Astron. Astrophys. 361, 407-414 (2000)

1. Introduction

In order to understand the history of galaxy formation and evolution, surveys at far-infrared (FIR) and submillimeter wavelengths are essentially important, since a large portion of star formation activity in the universe may be hidden by dust, prohibiting optical and near-infrared studies due to the enormous extinction. The recent detection of the far-infrared Cosmic Infrared Background (CIB) radiation by COBE, which appears to have comparable brightness to the total intensity in deep optical counts from the Hubble deep field (Fixen et al. 1998; Hauser et al. 1998; Lagache et al. 2000), indicates that the infrared-bright galaxies are responsible for roughly half of the energy released by nucleosynthesis. Various discrete source surveys are also being pursued from mid-infrared to submillimeter wavelengths. For example, source counts at 15 µm (Aussel et al. 1998; Altieri et al. 1998; Elbaz et al. 1999), at 170 µm (Kawara et al. 1998 (Paper I); Puget et al. 1999), and at 850 µm (Blain et al. 1999) have been reported. Using both CIB and these new number count data, modeling of the cosmic star formation history has been attempted by many authors (Guiderdoni et al. 1998; Dwek et al. 1998; Rowan-Robinson 1999; Tan et al. 1999; Ishii et al. 1999). All of them require a strong evolution in the star formation rate as we look back to high redshift (more than 10 times larger at ).

Although the FIR deep survey is now considered a key observing method for the exploration of the "optically dark side" of the star formation history of galaxies (Guiderdoni et al. 1997), the detectivity of 1m-class space FIR telescopes is likely to be limited by the noise due to the fluctuation of the IR cirrus (Low et al. 1984), emission from the interstellar dust, even at high Galactic latitude (Helou & Beichman 1990). Hence a detailed study of the IR cirrus fluctuations is highly important, especially for the planning of the deep surveys intended with forthcoming IR space telescopes such as ASTRO-F(IRIS) (Murakami 1998) and SIRTF (Fanson et al. 1998).

The spatial structure of the IR cirrus at 100 µm as measured by IRAS was extensively studied by Gautier et al. (1992) and Abergel et al. (1996). From a Fourier analysis of the brightness distribution, Gautier et al. found that the Power Spectral Density (PSD) of the brightness fluctuation at 100 µm follows a power-law function of the spatial frequency with an index of about 3 below the spatial frequency corresponding to the IRAS beam size (about ). They also found that the PSD is proportional to , where is the mean brightness of the IR cirrus. The ISOPHOT (Lemke et al. 1996) onboard the ISO (Kessler et al. 1996) is capable of observing the IR cirrus at wavelengths longer than 100 µm, with a better spatial resolution than that of IRAS. Herbstmeier et al. (1998) analyzed the spatial characteristics of four fields measured by ISOPHOT, and obtained similar power-law spectra for relatively bright cirrus regions. Recently Lagache & Puget (2000) have made a power spectrum analysis of the 170 µm image of the Marano 1 field, and found a significant excess at , which they attributed to the fluctuations due to unresolved extra-galactic sources.

In this paper we investigate the characteristics of FIR brightness fluctuations in the Lockman Hole, a region with a uniquely low HI column density (Lockman et al. 1986), and thus minimal IR cirrus contribution to the fluctuations. Therefore the fluctuations in the Lockman Hole are likely to be dominated by faint, distant galaxies (Helou & Beichman 1990; Herbstmeier et al. 1998) and the fluctuation analysis of FIR images will provide unique information on the number counts of infrared galaxies even below the source-confusion limit, thus constraining the parameters characterizing the number count models.

This paper is organized as follows: Sect. 2 briefly describes the observations and data processing. Sect. 3 explains the power spectrum analysis and examines the contribution by the IR cirrus fluctuations and also describes a simulation of the images and the PSDs. Sect. 4 describes the constraints on the galaxy number counts. The nature of the sources responsible for the fluctuations is discussed in Sect. 5, and Sect. 6 gives the conclusions.

© European Southern Observatory (ESO) 2000

Online publication: October 2, 2000
helpdesk.link@springer.de