WPCJ 2iBVYZ#|;olbook&)o=3Uo P['C&P"m^ANoϨANN[ANAANN䨨A¨ܜNANANAAAϏ[Nu[A[NANN'NNNNNNNNNNAϨAAAAAAAAAAAAAAAAAAAAܨ[[[NNܶuuuA[NNRANVUÂNN8uuNNAuNooee3o< "* R-#:s2PkC XP# # :s2PkC XP# 2nd tier (1)EHPA-Heading for 2nd-tier sectionM̸(' P-#:x2p}wC[X#x  P-x#:x2p}wC[X#BulletEHPA-Bullet (Schoolbook) .1 J-`g"Table TextEHPA-Formats text in Pechan tables ;'d J(#`2PkCP##&U PyQՈ&P#2 N 9 G1st tier (A)ing (A)EHPA-Heading for 1st-tier section D٠(' P-#:x2p}wC[X#  P-#:x2p}wC[X#a5TechnicalTechnical Document Style )WD (1) . a6TechnicalTechnical Document Style )D (a) . a2TechnicalTechnical Document Style<6  ?  A.   2!'t 6!a3TechnicalTechnical Document Style9Wg  2  1.   a4TechnicalTechnical Document Style8bv{ 2  a.   a1TechnicalTechnical Document StyleF!<  ?  I.   a7TechnicalTechnical Document Style(@D i) . 2#!vt"p"kZ#a8TechnicalTechnical Document Style(D a) . a8DocumentgDocument Style StyleXX` `  ` a4DocumentgDocument Style Style . a6DocumentgDocument Style Style GX  2&k#b$v%}%a5DocumentgDocument Style Style }X(# a2DocumentgDocument Style Style<o   ?  A.  a7DocumentgDocument Style StyleyXX` ` (#` a1Right ParRight-Aligned Paragraph Numbers:`S@ I.  X(# 2(I&&'?(a2Right ParRight-Aligned Paragraph NumbersC @` A. ` ` (#` a3DocumentgDocument Style StyleB b  ?  1.  a3Right ParRight-Aligned Paragraph NumbersL! ` ` @P 1. ` `  (# a4Right ParRight-Aligned Paragraph NumbersUj` `  @ a. ` (# 2V,&) )!*"}+a5Right ParRight-Aligned Paragraph Numbers_o` `  @h(1)  hh#(#h a6Right ParRight-Aligned Paragraph Numbers h` `  hh#@$(a) hh#((# a7Right ParRight-Aligned Paragraph Numbers!pfJ` `  hh#(@*i) (h-(# a8Right ParRight-Aligned Paragraph Numbers"yW"3!` `  hh#(-@p/a) -pp2(#p 2/#,$T-% .&.a1DocumentgDocument Style Style#Xqq   l ^) I. ׃  SourceEHPA-Sources for tables$Mv' eA#|M2PkC酐P#SOURCE:#|M2PkC酐P#Helv. Bold 10EHPA-Changes type to Helv. Bold 10 pt.%>o* ~J(#J2PkQFP# #J2PkQFP# 4th tier (i)EHPA-Heading for 4th-tier section&V3' P-#:x2p}wC X#g   P-g#:x2p}wC X#2m5'0(2)13*q3ChapterheadEHPA-Chapter title':Br= yxdddy   `<#2p}wC0~ #   @yxdddjxy    `<#2p}wC0~ #Bullet ListEHPA-Indented Bullet List(. J-`g(# Notesz"EHPA-Formats notes below table)L' eAx#|M2PkCՐP#NOTES:#|M2PkCՐP#ResumeEHPA-Pechan staff resume format 09/03/91* . Q-# a\  PCz P# (08@P 0 pP02}<+5,6-Ni9.;3rd tier (a)EHPA-Heading for 3rd-tier section+V\(H' P-#:x2p}wC[X#  P-#:x2p}wC[X#LetterEHPA-Pechan letterhead letter,uqx- J(    @Page Page @ h  (08@X`hp x  p#&U PyQՈ&P#  hhp p . ^x x  !P  pp  D3 1, 4D Appendix Covr PageEHPA-Formats text for appendix cover sheet-f= Tyxdddy T  `<T#2p}wC0~ # T  @TyxdddjyT#2p}wC0~ #NoteEHPA-For single note below table.G' dB #;2PkQdP#NOTE:X#;2PkQdP#2@/<0&=1>2?SourcesEHPA-Formats references at bottom of table/Q$* f@x#;2PkQ-%P# SOURCES:h  #;2PkQ-%P# Table TitleEHPA-Formats table titles in Pechan reports0ZM2' P2}#:x2p}wC[X#   }  P2}#:x2p}wC[X#ReferencesEHPA-Pechan references1gc!1 J-#&U PyQ&P#`gP4(#  J-#&U PyQ&P#Table Title @6cEHPA-Suppresses headers/footers & changes fontW.E2:'#*s\  PXP#   #*s\  PXP#2C3n@4X(A5B6}&CBibliogrphyEG5nBibliography0MQW 0G5nLQI 0G312 Tab/Fig/Refs5ـEHPA-Formats Tables & Figs & Refs headings11#[M4:# 2pQ # yxdddy  #&h P Q&P#Helv. Bold 1K&8tEHPA-Changes font to Helv. Bold 128t ?XI -5'' lM2P   lM2P  a1Bullet ListBullet List6$ 27F7C8]D9D:Ea2Bullet ListBullet List7/` ` ` a3Bullet ListBullet List8:` ` `  a4Bullet ListBullet List9E` ` `  a5Bullet ListBullet List:P  ` ` ` hhh 2J;iF<G=G>BHa6Bullet ListBullet List;[  ` ` ` hhh a7Bullet ListBullet List` hp x (# X` hp x (#2V`K zNX7P`S"m^56=ii,GG\o3=33iiiiiiiiii33ooo\QwzzGoGo\,pwfziGp~DAzA~pzw_\N~pppbGoGo=3============pDpppppfiiiiQDQDQDQD~pppp~~~~ppppppzpppffffziiiipppppp~QDQDQDQDwzAAAA~~~~pp___z\z\z\z\NN~~~~~~pzbzbzbzA~_z\Nppzp~No3\6\\\iiiCCi?i?,XX\==\\~~,X?ooee3o<\ `_"m^3=Iff%==\o3=3offffffffff33oooQzKpzzz~~z=o=o\%ifQpQ=bp=:f=p\ifQQAp_\\U=o=o=3============f=iiiiiQQQQQK=K=K=K=p\\\\pppp~\i\\~\\ziiiiQQQQpQQQQbbbbbbpK=K=K=K=pfz=z=z=z=pppp\\QQQzQzQzQzQ~A~Apppppp~\zUzUzUpz=pQzQ~A~\~\p\pNo3w\=QNNfffMDf=f=3GG\==\\pp%G=ooee3o<:(5,C:>5>:51:5L551&,,,(,,,,(C,,,,(,(:(((/P 15,5,5,5,5,PG:(5,5,5,5,:,>,>,>,>,:,:,:,:,5(5,>1>,5(:,5,5,5,5,:(:(:(:(:,5,5,5,5,>,>,>,>,>,>,:,(5(,,,,:,:,:,:,>,>,PL:::5(5(5(5(11:,:,:,:,:,:,L:5(1(1(1(:,,:,:5(15(5(:,>,:,N+,1,,,,,CC,;;,C,P,,P((PP"m^.1A]]"88Se.8..]]]]]]]]]]..eeeJ{xxxxoD]ooxioxuuf8e8eS"]]J`S8Zf51c5fS`]JMAfZZZP8e8e8.888888888888`5x]x]x]x]x]xJxSxSxSxSD5D5D5D5fSSSSffffuZx]SSuZSo`x]x]x]xJxJxJxJ`xSxSxSxSZZZZZZfD5D5D5D5]co5o5o5o5ffffSSxJxJxJiMiMiMiMoAoAffffffuZfPfPfP`o5fxJiMoAuZuZ`SfNe.eS1JGG]]]82]8{{]8"AA]++SSff"A8ee\\w.e77\\\wwweeeCe.wR)EreewwwwIeenR\\\wwwxio\eEfRfIfRxe|W87y\r\fxWlRx\\]\ceIfIs`Wx\rriIe77\``rigewiiiiiiiiiiiiiiiiiiiIIIIIIIeeeeeeeeeeeeeeeeeeee777777777777\\\\\\\````````````rrrrrrrrrrrrrrrrrrrrxfx8xs"m^'88SR!!2=!8888888888===,JHHHNHCNS)8NC^RNCNH?CRHbFF=!=!=288,92!6= ; Y=298,.'=6N660!=!=!!!d!!!!!!!!!!9 H8H8H8H8H8dPH,H2H2H2H2) ) ) ) R=N2N2N2N2R=R=R=R=F6H8N2N2F6N2C9H8H8H8H,H,H,H,N9H2H2H2H2N6N6N6N6N6N6S=) ) ) ) 8N;C C C C R=R=R=R=N2N2dSH,H,H,?.?.?.?.C'C'R=R=R=R=R=R=bNF6=0=0=0N9C R=H,?.C'F6F6N9N2R=N==2,++888!SS8!JJ8S!''8d22d==d'd!==77G=!!777GMGG=cc<;o;rfolWSGrfffZAeAe8.888888888888f>fffff]````J>J>J>J>rffffrrrrxffffxffofff]]]]o````ffffffrJ>J>J>J>lox;x;x;x;rrrrffWWWoSoSoSoSxGxGrrrrrrxfoZoZoZox;rWoSxGxfxfofrNe.}S1SSS```==`9}}`9(PPS88SSrr(P9ee\\w.e77\\\wwweeeCe.wR)EreewwwwIeenR\\\wwwxio\eEfRfIfRxe|W87y\r\rxWlRx\\]\ceIfIs`Wx\rriIe77\``rigewiiiiiiiiiiiiiiiiiiiIIIIIIIeeeeeeeeeeeeeeeeeeee777777777777\\\\\\\````````````rrrrrrrrrrrrrrrrrrrrxfx8xs"m^..;]]o%88Aa.8..]]]]]]]]]]..aaa]ooxxofx.So]xoxofxooof...N]%]]S]].]]%%S%]]]]8S.]SxSSS8+8a8.888888888888f.o]o]o]o]o]xSo]o]o]o]........x]]]]]x]x]x]x]oSo]f]oSx]o]o]o]o]xSxSxSxSx]o]o]o]o]]]]]]]x].....%..SoS]%]%]%]%x]x]x]x]]]x8x8x8oSoSoSoSf.f.x]x]x]x]x]x]xoSfSfSfSx]]%x]x8oSf.oSoSx]]x]N:.Z]8f]]]]]>=]8{{]8 88]88]]SS%88"m^.8C]]"88Se.8.e]]]]]]]]]]..eeeJ}uxxxoDf|oo|oruuro8e8eS"`]JfJ8Zf85]8fS`]JJ;fWSSM8e8e8.888888888888]8u`u`u`u`u`xxJxJxJxJxJD8D8D8D8fSSSSffffrSu`SSrSSo`u`u`u`xJxJxJxJfxJxJxJxJZZZZZZfD8D8D8D8f|]o8o8o8o8ffffSS|J|J|JoJoJoJoJr;r;ffffffrSoMoMoMfo8f|JoJr;rSrSfSfNe.lS8JGG]]]F>]8}}]8.AAS88SSff"A8ee\\w.e77\\\wwweeeCe.wR)EreewwwwIeenR\\\wwwxio\eEfRfIfRxe|W87y\r\fxWlRx\\]\ceIfIs`Wx\rriIe77\``rigewiiiiiiiiiiiiiiiiiiiIIIIIIIeeeeeeeeeeeeeeeeeeee777777777777\\\\\\\````````````rrrrrrrrrrrrrrrrrrrrxfx8xs"m^!/88YH!!':!8888888888!!:::=bHHHHC=NH8H=SHNCNHC=HC^CC=!!:88=8=8!==8Y===='8!=8N882'':!!!d!!!!!!!!!!=H8H8H8H8H8dYH8C8C8C8C8H=N=N=N=N=H=H=H=H=C8H8N=N=C8H=C=H8H8H8H8H8H8H8H=C8C8C8C8N=N=N=N=N=N=H=8H8====H=H=H=H=N=N=d^H'H'H'C8C8C8C8=!=!H=H=H=H=H=H=^NC8=2=2=2H==H=H'C8=!C8C8H=N=H=N#88!=88888%%SS8!JJ8S!228d!!88d==d2d!23vmYo`qvt"m^.8O]]x.88Aa.8..]]]]]]]]]]88aaafxxxxofx.]xfxoxofxooof8.8a].]f]f]8ff..].ffffA]8f]]]SA/Aa8.888888888888f.x]x]x]x]x]x]o]o]o]o]........xfffffxfxfxfxfo]x]ffo]xfofx]x]x]x]x]x]x]xfo]o]o]o]ffffffxf........]x]f.f.f.f.xfxfxfxfffxAxAxAo]o]o]o]f8f8xfxfxfxfxfxfo]fSfSfSxff.xfxAo]f8o]o]xffxfN:.]]8f]]]]]>=]8{{]8(SS]88]]ff.S8"m^$88YC!!':!8888888888:::8fCCHHC=NH2C8SHNCNHC=HC^CC=/888288882S8888!282H222!!:!!!d!!!!!!!!!!=C8C8C8C8C8dYH2C8C8C8C8H8N8N8N8N8H8H8H8H8C2C8N=N8C2H8C8C8C8C8H2H2H2H2H8C8C8C8C8N8N8N8N8N8N8H82C28888H8H8H8H8N8N8d^H!H!H!C2C2C2C2==H8H8H8H8H8H8^HC2=2=2=2H88H8H!C2=C2C2H8N8H8N#68!=88888%%SS8!JJ8S!!!8d!!88d22d!d!"m^ !%??\^++7C%??????????CCC7RSVV\SO\`1GZOl\\S\ZIO\SlOOI+C+C7CG=I?+CK)'I'jKCIG97/KCbCC;+C+C%%%n%%%%%%%%%%C)SCSCSCSCSCl`V=S?S?S?S?1)1)1)1)\K\C\C\C\C\K\K\K\KOCSC\C\COC\CSISCSCSCV=V=V=V=\IS?S?S?S?\C\C\C\C\C\C`K1)1)1)1)GZIO'O'O'O'\K\K\K\K\C\CndZ9Z9Z9I7I7I7I7O/O/\K\K\K\K\K\KlbOCI;I;I;\IO'\KZ9I7O/OCOC\I\C\KNCR7!777???((__?&RR?_&557n%%77nKKn5n&CC<YJA2p}wC d*w=5& wX pTC&LAYJBo2PkC P&UC% 4C P['CJP)o=3Lioe xzC&X&%e8.4e P['CP &M=!= P['C,Pd'l80&lX pTC%e8.LTee xzCXLd!M,%Bd|M2PkCP L<.Bd.2PkCPL~)`8.B`2PkCP'd8.Xd2p}wCLL:!BT:2PkC,PM<!1F<2p}wC,L q%V2*B֬V2PkCP s#Z2*ZZ2p}wCd TH% &4HX pTCJLD4B42PkCP E6D62p}wC \H(!oVH2p}wChfP,%j|P2p}wC"m^%,?JJw`%,,4N%,%%JJJJJJJJJJ,,NNNQ````YQh`%J`Qo`hYh`YQ`Y~YYQ,%,NJ%JQJQJ,QQ%%J%wQQQQ4J,QJhJJC4%4N,%,,,,,,,,,,,,Q%`J`J`J`J`Jw`JYJYJYJYJ%%%%%%%%`QhQhQhQhQ`Q`Q`Q`QYJ`JhQhQYJ`QYQ`J`J`J`J`J`J`J`QYJYJYJYJhQhQhQhQhQhQ`Q%%%%%%%%J`JQ%Q%Q%Q%`Q`Q`Q`QhQhQ~`4`4`4YJYJYJYJQ,Q,`Q`Q`Q`Q`Q`Q~hYJQCQCQC`QQ%`Q`4YJQ,YJYJ`QhQ`QN/%JJ,QJJJJJ11ooJ,bbJo, CCJ,,JJQQ%C,&)o=3Uo P['C&P:SF92p}wC/xC8!:x2p}wCX>YJA2p}wC d*w=5& wX pTC&LAYJBo2PkC P&UC% 4C P['CJP)o=3Lioe xzC&X&%e8.4e P['CP &M=!= P['C,Pd'l80&lX pTC%e8.LTee xzCXLd!M,%Bd|M2PkCP L<.Bd.2PkCPL~)`8.B`2PkCP'd8.Xd2p}wCLL:!BT:2PkC,PM<!1F<2p}wC,Lq%V2*B֬V2PkCPs#Z2*ZZ2p}wCdTH% &4HX pTCJLD4B42PkCPE6D62p}wC\H(!oVH2p}wChfP,%j|P2p}wC6NAq2p}wC6 -5 -6 -5 -5 -7 -5 -6 -4 -7 -3 -8 -3 -7 -1 -9 0 -8 20 D -5 11 -5 10 -4 11 -2 13 1 -1 12 0 12 0 12 1 12 3 11 3 12 4 11 4 11 6 11 6 10 7 10 8 9 8 9 9 8 9 7 20 D 6 2H}'Report#&U PyQU&P#X` hp x (#%'0*,.8135@8: 1st tier (A)#(#  J #o P['CU&P#Given the number of sources in each cell and information about unit costs and emission control efficiencies due to each of the selected control actions, emissions and costs corresponding to each of the cells can be calculated. Expected visibility can then be calculated from projected emissions. The preceding section provided an overview of the key data inputs to each of these computations of the IAS, and information about the data sources. This section details the actual IAS equations that are applied when an IAS case is "run."  PJ ̸2nd tier (1)#:x2p}wC!X#1. Emissions Equations#l2nd tier (1)#(#  J #o P['CU&P#In the IAS, the relationship between emissions, selected control actions, and sectoral demand is modeled using an emission equation summarized in this section. For each IAS source category, existing and new capacities are represented by separate cells to capture control option and emissions rate differences for existing and new sources. Each existing cell will have one associated "new" cell, the capacity of which will be estimated to grow in sufficient quantity to supply all of the associated sectoral demand that cannot be met by the capacity of sources that were in place in 1990 and which have not yet been retired as of the time period in question. Sectoral demand is met by existing (i.e., in place as of 1990) capacities first, then if the existing sources (after accounting for cumulative retirements) cannot meet all of the sectoral demand, new source capacity is added. Unless otherwise specified by the user, new demand for a cell is met by new capacity in the same IAS Source Region as the existing capacity. The following emission equation is used to estimate emissions from that part of demand that can be met from sources that were existing at the start of the simulation period (i.e., 1990). For each cell, c, and each pollutant, p, within that cell, emissions estimated for each year, t, are:),**Ԍ J ԙExisting Source Emissionc.p.t. =  J `1990 Emissionc.p/(fraction emissions remaining given 1990 existing controlsc.p) * (#  J `(fraction emission remaining with assumed control optionc.p.t) * (#  J` `intensity indexc.t, *(#  J8 `MIN {growth index c.t,, or(#  J (1,/1990 utilization fractionc)*(1cumulative fraction retiredc.t)*(capacity factorc.t) where:  J `1990 Emission is the 1990 emission for that cell and that pollutant in the 1990 emission inventory. This value is stored in the CBE0 table of the IAS.(#  J `fraction emission remaining given existing controls indicates the degree to which emissions in 1990 have already been reduced by existing controls in place, compared with the case where the same cell might have had no controls in the base year. If no controls are in place in a given cell as of 1990, then the value of this parameter will be 1. If the controls in place in 1990 remove 40 percent of the emissions of a particular pollutant, then the value for that cell and that pollutant will be 0.6. These fractions are taken from the COE0 data for the control option assumed to be in place in 1990, as specified in the LBF0 table.(#  J `fraction emission remaining with assumed control option is the control efficiency  J (relative to no controls at all, not relative to 1990 control levels) that is associated with the specific control option assumed to be in place in each forecast year, t. The control options assumptions are stored in the "Selected Control Actions (SCA") results table that is an output of the CADM computations (Step I of the IAS scenario computations). The fraction emissions remaining for each control option is found in the COE0 file.(#  Jx `intensity index is a variable representing any change in emission or sectoral demand not predicted by the economic model, or technological change. For example, the index may incorporate shifts in energy use behavior that are not motivated by economic conditions, but rather the result of a voluntary program. These values appear in the SDI0 data file, but unless the user adjusts them, they are always 1.0.(#  J `growth index is the REMIbased sectoral activity level in any year relative to that activity measured in 1990. Thus, in 1990, this will be 1.0, and will be greater than 1.0 in years where economic activity levels are forecasted to be higher than in the initial year; and less than 1.0 in years where economic activity levels are forecasted to be lower than in the initial year. These assumptions are in the SDI0 data.(#  J" `1990 utilization fraction is an input reflecting capacity utilization conditions for technologies in that cell in the 1990 base year. (New capacity will not be built until demand and/or retirements grow enough to consume any underutilized existing capacity.) These values will be equal to 1.0, except for source sectors for which 1990 is deemed to be a year of substantial underutilization. These values are found in the SCL data.(# ',**Ԍ J `cumulative fraction retired is the fraction of the existing capacity in place in 1990 that continues to emit air pollutants at the base year (1990) emission rate. These inputs are found in the CRT data.(#  J` `capacity factor is a constraint to be applied to note whether certain types of capacityinplace in a cell are fully utilized. The net effect of reducing the capacity factor is to shift utilization from existing to new sources within the cell where this occurs. These inputs are found in the CRT data.(# Once the existing capacity is fully utilized, new sources are needed to meet expected activity levels. A new capacity cell is added when the growth index for the corresponding existing source cell is greater than:  J  5(1/1990 utilization fractionc)*(1cumulative fraction retiredc.t). Using the subscript c to denote the same cell definition as in the associated existing source case, the emissions from the associated new capacity cell will be:  J0 New Source Emissionc.p.t =  J `1990 Emissionc.p/(fraction emissions remaining given 1990 existing controlsc.p) *(#  J `MIN (fraction emissions remaining with assumed new source control optionc.p),(#  J ` (fraction emissions remaining with assumed existing source control optionc.p.t) * (#  Jh ` intensity indexc.t) * {growth indexc.t _(#  J@ `   '[(1/1990 utilization fractionc) *(1cumulative fraction retiredc.t)*(#  J `   '(capacity factorc.t)] }(# Note that the MIN operator in the new source emissions equation ensures that the new sources will always be at least as well controlled as its existing capacity counterpart. In the baseline, new sources are always at least as highly controlled as existing sources. However, as more and more stringent control options are selected, new source requirements may become less stringent than what is being assumed for the existing source. When this happens, the IAS also applies that level of control to the new source assumption when estimating emissions, effectively ensuring that new sources are always  J controlled first. This operation is performed on a pollutant by pollutant basis, so that if the new source is still more controlled with respect to a different pollutant, that higher  J` level of control is not undone. Overrides of the new source control assumptions are not reflected in a scenario's cost estimate. The total emissions from a given IAS source category will be the sum of new and existing source emissions for that source category, based on the two equations above.  Pp# ̸2nd tier (1)#:x2p}wC!X#2. Control Costs#2nd tier (1)#(#  JB% #o P['CU&P#The IAS displays three separate projected cost results for each cell in each year. They are the levelized capital cost, the operating cost and the total levelized cost associated with the control(s) in place in a given year. Note that all of these costs, which are  J' viewable through the Projected Cost screen in the IAS, represent the levelized annual costs that are associated with each discrete forecast year. This section explains how measures of cost are calculated for each source cell within the IAS.z),**Ԍ J ԙ(1) The capital cost for a cell measures the cost of installing any control which is  J installed after 1990.;X F@ GFootnote (08@ (08@HPX!"#e P['C4P#ۍBase year $ = 1990. (Note that the fact that REMI forecasts are in $1987 is not problematic to the IAS because all REMI results used in IAS calculations have first been converted to an index of economic activity relative to 1990, and hence are unitless.); Capital costs are levelized over the years in which the control will be effective (and accounting for the levelization factor assumed). This levelized cost is the value reported by the IAS as the capital cost incurred by a  J` cell. The factor used to levelize the capital costs may be changed by the user.^` F GFootnote (08@ (08@HPX!"#e P['C4P#ۍThe levelization factor used to assess capital costs on an annual basis over the useful life of a given control option can be set by the user via the Financial Assumptions Button in the Control Scenario Screen in the IAS user interface. By default, this discount rate is currently set to 19 percent, which is intended to reflect a 7 percent real discount rate plus account for added variable  F capital costs (e.g., interest, insurance, taxes) that would increase direct capital expenditures by almost 100 percent.^(#  J (2) The operating cost measures the additional ongoing expenses incurred because of the control action. This includes the costs of maintaining and servicing a control technology, in addition to any change in operating efficiency which may arise because of the control action. Operating costs may be negative if there are substantial efficiencies or fuel cost savings.(#  J (3) The combined cost within the IAS is the sum of the levelized capital cost and the operating cost. It is an annual value for the modeled year in question, not a total over the entire decade represented by the modeled year.(#  K  Levelized Capital Costs. To update the capital cost for the modeled year in which a control is put in place, the IAS adds the "incremental" levelized capital cost of the control action to any levelized capital costs which are already in place in the cell. (The "incremental" capital cost that is applied to the costs reported by the IAS is calculated by subtracting the capital cost of the control action(s) in place in the preceding modeled year from the capital costs of the new (combined) control action being added as a retrofit in the current modeled year). The incremental capital cost of a retrofit action is then adjusted by the number of sources in the cell and the percentage of sources that have retired: `Adjusted Capital Cost =(# ` Capital Cost * Number of sources in the cell *(1 % retired at date of installation)(# Using this adjusted value, the cost is then levelized across the remaining useful life of  JQ the production source that is being retrofitQ`  FQ$ GFootnote (08@ (08@HPX!"#e P['C4P#ۍThe average remaining useful life of the facility is found in the master source cell list (SCL) data that was provided by ANL., using the levelization factor specified for the case. `Levelized Capital Cost = Adjusted Capital Cost * (L / 1+L) / (1 1 / (1+L)")(#  ,**Ԍ`where: `L = levelization factor assumed(# `n = remaining useful life of the facility being controlled(# ` The resulting levelized capital cost is applied to each forecast year from the first year the retrofit option is put in place, until the end of the useful life assumed in the levelization formula. If there are already levelized capital costs in that cell for that year (e.g., due to a control action taken in a previous period, but still being amortized in this  J forecast period), these incremental costs are added to them.  Kp  Annual Operating Costs. The annual operating cost associated with the control action selected for a cell for each forecast year is directly calculated from the unit cost input table for the control action in question, since there is no need to account for a legacy of costs of preceding years' decisions, as is the case with the capital costs. Also, operating costs are expressed in annual cost terms. However, the annual cost is adjusted based on the number of sources in the cell, the percentage of sources that have retired, and the capacity factor specific to that cell: `Levelized Operating Cost = Operating Cost * Number of sources in the cell *(# ` (1 % retired in that year) * capacity factor(# The resulting operating cost is applied to each forecast year that a control option remains in place.  KA  Numerical Example. The cost reported by the IAS considers the cost of maintaining any control which is in place in a given year, as well as the cost of installing any control  J technology not in place as of 1990. For example, if the IAS baseline has a control option  J A, which is then retrofit in an EMS with additional control option to achieve a control A  J and B option in 2010, the costs are calculated as follows: The baseline specifies control A in each year for a particular cell:  J Year  ' .Baseline Control  J 1990  ' .control A  J 2000  ' .control A  J 2010  ' .control A  Jb 2020  ' .control A  J: 2030  ' .control A  J 2040  ' .control A And the IAS retrofits control A with control B in the year 2010:  Jr# Year  ' .EMS Control  JJ$ 1990  ' .control A  J"% 2000  ' .control A  J% 2010  ' .control A and B  J& 2020  ' .control A and B  J' 2030  ' .control A and B  J( 2040  ' .control A and B Z),**ԌThe cost data for that cell (in the COC0 file) might indicate, for example, that control A has a onetime capital cost of $1,000 per capacity unit and an annual operating cost of $100 per capacity unit, while control (A and B) has a onetime capital cost of $1,500 per capacity unit and an annual operating cost of $130 per capacity unit. If there are 10 units of unretired capacity (e.g., 100 "megawatts [MW"]) in the cell in question, then the capital costs for applying the controls in this cell would be $100,000 and $150,000. The only costs assigned to the baseline scenario are the annual operating cost of $10,000. No capital cost is assigned to the baseline in this example, because control A has been in place as of 1990. (Cells that have controls installed after 1990 do reflect the capital costs of those controls, so there are some capital costs associated with the baseline, which need to be netted out of any EMS costs after the runs are completed.) Thus, the baseline costs for this cell would be: Y ddx !dd ""dDY  ""Year "=R Baseline Control "`Capital Cost "Operating Cost ""1990 "Acontrol A g$0 t/$10,000 ""2000|"Acontrol A|h0|t10,000 ""2010`"Acontrol A`th0`t10,000|""2020D"Acontrol ADth0Dt10,000`""2030("Acontrol A(th0(t10,000D""2040 "Acontrol A th0 t10,000( For the EMS costs, the annual operating cost shifts from $10,000 in the baseline to $13,000 starting in 2010. (Costs may decline over time with retirements of the cell's capacity.) However, the computation of the capital costs for the control scenario is more complex. When the IAS shifts to Control (A and B), it keeps track of what control level it begins with (i.e., from Control A), and nets that part of the control costs out of the capital  J cost for the combined option. The incremental capital cost in the year where a control  J action is changed (i.e., that which is actually incurred in 2010) is the difference between the capital cost for Control A and B ($150,000) and the capital cost for Control A ($100,000), or $50,000. The assumption is that the $100,000 capital cost for the Control A portion of the Control (A and B) option was already paid for when the control was first installed prior to 1990. This incremental capital cost of $50,000 is then levelized according to the formula above, perhaps resulting in an annual cost of $8,000. The remaining useful life of the existing facilities being retrofit may be 20 years, in which case, the IAS will reflect this $8,000 in 2010 and again in 2020. (Note that the total capital cost is not simply the sum of the costs shown in the two forecast years, but also includes another 18 years in between the forecast years, where the $8,000 cost would also be incurred. The costs presented for a forecast year in the IAS are an annual cost, just presenting costs at a single point in time.) Thus, the EMS's costs for this cell would be: ^ !dd ""dD Add L&""TD^ (""Year0'"@ EMS Control0'"`Capital Cost0'"Operating CostL&""1990("Acontrol A(i$0(t/$10,0000'""2000("Acontrol A(j\0(t10,000(""2010)">y control A and B)tf8,000)t13,000(""2020*">y control A and B*tf8,000*t13,000),**)""2030">y control A and Btj\0t13,000""2040">y control A and Btj\0t13,000  P  ٠1st tier (A)#:x2p}wC!X#C.`SCENARIO ANALYSIS AND THE CADM#i5 1st tier (A)#(#  Jr #o P['CU&P#The IAS is designed around the concept of examining five types of EMSs, and the control actions that would result from each. Two of the scenarios represent bounding scenarios. The "Baseline Forecast Scenario (BFS") represents the control actions that would be undertaken under current regulations for each of the cells in each of the years considered by the IAS, and the "Maximum Management Alternative (MMA") scenario represents the highest emission reductions achievable in each cell by retrofit control options available in the IAS data set. For other EMSs, which are intermediate between the BFS and MMA, the IAS starts with the Baseline control actions in place, then adds  J2 additional controls to each controllable cell as specified by the Control Scenario. All regions except for Canada, Mexico, Washington, Montana, and Texas contain controllable cells. Between the two bounding scenarios lie three possible types of EMSs: a Technology Specification Scenario, an Emission Cap Scenario, and a Visibility Standard Scenario. The IAS is designed to permit the user to select control actions using these three different strategies, or a combination of the strategies.  J (1) The user may specify control levels beyond the assumed minimum control level of the baseline for any of the controllable cells in any of the post2000 forecast years. If a cap is not also specified, then the IAS will treat this as the full control scenario specification, and will produce an emissions, visibility and cost projection consistent with controls specified by the user.(#  J (2) The IAS can select those control actions to meet an emission cap at the lowest  J cost. A Regional Cap definition includes the Market Region that must meet the cap, the pollutant measure to be capped, the allowable level of emission of that pollutant measure, and the year in which the cap is to be achieved. Using the leastcost algorithm of the CADM, the IAS simulates the idealized outcome of a marketpermit system under a regional cap. Although the terminology implies that emissions permit trading would be involved, this is not the only way that the results could be interpreted. They could also be thought of as an indication of what the leastcost method would be for specifying controls to achieve emissions levels consistent with those of a cap.(#  J"! (3) The IAS's CADM can identify and select those control actions that would meet a VAQ standard at a given receptor at the lowest cost. Standards for a visibility goal are expressed in units of annual average light extinction. It is possible for the user to identify specific IAS Source Regions that would be required to take control actions as well as those that would be exempted from action.(# By applying combinations of the Technology Specification, Regional Cap, and VAQ Goal, one can create many different control scenarios that are intermediate between the Baseline and the Maximum Management scenarios using the IAS. Each scenario can be saved as a new matrix of control actions independent of the EMS from which the control(,** actions were generated. That new scenario can then be used as a starting point for further modifications. The IAS computations used in each of the three types of scenario specification are documented in more detail in the rest of this section. However, it is worth noting here that the IAS has a range of other parameters that can be used to set up more specific forms of scenarios. For example, clean air corridors, exemptions of particular source groups from cap requirements, etc., are user options.  P ̸2nd tier (1)#:x2p}wC!X#1. Specifying Minimum Control Actions#uE2nd tier (1)#(#  Jj #o P['CU&P#Since the output of the CADM is a list of specific control actions, a list of minimum control actions entered as input to the CADM (e.g., in commandandcontrol format) will result in the same list being returned as output, unless other run options such as caps are simultaneously specified. When defining control technologies in place, their efficiencies, and their costs must be cellspecific, meaning that within a source sector, these assumptions can be different for different facility sizes or regions. This accommodates the need to allow for regionspecific baseline assumptions, and allows for different treatments of regions in EMSs. This, in turn, implies that retrofit options need to be considered on a cell by cell basis. To ensure data integrity in the specification of a commandandcontrol strategy, the IAS user  J interface allows the user to select only from those retrofit options that are available given  J what is already in place.  P: ̸2nd tier (1)#:x2p}wC!X#2. Regional Caps#J2nd tier (1)#(#  J  #o P['CU&P#The result of a regional cap simulation is a list of control actions that meet the cap with least cost. The first step in defining such a cap is to determine the set of IAS Source Regions that would be included in the cap calculations. This is termed a "Market Region," since the cap logic produces the leastcost method of achieving a cap, which simulates a system of tradable permits. To generate the list of selected control actions, the emissions in the Market Region to which the cap will be applied are computed given the starting point of the minimum specified control actions that will be assumed to be in place. (If there is no direct user specification, the starting point for controls is the set of baseline control actions.) If a region's emission level exceeds the cap, the most costeffective control option available within the Market Region is added. ("Effectiveness" is measured with respect to the pollutant measure being capped.) This process of calculating expected emissions, then comparing it to the cap, continues until the cap is met {or the most stringent control actions are in place in all the cells of that Market Region. Control  J" actions that can be selected are only those retrofits that will reduce emissions further than those associated with the minimum control actions specified. (The COX0 table assures this restriction.) There is no change in the rate of retirements and introduction of new sources, nor reassignment of new sources to alternative source regions assumed in achieving a regional cap. However, it is possible for the user to exclude specific source sectors from the cap logic, and to limit the locations for new source capacity. (,**ԌIf there are multiple caps to be processed (e.g., several independent Market Regions with their own caps, or caps applied to multiple years for the same Market Region), these caps are processed individually and sequentially. The first caps simulated are those in 2010, then those in 2020. It does not matter what order the different Market Regions are processed in within a forecast year, since their control decisions are independent. However, the IAS processes the Market Regions in a given forecast year in the order in which they are specified on the PC screen by the user. The specific steps of the Regional Cap processing by the CADM for each cap are listed below: `For each time period (working from 2010 to 2040), and for each trading region that has an emission cap defined in that year:(#  J (1) Generate a list of IAS source regions that are associated with the defined regional cap Market Region.(#  J (2) Calculate the total emissions of the relevant pollutant measure in the Market Region that would be generated in that year by all source categories in the relevant source regions, given the "current" control actions. (The current control actions are the minimum control actions specified for that year, or superseded by the control action chosen for the preceding time period, if processing for a year later than 2010.)(#  J (3) If the total emissions in step 2 are greater than the specified emission cap for the Market Region, generate a list of retrofit control actions available for all the cells in the Market Region, given their list of the current control levels (i.e., by reference to the COX0 data).(#  J (4) Compute a ratio for each control action identified in step 3 by dividing the emission reduction due to the control action by the incremental cost of the control action. Sort the list of control actions by emissionovercost ratios.(#  J( (5) As long as the regional cap is less than the projected emissions for the Market Region and the list of available control actions is not empty, perform the following steps:(#  J  (i)h  'Add the next control technology from the list.(#  J`  (ii)h  'Update the expected emissions for the region.(#  J8  (iii)h  'Determine if the cap has been met.(#  J (6) When the regional cap is met, the control actions selected for this time period are defined by the list of control actions generated in step 5 to reach the cap. If the list of available control actions has been exhausted but the regional emission cap is still less than the projected emissions, the list contains the maximum available control actions for the specified cells.(#  J% (7) Perform steps 1 to 6 for the rest of the Market Regions defined in this forecast year.(# ',**Ԍ J (8) Move to the next forecast year and start with the first Market Region listed for that year. Use as the minimum control actions, the results for that Market Region from the previous year's cap simulation.(#  P` ̸2nd tier (1)#:x2p}wC!X#3. VAQ Goals#_2nd tier (1)#(#  J2 #o P['CU&P#The CADM for attaining VAQ goals follows a similar set of steps to those described for regional caps except that a VAQ goal can only be set for a single receptor in each IAS case. (This is because the actions taken for one receptor are interdependent with the actions that might be taken at a different receptor. Thus, VAQ caps for multiple receptors are not possible in the way that multiple Market Regions are.) To calculate a list of control actions needed to meet a VAQ goal in a specific receptor location, the CADM first calculates the difference between visibility and the visibility goal for the receptor. If estimated visibility does not meet the goal, the most costeffective control action available will be added. The expected light extinction is then recalculated, and further control actions added in leastcost order, as necessary. This continues until the VAQ goal has been met, or the most stringent controls have been applied. For a receptor with a VAQ goal, the CADM does the following:  J (1) Calculate the expected light extinction at the receptor, given the control actions in place.(#  Jb (2) Generate a list of available control actions sorted according to the ratio of light extinction improvement at that receptor to the change in cost.(#  J (3) Add the next control technology in the order to the list of selected control actions.(#  J (4) Calculate the projected visibility for the receptor with the new control action in place.(#  J" (5) Return to step 1, unless the visibility goal has been met, or all control actions have been applied.(#  P ̸2nd tier (1)#:x2p}wC!X#4. General Summary of CADM#g2nd tier (1)#(#  J| #o P['CU&P#Thus, the IAS CADM will preferentially assign the cheapest technologies (in the sense of most costeffective) to achieve a cap or goal. The general approach is that the CADM rankorders all of the options feasible as retrofits to the starting set of control options (e.g., from the baseline controls, or alternative technology specification) according to each option's ratio of incremental levelized cost of a control option to incremental improvement towards the cap or goal. It always selects the most cost effective options cost first, until the objective is met. If an alternative control option for a cell is reached where a more costeffective option has already been selected via this process, it determines whether the less costeffective option would actually reduce emissions (or light extinction) further than the option already selected. If it would, then the less costeffective option is selected, and the more costeffective option is deleted. (,**ԌAs long as the specified regional cap or VAQ objective is not met and options exist which provide greater total emissions reductions, the CADM will continue choosi