5.1. 3-200/xm spectral energy distribution
The spectral energy distribution of the zodiacal light is dominated by scattered sunlight at A < 3.5 /xm and by thermal emission from dust grains at longer wavelengths. The thermal spectrum, which is of primary interest for ISO, cannot be observed from the ground. Starting in 1971 , broad-band photometry from rockets and balloons were performed in order to define the general shape of the zodiacal light in the thermal infrared [31-33]. The most extensive data sets are derived from the satellite missions IRAS and COBE [4,5].
ISO observed the spectrum of the infrared sky brightness by performing multi-filter photometry with ISOCAM at A < 16 /xm and by ISOPHOT at wavelengths up to 200/xm. In Fig. 7 we plot the broad band spectrum measured towards a dark position at A —A0 = 90° and ¡3 — 0° using the absolute photometric mode of ISOPHOT and the calibration in PIA V7. The mid-infrared ISOPHOT-S spectrum measured at the same position and date is overplotted with a solid line. The discrepancy between the DIRBE data, the ISOPHOT-S spectrum and the broad-band photometry reflects the unot yet finalised status of ISOPHOT (PIA V7) extended source calibration. Nevertheless, the figure reveals the main advantage of the ISOPHOT zodiacal light observations: the good filter coverage at the most interesting mid-infrared and A > 100/xm wavelengths, including mid-infrared spectrophotometry. With this spectral information one can try to decompose the main components of the infrared sky: zodiacal light, galactic cirrus, and extragalactic background light using their different spectral characteristics.
One of the most important contributions of ISO to zodiacal light studies is the mid-infrared (6-16 /xm) spectrophotometry, where the brightness of the zodiacal light increases by two orders of magnitude. Before the ISO mission only one rocket measurement of the mid-infrared spectrum had been attempted . Reach et al.  observed the spectrum at one sky position with the ISOCAM circular variable filters from 5-16 /xm. They found that the spectrum is remarkably well-fit by a Planck curve of T=262 K, and that there are no indications of sharp lines. They also found a good agreement with the COBE/DIRBE broad-band photometry at 4.9 and 12^m.
With ISOPHOT we also observed the mid-infrared spectrum, although the spectral coverage of the spectrophotometer ISOPHOT-S is limited to the slightly smaller 5.9-11.7 ¿zm range. We collected 27 high quality ISOPHOT-S measurements distributed over the sky region accessible by ISO (Fig. 8a). Some positions are identical with those where ISOPHOT multi-filter photometry was also performed. We defined six sky segments by considering the symmetry of the zodiacal light (outlined by long-dashed lines in Fig. 8a), and created template spectra by averaging the individual spectra of each segment. The six templates are plotted on the left hand side of Fig. 9. The templates can be well approximated by Planck curves whose temperatures show systematic variations along the sky: increasing temperature with the ecliptic latitude and decreasing temperature with the distance from the sun (see Figure 86). The residuals of the blackbody fits, plotted on the right hand side of Fig. 9, do not show any definite spectral features like PAHs or silicate bands. The lack of a strong 10/xm feature in the observed spectrum, similar to in
Was this article helpful?