An I-band calibration of the SBF method at blue colours

6S.Mieskeetal.:SBFcalibrationatblue

colours

Fig.2.TwoimageexcerptsfromthesamechipandexposureclosetotwoinvestigateddEs,illustratingthevariabledegreeoffringing.Intheleftthumbnail,FCC194isatthebottomright.Intherightthumbnail,FCC196isatthebottom.Bothimageshavethesameintensitycuts.Notethefringingpatternsintherightthumbnail,whicharenotpresentintheleft

one.

Fig.3.ColourmagnitudediagramofthegalaxiesenteringourSBFanalysis,i.e.thoseoneswithanSBFS/N>3,seealsoFig.5.SmallbluecirclesindicatecoloursderivedfromthepresentIMACSdataset.SmallredasterisksindicatecoloursfromtheWFCCDdatasetofHilkeretal.(2003).Largeencir-cledasterisksisthemeanofbothvalues.TheVmagnitudeisfromtheWFCCDdatasetforallgalaxies.

TheerrorofP0wasestimatedintwoways:

1.basedontheMonteCarlosimulationspresentedinMieskeetal.(2003a).Forthemedianseeingofourdata(0.75′′),inter-polationofthesimulationresultsfor0.5and1.0′′forFornaxdEsand1hourintegrationyieldsthefollowinguncertainties:0.37magintherange 11>MV> 13mag,0.27magintherange 13>MV> 14.5mag,and0.20magintherangeMV< 14.5mag.Notethatthesimulationerrorsarecalcu-latedforimagesofthesamepixelscalethantheoneusedhere.ThefainterVLT-FORSzeropointsassumedinthesimulationsarepracticallycompensatedbythelargerintegrationtimesofourMagellan-IMACSdata.

2.fromthescatterofthe ttedvaluesforP0whensubtractingthethreedi erentskybackgroundpowerspectra.Thiswasanimportantdouble-checkoftheerrorsadoptedfromthesimu-lations,becausethefringingoftheIMACSdatarepresentanadditionaluncertaintysource,apartfromthephotonnoiseandbackgroundgalaxy uctuationswhichhadbeenincludedinthesimulations.

ThemaximumofbotherrorestimateswasadoptedaserrorinP0.Forsevenoutofthe25dEsinvestigated,thescatterfrombackgroundsubtractionwaslargerthantheMonteCarloesti-mate,seeTable1.Theresultingmeanerrorof(m M)is0.41magwhenadoptingcalibrationcaseA).Therearefourgalax-iesintherange 16.5<MV< 11.5magwhichwereim-agedtwiceinthecourseofoursurveysincetheywerelocatedintheoverlapofadjacent elds.Thedi erencesin(m M)betweenthosedoublemeasurementsrangedbetween0.10and0.35mag.Thisisencouraginglylow,indicatingthatouruncer-taintyestimatesareontheconservativeside.

3.2.1.ErrorestimateforSBFdistances

Theerrorof(m M)hastwoequallyimportantcontributions:uncertaintyin(V I)0andP0.

Theuncertaintyin(V I)0of0.058mag(seeabove)translatesintoa0.26magdistanceuncertaintyinthecaseoftheempiri-callycalibratedslope4.5.

3.3.Correlationbetweenseeingand

mIasa

functionofseeing,seeFig.6.Thosetwoobservablesshouldbeindependentofeachother.However,wedo nda2σsig-