4.3 Luminosity

The principal energy source of PGs is the release of binding energy in nuclear reactions at the centres of hot young stars. A crude estimate of the luminosity of a young galaxy can be obtained by simply considering the energy required to convert a fraction Z of hydrogen into metals, based on the metalicities of old stars in the Milky Way, for which Z 0.01. A PG of mass 10¹¹ M, with a star-forming phase lasting 1 Gyr (equivalent to a SFR of about 100 M yr^-1) has a net bolometric luminosity 6 x 10³⁷ W. This makes PGs 10-100 times brighter than nearby bright spiral galaxies in which stars are being formed at a rate of a few M yr^-1. At z = 5 this corresponds to an optical (4500 Å) flux of 10^-18 W m^-2 - at least 100 times fainter than the terrestrial night-sky background at these wavelengths; while in the near-infrared (1-5 µm) the Earth's atmosphere is ~ 1000 times brighter than at optical wavelengths, making PGs at least 10⁵ fainter than the sky! For these reasons PGs are unlikely to be masquerading in existing catalogues of bright galaxies and can only hope to be detected by the largest telescopes.

4.4 Spectral Energy Distribution

In recent years significant progress has been made in reliably modelling the evolution of stellar populations in galaxies (e.g. Charlot et al. 1996). The basic approach, known as ``isochrone synthesis'', involves computing the evolutionary tracks of stellar populations for an instantaneous star burst (i.e. with no age dispersion) incorporating a finite rate of star formation. In this way the distribution of stars of various masses and ages can be modelled smoothly with time and isochrone synthesis models are found to reproduce well the observations of stellar populations in nearby galaxies if they formed more than a few Gyrs ago. This technique has been widely used to predict the major spectral characteristics of young galaxies 1 Gyr after formation. Figure 3 shows the resultant spectral energy distributions for a galaxy forming stars at a rate of 1 M yr^-1 observed after ellapsed periods ranging from 4 x 10⁷ yrs to 10¹⁰ yrs. The essential characteristic of Figure 3 is that genuinely young galaxies radiate an almost constant energy density from 0.1-2.0 µm. The rise in the relative emission at longer wavelengths as the stellar population ages is due to the death of massive hot stars, whose lifetime (t) varies with mass (M) as t (M/M)^-2 x 10¹⁰ yrs, which evolve into cooler red giant stars and supernovae. One important consequence of the flat spectral energy distribution is that the measured optical or near-infrared flux of a PG is a direct measure of the instantaneous SFR within the system.

Figure 3. The spectral energy distribution expressed as power per unit frequency interval computed from stellar population synthesis code (see text) for a galaxy with mass 1011 M and a constant SFR of 1 M yr-1 observed after a time (yrs): 4 x 107 (bottom curve), 108, 3 x 108, 5 x 108, 6 x 108, 109, 2 x 109, 3 x 109, 5 x 109, 8 x 109, 1010 (top curve). In these models Lyman at 1215 Å is shown in absorption. No extinction due to dust has been assumed (see Figure 5).

In addition to the stellar continuum emission, young star forming regions within the Milky Way and other nearby galaxies exhibit intense line radiation, principally the hydrogen recombination lines of Lyman at 1215 Å and the Balmer line H at 6562 Å. The Lyman line is shown in absorption in Figure 3, but in star-forming regions these lines are seen in emission associated with the ionised hydrogen in the surrounding gas which constitutes the star-forming nebulae. The intensity of these lines is a measure of the ambient ionizing UV flux from young hot stars and can also be used as a tracer of star formation to estimate the SFR in an independent way from the spectral energy distribution (Kennicutt et al. 1987).

In conclusion, although predictions are sketchy, there are certain characteristics which PGs are likely to have that one can highlight: PGs should not be rare objects and can be identified in the optical or near-infrared by their flat continuum emission or intense line radiation. On the other hand, despite being intrinsically luminous objects, at least compared to normal galaxies, the likely formation redshift of PGs dictates that they are almost certainly faint and possibly of low-surface brightness, making them very difficult to detect.