Page 1 PROPOSAL FOR HUBBLE SPACE TELESCOPE OBSERVATIONS ST ScI Use Only ID 4272c Report Date: 09-May-96:19:30 Version: ********** Check-in Date: ********** 1.Proposal Title: SUPERSONIC CHROMOSPHERIC WINDS CYC3-MED ------------------------------------------------------------------------------------ 2. Scientific Category 3. Proposal For 4. Proposal Type 5. Continuation ID COOL STARS GO Sub Category STELLAR ATMOSPHERES ------------------------------------------------------------------------------------ 6. Principal Investigator Institution Country Telephone Andrea K. Dupree SMITHSONIAN ASTROPHYSICAL OB USA ------------------------------------------------------------------------------------ 7. Abstract We have discovered supersonic wind velocities in the chromosphere of the hybrid supergiant star: Alpha Aquarii (HD 209750; G2Ib). These mass motions in excess of the photospheric escape velocity are evident in the profile of the He I 10830A line, and confirmed by our model calculations. Such supersonic motion of a wind profoundly impacts the velocity profile and energy requirements of stellar winds and subsequent mass loss. We are proposing HST spectroscopic observations of Alpha Aqr in order to obtain line profiles and velocity shifts of transition region lines to identify the emitting region, to define the wind acceleration profile, and to settle long-standing controversies about the wind temperature. ------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------ 9. Est obs time (hours) pri: 7.25 par: 0 10. Num targs pri: 1 par: 0 ------------------------------------------------------------------------------------ 11. Instruments requested: HRS ------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------ Page 2 I. GENERAL FORM Proposal 4272c PI: Andrea K. Dupree Proposal Title: SUPERSONIC CHROMOSPHERIC WINDS CYC3-MED ------------------------------------------------------------------------------------ 1. Proposers: Proposers Institution Country ESA ------------------------------------------------------------------------------------ Pi Andrea K. Dupree SMITHSONIAN ASTROPHYSICAL USA OBSERVATORY Barbara A. Whitney SMITHSONIAN ASTROPHYSICAL USA OBSERVATORY Dimitar D. Sasselov SMITHSONIAN ASTROPHYSICAL USA OBSERVATORY ------------------------------------------------------------------------------------ 3. Description of proposed observations. We propose to use the GHRS at medium resolution, G160M grating in order to measure the line widths, line positions, and flux of 4 transitions in the spectrum of Alpha Aquarii: N V (center at 1240.8A), C II (center at 1335.1A), C IV (center at 1549.5A), He II (center at 1652.0A). A wavelength calibration will be taken before each spectral line observation. Page 3 ------------------------------------------------------------------------------------ 4. Justification of need for HST observations. Ultraviolet spectra are required to study the transition regions and winds of cool stars because the important resonance transitions at these temperatures occur in the ultraviolet. The capabilities of HST are required for the spectral resolution necessary to identify the He II line and to determine the line profiles of the transition region lines. Spectra from IUE have been obtained at low dispersion (about 6 Angstrom resolution), and fluxes can be extracted in cases where blends are not present. However the region near the He II feature also contains transitions of Si I, S I, and C I, and no certain identification can be made from low dispersion spectra. Moderate dispersion is necessary to detect the He II line. The line profiles are needed of the resonance transitions: C II, C IV, and N V. We have pushed IUE to the limit, obtained a 13 hour exposure on Alpha Aqr, but these lines, or even faint smudges at the appropriate positions, are not visible on the photowrite image nor on a digital display. The line profiles, accessible only with HST, are needed to substantiate and extend the velocity profile to higher temperatures. Alpha Aqr has been observed with IUE frequently. After defining the class of hybrid stars (Hartmann, Dupree, and Raymond 1980), we were the first to note changes in the Mg II profile using IUE data (Dupree and Baliunas 1979). Others have also remarked on the variability of transition region line fluxes and identified a period of variation (Broscius, Mullan, and Stencel 1985). Alpha Aqr is included in the SAO- Mt. Wilson HK Program, and has been monitored from 1983 to the present time in the Ca II H and K lines near lambda 3900 which act as the tracers of photospheric magnetic fields. A search for periodicities in a 5-year time series revealed (Rao et al. 1993) that both long (approximately 400 days) and short (approximately 75 days) period variations are present. The longer periods may be ascribed to rotation; the shorter period may reflect episodic behavior, possibly related to pulsation. Because the discovery of a long period demonstrates the presence of rotation markers on the surface, Rao et al. (1993) conclude that magnetic structures are present in the atmospheres of these supergiant stars. This offers the possibility that magnetic loops are the source of the C IV emission; this observation also makes plausible the channeling and acceleration of a stellar wind, and the potential for heating and momentum deposition by a magnetically associated process such as Alfven waves. The short-period variations may be connected with the rapid acceleration we have found from modeling the He I lambda 10830 line discussed in Item 1, above. To follow up on the short-term atmospheric variability of Alpha Aqr, we have carried out an IUE program to monitor this star every ten days for a period of 70 days in 1991. Fluxes of chromospheric and transition region lines were obtained as well as Mg II line profiles. Simultaneously, we obtained spectra at H alpha, Ca K (from Oak Ridge Observatory, F. L. Whipple Observatory, David Dunlop Observatory), He I (lambda 10830) (from the Canada-France-Hawaii Telescope) and HK fluxes (from Mt. Wilson), giving us a solid idea of the time variability of the atmosphere. While these spectral features provide enough information on the time-dependent dynamics of the atmosphere, the HST line profile observations are the crucial physical constraint to the mechanisms responsible for that dynamics. For practical purposes, the HST observations should ensure the uniqueness of our atmospheric modeling. We propose to observe 1 star: Alpha Aquarii (HD 209750; G2 Ib; V= 2.90; B-V= +0.98). Fluxes and line profiles will be obtained using the GHRS, Grating 160M at 4 different grating positions. Total required integration times will be accomplished with 15 minute segments. The flux in three of our target lines, C II, C IV, and N V, has been measured (an upper limit exists for He II) from low dispersion IUE spectra (Hartmann et al. 1980, 1982). An average of the two values was taken. These fluxes represent the flux in the multiplet, so we have assumed optically thin ratios (2:1, for the Li-like series; and according to statistical weight for C II) to estimate the flux in the strongest component of the multiplet. It is also necessary to estimate the line width, which we take as 1 Angstrom, based on measures of transition region line profiles in a hybrid giant, Alpha TrA (Hartmann et al. 1981), and confirmed by examination of S I and Si II profiles in Alpha Aqr from our high dispersion IUE spectrum. The sensitivity for the HRS Grating G160M, 2'' Aperture, at the wavelengths of interest is given (GHRS Instrument Handbook, Version 3.0, Jan. 1992, Table 4-13, p. 45) for the combined OTA + aperture + HRS system. To estimate the exposure time to achieve a signal-to-noise of about 20, under conditions when neither scattered light nor dark count is important, Equation 4-5 in the GHRS Instrument Handbook (3.0) applies: R^2 = s n_s t, where R is the signal to noise ratio, n_s is the number of bins (n_s =1), s is the count rate from the star at the Earth, and t is the integration time. We summarize the parameters for the four observations in the Table below and an evaluation of the necessary time for each one. A total of 7.25 hours of primary integration time results. Line lambda(Angstrom) Obs. Flux S^dagger Count Rate Time erg cm^-2 (cts/diode/s) (hrs) s^-1 Angstrom^-1 N V 1238.8 1.3 x 10^-13 4.0 5.3 x 10^-2 2.10 C II 1334.5 1.3 x 10^-13 6.0 7.8 x 10^-2 1.40 C IV 1548.2 2.0 x 10^-13 4.4 8.8 x 10^-2 1.30 He II 1640.5 <1.0 x 10^-13 4.6 4.6 x 10^-2 2.45 dagger: G160M grating Sensitivity: 10^11 counts diode^-1 /erg cm^-2 s^-1 Angstrom^-1. Page 4 ------------------------------------------------------------------------------------ 5. Description of special scheduling requirements. Please schedule between May 10 and November 1, 1993 so that IUE can be used simultaneously. Advance notice would help us organize the simultaneous ground- based spectroscopy. ------------------------------------------------------------------------------------ 6. Description of special calibration exposures. Wavelength calibrations necessary before each set of spectral line observations in order to obtain precise Doppler shifts of emission features. Page 5 ------------------------------------------------------------------------------------ 7. Data reduction and analysis plans. The ultraviolet spectra from HST will be reduced at the Solar and Stellar Division Computer Facility at the Center for Astrophysics which has established several standard spectroscopic reduction programs (IRAF and IDL) in addition to codes especially developed to reduce and analyze IUE ultraviolet spectroscopic data. Spectral deconvolution will be carried out to recover from the spherical aberration of the primary mirror of HST. We are aware of the procedures being tested both by the GHRS team and scientists at the STScI. In addition, we plan to consult with other scientists in the SSP Division here who have much experience in deconvolution of images and spectra both for speckle imaging and interferometry. We will acquire spectra with S/N of 20 or more to make this possible. Line fluxes and line centers will be determined using the HRS calibration and the various routines available in these reduction packages. In the past, we have routinely reduced IUE as well as MMT and FLWO 1.5-m echelle spectra with these packages. The optical data (H alpha, Ca II, He I) will have a first reduction at the observatories where they are acquired and then will be incorporated into the analysis at the Center for Astrophysics. Our goal is to find what physical mechanism heats these hybrid atmospheres, where in the atmospheres acceleration occurs, and which physical mechanisms produce winds in luminous cool stars. We believe that the answer to ``where ?'' is to be found in the observational data. We shall compute detailed synthetic spectra (line profiles) from time-dependent atmospheres to match the data. Our atmospheric models will be neither in radiative nor in dynamical equilibrium, and will assume spherical symmetry. The first models thus constrained by the available observations will be used to define the physics of wind acceleration. Our ability to treat explicitly different physical mechanisms and to compute synthetic spectra will be used to make predictions about the behavior of certain lines and to find sensitive spectral features for model verification. As a result we shall reject or accept a certain model. The rejected models (due to lack of a unique solution or due to missing physics) will be initialized again. The accepted models will hopefully offer answers to the` `which mechanisms?'' question. A semi-empirical spherically symmetric atmosphere built by the code PANDORA (Avrett and Loeser 1992) begins the process. This code solves the non-LTE radiative transfer equations, including the effects of expansion and spherical geometry on the line source function, as well as the emergent profiles. Parameters of the atmosphere are constrained by observed fluxes and profiles of chromospheric lines such as H alpha, Ca K, He I and Mg II, and transition region lines measured by IUE. An Alpha Aqr model has been constructed (Dupree et al. 1992) and is necessary for interpretation of profiles rather than plane-parallel static models (Harper 1992). This initial atmosphere becomes the input for the code HERMES (Sasselov and Raga, 1992), which computes a time-dependent model of the atmosphere. HERMES is a Lagrangian hydrodynamic scheme of second order (Godunov technique) with iterative and simultaneous non-LTE radiative transfer calculations by a version of the code MULTI (Carlsson, 1986; Uitenbroek, 1990). The output of HERMES contains time-dependent line profiles for a number of species which can be matched to observed spectra. The software described in our overall plan above is already developed. At this point, we plan to use linearized theory for Alfven wave dissipation, and to include the effects of dust as a momentum transfer term with a prescribed cross section. Our method and HERMES will easily allow further sophistication of the input physics. Page 6 ------------------------------------------------------------------------------------ 8. Additional comments or special requests. We will make coordinated spectroscopic observations from the ground (Ca II, H alpha, He I), and will apply for time on IUE to obtain transition region fluxes and line profiles of the Mg II emission. Because IUE has Beta-angle constraints on its target, scheduling between May 10 and November 1 is optimum. As much advance notice as possible would be helpful. The Mt. Wilson HK program will continue to follow Alpha Aqr through 1993 so that its variations are well documented. ------------------------------------------------------------------------------------ 9. Description of previous HST work. Program No. 2693: A Search for Mass Loss from Two Red Giants in NGC 6752, (A. K. Dupree, PI). This small (< 10 hrs) program obtained Mg II profiles for two red giants, A31 and A59, in the globular cluster NGC 6752. Observations were made in April 1992. The data were received in May 1992, and we are in the process of reducing it along with the contemporaneous ground-based spectra. This program is not related to the present proposal. ------------------------------------------------------------------------------------ 10. Resources to be supplied by investigator's institution(s). The Smithsonian Institution provides a full (12-month) salary and benefits for A. K. Dupree, the Principal Investigator. The Smithsonian Institution also provides a full (2-month) salary and benefits for D. D. Sasselov who is a Harvard -Smithsonian Center for Astrophysics PostDoctoral Fellow until the Summer of 1993. Computer facilities, and the people responsible for system operations in the Solar and Stellar Physics Division of the Center for Astrophysics have been provided by the Smithsonian Institution. Our Division maintains a UNIX-based Local Area Network of SUN Workstations. The SSP Divisional Computer System currently is comprised of 4 SUN4 servers providing disk services to approximately 26 client workstations. The workstations provide text processing and connectivity to divisional scientists. They are also used for running code that does not require large memory or disk space and which does not require a fast SUN machine to complete a run in a reasonable amount of time. Our proposed model analysis of the HST data will require a faster, dedicated machine, and more disk space than is now available. We hope to acquire some of this hardware as a part of our Cycle 1 HST Program. Standard reduction and software packages (IRAF, IDL, MAXIMA, MONGO, AIPS, ...) have been provided by the Smithsonian Institution, and are maintained and updated by the system staff as necessary. However, for specific applications of interactive reduction, deconvolution, analysis, and model-building pertaining to our proposal, we must appoint people using funds from outside grants. Similarly, postdoctoral scientists, and students from Harvard University and elsewhere must be hired with grant funds. ------------------------------------------------------------------------------------ 11. Address Information Name: ANDREA K. DUPREE Category: PI Institution: Smithsonian Astrophysical Observa Address: SMITHSONIAN ASTROPHYSICAL OBSERVATORY 60 GARDEN STREET, MS-15 City: CAMBRIDGE State: MA Zip Code: 02138 Country: USA Telephone: 617-495-7489 Telex (or e-mail): 921428 SATELLITE CAM ------------------------------------------------------------------------------------ TARGET LIST a) Fixed Targets ID = 4272c [ 7] ------------------------------------------------------------------------------------------------------------------------------------ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 ------------------------------------------------------------------------------------------------------------------------------------ Tar| Target | Target | Target |Coord | Radial |Acqui|FLX| Flux data No | Name | Description | Position |Eqnx | Vel. |Prblm|REF| | | | | | | | | ------------------------------------------------------------------------------------------------------------------------------------ 1 HD209750 A,138 RA = 22H 05M 46.93S +/- 2000. V = +8 1 V=2.90,TYPE=G2IB ALPHA-AQR 0.007S, 2 B-V=+0.98 GSC-5224- DEC = -00D 19' 10.9" +/-0.1" 1806 ------------------------------------------------------------------------------------------------------------------------------------ EXPOSURE LOGSHEET ID = 4272c [ 8] ------------------------------------------------------------------------------------------------------------------------------------ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |10 | 11 | 12 |13 |14| 15 ------------------------------------------------------------------------------------------------------------------------------------ Line | Seq | Target |Instr | Oper. | Aper |Spectral|Central| Optional |Num| Time | S/N |Flx|Pr| Special Number | Name | Name |Config| Mode |or FOV |Element |Waveln.| Parameters |Exp| |Rel. Time|Ref| | Requirements ------------------------------------------------------------------------------------------------------------------------------------ 1 HD209750 HRS ACQ 2.0 MIRROR-A2 BRIGHT=RETURN 1 1.8S 1 CYCLE 3 MAP=DEF ONBOARD ACQ FOR 2 SEARCH-SIZE=3 GROUP 1-10 WITHIN 30H Comments: STEP-TIME=200MS ------------------------------------------------------------------------------------------------------------------------------------ 2 HD209750 HRS ACQ/ 2.0 MIRROR-A2 1 163.2S 1 CYCLE 3 PEAKUP ONBOARD ACQ FOR 3- 10 Comments: STEP-TIME=1.6S ------------------------------------------------------------------------------------------------------------------------------------ 3 WAVE HRS ACCUM SC2 G160M 1651.0 STEP-PATT=3 1 DEF 1 CALIB FOR 4 NO SLEW SEQ 3-4 NO GAP CYCLE 3 ------------------------------------------------------------------------------------------------------------------------------------ 4 HD209750 HRS ACCUM 2.0 G160M 1651.0 STEP-PATT=5 15 10M 1 CYCLE 3 ------------------------------------------------------------------------------------------------------------------------------------ 5 WAVE HRS ACCUM SC2 G160M 1549.5 STEP-PATT=3 1 DEF 1 CALIB FOR 6 NO SLEW SEQ 5-6 NO GAP CYCLE 3 ------------------------------------------------------------------------------------------------------------------------------------ 6 HD209750 HRS ACCUM 2.0 G160M 1549.5 STEP-PATT=5 8 10M 1 CYCLE 3 ------------------------------------------------------------------------------------------------------------------------------------ 7 WAVE HRS ACCUM SC2 G160M 1335.1 STEP-PATT=3 1 DEF 1 CYCLE 3 CALIB FOR 8 NO SLEW SEQ 7-8 NO GAP ------------------------------------------------------------------------------------------------------------------------------------ 8 HD209750 HRS ACCUM 2.0 G160M 1335.1 STEP-PATT=5 8 10M 1 CYCLE 3 ------------------------------------------------------------------------------------------------------------------------------------ 9 WAVE HRS ACCUM SC2 G160M 1233.0 STEP-PATT=3 1 DEF 1 CYCLE 3 CALIB FOR 10 NO SLEW SEQ 9-10 NO GAP ------------------------------------------------------------------------------------------------------------------------------------ 10 HD209750 HRS ACCUM 2.0 G160M 1233.0 STEP-PATT=5 13 10M 1 CYCLE 3 ------------------------------------------------------------------------------------------------------------------------------------ Summary Form for Proposal 4272c [ 9] Item Used in this proposal ------------------------------------------------------------------------------------------------------------------------------------ Configurations HRS ------------------------------------------------------------------------------------------------------------------------------------ Opmodes ACQ ACQ/PEAKUP ACCUM ------------------------------------------------------------------------------------------------------------------------------------ Optional Parameters BRIGHT=RETURN MAP=DEF SEARCH-SIZE=3 STEP-PATT=3 STEP-PATT=5 ------------------------------------------------------------------------------------------------------------------------------------ Proposal for GO ------------------------------------------------------------------------------------------------------------------------------------ S/C Hours 7.25 ------------------------------------------------------------------------------------------------------------------------------------ Scientific Category COOL STARS ------------------------------------------------------------------------------------------------------------------------------------ Scientific Sub-category STELLAR ATMOSPHERES ------------------------------------------------------------------------------------------------------------------------------------ Special Requirements CYCLE 3; ONBOARD ACQ FOR 2; GROUP 1-10 WITHIN 30H ONBOARD ACQ FOR 3-10; CALIB FOR 4 NO SLEW; SEQ 3-4 NO GAP; CYCLE 3 CALIB FOR 6 NO SLEW; SEQ 5-6 NO GAP; CALIB FOR 8 NO SLEW; SEQ 7-8 NO GAP CALIB FOR 10 NO SLEW; SEQ 9-10 NO GAP ------------------------------------------------------------------------------------------------------------------------------------ Spectral Elements MIRROR-A2 G160M ------------------------------------------------------------------------------------------------------------------------------------ Target Names HD209750 ALPHA-AQR GSC-5224-1806 ------------------------------------------------------------------------------------------------------------------------------------