Current developments and future plans (Zodiacal Dust Cloud)

In this topic, I will briefly review the experiments currently planned, or in operation, in Solar System Astronomy: rocket-borne, space-borne, and ground-based observations, which are likely to throw new light on the structure and nature of the Sun’s dust cloud. I will also touch on present unresolved issues which warrant attention, and sketch plans for future observations which Dr. Garik Israelian and I plan to make, to extend and refine the Doppler radial velocity measurement technique. We hope to map dust kinematics to add new detail to the latest picture of the Zodiacal Cloud, as revealed by IRAS, COBE and ISO.

Current Zodiacal Light research CIBER

The Cosmic Infrared Background Experiment (CIBER) (Bock et al, 2006), is a rocket-borne instrument designed to search for signatures of primordial galaxy and star formation in the cosmic near-infrared extra-galactic background, in the wavelength region 0.8 to 2 microns. Its launch, funded by NASA, is scheduled for September 2007. It is hoped that CIBER will settle the question of whether the DIRB really is the echo of the first stars and galaxies. CIBER consists of a wide-field two-colour camera, a low-resolution absolute spectrometer, and a high resolution narrow band imaging spectrometer. It will trace the Zodiacal Light with the low resolution spectrometer, using the intensity of scattered Fraunhofer lines, to provide an independent measurement of Zodiacal emission, and a new check on the Kelsall et al Zodiacal dust models, which were used to compare with the results of the Diffuse Infrared Background Experiment (DIRBE) of the Cosmic Background Explorer (COBE). It is always necessary to subtract an accurate estimate of the intensity of ZL emission in order to map the Galactic contribution and the level of emission from elsewhere – the Infrared Background. Models of the levels of ZL contribution are still so uncertain that estimates of the IRB are, at present, tentative. Being essentially a low resolution experiment, CIBER will not be able to do spectroscopy, as such, of the ZL. But the CIBER team hope to accurately assess the contribution of local (z = 1-3) galaxies, and hence to determine the excess unaccounted for. In this residual radiation, CIBER will be searching for the newly-predicted signature of high luminosity primordial stars and mini-quasars – a red-shifted Lyman cut-off feature between 0.8 and 2 microns.


It will be interesting to see what view the Planck experiment, due for launch in 2008, will have of the Zodiacal Light. The satellite will be put into a Lissajous orbit around the Sun-Earth libration point L2, approximately 1 AU from the Earth, which will give it a significantly different perspective on the Zodiacal Cloud. And, ‘seeing’ at wavelengths over 300 microns, it is possible that the ZL will look very different in shape. Planck will construct all-sky maps in nine frequency bands, covering the range 30 GHz to 857 GHz, with relatively high sensitivity and resolution (Maris et al, 2006). According to the Planck ‘Bluebook’ (ESA, 2005), Planck is expected to detect Zodiacal Light at 857 GHz and possibly 545 GHzd; at these wavelengths the Zodiacal emission is, until now, poorly constrained by observations, and its level uncertain by more than a factor of two. At lower frequencies than this, levels of emission from the ZL are expected to be very low. It is hoped that accurate mapping of the dust emission in this wavelength region will give information on the geometrical distribution of the largest (up to 1 cm) grains in the dust cloud, in three dimensions, and as a function of season. The detection of thermal emission at wavelengths as long as this makes Planck sensitive to dust colder than has been previously mapped; finding such emission could imply the existence of an extension of the Zodiacal Dust Cloud into the outer reaches of the Solar System, perhaps into the Kuiper Belt. Of course, even if the cloud only extends to the asteroid belt, there would be Rayleigh-Jeans radiation from warmer dust at 350 microns. The signature of colder, more distant dust would be an enhanced intensity of submil-limetre emission. The possibility of Planck detecting the IRAS dust bands does not appear to have been discussed; however it is hoped that it will be able to map the cometary dust trails reported by IRAS.


The Spitzer orbiting infrared telescope was named after the first astronomer (Lyman Spitzer in 1946) to suggest putting a telescope in space. It has a sensitivity 1,000 times better than ISO, and has full wavelength coverage in the mid-IR (Werner et al, 2004). The "First Look Survey" of Spitzer (FLS) has investigated the mid-IR Background brightness in the ecliptic plane. The sensitivity of Spitzer is background-limited, and dominated by Zodiacal Light at all but the longest wavelengths and lowest Galactic latitudes. Spitzer carries a short wavelength Infrared Array Camera (IRAC), which provides absolute photometry, relative to the brightness of a dark sky patch, at a range of four wavelengths spanning 3.6 to 8 microns. This instrument is sensitive to emission from evolved galactic systems, and has produced many beautiful multi-’colour’ images of many nearby galaxies. The 24 micron long-wavelength Multiband Imaging Photometer for Spitzer (MIPS) has provided absolute photometry of ZL emission. The measured ZL background was found to be within 5 per cent of Spitzer model predictions at 24 microns derived from COBE/DIRBE data. The total background brightness at 8 microns was 36 per cent brighter than the model predictions. However, this is thought to be consistent with known intrinsic variations of the ZL as a function of viewing geometry (Meadows et al 1997).

Recent reports of analysis of Spitzer data (Reach 2005) indicate that it is indeed achieving success in the pursuit of refinement in the known structure of the Zodiacal Cloud. They say, "Using the new infrared spectral and imaging capabilities of the Spitzer space telescope, we find dust trails from short-period comets, new structures in the zodiacal light, weak 9-11 micron silicate spectral features from some comets and the zodiacal light, and the color temperature of cometary nuclei, comae, and dust trails. The high frequency of dust trail detections for short-period comets suggests they are a generic phenomenon and that most such comets lose mass primarily in the form of meteor-sized particles. The color temperature of the trails is higher than a grey-body, while their dynamics suggest they must be of order mm-sized, so they must maintain temperature gradients across their small surfaces. The zodiacal light silicate feature requires a contribution from small amorphous silicates, though the continuum requires larger particles." This quote has direct relevance to the work I hope to undertake in the coming year.

Objectives of our proposed new work in Zodiacal Light

Doppler measurements on scattered light at optical wavelengths

Dr. Garik Israelian (of the Institute of Astrophysics, La Laguna, Tenerife) and I believe it would be very instructive to look at Doppler shifts in regions away from the ecliptic plane, and see if we can detect a higher proportion of retrograde material. We have seen that high radial velocities might also be an indication of a significant Interstellar component. If we can obtain good enough quality spectra in a slice perpendicular to the ecliptic, including the ecliptic pole and the Gegenschein, it should be possible to choose between these hypotheses.

We have seen that the IRAS observations, interpreted by Low et al (1984), indicated the existence of dust bands, producing ‘shoulders’ superposed on the smooth photometric shape of the traditionally known Zodiacal Light. We propose to look for the dust bands’ dynamic signature, by measuring Doppler shifts in a slice of the ZL across the ecliptic plane, at an intermediate elongation, again using a ground-based optical and mid-infrared spectrometer. A sudden change in radial velocity at some point in this slice would be confirmation of the existence of such a discrete component in the cloud, and would offer the first opportunity to determine its motion; this could then be compared with the motion of the putative parent asteroidal families.

As for instrumentation, with the benefit of 30 years’ hindsight, it seems to me that, as far as maximising flux is concerned, we, that is, all the F-P radial velocity observers of the time, missed the obvious. The Fabry-Perot instrument was used because of its high throughput at a relatively high resolution, but, by using the pre-monochromator, we actually threw away 3,000 A of information in the visible alone, all of which contained absorption lines shifted by the same amount, and all of which therefore potentially could have given us more information on radial velocity. In admitting only actually 6 A spectral range, we rejected 99.8 per cent of the available useful signal. In addition, we used only a small coelostat to collect light. The justification was that we wanted to look at a large area on the sky to achieve large flux-levels, and it would have been hard to find a telescope with a comparable field of view, of one degree, on the sky. But in fact larger flux levels can be obtained with large mirrors with a lesser field of view, more than compensating for this gain. In future I would like to correlate the modification of the whole spectral range available, and thus vastly improve the signal-to-noise in the observed spectra. Moreover, with modern echelle technology, there is every reason to employ the largest collector of light available to feed the spectrometer in future. These two improvements ought to allow not just comparison of an ‘average shift’ with theory, but detailed comparison of multiple spectral line shapes with theoretical models, yielding a much more thorough and accurate mapping of radial velocities. Exploratory experiments will be made in July 2007 with an echelle spectrometer attached to a 3.5 m telescope to test the validity of these arguments and see what quality of results can be obtained.

Recently, Bernstein, Freedman and Madore (2002) detected extragalactic background light using an echelle spectrograph at the 2.5 m Du Pont telescope in Las Campanas (Chile). These authors have developed a technique that uses the strength of the Zodiacal Fraunhofer lines to identify the absolute flux of the Zodiacal Light in the complex night sky spectrum. The analysis is much simpler if we are only interested in the Doppler shifts of Zodiacal Fraunhofer lines.

Dr. Israelian and myself are planning observations with the SARG spectrograph located at the Nasmyth focus of the 3.5 m Galileo telescope in La Palma, Canary Islands. SARG is a high efficiency echelle spectrograph designed for the spectral range 370 to 900 nm, and for resolution from R=29,000 up to R=164,000. Both single object and long slit (up to 30 arcsec) observing modes are possible. A dioptric camera images the cross-dispersed spectra onto a mosaic of two 2048 by 4096 CCDs. We plan to use a long slit mode (aperture 5, providing an area 1"07 by 26"7 arcsec) which provides a resolving power of 43,000.

To minimize read-out noise, we will bin the data on the chip in intervals of 4 pixels in the spatial direction, and average over the full extent of the slit in the data reduction. We will use 30 minute exposures which are expected (given the results of Bernstein et al 2002 and the Exposure Time Calculator of SARG) to provide a signal-to-noise ratio between 60 and 80 per resolution element. We shall use a cross disperser CD1 (600 g/mm, 4840 angstrom, order separation 10.7 arcsec) which provides spectra in the range 3,600-5,140 A in 50 orders. This spectral window is not severely contaminated by sky glow (Bernstein et al 2002), being in the blue part of the spectrum, and we hope to be able to observe ZL lines directly in these spectra without insurmountable problems in sky subtraction.

The solar Fraunhofer lines are clearly visible in this part of the spectrum and they will be Doppler-shifted due to the scattering by dust particles. We will use the Solar High Resolution Spectral Atlas of Kurucz et al to identify and study the Fraunhofer lines. We are not interested in absolute flux determinations. This experiment will allow us to increase the precision of Doppler measurements due to the fact that we will be observing many Fraunhofer lines simultaneously. We aim to observe the ZL at various latitudes and elongations, and will study the velocity fields of the dust bands in visible light. Our study will allow us to constrain models of the ZC, and dust band formation.

Doppler shifts of emission lines in the infrared

Investigating the ISOCAM spectra, Reach et al (1997) found a hint of a 9-11 micron feature, and proposed that the particles producing the ZL are composed of silicates similar to those found in the tail of P/Halley. The mid-infrared spectra of many comets show structured 9-11 micron emission bands, and the spectrum of the dust cloud around Beta Pictoris has been found to exhibit a comparable structure. The shape of the ZL silicate emission line appears to be similar to those of comet Halley and the cir-cumstellar dust around Beta Pictoris, but the ZL silicate feature is superposed on a relatively brighter continuum. It is very important to confirm the detection of the silicate line at 9-11 microns. Further work with Spitzer may reveal more structure in this feature. We propose to look at it from the ground.

We plan observations of the silicate feature with VISIR on the 8.2 m VLT (UT3) in Chile. Using a high resolution mode (providing resolution 25,000 at 10 microns), we will try to observe in the 8-12 micron atmospheric window. We propose to use a long slit mode: a slit width of 1 arcsec, and slit length 32.3 arcsec. The sky (airglow) lines will be subtracted by the "dithering" and "nodding" method (see for example, the Gemini Near Infrared Spectrograph website 2007).

These observations will allow us:

1) to verify the presence of the silicate emission feature.

2) to study the line profile and its variability.

3) to study the silicate feature in ZL dust bands.

4) for the first time, to attempt Doppler shift measurements on an emission line, if there is enough fine structure in the feature to allow this.

We feel that, in the future, the kinematic Doppler-shift signatures of moving dust particles will provide a complimentary source of information, alongside continuing photometry, on the structure of the Zodiacal Cloud: this will lead to more confidence not only in our understanding of interplanetary dust and its origins, but also in our assessment of the emissions from beyond – from Interstellar, Intergalactic, Cosmic, and Background sources.

Almost 35 years after my first experiences with a Fabry-Perot spectrometer aimed at the Zodiacal Light, there are still excellent reasons to continue, with ever improving techniques, to map Radial Velocities in the Zodiacal Dust Cloud.

Concluding remarks


The results of my work in 1970-4 were, as planned, the first comprehensive survey of Doppler shifts in the Zodiacal Light, covering elongations in the Ecliptic plane from 25° East to 25° West. The spectra and plots of wavelength shift versus elongation stand as a body of work which continues to be cited by those who have followed in the field. There is fair agreement between my results and those of East and Reay, and Reynolds et al, who both worked later with higher power equipment. The differences between the bodies of data still pose questions which further observations will be able to clear up: the question of whether there are variations over time in the kinematic structure of the ZL and if so, whether they are seasonal, and apparent, caused by the Earth’s journey around the Sun, perhaps through a continuous flow interstellar component of dust, or if they reflect actual structural changes in the Zodiacal Dust Cloud, perhaps due to asteroidal events or visits from comets.

My conclusions at the time were that an emission component at MgI existed, generated in the Earth’s atmosphere, that the Zodiacal Cloud consisted of dust particles of the order of 1 micron, in orbit around the Sun, mainly in a prograde sense, and that the morning-evening asymmetry we found might be due to components of dust NOT in orbit around the Sun. We found no evidence for contra-rotating dust, but at the resolution used I do not consider that the presence of a small percentage of such dust is ruled out. The recession of the dust in the area of the Gegenschein was, and is, an unsolved mystery. This effect was confirmed by East and Reay, but not seen at all by Reynolds et al. So it is not entirely clear whether it really exists or not, or if it sometimes exists! The results of Fried are not congruent with any of the other three sets of data, but apart from these, the rest of my, and our, conclusions of 1974 remain supported by all other subsequent experiments.


Often it is hard, among the wealth of technical details published on such a subject, to form, in the head, a clear picture of the physical shape of the Zodiacal Cloud – the kind of graphic visual we would like to present to our children, and those who follow us. Figure 5.1, from Stark (2006), is a rare attempt in the published literature to provide a geometrical visualisation of the structure of the dust cloud.

From Stark (2006). A geometrical representation of the 1998 DIRBE model of the Zodiacal Cloud. The red spheroid represents the smooth diffuse component (no structure attempted); the blue rings are the three (presumed asteroidal) dust bands; the green ring is the circumsolar Earth-resonant ring, showing a 'Blob' following the Earth (coloured blue).

Figure 5.1 From Stark (2006). A geometrical representation of the 1998 DIRBE model of the Zodiacal Cloud. The red spheroid represents the smooth diffuse component (no structure attempted); the blue rings are the three (presumed asteroidal) dust bands; the green ring is the circumsolar Earth-resonant ring, showing a ‘Blob’ following the Earth (coloured blue).

The figure is diagrammatical, but clearly illustrates the three dust bands, showing them as flat rings, and makes no attempt to show the ‘hollow torus’ indicated by evolutionary models.

It is highly likely that this representation will have to be updated, as new and better measurements are achieved.

The final three figures which follow are ‘artist’s impressions’. Figure 5.2 is a representation of how the consensus of Zodiacalists pictured the Sun’s dust cloud in 1970 – quite smooth and diffuse, roughly lenticular in shape, symmetrical about the ecliptic plane, its density decreasing quite quickly beyond the orbit of Mars. Figures 5.2 and 5.3 are speculative visualisations of the information currently available in 2007.

In Figure 5.3, the various currently accredited components of the dust cloud are illustrated: again a relatively smooth continuous ellipsoidal (lenticular) main cloud, perhaps spreading a little less in the direction perpendicular to the ecliptic than the Fan Model suggests, but now showing significant structure. Each inner planet has a focussing resonant effect, producing a system of circum-solar rings at the radii of the planets – and corresponding depleted areas between the planetary orbits. The innermost rocky planets also produce ‘blobs’, similar to, but smaller than the blob which has been observed following the Earth, and they have an overall effect, over time, of producing ‘lumpiness’ in the structure (not yet observed). But most spectacularly, the smooth dust cloud is now flanked by discrete inclined dust bands, originating from collisions in the asteroid belt. This visualisation (Figure 5.3) follows the interpretation of Stark (2006) in Figure 5.1, the IRAS observed photometric irregularities each being attributed to a separate narrow dust band. Some asteroids can be seen, including some of the families which have produced the dust bands, outside the orbit of Mars. There is an additional faint huge, cool dust ring, out towards the Kuiper belt, again, not yet seen! The angles of inclination of the rings are depicted schematically only.

Figure 5.3 shows an alternative and probably more plausible interpretation of the IRAS dust band discoveries. Models developed by Dermott et al (private communication) show how a single collision between large asteroids can initially produce a band of dust debris distributed along the path of the ‘parent’ bodies, similar to those shown in Figure 5.3. But over a few million years the models show the individual orbits of the particles precessing at various rates, until eventually the major axes of their elliptical orbits are spread randomly around the Sun, producing a smooth, fat toroidal distribution. The orbits of these particles may be thought of as a component in the ecliptic plane, plus a synchronous sinusoidal "North-south" movement. Hence, viewed from the Earth, which is situated inside the torus, the dust particles spend more time at the edges of the ‘doughnut’ than inside, so we see peaks in optical depth at the top and bottom edges of the torus. Evidently these peaks would be symmetrically placed above and below the ecliptic, which, in the case of the 10° IRAS bands, is certainly what is observed. Figure 5.4 shows one such torus, a highly evolved remnant of an asteroidal collision. The complete picture may require the addition of at least two more toroidal components.

Perhaps some combination of Figures 5.3 and 5.4 would truly represent the appearance of the complex circum-solar dust cloud that will be eventually photographed from a remote space craft in the future. In addition to the asteroidal debris, a comet is seen approaching the Sun, a reminder that quite possibly the major fraction of the dust is actually due to comets. Indeed, to this picture should probably be added a whole fine tracery of cometary dust trails, the ejecta which follow the orbital paths of the comets on their journeys to and from the Sun. And, just possibly, we should add a fairly uniform ‘background’ of dust flowing through the Solar System, a consequence of the Sun’s journey through interstellar space; one wonders how much of a contribution this might be making to current estimates of the Diffuse Infrared Background. If we could accurately map Doppler shifts in this residual drifting component, we would know.

Perhaps in the near future, kinematic studies of the material around us will resume their rightful place alongside photometry, in unravelling the secrets of the formation and evolution of the Zodiacal Dust Cloud.

Artist's impression of how the Zodiacal Cloud was perceived in 1970.

Figure 5.2 Artist’s impression of how the Zodiacal Cloud was perceived in 1970.


Artist's impression of the currently perceived Zodiacal Cloud, seen, enhanced, from the orbit of Uranus. The Sun is omitted.

Figure 5.3 Artist’s impression of the currently perceived Zodiacal Cloud, seen, enhanced, from the orbit of Uranus. The Sun is omitted.


Showing an impression of an evolved toroidal dust band resulting from a single asteroid collision.

Figure 5.4 Showing an impression of an evolved toroidal dust band resulting from a single asteroid collision.

Next post:

Previous post: