2011 Fehr Peers Downtown Streetcar Ridership Study

8/3/2019 2011 Fehr Peers Downtown Streetcar Ridership Study

1/17

Fehr&Peers Page|1 October4,2011

TechnicalMemorandum

Date: October4,2011

To: JimHecht,HDR

From: Fehr&Peers

Subject: PreliminaryEstimatesofRidershipfortheDowntownLosAngelesStreetcar

1.0 DescriptionofMethodology

Traditionalmethods of forecasting transit ridership often employ regional travel demandmodels to

predict ridership. Suchmodels are relatively unresponsive to changes in station-level land use and

transitservicecharacteristics.InthecaseofLosAngeles,thelargesizesofthetrafficanalysiszonesin

theMetrotraveldemandmodelprecludedetailedlandusedatacollectionanddifferentiationatthe

station-level.Furthermore,theMetrotraveldemandmodeldoesnothaveastreetcarcalibratedmode,

meaningitsusewouldrequirecalibrationandvalidationofanewstreetcarmodeoftravelinthemodel.

Alternativelythestreetcarwouldhavetobeclassifiedusingamodethatcurrentlyexistsin themodel

suchasbusorlightrail,whichhavedifferentridershipcharacteristicsthanstreetcar.Therefore,forthe

purposesoftheAlternativesAnalysis(AA)Study,wehavechosennottousetheMetrotraveldemand

modelforforecastingDowntownLosAngelesStreetcar(LAStreetcar)ridershipandinsteaduseaDirect

RidershipModelcalibratedandvalidatedtostreetcarforridershipforecasting. DirectRidershipModels(DRMs)aredirectlyandquantitativelyresponsivetolanduseandtransitservice

characteristicswithintheimmediatevicinityandwithinthecatchmentareaoftransitstations.Theycan

predictridershipatindividualstationsbasedonlocalstationareaandsystemcharacteristics.DRMsare

basedonempiricalrelationshipsfoundthroughstatisticalanalysisofstationridershipandlocalstation

characteristics.

The effects of station-level variables are expected to be highly significant in accurately forecasting

streetcarridership.Whilestreetcarsystemsareusedfortraditionalcommutetrips,ourresearchwith

transitagenciessuggeststheymoreoftenprovideaccessandcirculation fordowntownorcitycenter

areas.Theyservetouristneedsandoftenduplicateexistingtransitserviceprovidedbybus.Thus,itwasexpected that individual station-area characteristics greatly affect boardings and overall ridership

projections.Recognizingthatvariablesaffectingstreetcarridershiparedifferentthanthoseforregional

rail systems, the basis for analysis draws from the characteristics of existing streetcar systems in

Portland, Seattle, and Tacoma. These systems were chosen because they are most similar to the

proposedLAStreetcar.


2/17


1.1DataCollection

RidershipdatawascollectedforthePortland,SeattleandTacomastreetcarsystemsatthesystemlevel,

andwhereavailable,attheindividualstationlevel.Variablescollectedatthesystemlevelincluderoute

length,openingyear,frequencyofservice,traincapacity,fare(includingpresenceoffreestations),and

transferpolicy. At the station level, datawere gathered for the area within a quarter-mile1 of the

station and included intersecting transit, retail andgeneral employment density, householddensity,

streetconnectivity,distancebetweenstations,numberofhotelsandnumberofspecialeventscenters.

Table1showsthecharacteristicsofthecomparativesystems.Table2showsthedatacollectedatthe

corridorlevel.Table3showsdatacollectedatthestationarealevel.

TABLE1ResearchedStreetcarSystems

System Route

Length

(both

directions)

Number

of

Stations

Fare Weekday

Spanof

Service

Peak

Weekday

Frequency

(trains/hour)

Daily

System

Boardings

Daily

Boardings

perMile

Portland

StreetcarSystem

8.0Miles 47 $2.10

(Freeatmajority

of

stops)

5:30AM

11:30PM

4.6 11,700 1,460

Seattle

SouthLake

UnionLine

2.6Miles 12 $2.50 6:00AM

9:00PM

4.0 2,300 880

Tacoma

Streetcar

System

2.7Miles 8 Free 5:20AM

10:10PM

5.0 2,900 1,070

TABLE2CorridorLevelDataCollectionVariableClass VariableList

Ridership AverageWeekdayBoardings,PeakMonth AverageWeekendBoardings,PeakMonth AverageWeekdayBoardings,Off-PeakMonth AverageWeekendBoardings,Off-PeakMonth

ServiceCharacteristics RouteLength(miles,countingbothdirections) OpeningYear AMPeakFrequency(minutes) PMPeakFrequency(minutes)

DailyAverageFrequency(minutes) TrainCapacity IntersectingFeederBusesDuringOperatingHours IntersectingFeederTrainsDuringOperatingHours

1TheStreetcarDRMtreatsallemploymentandhouseholdswithinamilewalkequallyanddoesnotestimatea

captureratewithinthewalkthatdecreasesbydistancefromthestop.


3/17


VariableClass VariableList

NumberofStopsonStreetcarLine RegularFare TransferAccepted PassesAccepted

Population&Employment(withinmileofcorridor)

RetailEmployment Non-RetailEmployment JobMix(Retail/Non-RetailRatio) EmploymentDensity Households HouseholdDensity Jobs/HousingRatio

TABLE3StationLevelDataCollection

Variable Description Source

Boardings Averagedailyboardings. TransitAgency

Alightings Averageweekdayalightings. TransitAgency

StartofLine Binaryvariableindicatingstationisthefirststopon

theline(0/1).

Fehr&Peers

Intersections Numberofintersectionswithinaquartermileofthe

station.Limitedaccesshighwaysandrampsnot

included.

Fehr&Peers(Calculated

inGIS)

StopstoTerminus Numberofstationsuntiltheterminusoftheline.(A

measureofhowmanydestinationsareaccessible)


inGIS)

Buses Numberofintersectingbuseswithinoneblockofthe

station.

TransitAgency

NumberofFeeder

Trains

Numberofdailytrainsonintersectingraillines

withinoneblockofthestation.

TransitAgency

RailAccess Binaryvariableindicatingthestationhasatransferto

arailline(0/1).

Fehr&Peers

Free/Paid Binaryvariableindicatingwhetherthestationisfare

freeorpaid(0/1).

TransitAgency

DistancetoNearest

Station

Distancetoclosestdirectionalstation.(Closer

spacedstationshaveasmallercatchmentareathan

furtherspacedstations)


inGIS)

RetailEmployment Numberofretailemployeeswithinmileradiusof

station.

U.SCensusBureau

LongitudinalEmployer

HouseholdDynamics

Non-Retail

Employment

Numberofallother(non-retail)employeeswithin

mileradiusofstation.

U.SCensusBureau


HouseholdDynamics

RetailMix Ratioofretailemployeestonon-retailemployees

(RetailEmployment/Non-RetailEmployment).

U.SCensusBureau


HouseholdDynamics

TotalEmployment Totalnumberofemployeeswithinmileradiusof

station.

U.SCensusBureau


HouseholdDynamics


4/17


Variable Description Source

Households Numberofhouseholdswithinmileradiusof

station.

2010Census

JobsHousingBalance Numberofjobsperhouseholdwithinmileradius

ofstation(TotalEmployment/Households).

2010CensusandU.S

CensusBureau


HouseholdDynamicsUrbanDensity SumofRetailEmploymentandHouseholdswithin

mileradiusofstation.

Fehr&Peers

Hotel Binaryvariableindicatingthatahotelislocatednear

thestation(0/1)

Fehr&Peers

Center Binaryvariableindicatingthataspecialeventscenter

islocatednearthestation

Fehr&Peers

CenterSize Avariablethatcapturesthemagnitudeofthespecial

eventscenter,basedonsquarefootage

Fehr&Peers

1.2DirectRidershipForecasting

ThestationleveldatacollectedfromPortland,SeattleandTacomawereusedtoperformordinaryleast

squares (OLS) regression analysis to predict daily boardings per station. This analysis is based onempirical relationships found through statistical analysis of station ridership and local station

characteristics.Multiple iterationsof all collected data were tested in theregressionmodel, but the

variablesthatenteredintothedirectridershipforecasting(DRF)modelassignificantwerethefollowing:

UrbanDensityameasureofretailintensityandresidentialdensityofthestationarea NumberofFeederTrainsameasureofthemagnitudeofregionaltransitconnections StartofLinethisvariableonlyappliestobi-directionallineswheremorepassengerstendto

boardatthefirststation,butdoesnotapplytoloopsystems

CenterSizeameasureofthemagnitudeofaspecialeventscenterservedbythestation Free/Paidabinaryvariableindicatingwhetherthestationisfreeorpaid(butnotproviding

informationabouttheleveloffare)

TheR2valueofthemodelis0.56whichrepresentsafairlyhighgoodnessoffit.Oneofthelimitstothe

model is the limited number ofbuilt streetcar systems in the USand thus limiteddata availability.

Although intuitivelymore variables thanthose included in themodel influence ridership,dueto the

limiteddataavailabilitywewereunabletodistinguishstatisticallysignificantrelationshipsbetweenall

variables.Thatbeingsaid,wewereabletofindsignificantrelationshipsbetweenboardingsandseveral

stationlevelvariablesinordertocreateastatisticallysignificantmodelwithagoodfit.

Another factor that could affect goodness of fit is the close spacing of stops along streetcar lines.

StreetcarlinessuchastheonesstudiedinPortland,SeattleandTacomatendtohavecloserstopspacingthanregionaltransitsystemssuchasbusesandlightrail.Duetotheclosestopspacing,thecatchment

areaofsomestopsmayoverlap.Thiscouldresultinvariationamongthepredictedvaluesatthestation

level,butshouldnothaveastrongimpactonthepredictedvaluesatthecorridorlevel.

DataforeachofthesefivevariableswerecollectedforeachpotentialstopalongtheLAStreetcarline.

These variables were used to predict daily boardings at each station and were summed for each


5/17


configuration to estimate daily boardings along the line. The expected system boardings are

summarizedinSection3.0.

1.2.1UrbanDensity

Urbandensityisasumofretailemploymentandhouseholdswithinmileofthestation.Moststations

alongtheproposedLAStreetcarcorridorhaveeitherhighhouseholddensityorhighretailemployment

density,butfewhaveboth.Theareaswiththehighesthouseholddensityinclude6thandBroadway,and6

thandHill.Theareaswiththehighestretailemploymentdensityarearound9

thandBroadway,11

thand

Figueroa, and 6th and Hill. Special attention was paid to areas along the corridor where retail

employmenthasbeenchangingoverthepastfiveyears.TheDowntownCenterBusinessImprovement

Districtwasconsultedtoinsurethataccurateretailemploymentcountswereusedasinputsintothe

model.Theseareasincluded:

7th+Figthatwillbereopenedin2012asFigat7 thwithTargetastheanchorstore MacysPlaza DowntownLosAngelesRalphs

1.2.2NumberofFeederTrains

Numberoffeedertrainsisthedailynumberoftrainsonintersectingraillineswithatransfertothestreetcarstop.Severalraillineswouldhavetransfers totheproposedLAStreetcarline.These include

theMetroBlueLine(withstopsatPicoand7th/MetroCenter),Red/PurpleLine(withstopsat7

th/Metro

Center,PerishingSquareandCivicCenter),GoldLine(whichstopsatUnionStation),MetroLink(which

stopsatUnionStation)andtheExpoLine(openingin2012withstopsatPicoand7th/MetroCenter).

Sinceatsomelocations severalLA Streetcarstops arelocated near oneMetro station,the expected

numberofriderstransferringfromaMetrolinetotheLAStreetcarweredistributedamongtheclosest

streetcarstations.

1.2.3StartofLine

AnalysisofthePortland,SeattleandTacomastreetcarsystemsfoundthatahighnumberofpassengers

wereboardingatthestartofthelinestation.ThisdoesnotapplytotheLAStreetcarsinceitwilloperateinaloopandthereforehasnoendofthelinestation.However,inthecasethatabi-directionallineis

proposed,thisvariablewillbeapplied.

1.2.4CenterSize

Thisvariabletakes intoaccountspecialeventscenterswhichareservedbythestreetcarlineandalso

considersthesizeofthecenteranditsimpactonridership.Thespecialeventscentersconsideredalong

theLAStreetcarcorridorincludethe following: LAConventionCenter,StaplesCenter,LA Live,MOCA,

DisneyTheater,CityHall,OlveraStreet,andLittleTokyo.Broadwaybetweenapproximately3rdand7

th

Streetswas givenconsiderationasa special generator due to its regionaldraw.However, since the

primarydrawisshopping,theUrbanDensityvariableeffectivelycapturestheridershipresultingfroma

shoppingricharea.Inthemodel,theCenterSizevariableisreservedforuniquetripgeneratorsthat

cannotbeexplainedbyemploymentor residentialpopulationalone.SimilartotheNumberofFeederTrainsvariable,sincesomecenterscanbeservedbymultiplestreetcarstops,theexpectednumberof

streetcarridersgeneratedbyeachcenterwasdistributedamongthecloseststreetcarstops.

1.2.5Free/Paid

Alongthestreetcar systemsstudied,whether ornota stationwas freeor paidwas foundtohavea

significantimpactonridership.FortheLAStreetcaralternativesdiscussedinSections2and3,allLA


6/17


Streetcarstopsareassumedtobepaid.However,forthesensitivityanalysisinSection4,allLAStreetcar

stopsareassumedtobefree.

1.3ForecastYear

The forecast year for the ridership estimates is 2015 (expected opening year). We did not assume

additional development (housing, retail) beyond the 2010 data due to the models use ofparcel by

parcelchangeswithinamilewalkshedwhengrowthforecastsarefarmoreaggregateinnature.The

MetroExpo Line(Phase I)is expected toopen in2012 and Feeder Trains atthePico and 7th/Metro

Center Stations were adjusted to reflect this upcoming service. All other regionally significant rail

projectsareexpectedafter2015andwerenotincludedinthemodel.

1.4PostModelProcessingConsiderations

Several post-model adjustments were considered to account for variables not included in the DRF

model.Thesearedescribedbelow.

1.4.1FrequencyofService

Wedidnothaveenoughdatatomakea statistically significant relationshipbetweenfrequency and

boardings,butwe haveidentifiedthis variableas an important factordistinguishingtheLA Streetcarfromtheotherstreetcarsystemsstudied.Amongthestationsstudied,theaveragepeakhourfrequency

is3.6trainsperhour.However,theLAStreetcarisplannedtohave8.6trainsperhourduringpeak

hours(onetrain every7 minutes). Althoughno research exists thatprovidesanelasticityvalue for

frequencyofservicespecifically forstreetcarsystems,according toTCRPSynthesis 66,NewYorkCity

reported anelasticity value of 0.2 for transitservicefrequency.Consideringthe increased frequency

betweenthesystemsstudiedandtheLAStreetcarsystem,applyingthiselasticityvaluewouldresultin

anexpected ridership increase of28%.This increase can be applied asa post-modelprocessto the

resultsoftheDRFmodeltoaccountforincreasedridershipduetothemorefrequentserviceoftheLA

Streetcarthansystemsstudied.

1.4.2SpanofService

Anothervariableconsideredforpostmodelprocessingwas spanofservicesincetheLA Streetcarwill

operated later than thethreestreetcarsystemsstudied.TheLA Streetcaris plannedtooperatefrom

6:00to12:00AMMondaythroughThursday.Thisisaspanof18hours.Theaverageweekdaytimespan

ofthestationsstudiedis17.3hours.ThistimespaniscomparabletotheLAStreetcar.Furthermore,

sincethepeakboardinghoursofstreetcarsystemsareduringtheday,extendingservicehoursintothe

night is not expected to have a significant impact on ridership. Late night boardings along the LA

Streetcar line are likely to take place at special events centers and these boardings are already

accountedforinthemodelthroughtheCentersvariable.Therefore,nopost-modelprocessingforspan

ofservicewasperformed.

1.4.3InteractionswithLADOTDASHService

Oneconsiderationwaswhethersomelevelofridershipaccountedforinthemodelwouldactuallytakeplace on the LADOT DASH,Metro Bus, orMetro Rail services and therefore require a post-model

adjustment.However, theLA Streetcarroute structure (clockwise loop anddestinations served), and

typeofservice(streetcarversusshuttleorHRT)suggeststhattheLAStreetcarwouldbecomplementary

toexistingandplanneddowntowntransitservice.WhiletheproposedLAStreetcaralignmentsduplicate

orparallelportionsofLADOTDASHRoutesB,D,andF,noneoftheseroutescouldbefullyreplacedby

theLAStreetcar.Further,opportunitydoesexist toreconfigureLADOTDASH toeliminateduplication

withLAStreetcarbymodifyingroutestructure.Therefore,postmodeladjustmentsdonotincludedirect


7/17


transfer of LADOTDASH ridership to the LA Streetcar since they are expected to complement one

another. Transit mode choice (LA Streetcar versus shuttle, Metro Bus, orMetro Rail) is effectively

capturedintheindependentvariablesusedintheDRFmodel.

1.4.4ConsiderationofCommuter(Express)Buses

ConsiderationwasgiventowhethertheLA Streetcarcouldprovidea first/lastmileconnectiontothe

commuterbusesthatprovideaccesstodowntownemploymentfromsuburbanlocations(suchasMetro

SilverLine,BigBlueBusRapid10,orLADOTCommuterExpress).SincetheDRFmodeldoesnotincludea

variabletoaccountforthistypeofservice,apostmodeladjustmentwasexplored.Stoplocationsforthe

commuter services were identified anddetermined to provideenough coverage (based on distance

betweenstops)withindowntownthattransferstoothertransitlineswouldnotbeneededtocomplete

firstmile/lastmiletripswithindowntownLosAngeles.Thisfindingallowsustoconcludethatnopost

modeladjustmentwasdeemednecessary.

1.4.5ConsiderationofaDowntownNFLStadium

AnNFLstadiumiscurrentlyplannedfordowntownLosAngelesandwouldbelocatedonthecurrentsite

oftheWestHalloftheLosAngelesConventionCenter.Itwouldseat68,000forfootballandcouldbe

completed by 2015. How the stadium and its related activity could affect streetcar ridership wasconsidered as a post model adjustment. However, we believe that the Centers variable effectively

captures the ridership attributed to the Staples Center/LA Live/Convention Center destination for a

typicalday.ThemodelassumeseventsatStaplesCenterandLALiveandConventionCenter.Itwouldbe

aspecialcircumstancetohaveeventsatallfourcentersonthesamedayandwouldrepresentatypical

ridership; therefore no furtheradjustmentwasnecessary forpredicting typicalstreetcar ridership.A

discussionofhowridershipcouldbeaffectedduringspecialeventdaysisincludedinSection6.0.

1.4.6ConsiderationofAngelsFlight

AnglesFlightisfunicularrailwayconnectingHillStreetandCaliforniaPlazainBunkerHill.Eachridecosts

$0.25 and the railway averages about 2,100 weekday boardings per day.While this service is well

utilizedby officeworkers traveling fromBunker Hill totheBroadwayareaof downtown, itis also a

tourist attraction and the proposedmidblock stop on Hill Street between 3rd Street and 4th Street(Alternatives3,4,5,6&7only)wouldprovideadirectconnectiontothefunicular.Nearbystopsat3

rd

Street&Grand Avenue (Alternatives1&2only), 3rdStreet&Broadway,and4

th Street&Broadway

wouldalsoprovideaccess.Itisestimatedthatapproximately30%ofaverageweekdaydailyboardings

areattributabletotourists.ConsideringalikelysynergybetweentheuseofAngelsFlightandstreetcar

amongtourists,a postmodeladjustmentequalto1/3rdof projecteddailytouristboardingsonAngels

Flight(walkandothermodesofaccesswouldaccountforthemajorityoftripstoandfromthefunicular)

wasdistributedproportionallyamongthefourproximatestops.

2.0 Alternatives

Fehr&PeersusedthealternativesandstopspreparedbyHDRtoestimatethetotaldailyboardingsat

eachstop,summedtoreflecttheboardingsforeachline.Theridershipforecastsinthissectionare

basedonthefollowingkeyoperatingcharacteristics.

HoursofOperation:6:00amto12:00midnight Headways:7minutepeakand10-15minuteoff-peakservice Fare:$1.50(Section4providesanassessmentofreducedfareandfare-freeservice)


8/17


Thesevenconceptalternativesincludethefollowingalignmentconfigurations:

Alternative1:A4-B3-C2 Alternative2:A4-B3-C3 Alternative3:A6-B1-C2 Alternative4:A6-B1-C3 Alternative5:A7-B1-C2 Alternative6:A7-B1-C3 Alternative7:A6-B1-C4

3.0 ResultsandDiscussion

The results of the DRF for expected daily ridership and performance (boardings per mile) of each

alternativearesummarizedinTable4:

TABLE4.DailyRidershipbyAlternativefromDRFModel

AlignmentConfiguration #ofStops

Length(Miles) SystemBoardings AverageperMile

Alternative1 25 3.83 5,430 1,420

Alternative2 21 3.32 4,380 1,320

Alternative3 30 4.29 5,910 1,380

Alternative4 26 3.78 4,870 1,290

Alternative5 34 5.16 6,720 1,300

Alternative6 30 4.65 5,680 1,220

Alternative7 25 3.79 5,000 1,320

Afterpost-modelprocessingwasappliedtoaccountforthehighfrequencyoftheLAStreetcarlineand

link to Angels Flight, the following expected daily ridership values were determined, summarized in

Table5:

TABLE5.DailyRidershipbyAlternativefromDRFModelwithPost-ModelProcessing

AlignmentConfiguration #ofStops

Length(Miles)

SystemBoardings AverageperMile

Alternative1 25 3.83 7,160 1,870

Alternative2 21 3.32 5,820 1,750

Alternative3 30 4.29 7,780 1,810Alternative4 26 3.78 6,440 1,700

Alternative5 34 5.16 8,810 1,710

Alternative6 30 4.65 7,480 1,610

Alternative7 25 3.79 6,610 1,740


9/17


Alternatives1&2areprojectedtogeneratethehighestboardingspermileofthesevenalternatives.

Thisisduetobetteraccesstomajortripgeneratorssuchasmajorhotels,MOCA,andthe7th/Metro

CenterStation.Shownbelowareprojecteddailyboardingsbystop.Thesealternativesareprojectedto

generate fewersystemboardingsthanAlternatives3,4&7whichismainlyaccountedforbythefact

that Alternatives 1&2havefewer stops.Thestopswith thehighest projected ridership arelocated

aroundtheStaplesCenter/LALive/ConventionCenter.

0100

200300400500600700800900

1000

2nd&Grand(6)

1st&Broadway(8)

2nd&Broadway(9)

3rd&Broadway(11)

4th&Broadway(13)

5th&Broadway(15)

6th&Broadway(16)

7th&Broadway(21)

8th&Broadway(26)

9th&Broadway(27)

Olympic&Broadway(37)

11th&Broadway(36)

Pico&Broadway(38)

Pico&Olive(45)

Pico&Hope(46)

Hope&12th(47)

Hope&11th(48)

11th&Figueroa(33)

Olympic&Figueroa(32)

Flower&9th(31)

Grand&9th(30)

8th&Grand(42)

7th&Grand(43)

6th&Grand(44)

3rd&Grand(20)

DailyBoardingsperStaonforAlternave1


10/17


Alternatives3,4&7generatemoreridershipthanAlternatives1&2withfeweraveragedailyboardings

per mile. They contain five more stops (4 stops for Alternative 7) than their respective C2/C3

counterpartfromAlternatives1&2.

0

100

200

300

400

500

600

700

800

900

1000


01002003004005006007008009001000

2nd&Grand(6)

1st&Broa

dway(8)

2nd&Broa

dway(9)

3rd&Broad

way(11)

4th&Broad

way(13)

5th&Broad

way(15)

6th&Broad

way(16)

7th&Broad

way(21)

8th&Broad

way(26)

9th&Broad

way(27)

Olympic&Broad

way(37)

11th&Broad

way(36)

Pico&Broad

way(38)

Pico&O

live(45)

Pico&H

ope(46)

Hope&12th(47)

Hope&11th(48)

11th&Figueroa(33)


Flower&

9th(31)

Grand&

9th(30)

9th&O

live(29)

9th&

Hill(28)

8th&

Hill(25)

7th&

Hill(22)

6th&

Hill(17)

5th&

Hill(14)

4th&

Hill(12)

2nd&

Hill(10)

1st

&Hill(7)



11/17


Alternatives5&6generatethehighestsystemridershipofthesevenalternatives.Whiletheygenerate

loweraveragedailyboardingspermilethantheotheralternatives,theyhavethehighestnumberof

stationsamongthesevenalternatives. Alternatives5& 6add stopsatCityHall,OlveraStreet/Union

StationandLittleTokyo(+2more)andincludemoretripattractors(OlveraStreet/LittleTokyo)andan

additionalrailconnection.

0100200300400500600700

800900

1000

2nd&Grand(6)

1st&Broadway(8)

2nd

&Broadway(9)

3rd

&Broadway(11)

4th

&Broadway(13)

5th

&Broadway(15)

6th

&Broadway(16)

7th

&Broadway(21)

8th

&Broadway(26)

9th

&Broadway(27)

Olympic

&Broadway(37)

Broad

way&11th

(35)

11th&Grand(34)

11th

&Figueroa(33)


F

lower&9th(31)

Grand&9th(30)

9th&Olive(29)

9th&Hill(28)

8th&Hill(25)

7th&Hill(22)

6th&Hill(17)

5th&Hill(14)

4th&Hill(12)

2nd&Hill(10)

1st&Hill(7)


0

100

200

300

400500

600

700

800900

1000

2nd&Grand(6)

1st&Broadway(8)

2nd&Broadway(9)

3rd&Broadway(11)

4th&Broadway(13)

5th&Broadway(15)

6th&Broadway(16)

7th&Broadway(21)

8th&Broadway(26)

9th&Broadway(27)

Olym

pic&Broadway(37)

Br

oadway&11th

(35)

11th&Grand(34)

11th&Figueroa(33)


Figueroa&9th(49)

Figueroa&8th(50)

Figueroa&7th(51)

Hope&7th(52)

7th&Hill(22)

6th&Hill(17)

5th&Hill(14)

4th&Hill(12)

2nd&Hill(10)

1st&Hill(7)



12/17


Amajorfactorinthedifferenceinridershipcanbeexplainedbythenumberofstops.Thehigherthe

number of stops (assuming the stop is located near factors that influence ridership) the higher the

ridership. Forexample, theprimaryreasonwhy theC2 alignmentshave higherboardings than C3 is

numberofstations:C2stopsfivetimesbeforereachingFigueroa/11thandC3justonce.C2alsoprovides

superioraccesstotheBlueLineandConventionCenter.

0

100

200

300

400

500

600700

800

900

1000

Olvera&

LosAngeles(1)

Temple&

LosAngeles(2)

1st&

LosAngeles(5)

1st

&Broadway(8)

2nd

&Broadway(9)

3rd&

Broadway(11)

4th&

Broadway(13)

5th&

Broadway(15)

6th&

Broadway(16)

7th&

Broadway(21)

8th&

Broadway(26)

9th&

Broadway(27)

Olympic&

Broadway(37)

11th&

Broadway(36)

Pico&

Broadway(38)

P

ico&Olive(45)

P

ico&Hope(46)

H

ope&12th(47)

H

ope&11th(48)

11th

&Figueroa(33)

Olympic

&Figueroa(32)

Flower&9th(31)

G

rand&9th(30)

9th&Olive(29)

9th&Hill(28)

8th&Hill(25)

7th&Hill(22)

6th&Hill(17)

5th&Hill(14)

4th&Hill(12)

2nd&Hill(10)

1st&Hill(7)

LA

Mall&Main(4)

City

Hall&Main(3)


0100200300400500

600700800900

1000

Olv

era&LosAngeles(1)

Tem

ple&LosAngeles(2)

1st&LosAngeles(5)

1st&Broadway(8)

2nd&Broadway(9)

3rd&Broadway(11)

4th&Broadway(13)

5th&Broadway(15)

6th&Broadway(16)

7th&Broadway(21)

8th&Broadway(26)

9th&Broadway(27)

Olym

pic&Broadway(37)

B

roadway&11th

(35)

11th&Grand(34)

11th&Figueroa(33)

Oly

mpic&Figueroa(32)

Flower&9th(31)

Grand&9th(30)

9th&Olive(29)

9th&Hill(28)

8th&Hill(25)

7th&Hill(22)

6th&Hill(17)

5th&Hill(14)

4th&Hill(12)

2nd&Hill(10)

1st&Hill(7)

LAMall&Main(4)

CityHall&Main(3)



13/17


Twofactorsinfluencingridershipareretailjobsandhouseholdswithinmilewalkofastop.Whilethe

11th/Grandstopwouldprovidebetteraccesstothenewcondotowersadjacenttotheintersectionthe

11th/Figueroaand11

th/Picoarewithinamileof11

th/Grandandcapturemostofthepopulation.

4.0 PerformanceComparison

The projected opening day performance of the LA Streetcar (in terms of boardings per mile) was

comparedtoexistingandplannedstreetcarsystems,LRT/BRTinLosAngeles,existingDASHservice,and

existingbusservicealongBroadwaywithinthecorridor(southboundonly)wheretheLAStreetcarwould

operate.

First,acomparison(Table6)wasmadetoactualopeningmonthridershipofthePortland,Tacoma,and

SeattlestreetcarlinesandprojectedopeningmonthridershipoftheplannedCharlotte,SaltLakeCity,

Tucson,andAtlantaStreetcar lines.TheLA Streetcaroutperformsbothactualandprojectedopening

monthridershipofthesesystems.

TABLE6.ComparisontoOpeningMonth/ProjectedRidershipofExisting/PlannedStreetcarSystems

*OpeningMonthActual

**ProjectedOpeningDay

AlignmentConfiguration Length(Miles)


Alternative3 4.29 7,780 1,810

Alternative4 3.78 6,440 1,700

Alternative7 3.79 6,610 1,740

Portland(StarterLine) 4.8 4,982 1,040*

Tacoma 2.7 2,170 800*

Seattle 2.6 1,316 510*

Charlotte(Planned) 2.8 1,500 540**

SaltLakeCity(Planned) 4 3,000 750**Tucson(Planned) 3.9 3,600 920**

Atlanta(Planned) 2.6 2,600 1,000**


14/17


Second,acomparison(Table7)wasmadetoactualopeningmonthridershipofthefourexistingLRTand

oneBRTlinesinLosAngeles.TheLAStreetcaroutperformsactualopeningmonthridershipinboardings

permileoftheselines.

TABLE7.ComparisontoOpeningMonthRidershipofMetroLRT/BRT

Source:Metro

Third,acomparison(Table8)wasmadetoexisting(2011todate)ridershipofthefivebestperforming

downtownDASHroutes.TheLAStreetcaroutperformsexistingridershipoftheseroutes.

TABLE8.ComparisontoExistingDASHRidership

Source:LADOT



Alternative3 4.29 7,780 1,810

Alternative4 3.78 6,440 1,700

Alternative7 3.79 6,610 1,740

OrangeLine 28 16,360 580

GoldLine 27.4 18,364 670

GoldLineEastsideExtension 12 7,156 600

GreenLine 40 15,800 400

BlueLine 44 30,800 700



Alternative3 4.29 7,780 1,810

Alternative4 3.78 6,440 1,700

Alternative7 3.79 6,610 1,740

DASHA 6.1 3,886 640DASHB 6.7 3,525 530

DASHD 7.5 4,081 540

DASHE 6.3 7,352 1,170

DASHF 7.2 3,306 460


15/17


Fourth,a comparison (Table9)wasmadeto existing (July2011)ridershipof thesix best performing

Metro bus routes on Broadway between Cesar Chavez Ave and Pico Boulevard. The LA Streetcar

outperformsexistingridershipoffourofsixoftheseroutes.Thehighperformanceofthetworoutes

thatoutperformtheLAStreetcarislikelyduetotheconcentrationofboardingsalongtheBroadway

corridorandcorrespondingdispersedalightingsalongtheconsiderablelengthofeachbusroute.

TABLE9.ComparisontoSouthboundBroadwayBusRidership(CesarChavezAvetoPicoBoulevardOnly)

Source:Metro

5.0 FareElasticity

A sensitivity test of a free versus reduced fare versus paid scenarios was completed to better

understandtheresults,differencesinalternatives,andhowtheycomparetoothersystemtypes.

BoththeTacomastreetcarsystemandsignificantportionsofthePortlandstreetcarsystemarefarefree.

ThevariableFree/Paid was found tohave a statistically significant influence on ridership in theDRF

model.TotesttheimpactofapplyingafarefreesystemtotheLAStreetcar,acomparisonofridershipestimatesof free versuspaid systemswasconductedusing theFree/Paidvariablein themodel.The

resultscan beseen inTable10. Operating theLA Streetcaras a completely fare-free systemwould

increaseexpectedridership49to57%overasystemwhereafareischarged.

WhiletheDRFmodelcandistinguishbetweenfreeandfullfaresystems,itdoesnotmeasurechangesin

the level offare. Rather thanchargingthefullMetro fareof$1.50per ride, one considerationis to

chargeareducedfareof$0.50perride.Awidelyacceptedvalueforfaretoridershipelasticityis-0.42

(although this value isnotstreetcar-specific). According to this elasticityvalue, a fare decrease from

$1.50 to$0.50perride would result ina 27% increasein ridership.Theridership estimates under a

$0.50farescenarioarealsosummarizedinTable10.

2http://onlinepubs.trb.org/onlinepubs/tcrp/tcrp_rpt_95c12.pdf



Alternative3 1.45 2,480 1,710

Alternative4 1.24 2,360 1,900

Alternative7 1.24 2,360 1,900

MetroLine30 1.37 2,008 1,470

MetroLine40 1.37 2,985 2,180

MetroLine45 1.81 4,020 2,220

MetroLine730 1.45 792 550

MetroLine740 1.45 1,204 830

MetroLine745 1.45 1,780 1,230


16/17


TABLE10.FareLevelSensitivityTesting

Paid

($1.50perRide)

ReducedFare

($0.50perRide)

Free

AlignmentConfigurationLength

(Miles)

System

Boardings

Avgper

Mile

System

Boardings

Avgper

Mile

System

Boardings

Avgper

Mile

Alternative1 3.83 7,160 1,870 9,090 2,370 10,670 2,790

Alternative2 3.32 5,820 1,750 7,390 2,230 8,770 2,640

Alternative3 4.29 7,780 1,810 9,880 2,300 12,000 2,800

Alternative4 3.78 6,440 1,700 8,180 2,160 10,100 2,670

Alternative5 5.16 8,810 1,710 11,190 2,170 13,600 2,630

Alternative6 4.65 7,480 1,610 9,500 2,040 11,700 2,520

Alternative7 3.79 6,610 1,740 8,390 2,210 10,120 2,670

6.0 OtherFactorswithPotentialtoPositivelyInfluenceRidershipofLAStreetcar

6.1ConsiderationofOfficeEmploymentDensity

Anotherconsideration in theridership forecastingprocess is that officeemploymentdensityismuchhigher in downtown Los Angeles than in the cities included in the model. For example, the office

employmentdensityalongthePortlandstreetcarlineis14,000officejobspersquaremilewhileforthe

area along theproposed LAstreetcar is40,000 office jobsper squaremile.3 Fortheexistingmodern

streetcar lines in the U.S., office employment is not a significant predictor of ridership. Retail

employment,specialgenerators,connectionstohighcapacity transitlines,andfarestructurearekey

drivers of streetcar ridership. The existing modern streetcar lines in the U.S. primarily serve retail,

tourist,andhome-basedothertrips,whilehome-basedworktrips(eithernon-linkedorlinked)donot

haveastrongimpactonridership.Ifcommutetripsbyofficeworkersprovetobeamoresignificant

market for theLAStreetcar, thehigherdensityof downtownLosAngelesemployment comparedto

othersystemscouldresultinanincreaseinridershipabovecurrentprojections.

6.2ConsiderationofSpecialEvents

The model results show the ridership estimates for a typical day. However, special events could

generateperiodicpeaksinridership.Forexample,iftheNFLstadiumproposedtobebuiltnexttoLA

Liveweretobefilledtocapacity,thiswouldmean68,000peopletravelingtoandfromtheareawithina

rather narrow window of time. The Mobility Group estimated that 15 percent of fans will attend

weekendgamesbytransit.4SincestationsfortheMetroBlue,RedandPurplelinesarelocatednearby,it

isexpectedthatthemajorityoftransitriderswouldusetheselinesforahome-basedtrip.Weestimate

that12percentoffanswoulddesiretousethestreetcartotravelfromthestadiumonagameday.

Thiswouldrepresent680 1,360 additionalboardingsona gameday,which isan increasein daily

ridershipof1020percent.Thesefansleavingthestadiumafteragamewouldprimarilybeboardingat

the11thStreet&FigueroaStreetstoportheOlympicBoulevard&FigueroaStreetstop,whichalready

havesomeofthehighestlevelsofexpectedridershipofanystop.Ifeachstreetcarhasacapacityof100

3 Source:LEHD,http://lehdmap.did.census.gov/.Officejobs aredefined as jobsin thefollowingNAICS industry

sectors: information; finance and insurance; real estate and rental and leasing; professional, scientific, and

technicalservices;managementofcompaniesandenterprises;publicadministration.

4http://www.dailybreeze.com/latestnews/ci_18980892


17/17


riders, it would take 7-14 vehicles to accommodate all of the streetcar riders. Typical peak hour

frequencywillbeonestreetcarevery7minutes.Iftwosparevehiclesareaddedtoservespecialevents,

as is indicated in the Alternatives Analysis, the frequency can be increased to one vehicle every 5

minutesand40seconds.Thiswouldmeanitwouldtake4080minutestoserveallofthefanswanting

toboardthestreetcarafteragame.Itisunlikelythatfanswillbewillingtowaitmorethan30minutes

forastreetcar,sotheactualincreaseinridershiponagamedaywouldlikelybeonthelowerendofthe

estimatedrangegiventhecapacitylimitsofthestreetcarline.Otherrecurringspecialeventsincludethe

DowntownArtWalkandstreetfairs.

Documents

2011 Fehr Peers Downtown Streetcar Ridership Study