Title The Heart of Eta Carinae Pi S. M. White Time 1. Name of program and authors: The Heart of Eta Carinae S. M. White, J. Chapman and B. Koribalski 2. One short paragraph with science goal(s) Eta Carinae is a famous southern system consisting of two very massive stars in a 5.5-year-period binary orbit. The primary is a luminous blue variable (LBV) with a mass loss rate of .001 solar masses per year in a wind at 500 km/s. The secondary has never been seen, but is believed to be a very hot luminous star supplying ionizing photons to the dense LBV outflow. The binary orbit is highly elliptical, and every 5.5 years the system undergoes an eclipse in X-rays and radio emission. The millimeter emission from this system consists of many Janskies of (varying) thermal continuum from a subarcsecond-size source together with very strong masing recombination lines. Scientific goals of the millimeter continuum observations will be to resolve the heart of the binary system and study the interaction of the two components and the outflowing gas. At 3 mm wavelengths the smallest structure seems to be about 0.2 arcseconds, but in the higher ALMA bands the optically thick surface in the outflow will be smaller, particularly at the flux minimum at periastron, allowing more detailed mapping of the gas. The strongest masing recombination line is now identified with the "Weigelt blobs", relatively slowly moving dense ejecta with odd ionization states lying outside the binary orbit (by about 0.3 arcseconds) and irradiated by the hot stars. This recombination line maser is very different from the accretion disk maser in MWC 349, and measurements of transitions at different n and their changes with the 5.5 year cycle as the radiation field illuminating the gas changes will allow us to investigate physical conditions in the gas in detail. The non-masing recombination line emission is also strong and can be mapped by ALMA at high spatial resolution over a broad velocity range (hundreds of km/s), allowing us to investigate motions of gas in the outflow from the system with much better spatial resolution than is presently possible. 3. Number of sources: 1 4. Coordinates: 4.1. Eta Carinae - 10:45:03, -59:41 4.2. Moving target: no 4.3. Time critical: yes: the system has a 5.5 year cycle and both continuum and line fluxes and the spatial morphology change by large factors over this timescale. 5. Spatial scales: 5.1. Angular resolution (arcsec): .005" is the size of the binary orbit. The size of the optically thick surface is somewhat larger at lower ALMA frequencies. There is sufficient flux for ALMA to map Eta Car on all spatial scales, although it may be resolved out on the longest baselines, particularly at apastron. We re-emphasize that the relevant angular scales change with orbital phase. 5.2. Range of spatial scales/FOV (arcsec): For the core region a single pointing should be adequate. The spatial scales range from the optically thin outer Homunculus nebula, at 18 arcsec across, through the Little Homunculus, about 3 arcsec across and prominent at microwaves, down to the Weigelt blob separation of 0.3 arcsec and then the binary orbit dimension of 0.005 arcsec. 5.3. Required pointing accuracy: Mosaic should not be needed but high dynamic range imaging will be achieved with Eta Car: the source itself can be used for reference pointing, so no extreme pointing precision is required of the antennas for this project. 6. Observational setup 6.1. Single dish total power data: no 6.2. Stand-alone ACA: beneficial but not necessary 6.3. Cross-correlation of 7m ACA and 12m baseline-ALMA antennas: beneficial but not necessary 6.4. Subarrays of 12m baseline-ALMA antennas: no 7. Frequencies: 7.1. Receiver bands: all. Source is optically thick so different bands penetrate to different depths in the source. The source flux rises with frequency. 7.2. Lines and Frequencies (GHz): Continuum and recombination lines: there are at most 1 or 2 H-alpha lines per band, so for the purposes of the initial study we can limit ourselves to one per high-priority band: H39a 106.737 GHz H30a 231.901 GHz H26a 353.623 GHz H21a 662.404 GHz 7.3. Spectral resolution (km/s): The narrow masing recombination line is at +60 km/s and is about 20 km/s wide. For it, a resolution of 2 km/s is ultimately desirable. The broad non-masing recombination line from the outflowing gas is known to be over 1000 km/s wide (emission over the range from -200 to +1200 km/s has been seen in microwave observations; at 3 mm emission is seen over at least -50 to 350 km/s). 10 km/s resolution should be sufficient for it. With the initial one-quadrant correlator, mode 71 will cover 2000 km/s at 10 km/s resolution in band 6: this means that we would only have limited resolution for the narrow masing line, but it should be sufficient to separate the narrow line from the continuum and broad line in order for it to be mapped separately. 7.4. Bandwidth or spectral coverage (km/s or GHz): require 1500 km/s overall, but a narrower high-resolution window can be placed on the masing line when more correlator quadrants are available. 8. Continuum flux density: at 3 mm wavelength the flux ranges from 2.2 Jy at periastron to about 30 Jy at apastron. 8.1. Typical value (Jy): 10 Jy 8.2. Required continuum rms (Jy or K): .001 Jy 8.3. Dynamic range within image: 10000 8.4. Calibration requirements: absolute 1-3% for time variability studies repeatability 1-3% relative 1-3% 9. Line intensity: the narrow masing line ranges from 1 Jy with 20 km/s width at periastron to 15 Jy with 20 km/s width at apastron. The broad component is about 5 Jy at apastron. 9.1. Typical value (K or Jy): 5 Jy 9.2. Required rms per channel (K or Jy): .02 Jy 9.3. Spectral dynamic range: 1000 9.4. Calibration requirements: absolute 1-3% repeatability 1-3% relative 1-3% 10. Polarization: no (as far as we presently know) 10.1. Required Stokes parameters: I 10.2. Total polarized flux density (Jy): 10.3. Required polarization rms and/or dynamic range: 10.4. Polarization fidelity: 10.5. Required calibration accuracy: 11. Integration time for each observing mode/receiver setting (hr): a full track for sampling of all spatial scales to achieve high dynamic range imaging. If we alternate between the 4 high-priority bands within a single track then 10 hours per track is needed. 12. Total integration time for program (hr): for time variability 1 track every 6 months away from periastron, at 20 hours per year. For the "spectroscopic events" that will be the subject of large multiwavelength campaigns at periastron, one track every 2 months for 10 months around periastron (5.5 year orbit). 13. Comments on observing strategy : Primary target for high-spatial-resolution and high-dynamic-range capabilities of ALMA since it has a large flux in a compact region, appropriate for self-calibration. Emission over a range of spatial and flux scales requires high dynamic range. Continuum subtraction an issue for imaging of line emission.