000
  16.000
  16.000
  16.000
  16.000
  16.000
  16.000
  16.000
  16.000
  16.000
  16.000
  16.000
 150.000
 150.000
 150.000
 140.000
 140.000
 140.000
1022.000
1022.000
1022.000
1200.000
1200.000
1200..000
  24.000
  17.000
13000
13000
  100
  100
  100
  100
  100
13000
13000
13000
13000
13000
  100
  100
  100
  100
  100
  100
  10U
  100
13000
13000
13000
13000
13000
13000
  100
  100
  100
  100
  100
  100
  100
  100
  100
13000
  100,
  100
13000
  100
  100
13000
  100
  100
13000
  100
  100
  100
  100
.000
.000
.000
.000
• oou
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.QUO
.000
.000
.000
.1)00
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000

-------
 sec
EMISS. FACTOH   HEAT CONTENT SCREEN
                                    BOUND
10100902
 01009U3
 0101101
 0101102
:0101103
10200101
10200102
10200103
10200104
10200105
10200106
10200107
10200199
10200201
10200202
10200203
 0200204
 0200205
 0200206
 0200207
10200208
10200209
10200210
10200211
10200212
10200213
10200214
10200299
10200301
10200302
10200303
10200304
10200305
10200306
10200307
10200308
10200309
10200310
10200311
10200312
10200313
10200314
10200315
10200316
10200401
10200402
10200403
10200501
10200502
0.5(50
0.580
0.250
0.250
0.090
0.4UO
O.OVO
0.4UO
0.690
O.J30
0.120
0.090
1.360
0.020
1.660
0.680
0.600
0.660
1.300
0.820
0.680
0.270
0.270
0.820
0.270
0.140
0.680
0.810
u.bio
1.060
0.810
0.810
0.810
0.810
0.610
0.810
0.8 10
0.810
0.810
0.610
0.810
0.810
0.810
0.410
0.410
0.410
0.410
0.150
                    17.000
                    17.000
                     8.000
                     8.000
                     M.OOO
                    26.000
                    26.000
                    26.000
                    26.000
                    26.000
                    26.000
                    26.000
                    26.000
                    22.000
                    22.000
                    22.000
                    22.000
                    22.000
                    22.000
                    22.000
                    22.000
                    22.000
                    2?.000
                    22.000
                    22.000
                    22.000
                    22.000
                    22.000
                    16.000
                    16.000
                    16.000
                    16.000
                    16.000
                    16.000
                    16.000
                    16.000
                    16.000
                    16.000
                    16.000
                    16.000
                    16.000
                    16.000
                    16.000
                    16.000
                    50.000
                    50.000
                    50.000
                    40.000
                    40.000
 100
 100
 100
 100
 100
1500
150U
1500
1500
 500
 500
 500
 500
 500
1500
1500
1500
1500
1500
1500
1500
1500
1500
1500
1500
1500
1500
1500
1500
1500
1500
1500
 500
 500
 500
 500
 500
 500
 500
 500
1500
1500
1500
1500
 500
 500
 500
 500
: 500
.000
• uoo
.000
,000
.000
.000
.000
.000
.000
.000
.000
• QUO
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000

-------
 sec
      EMISS. FACTUK   HEAT CONTENT. SCKEFN UPHEK BOUND
10200503
10200601
10200602
10200603
10200702
10200703
10200704
10200705
10200706
10200707
10200708
10200709
L0200799
10200802
10200803
10200901
10200902
10200903
10201002
10201003
10201101
10201102
10201  "
20100
2U100.
20100201
20100202
20100301
20100302
20100401
20100501
20100601
20100701
20200101
20200102
20200201
20200202
20200301
20200401
20200402
20200501
20200601
20200701
20200801
20200802
30100101
30101001
30101002
30101003
03
01
02
0.150
0.301
0.230
0.120
0.230
0.120
0.120
U.120
0.120
0.210
0,210
0.210
0.250
0.640
0.260
0.580
0.580
0.560
0.120
0.120
0.250
0.250
0.250
0.340
2.7UO
0.340
3.700
2.640
0.340
0.340
0.340
0.340
0.340
0,340
2.700
0.340
3.700
0.820
3.350
0.3AO
0.340
0.340
0.340
0.340
3,700
5.500
5.500
4.000
2.000
 140.000
1022.000
1022.000
1022.OOU
1200.000
1200.000
1200.000
1200.000
1200.000
1200.000
1200.000
1200.000
1200.000
  24.000
  24.000
  17.000
  17.000
  17.000
  90.000
  90.000
   b.ooo
   8.000
   b.ooo
 140.000
 140.000
1022.000
1022.000
 138.000
 138.000
 150.000
 130.000
 138.000
1200.000
 140.000
 140.000
1022.000
1022.000
 110.000
 138.000
 138.000
 150.000
 130.000
 138.000
1200.000
1200.000
   1.000
   1.000
   1.000
   1.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
1500.000
  10.000
   0.700
   1.500
   0.100
   0.700
  10.000
   0.0
   0.0
   0.0
   0.0
  10.000
   0.700
   1.500
   0.100
   0.700
   0.. 700
  10.000
   0.0
   0.0
   0.0
   0.0
   0.0
   0.0
   0.0
   0.0
   0.0

-------
 sec
       EMISS. FACTOR   HEAT CONTENT SCHEEN UPPEK BOUND
30101004
3010:
3010
2010:
3010
3010J
005
301
303
304
30101305
30101306
30101307
30101308
30101399
30600101
30600102
30600103
3060010*
39000199
39000201
39000202
39000203
39000204
39000206
3900020f
39000208
39000209
39000299
39000401
39000402
39000403
39000404
3900040$
39000406
39000407
39000408
39000409
39000410
390004H
3900043U
39000431
39000432
39000450
39000451
39000452
39000499
39000501
39000502
39000503
39000504
39000505
39000506
5.500
5.500
5.500
4.5UO
5.000
0.200
5.500
5.500
5.500
U.2UO
5.500
0.470
0.230
0.470
0.230
0.700
U.ttOO
U.BOO
O.bOO
U.bOO
U.bUO
0.600
O.UOO
O.BOO
U.bOO
O.bOO
o.uoo
O.bOO
o.boo
o.boo
O.dOO
O.bOO
o.boo
0.300
U.bOO
O.bOO
O.bOO
O.bOO
o.boo
O.bOO
0.800
o.boo
O.bOO
o.boo
O.bOO
0.800
O.bOO
0,800
u.boo
   1.000
   l.UOO
   1.000
   1.000
   l.UOO
   1.000
   1.000
   1.000
   1.000
   1.000
   1.000
6300.000
   1.022
 150.000
1022.000
  26.000
  22.000
  22.000
  22.000
  22.000
  22.000
  22.000
  22.000
  22.000
  22.000
  50.000
  5U.OOO
  5o*ooo.
  50.000
 150.000
 150.000
 150.000
 150.000
 150.000
 150.000
 150.000
 150.000
 150.000
 150.000
 150.000
 150.000
 150.000
 150.000
 140.000
 140.000
 140.000
 140.000
 140,000
 140.000
  0.0
  0.0
  0.0
  o.u
  0.0
  0.0
  0.0
  o.u
  0.0
  0.0
  u.u
  0.024
391.400
  2.667
  0.391
 35.000
 37.000
 40.0.00
 30.000
 4U.OOU
  0.360
 40.000
 40.000
 4U.OOO
 4U.OOO
  1.000
  5.500
  4.300
 10.000
  U.120
  0.060
 10.000
  5.000
 10.000
 10.000
 10.000
 10.000
 10.000
  0.000
  u.ooo
 ,0.000
  0.000
 10.000
  1.000
  5.800
  4.600
 10.000
  0*125
 10.600

-------
 sec
 FACTO*   rtfaT CONTENT SCREEN UPH£H WOUND
39000507
39000508
39000509
39000510
39000511
39000530
39000b31
39000b32
39ooobbo
39()00bbl
39000552
39000599
39000601
39000602
39000603
39000604
3900060b
39000606
39000607
39000608
39000609
39000610
39000611
39000630
39000631
39000632
39U00650
39000651
39000652
39000699
39000701
39000702
39000799
39000801
39000899
39000999
39001099
50100101
boiooioa
SOlOObOb
b0100506
50lOOb07
b0100599
5019000*
5019000b
50190006
50190010
50300101
50300102
U . 8 0 U
0.800
0.800
0.800
0.800
0.800
0.800
0.800
0.800
u.aoo
0.80U
0.800
0.700
0.700
0.700
0.700
0.700
0.700
0.700
0.700
0. 700
0.700
U.700
0. 700
0. 700
0.700
0.700
0.700
0.700
0.700
0.230
0.230
0.230
0.2e>0
0.260
O.b80
0.120
0.200
0.200
0.300
0.500
0.500
0.500
0.410
0.410
0.300
0.120
0.300
0.200
 140.000
 140.000
 140.000
 140.000
 140.000
 140.000
 140.000
 140.000
 140.000
 140.000
 140.000
 140.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1022.000
1200.000
1200.000
1200.000
  24.000
  24.000
  17.000
  90.000
  10.000
  10.000
  10.000
  10.000
  10.000
  10.000
 150.000
 140.000
1022.000
1200.000
  10.000
  10.000
1 0.000
 5.200
10.000
10.000
10.000
10.000
10.000
10.000
 .0.000
 0.000
 ,0.000
 0.000
 0.125
 U.800
 0.700
 1.000
 0.020
 0.010
 1.000
 0.700
 1.000
 1.000
 1.000
 1.000
 1.000
 1.000
 1.000
 1.000
 1.000
 1.000
 0.900
 1.000
 1.000
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 0.0
 c.o
 0.0

-------
 sec        EMISS. FACTOR   HEAT CONTENT SCHEEN UPPE« BOUND


50300103          0.100         10.000          0.0
50300104          0.5UO         10.000          0.0
50300105          0.100         1U.OOO          0.0
50200106          0.400         10.000          0.0
50300506          U.500         10.000          0.0
50300599          0.500         10.000          0.0
50390004          0.410        150.000          0.0
50390005          0.410        140.000          0.0
50390006          0.300        1022.000          0.0
50390007          0.250        1200.000          0.0
50390U10          0.120         90.000          0.0

-------