Paul J Francis , Greg M. Wilson , Bruce E. Woodgate, PASA, 18 (1), in press.
Next Section: Conclusions
Title/Abstract Page: On the Gas Surrounding
Previous Section: The Nature of the
Diffuse Ly
Emission
How can we observationally determine whether neutral gas structures like the ones hypothesised in Section 3 exist? One approach would be to find larger samples of high redshift galaxy clusters with background QSOs, but this would be enormously expensive in telescope time. Another approach is to search for diffuse Ly emission from the hypothesised neutral gas.
Even in the absence of photoionisation from young stars and/or AGN, the strong UV background in the high redshift universe should photoionise the outer layers of any neutral gas clouds in the early universe, and hence induce Ly emission. This emission has been modelled by Binette et al. (1993) and by Gould & Weinberg (1996) (see also Hogan & Weymann 1987). Predicted Ly surface brightnesses are very low: typically
or less. Nonetheless, if the emission covers a large area on the sky, the sensitivity can be increased by 
, where a is the area integrated over. This approach has been applied to emission from individual Lyman limit systems, and stringent upper limits placed on the surface brightness (Bunker, Marleau & Graham 1998).
We used results from these papers, together with our own modelling using MAPPINGS II, to predict the diffuse Ly flux from the hypothesised neutral gas structure around the 2142-4420 cluster. The biggest sources of uncertainty are the strength of the UV ionising background at this redshift (literature values disagree by a factor of 
), the spectrum of the UV background, and the geometry of the structure (which can also introduce a factor of uncertainty into the predicted diffuse flux). All predictions ignore dust obscuration: if any dust is present, then the Ly fluxes will probably be reduced by a large factor. We predicted that the diffuse flux per unit area F should lie in the range
.
Observations
We have made two attempts to detect diffuse Ly emission from the hypothesised neutral gas structure surrounding the 2142-4420 cluster.
The first attempt used the Double Beam Spectrograph on the Siding Spring 2.3m telescope. A long-slit spectrum was taken, with the slit centred on QSO 2139-4434, and the slit angle positioned so that it included QSO 2139-4433. The slit was 2
wide and 3
long, giving a total sky coverage of 360 square arcsec. Spectral resolution (2 pixels) was 1.1Å. 1800 second exposures of the cluster field were alternated with equal length exposures on randomly chosen control fields one degree away (in random directions).
A total on-field exposure time of 23,400 seconds was obtained, in dark photometric conditions, on the nights of 9th and 10th July 1999. The total control field exposure was identical. The night sky spectrum was found to be very stable throughout both nights, so on- and off-field exposures were combined, using inverse variance weighting to minimise sky noise. The night sky brightness at around 4110Å (the expected wavelength of Ly at the cluster redshift) was
.
No significant excess flux was seen at wavelengths of around 411 nm in the on-field image as compared to the off-field image. This enables us to place a 
upper limit on the diffuse emission from the cluster halo, if it fills the slit, of
.
The second attempted measurement was made with the Taurus Tunable Filter (TTF), on the Anglo-Australian Telescope. TTF is an imaging Fabry-Perot etalon system, with a resolution at 4100Å of down to 4Å, and a monochromatic field of view approximately 5 in diameter. All observations were taken in photometric conditions on the nights of 11th and 12th August 1999.
Charge shuffling was used to alternate, every 60 seconds, between images of the cluster field at the expected wavelength of Ly, and at a control wavelength 50 Å to the red. Observations were made at 4Å resolution, with central wavelengths of 4110 Å and 4114 Å. On-wavelength exposure times were 3000 and 3600 seconds respectively. The field of view was centred on the brightest of the Ly emitting galaxies: at 21:42:27.48-44:20:28.4 (J2000).
The difference images (on-wavelength minus off-wavelength) were very clean, with all continuum sources removed to high precision. These images were aligned and co-added, using inverse variance weighting. As no control fields were observed, we have no sensitivity to diffuse flux filling the field of view. We are sensitive to variations in the diffuse flux on smaller angular scales. On scales of
, we see no regions of excess flux, to a upper limit of
, at both wavelengths.
Summary
We failed to detect diffuse emission from any neutral gas around the 2142-4420 cluster. Our observations were not, however, deep enough to rule out the presence of even dust-free neutral gas. They do, however, demonstrate the darkness and stability of the sky background at these blue wavelengths.
Next Section: Conclusions
Title/Abstract Page: On the Gas Surrounding
Previous Section: The Nature of the
© Copyright Astronomical Society of Australia 1997