ebook img

Testing the forward approach in modelling beta Cephei pulsators: setting the stage PDF

1.6 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Testing the forward approach in modelling beta Cephei pulsators: setting the stage

Comm. in Asteroseismology Contribution tothe Proceedingsofthe 38thLIAC/HELAS-ESTA/BAG,2008 Testing the forward approach in modelling β Cephei pulsators: setting the stage 9 0 0 A.Miglio,J.Montalba´n, andA.Thoul 2 Institutd’AstrophysiqueetdeG´eophysique n Universit´edeLi`ege,All´eedu6Aouˆt17-B4000Li`ege-Belgique a J 4 Abstract 1 ] The information on stellar parameters and on the stellar interior we can get by studying R pulsating stars depends crucially on the available observational constraints: both seismic constraints (precision and number of detected modes, identification, nature of the modes) S and “classical” observations (photospheric abundances, effective temperature, luminosity, . h surfacegravity). Weconsiderthecaseofβ Cepheipulsatorsand,withtheaimofestimating p quantitatively how the available observational constraints determine the type and precision - ofourinferences,wesetthestageforHare&Houndexercises. Inthiscontributionwepresent o preliminary resultsforonesimplecase,whereweassumeas“observed” frequencies asubset r offrequenciesofamodelandthenevaluateaseismicmeritfunctiononadenseandextensive st gridofmodelsofB-typestars. Wealsocomparethebehaviourofχ2 surfacesobtainedwith a andwithoutmodeidentification. [ Session: Poster 1 v 2 7 Tools 0 2 In order to set the stage for Hare&Hound exercises, the following three main components . need tobedefined: 1 0 • Theoreticalpredictions: Thegridofmodelsweuseisbetadat(Thirion&Thoul 9 2006). Stellarmodelsandadiabaticfrequencies arecomputed, respectively,withcles 0 (Scuflaire et al. 2008a) and losc (Scuflaire et al. 2008b). The masses considered in : thegridspanthedomainbetween7.8and18.5M⊙,ametalmassfractionZ between v 0.01 and 0.025, an initial hydrogen mass fraction X = 0.70, and four values of the i X overshootingparameter(αov)intherange0-0.25. Frequencies oflow-orderoscillation modesofdegreeuptoℓ=2arecomputed formain-sequence models. r a • Observational constraints: we consider only seismic constraints, i.e. a subset of theoretical eigenfrequencies of amodel M0inthe grid. The effects of rotation on the oscillation modes are not considered in this first step, thus all modes are assumed to beaxisymmetric(m=0). Concerning thedegree of theobserved modes, weconsider thecasewhereℓisunknown aswelltheonewhereℓisavailableasaconstraint. TestingtheforwardapproachinmodellingβCepheipulsators: 2 settingthestage ℓ ν (µHz) 0 57.78 Table 1: Theoretical oscillation frequen- cies of model M0 considered as observa- 1 58.93 tional constraints, the uncertainty on the 1 80.29 frequenciesisassumedtobe0.1µHz. 2 39.46 • Merit function: Inorder tocompute ameritfunction ateachpoint ofthegrid,we useadoubleoptimisationproceduresimilartotheoneextensivelyadoptedinsdBaster- oseismology(seee.g. Brassardetal.2001 andCharpinetetal.2005). Foreachmodel inthegridwefindthebestglobalmatchbetween “observed”andtheoretical frequen- cies by using a standard χ2 merit function. Then we study the properties of good-fit modelslooking atminima inthe χ2 asa function of thestellarparameters/properties (e.g. location in an HR diagram, mass, central hydrogen mass fraction (Xc), mean density,...). First test Weconsider asseismicconstraints 4oscillation modesof themodel M0(see Table1), with frequenciesinthetypicaldomainofthepulsationmodesobservedinβCepheistars. Themain parameters and properties defining M0 are: M = 10M⊙, Xc = 0.2, αOV = 0, Z = 0.02, logTeff =4.34andlogL/L⊙=4.02. Wethencomputethemeritfunctiononasub-gridofmodelsofbetadat,whereαOVand Z arethe sameasinM0. The behaviour of χ2 formain-sequence models of different mass and evolutionary status isshown inFig. 1. Inthe casewhere ℓ isgivenasan observational constraintforallthemodes(rightpanels),theproperties andparametersoftheinputmodel M0 are easily recovered due to the appearance of an isolated global minimum in the χ2 function. The increase ofthe number of χ2 minima inleftpanels(ℓ isunknown) compared to right panels allows to estimate the loss of information when mode identification is not available. Nevertheless, even if no mode identification is available, the frequencies of 4 modesallowtoconstraintheparameterspaceinregionsclosetoχ2 minima. The results of this simple test show that the approach presented here is viable tool to determinethenumberandprecisionofobservationaldatarequiredtoconstraintheproperties ofβCepheipulsators. However,anextensiveandthoroughstudyisneededtoinvestigatethe effect ofconsidering both additional uncertainties (e.g. onparameters such asovershooting and initial chemical composition, on rotational splittings and identification of m) as well as other constraints (e.g. luminosity, Teff, logg, photospheric abundances, non-adiabatic constraints). Acknowledgments. A.M.andJ.M.acknowledgefinancialsupportfromtheProdex-ESACon- tractProdex8COROT(C90199). A.M.isaCharg´edeRecherches andA.T.isaChercheur qualifi´e oftheFRS-FNRS. References Brassard,P.,Fontaine,G.,Bill`eres,M.,etal. 2001,ApJ,563,1013 Charpinet,S.,Fontaine,G.,Brassard,P.,etal. 2005,A&A,437,575 Scuflaire,R.,Th´eado,S.,Montalb´an,J.,etal. 2008a,ApSS,316,83 Scuflaire,R.,Montalb´an,J.,Th´eado,S.,etal. 2008b,ApSS,316,149 Thirion,A.&Thoul,A.2006,ESASP-1306,383 A.Miglio,J.Montalba´n,andA.Thoul 3 ℓ not known ℓ known Figure1:Colour-codedχ2(logarithmicscale)inthreedifferentplanes:logTeff-logL/L⊙(upperpanels), logTeff-meandensity(middlepanels),andmass-Xc(bottompanels). Whitedotsrepresentthepositionof theinputmodelM0ineachparameterspace. Rightpanelsshowχ2computedincludingtheidentification ofℓasaconstraint,whereasinleftpanelsℓisconsideredunknown(ℓ=0,1or2).

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.