Sawtooth Software V6.4.6 Crack
Gaspard Xenos
SMRT stands for "Sawtooth Software Market Research Tools". It is an umbrella that incorporates various Sawtooth Software conjoint tools into a single shell. You may e-mail sup...@sawtoothsoftware.com if you have questions relating to information on this web page. We strongly suggest using the latest version of SMRT.
1 IntroductionIn recent years, there have been significant advances in our understanding ofpulsation in long period variable stars (LPVs). Particularly important isthe discovery of multiple period-luminosity relations for AGB stars in the LMCand the interpretation of all but one of these relations in terms of radialpulsation in low order modes (Wood et al. 1999). Another advance isthe realization of the importance of pulsation in the production of mass lossfrom red giants (e.g. Wood 1979; Bowen & Willson 1991; Hfner et al. 1996).One problem with the observational study of pulsation and mass loss in red giants is that nearly all the stars studied so far are field stars. This means thatwe have no way of directly determining the initial or current stellar mass.Any relations that are found for the field stars will bebroadened by the range in mass (age) and metallicity existing among the stars.For Galactic stars, but not LMC stars, we have the additional problem that the luminosity (distance) isgenerally poorly known. As a result of these uncertainties, precise comparison of observedperiods, luminosities, amplitudes and mass loss rates with theoreticalcalculations is complicated.A way around these problems is to observe pulsating red giant starsin star clusters where the giants have a common (initial) mass, compositionand distance. We can then readily compare theoretical and observed period-luminosity relations, and we can see how the mass loss rate depends onpulsation properties and quantities such as mass, luminosity and metallicity.The driving of mass loss by pulsation is a complicated process.Levitation of the stellar atmosphere caused by the pulsationleads to the formation above the photosphere of a cool and denseenvironment where dust grains form and grow efficiently. Radiation pressure on dust grains then combines with momentum in shock waves to drive the mass loss(e.g. Wood 1979; Bowen & Willson1991; Hfner et al. 1996). It is found that larger pulsation velocity amplitudes and higherluminosities drive a higher mass loss rate. One of the main purposes of this study is toobserve the pulsation velocity amplitudes in a good sample of redvariables in a star cluster so that a comparison of mass loss rate with pulsation velocityamplitude (and luminosity) can be made.The pulsation velocities of LPVs are best measured in thenear infrared since optical absorption lines measure velocities in the outer layers only. Theseoptical velocities are nearly always directed inward relative to the stellar center-of-mass(e.g. Wood 1979). The near-infrared lines reveal the deeper, larger amplitude pulsation.Monitoring of the velocity variations in field variablesusing infrared (1.6 m) CO lines has been carried out by Hinkle (1978), Hinkle et al. (1982, 1997) andLebzelter et al. (2000). These observations revealed velocity curves with amplitudes between 3 and 30 km s-1. The velocity curves of Miras are sawtooth-shaped and of large amplitude, withphases of line doubling around maximum light. The typical velocity curvesof semiregular variables (SRVs) are of smaller amplitude than in Miras and they are rather sinusoidal (i.e. continuous with no line doubling), althoughthey can reflect the partly irregular behaviour seen in the light curves.A summary is given in Lebzelter & Hinkle (2002).Theglobular cluster 47 Tuc (NGC 104) was chosen for this study sinceit contains the richest knowncollection of long period variables (LPVs). Properties of this cluster from the literature aresummarized in Sect. 2. Alongits giant branch, four Miras and more than 10 semiregularand irregular variables have been detected (Sawyer-Hogg 1973; Lloyd-Evans 1974). Several infrared searches for indications of mass lossfrom giant branch stars have also been reported (see Sect. 3).
2 47 Tuc and its variables47 Tuc is a prototype "metal-rich'' globular cluster as well as being one of the closest. Recently, Gratton et al. (2003) presented a very detailed study on the fundamentalparameters of 47 Tuc. They derive a metallicity of on the Carretta & Gratton (1997)scale, which is in agreement with the findings of several other groups (see references in Gratton et al.). A metallicitybased on the Zinn & West (1984) scale is given by e.g. Briley et al. (1995) with [Fe/H] = -0.76.An accurate distance to the cluster is still a matter of some debate. Values in the literature forthe distance modulus of 47 Tuc scatter by about 0.2 mag (for an overview see Gratton et al. 2003).Gratton et al. (2003) finda distance modulus of from main sequence fitting. All studies agree that there is very little reddening towards 47 Tuc ().Based on their estimates for the distance and metallicity Gratton et al. (2003) derived an age of 47 Tuc of Gyr. Hesser et al. (1987) derived a turnoff mass of 0.9 ,but used an age estimate of13.5 Gyr. However, with the age from Gratton et al. and the more recent theoretical isochrones from Bertelli et al. (1994) we again come to a turnoff mass between 0.86 and 0.9 .A large numberof photometric measurements in the blue, visual and near- andmid-infrared (Lee 1977; Frogel 1983; Montegriffo et al. 1995; Origliaet al. 1997; Ramdani & Jorissen 2001) establish a very accurate color-magnitude diagram ofthis cluster showing a well defined AGB. Luminosities and surfacetemperatures have been derived for the AGB stars (e.g. Whitelock 1986). Light curves of the LPVs V1 to V8 were measured by Arp et al. (1963). Several additional variables were detectedby Lloyd-Evans & Menzies (1973). A further study, including the variables V3 to V7, V11(=W12), V13 and V18, was presented by Fox (1982). The light variability of thesestars will be further discussed in Sect. 4.1.
3 Measurements of mass loss in 47 TucThere have been many attempts to measure the mass loss from AGB starsin globular clusters. Mass loss at the end of the red giant branch (RGB, also called the first giant branch or FGB) phasehas already been proposed to explain the observed gaps in the horizontalbranch of globular clusters (Soker et al. 2001; Schrder & Sedlmayr 2001).Cohen (1976) proposed the existence of circumstellar shells (and hence mass loss)around first giant branch stars in globular clusters to explain emission componentsin the Hline. Bates et al. (1990) summarized different studies onHlines of globular cluster stars and found a further indication forcircumstellar shells from the Na D line profiles. Lyons et al. (1996)discussed mass motions in the atmospheres of 63 red giants from five different globular clusterswith the help of Hand Na D lines. Core shifts of these lines indicatemass flow. The lower luminosity limit for outflow from Hand Na D lines is and 2.9, respectively. Only part of the red giants showed an outflow. Standard deviations in the final masses of white dwarfs in globularclusters of the order of 0.1 mag may indicate a stochastic nature of the mass loss(e.g. see the discussion by Alves et al. 2000).Not only is mass loss known to be essential instellar evolution but the mass loss from red giants in globularclusters should result in intracluster gas and dust (Evans et al. 2003). Searches for the intracluster gas havebeen limited in their results. The low metallicity no doubt plays a role in reducing silicate emission, making dust detection difficult (Frogel & Elias 1988;Helling et al. 2002). However, findings from AGB stars inthe Large Magellanic Cloud (LMC) confirm that low metallicity stars reach high mass loss rates(e.g. Wood et al. 1992; van Loon et al. 1999).A few recent papers provide more detailed information on circumstellar shells around AGB stars in 47 Tuc. One of the most definitive results is by Ramdani & Jorissen(2001) who used the ISO satellite to measure the mid-infrared emission around6 AGB variables in 47 Tuc. Their findings, however, highlight the difficulty ofcorrelating pulsational properties with mass loss. Half of the variables,namely V5, V7 and V15, show no or only marginal 12 m-excess while the otherthree stars, V3, V11 and V18, do have a detectable excess flux at 12 m.The excess of V11 is rather small. V3 is a Mira and would be expected to show an excess.But the largest infrared excess was measured for V18, an irregularAGB variable of only modest luminosity. Origlia et al. (2002) suggested that the mass loss from red giantstars is episodic. They found an IR excess for V8 and no excess for V21(see also Sect. 5.3).Overall, some indication of the presence or absence of circumstellar materialexists for 12 AGB variables in 47 Tuc. A substantial infrared excess in V3, V11 and V18 was detected byGillett et al. (1988), but these authors did not find indications for an infrared excess in V5 and V13.An excess V-band polarization was found in V1 to V4, V6, V8 and V11 by Forte et al. (2002) indicating the existence of circumstellar material around these stars.Glass & Feast (1973) reported an L-band excess in V3.Frogel & Elias (1988) detected a 10 m excess in the four cluster variablesV1, V2, V3 and V4. They calculated total mass loss rates for these objects ranging between4.6 and yr-1. A 10 m excess was also detected in V1 and V3by Origlia et al. (1997). However theirestimated dust mass loss rates are about ten times less than those of Frogel & Elias (1988).Table 1:Properties of LPVs in 47 Tuc.
4 Observations and resultsWe selected 12 long period variables in the globular cluster 47 Tuc for which a period or at least some information on the variability type was given in thecatalogue of variable stars in globular clusters published by Clement et al. (2001). These stars are moderately bright infrared sources, with.Table 1 summarizes the properties of the stars.Time series of infrared spectra in the H band were obtained in 2001 and 2002 with the 74 inchtelescope at Mount Stromlo Observatory, Australia. The NICMASS detector, already successfully usedfor a preceding program at Kitt Peak (Joyce et al. 1998), was used at the Coud focus of the telescope.The standard infrared observation technique was used. Spectra of each star were obtained at two different slit positionsto allow sky subtraction. Resolution was set to R=37 000. We achieved a S/N ratio of 30 or better. An example spectrum is shown in the left part ofFig. 1 with an identification of some relevant spectral features.The spectral range covered a number of second overtone CO lines,some OH lines, and a few metallic lines.When 47 Tuc was visible at Mount Stromlo spectra were obtained approximately once a month. The observing program had an unexpected end when Mount Stromlo Observatory was destroyedby a bush fire in early 2003. Nevertheless time coverage and sampling of the 47 Tuc AGB variables,combined with a few additional spectra taken in the same wavelength region with the PHOENIX spectrograph at Gemini South, is sufficient for the investigation presented here.
Figure 1: Left panel: example spectrum of the star V11. Several prominent features are identified. Right panel: example of line doubling (CO 4-1, V3).Open with DEXTER
The bright stars Cet and Oph have been used as primary velocitystandards (Udry et al. 1999). Velocities of the variables were determined by a cross correlation technique,using the IRAF task fxcor. Typical velocity uncertainties, determined from multipleobservations of some stars in the same or consecutive nights, were found to be0.4 km s-1.To determineproper phases for the velocity curve, photometric measurements of all variables in our sample except V5(see below) were done as well. Data in the blue and red MACHO filters were obtainedwith the 50inch telescope at Mount Stromlo (also destroyed in the January 2003 fires). V and I measurementswere obtained with ANDICAM at the YALO telescope (before the MACHO dataset), with a few additional measurements from ANDICAM at the CTIO 1.3 m telescope (after the MACHO dataset). The different data sets have been combinedusing transforms based on about 40 nonvariable cluster stars ranging in V-I between 0.9 and 2.1. The MACHO blue filter has a mean wavelength close to that of the V filterso MACHO blue magnitudes can be reliably transformed to V. Long time seriesare therefore available in the V filter. No obvious phase shift relative to the otherfilters was noticed, but a conclusive result cannot be drawn from our data.From the V light curves we determinedthe time of the light maxima, phase zero.
Figure 2:Light ( upper panel) and velocity variations ( lower panel) for V1. Filled symbols in the lower panel indicate individual velocity measurements. For a better illustration of the velocity change data are repeated shifted by an integral number of periods forward and backward in time (open symbols). The period listed in Table 1 is used. The typical error bar for the velocity data is indicated.Open with DEXTER
Figure 3:Same as Fig. 2 for V2.Open with DEXTER
Figure 4:Same as Fig. 2 for V3.Open with DEXTER
Figure 5:Same as Fig. 2 for V4.Open with DEXTER
Figure 6:Velocity change of V5. Only part of the time of monitoring is shown. Symbols as in the lower panel of Fig. 2. No parallel light curve data exist for this star. A typical error bar for the velocity data is indicated.Open with DEXTER
Figure 7:Light ( upper panel) and velocity ( lower panel) variations for V6. A typical error bar for the velocity data is indicated.Open with DEXTER
Figure 8:Same as Fig. 7 for V7.Open with DEXTER
Figure 9:Same as Fig. 2 for V8.Open with DEXTER
Figure 10:Same as Fig. 7 for V11.Open with DEXTER
Figure 11:Same as Fig. 7 for V13.Open with DEXTER
Figure 12:Same as Fig. 7 for V18.Open with DEXTER
Figure 13:Same as Fig. 7 for V21.Open with DEXTER
4.1 Light curvesLight curves for most of the variables are shown in the upper panels of Figs. 2 to 13. Periods derived from our measurements were in most cases in good agreement with the values fromthe literature. In our data V4 shows a main period of 170 d in agreement withthe value of 165 d from Arp et al. (1963).Fox (1982) found a period of 82 days for V4 and suggested that this star switched between two modes. This type of multimode behaviour has subsequentlybeen found to be common in the light curves of LPVs (e.g. Wood et al. 1999).For V5, we unfortunately do not have sufficient usable data as this star was located in a bad area of the CCD. The Sawyer-Hogg value(1973) is 60 days, but different values can be found in the literature, ranging between 29 and70 days (Fox 1982). From our velocity data we rather favour a period of 50 days.The short periods of 52 and 100 days reported for V11 could not be confirmed in our data. Instead, the star shows some longperiod variation lasting more than 200 days (Fig. 10). Compared to the data presented by Fox (1982)the amplitude of the variation is smaller. This star appears to be another multimode pulsator. For V13, there is a hint of the catalogued 40 d period in our light curve,but a period of 90 d may also be present. Additionally, our light curve suggests a long period (Fig. 11) about 10 times the short period. This star appears to be yet another multimode pulsator.V18 was reported variable bothin the catalogue of Sawyer-Hogg (1973) and by Lloyd Evans (1974), while Fox (1982)found no variability in this star. We found photometric variability with an amplitude of about0.2 mag in this object (Fig. 12). Our short light curve would suggest a period of about 83 days. For V21, no period determination existed in the literature. Our data set(see Fig. 13) shows a period of 76 days.4.2 Velocity curvesVelocity curves were determined for all 12 stars of our sample. All stars show at least some velocity variability above our detection threshold. Line doubling was detected in three stars of our sample (Table 2).An example of line doubling is shown in the right part of Fig. 1.The velocity curves we derived are shown in the lower parts ofFigs. 2 to 13 together with the corresponding light change. The velocity curves can be roughly separated into threegroups: V1, V2, V3, V4, and V8 show velocity curves very similar in shape to those typical of Mirasfound in the solar neighborhood (Lebzelter & Hinkle 2001). The velocity amplitude of V1, V2 andV3 are all similar to nearby Miras, too. The other two stars show a similar shape but a clearlysmaller amplitude. While we confirm the Mira-like nature of these five variables (e.g. noted byWhitelock 1986), we note that the smaller velocity amplitude of the latter two starsis also accompanied by a smaller light amplitude.The second group of variables, consisting of V5, V6 and V7 is comparable to local semiregular variables (Lebzelter & Hinkle 2002). Their amplitudes are much smaller than in the first group. V5 shows a very regular light change, while the other twoobjects are obviously not strictly periodic, in agreement with the semiregular nature of their light curves. The third groupconsists of V11, V13, V18, and V21. In all four of these stars, there is no obvious correlation between the lightchange and the velocity variations. In the case of V13, it cannot be ruled out that variations occur on a time scale similarto the short or even the long period (see above). For V11 our observations do not allow a clear picture of the light change, so a correlationis not possible. On the other hand, V18 and V21 do show a rather well defined light change, but velocity variabilityoccurs on a different time scale. The reason for that is not clear and longer spectroscopic and photometric time series willbe needed to understand this phenomenon.Table 2 summarizes the results on the velocity variations in the 47 Tuc AGB variables. The table gives thetotal velocity amplitude, the characteristics of the velocity curve and comments on the occurrence of line doubling.Table 2:Data on the AGB variables in 47 Tuc.