Giant Planet Spectra

6.7.1 General features of giant planet spectra: UV to microwave

At UV wavelengths, the atmospheres of the giant planets are optically thick due to Rayleigh scattering. Hence, at these wavelengths most of the light we see is Rayleigh-scattered sunlight, modified by the absorption of high-altitude haze layers. Towards the poles, auroral glow is also seen, especially for Jupiter. Superimposed on this spectrum are the absorption features of several gases that suffer photolysis in the upper atmosphere (discussed in Chapter 4). As can be seen in Table 4.4, as the wavelength increases, Rayleigh scattering rapidly becomes less important and thus at visible wavelengths sunlight may penetrate to, and be reflected from, the deeper cloud layers at several bar pressures. Towards the red end of the visible spectrum, weak vibration-rotation bands of methane (and also ammonia) appear, which become increasingly strong in the near-infrared. In the center of these bands, the atmosphere is optically thick and thus any light that is detected must be reflected from the upper cloud and haze layers of the atmosphere. Between the bands the atmosphere is optically thin and thus sunlight may be reflected from both upper and deeper cloud layers. Hence, analysis of the near-IR reflection spectrum is very important in determining the vertical cloud structure of the giant planets. The solar spectrum diminishes at longer wavelengths, and thus in the mid-infrared the spectrum is dominated by thermal emission from the atmosphere itself, modulated by the presence of numerous vibration-rotation (Section 6.3.4) absorption bands of several molecules. In the far-infrared the thermal emission spectra of the giant planets becomes dominated by collision-induced H2-H2 and H2-He absorption (Section 6.3.6) together with the rotational bands of several molecules (Section 6.3.3). At submillimeter and microwave wavelengths these sources of absorption become increasingly weak and thus thermal emission from the deep levels of the atmospheres may be detected at wavelengths other than near 1.3 cm, where ammonia has an inversion band (Section 6.3.5).

6.7.2 Near-IR and visible reflectance spectra

Estimates of the disk-averaged geometric albedos of the giant planets are shown in Figure 6.15. Data below 1 ^m is taken directly from the ground-based observations of Karkoschka (1994, 1998a). Above 1 ^m, data come from a variety of sources, which are listed in the figure. These spectra are formed by sunlight scattering off aerosol particles at different altitudes in the atmosphere. Sunlight scattering from deep clouds passes through a longer path of methane (and for Jupiter and Saturn, ammonia) before reaching the observer than sunlight scattering from haze layers in the upper troposphere. In the near-infrared the absorption of methane (and ammonia) is significant and thus only wavelengths close to the methane absorption minima of 1.05 ^m, 1.3 ^m, 1.6 ^m, 2.0 ^m, and 3.0 ^m may be scattered from deeper clouds, leading to a series of narrow reflection peaks at these wavelengths. Sunlight scattering of higher clouds suffers less methane absorption and thus the reflection peaks are broader, while sunlight scattering off upper-tropospheric and lower-stratospheric hazes suffers very little absorption and thus has very broad reflection peaks. Hence, by analyzing the shape of the observed reflection spectra, we can deduce the vertical cloud structure of the giant planet atmospheres. In addition, the particle size of clouds may be estimated from analyzing how the reflectivity of cloud layers varies with wavelength since we saw in Section 6.5.2 that the extinction cross-section of particles tends to zero at long wavelengths at different rates, depending on the particle size.

The main absorption bands of methane (shown previously in Figure 6.6) are clearly seen in Figure 6.15 for all four giant planets. At wavelengths greater than 1 ^m, the shape of the spectra tell us that the reflection spectra of Jupiter and Saturn are formed from reflections from a vertically extended cloud system, since a combination of reflections from deep, middle, and high cloud layers is required to simulate the observed spectra. In contrast, the reflection spectra of Uranus and Neptune have narrow peaks near the main methane passbands indicating that the dominant reflection comes from the main clouds at bar, although some sunlight is also reflected from the tropospheric and stratospheric hazes, particularly for Neptune. Towards the visible, the absorption of methane rapidly decreases, and the reflectivity of the small

Jupiter

Jupiter

WovclcngLh (gUIl)

WovclcngLh (gUIl)

Soturri

WovclcngLh (gUIl)

WovclcngLh (gUIl)

WovclcngLh (gUIl)

Neptune

WovclcngLh (gUIl)

Neptune

WovclcngLh fyn)

Figure 6.15. Measured and calculated geometric albedo spectra of the giant planets. Below 1 ^m, the spectra are those measured from the ground-based observations of Karkoschka (1994, 1998). The spectra above 1 ^m are from the following sources: Jupiter, Ga/i'/eo NIMS real-time spectra (Irwin et a/., 1998); Saturn, Cassini VIMS typical spectrum (Roos-Serote, 2008); Uranus, ground-based UKIRT observation (Irwin et a/., 2007); and Neptune, ground-based UKIRT observation (Irwin et a/.—unpublished).

WovclcngLh fyn)

Figure 6.15. Measured and calculated geometric albedo spectra of the giant planets. Below 1 ^m, the spectra are those measured from the ground-based observations of Karkoschka (1994, 1998). The spectra above 1 ^m are from the following sources: Jupiter, Ga/i'/eo NIMS real-time spectra (Irwin et a/., 1998); Saturn, Cassini VIMS typical spectrum (Roos-Serote, 2008); Uranus, ground-based UKIRT observation (Irwin et a/., 2007); and Neptune, ground-based UKIRT observation (Irwin et a/.—unpublished).

particles found in the tropospheric hazes increases leading to the increase in albedo observed for all four planets. The visible spectra of both ice giants can be seen to be significantly weighted towards the blue end, especially Neptune, which gives these planets their characteristic blue colors. Although this coloration arises partly from the red-absorbing nature of methane, which can be clearly seen, it is found that in addition the main cloud deck at bar must also be significantly red-absorbing, as was mentioned in Chapter 4.

The variation of reflectivity of a planet across the visible disk may also tell us something about the scattering properties and vertical distribution of cloud particles. The least ambiguous definition of reflectivity is the bidirectiona/ re^ectivitj /Mnction, or BDRF, defined as where I is the measured reflected radiance; m0 is the cosine of the solar zenith angle; F0 is the solar flux at the distance of the Earth; and D is the distance of the planet from the Sun in AU (Hanel et al., 2003). This formulation correctly approximates the fact that the flux of sunlight arriving per unit area of a horizontal surface depends on the cosine of the zenith angle. For a Lambertian reflecting surface (where the BDRF is the same in all directions and equal to a constant RL), Equation (6.77) may be rearranged as I = Rlm0F0/(^D2) or I = I0M0, where I0 is the maximum reflected radiance viewed when the Sun is directly overhead (m0 = 1). Hence, for a single Lambertian cloud layer, the observed reflectivity is expected to be limb-darkened with reflectivity decreasing towards the limb of the planet as m0 tends to zero. A cloud of particles which is optically thick and thus multiply-scattering is found to approximate well to a Lambertian reflector regardless of the phase functions of individual scatterers. However, the extended vertical distribution of clouds in the giant planets together with significant gaseous atmospheric absorption means that scattering conditions approach the single scattering limit at some wavelengths and significant departures from this simple limb-darkening rule are then observed. A common approximation to the limb-darkening curve for non-Lambertian cases is the semi-empirical Minnaert limb-darkening equation (Minnaert, 1941)

Was this article helpful?

0 0

Post a comment