New telescope imaging techniques improve the observation of ejecting stars

Stars and planets are formed by the gravitational collapse of interstellar clouds of gas and dust. When a cloud collapses, it spins up an equatorial disk around each forming star. As both the star and planets accrete from the disk, the star's magnetic field captures material from the inner disk. Some is ejected along its polar axis as a bipolar jet.

For extremely young stars—less than than 100,000 years old, and still embedded in their nascent gaseous cocoons—these jets are well known and called Herbig-Haro objects.¹ It is only recently that they have been found around stars older than a million years, when they have emerged from their cocoons and are clearing the disks around them.

We recently contributed to this breakthrough. Using the Hubble Space Telescope's Imaging Spectrograph² (STIS) in its coronagraphic mode, we were searching for the disk of a six million year-old star, performing visible and UV spectroscopy on the obtained images.^3,4 It was during this observation cycle that we unexpectedly saw an emerging jet. This was the first time one had been observed from a star this old. The phenomenon was only visible because we used a novel tunable imaging technique. While the Hubble Space Telescope (HST) can see disks and some jets using broadband imaging, the jets can be distinguished more clearly by exploiting their emission line spectrum: enhanced contrast can be obtained against sky and detector backgrounds, even while observing from the ground.

In our work, we are using the Goddard Fabry-Perot (FP) imager of the Apache Point Observatory's 3.5m telescope. Combining its tunable narrow band imaging with a coronagraph enables the observation of faint emission line objects close to a bright star.

We do this by tuning the FP imager to strong emission lines such as the 656.3nm Ha or 671.7nm [SII] lines, and then subtract off-line from on-line images. This procedure reveals the jets, which typically consist of quasi-periodic clumpy structures, as shown in Figure 1. We don't yet understand the clumping mechanism. Possible explanations include instabilities in the inner disk due to a planet with an elliptical orbit; spiral waves; or the star's magnetic cycle.

We can discriminate further by offsetting the wavelength to the Doppler shifts of the jet. Since the oncoming jet is blueshifted, and the receding jet is redshifted, this allows the orientation and the speed of the jet to be determined (typically 100–400km/s), as shown in Figure 2. Such tunable imaging is not yet available on the HST, nor on any other astronomical space-based imager.

Observing the same system from year to year, we can see brighter structures in the jets move out from the stars, and measure their movements. Assuming that the jet is polar and the disk equatorial, combining this proper motion with the distance to the stars enables us to solve for the complete orientation in space of the jet system, and consequently of the disk. This orientation information is useful for interpreting HST and other space observations, such as that in x-ray data from Chandra and IR data from Spitzer.

While there are more stars to be explored with the current instruments, we hope to be able to probe deeper by building additional capabilities into the coronagraphic FP imager. We have been exploring the requirements of future space missions for imaging and characterizing planets around other stars, and for measuring faint supernovae to refine our measurements of the acceleration of the universe.

Some of our developments include integral field spectrographs with two spatial and one spectral dimension, and the photon-counting imaging detectors needed to exploit the low sky backgrounds in space. Observing in three dimensions allows us to detect all the photons in all the wavelengths within range—over the jet velocity range and for several emission lines at once—in all directions surrounding the stars. This deepens the observation for a given observing time. Simultaneous observations also ensures that different wavelengths and image points are acquired under identical atmospheric conditions.

We plan to focus the telescope image on a microlens array, following the Tiger design by Bacon et al.,⁵ and that of Larkin et al. for the Osiris instrument on the Keck telescope.⁶ Here, each microlens produces a small focal spot with dark space in between, into which short spectra are dispersed.

To enhance the high spatial contrast required for coronagraphy, we will insert a pinhole array at the focus of the microlens array, and for the coronagraphic occulting spot, we will block several of the lenslets and pinholes. The pinhole array will form the entrance aperture to the spectrograph. We plan to switch between the FP imaging mode and the integral field spectroscopy (IFS) mode, depending on whether a wide field or deep spectroscopy of a small area is needed.

Finally, we believe that the same technologies that enable the detection of faint jets and the closer observation of stars—thus extending the age and mass range of the stars whose circumstellar regions can be probe—will also significantly improve the imaging capabilities of future NASA flight missions.