<1.4 x 10~9
Teanby et al. (2006)
Noll and Larson, 1991), which indicates rapid upwelling and vertical mixing. Indeed, the inferred stratospheric eddy-mixing coefficient profile is found by some studies (e.g., Atreya et al., 1999) to be much greater than that of Jupiter's. This mixing may arise from gravity waves generated in the convective troposphere. It is presumed that since CO is found, N2 may also be present in the stratosphere, although this is impossible to detect directly. However, at the time of writing, photochemical products such as HCN, which are observed in Neptune's atmosphere indicating N2, have not been observed in Saturn's atmosphere. In addition to the aforementioned disequilibrium species PH3 and CO, arsine, and germane are also observed in the Saturnian atmosphere, but at higher v.m.r.s (both are estimated to be
Figure 4.17. Observed and modeled abundance profiles in the atmosphere of Saturn.
Figure 4.17. Observed and modeled abundance profiles in the atmosphere of Saturn.
2.3 x 10~9) than seen in Jupiter's atmosphere, consistent with the generally increased abundance of heavy elements in Saturn's atmosphere than in Jupiter's. The latitudinal variation of PH3 was investigated by Fletcher et al. (2007a), who used Cassini CIRS data to show that at the 250 mbar level the abundance of phosphine is greater at the equator than at midlatitudes, suggesting upwelling at the equator, as was also seen for Jupiter by Irwin et al. (2004). This characteristic has also been seen with Cassini VIMS observations (Baines et al., 2005). At midlatitudes, Fletcher et al. (2007a) saw an anti-correlation between PH3 at 250 mbar and temperature, while polewards of 60°S, PH3 was seen to be depleted at 500 mbar, but reduced more slowly with altitude than at other latitudes, suggestive of subsidence and reduced PH3 photolysis rates in subpolar regions. Around the poles themselves (within 3°) Fletcher et al. (2008a) find the deep phosphine abundance to be significantly depleted implying strong local subsidence in the cores of polar vortices. We will return to the polar vortices of Saturn in Chapter 5.
For their analysis Fletcher et al. (2007a) assumed a phosphine profile that was well-mixed up to 600 mbar and then fell off with a defined fractional scale height above that due to UV photolysis (i.e., the scale height with which partial pressure decreases with height is set as a specified fraction of the pressure scale height). This fits well with the estimate of Bezard et al. (1984) from Voyager IRIS observations that the radiative-convective boundary (which gives a measure of the penetration depth of sunlight) on Saturn is in the 500 mbar to 600 mbar range and is also consistent with Cassini CIRS observations, which find the radiative-convective boundary to be in the 350mbar to 500mbar range, depending on latitude (Fletcher et al., 2007b). Furthermore, the assumed phosphine profile is consistent with the modeling calculations of Perez-Hoyos and Sanchez-Lavega (2006b), who find that due to haze absorption, sunlight (integrated over all wavelengths) may only penetrate to the 600 mbar level.
The variation of the para-H2 fraction, fp, with latitude in the upper troposphere was determined from Voyager IRIS measurements by Conrath et al. (1998). Considerable north/south asymmetry was seen, showing Saturn to be significantly affected by seasonal forcing. In addition, some regions with low upper tropospheric temperatures, such as at 60°S, were concluded to have low fp indicating upwelling, while other regions, having higher than average temperatures, such as 15°S, were seen to have higher f indicating downwelling. More recent estimates of the para-H2 fraction in the upper troposphere of Saturn have been determined from Cassini CIRS observations by Fletcher et al. (2007b). The ortho/para ratio is found to be approximately in equilibrium at all latitudes except for the equator, and latitudes polewards of 40°N. It is argued that at the equator this is due to rapid upwelling, which would fit with the observation of high abundances of PH3 (Fletcher et al., 2007a). Elsewhere, the high abundance of hazes in Saturn's atmosphere acts to effectively catalyze the ortho/para equilibration reaction at all latitudes except polewards of 40°N, where the abundance of larger tropospheric haze particles (which would appear to be more efficient catalysts of ortho:para equilibration) is observed to decrease (as we shall see in the next section, "Clouds and hazes''). In the stratosphere, Fouchet et al. (2003) used disk-averaged ISO/LWS observations to show that while fp had the same value at both the 1 mbar and 10 mbar levels, this value was higher than the tropopause value, but lower than the equilibrium value. They proposed that some catalysis occurs on lower stratospheric equatorial haze, which is observed to be confined below 10 mbar.
In the upper stratosphere, methane photochemistry again produces hydrocarbons, such as ethane and acetylene, whose abundances may be estimated from mid-IR spectroscopy (de Graauw et al., 1997; Greathouse et al., 2005; Moses et al., 2000; Sada et al., 2005). The mole fraction of acetylene has most recently been determined from Cassini CIRS observations (Howett et al., 2007) to be 2 x 10~7 at 2 mbar, increasing with altitude, and is seen to decrease towards the south pole by a factor of almost 2, similar to the observed distribution seen in Jupiter's stratosphere (Nixon et al., 2007). Similarly, the abundance of ethane is found to be 1.5 x 10~5 at 1 mbar and increases towards the south pole by a factor of 2.5 (Howett et al., 2007). Like Jupiter, the reason for these latitudunal differences is again believed to be due to the stratospheric circulation bringing hydrocarbon-rich air from equatorial high altitudes towards the poles and descending, with the acetylene abundance seen to drop as it decays faster than it can be enriched by this circulation.
In addition to the general micrometeoroid source of stratospheric oxygen discussed in Section 4.3.4, an additional source of stratospheric water in Saturn's atmosphere appears to be material falling from the rings onto the planet at specific latitudes (magnetically connected to rings). Although it is difficult to measure the latitudinal variation of water vapor, measurements have been made at UV wavelengths (Prange et al., 2006) and it is also possible to observe a decrease in hydrocarbon abundance at certain latitudes. This could be due to reactions between water and hydrocarbons which lead to other, so far undetected molecules. This may be an indirect signature of water from the rings.
Using an ECCM, and assuming all heavy elements are enriched to the levels similar to those previously described, Atreya et al. (1999) predict the same three main cloud decks to occur in Saturn's atmosphere as are expected in the Jovian atmosphere. However, since the Saturnian temperatures are lower than those of Jupiter the clouds are based at correspondingly deeper levels. Hence, Atreya et al. (1999) find that an aqueous-ammonia cloud (blending into a water ice cloud at higher levels) is expected to start condensing near 20 bar with a column density of 12,000 kg m ~2, followed by an ammonium hydrosulfide cloud at —6 bar with a column density of 200 kg m ~2, and an ammonia ice cloud at 1.8 bar with a column density of —100 kgm~2 (again these cloud densities are again greatly overestimated, but they do provide a guide for the relative maximum cloud thicknesses). Hence, part of the reason that the visible cloud features of Saturn have a lower contrast than those of Jupiter is that the main cloud decks of Saturn lie at considerably deeper pressures. A number of estimates of Saturn's cloud structure have been made from ground-based telescopes, HST, Voyager, and most recently Cassini observations, which are summarized in Figure 4.13. While to a first order the presumed ECCM cloud structure is roughly consistent with the measured abundances of water vapor and ammonia, observations suggest that, like Jupiter, the depletion in abundance of these volatiles occurs at deeper pressures and higher temperatures than predicted by a simple ECCM. For example, ground-based microwave observations by de Pater and Massie (1985) suggest that, globally, ammonia is depleted between 3 bar and 1.5 bar, which would push the NH3 cloud to higher altitudes and reduce its opacity. Indeed, cloud models of Saturn that are consistent with measured visible and near-infrared reflectance spectra place a thin haze layer at roughly 500mbar and a deeper optically thick cloud top at 1.5 bar (de Graauw et al., 1997). The other factor contributing to the low-contrast appearance of Saturn's belts and zones is that Saturn's pressure scale height is more than twice that of Jupiter. Assuming that the tropospheric haze has roughly the same number density in both atmospheres, the column abundance of haze above the ammonia condensation level is estimated to be 5x greater in Saturn's atmosphere than in Jupiter's leading to lower contrast of convective cloud features (Smith et al., 1981). The discrete cloud features that are observed appear to be the tops of active convection systems that push their way up into the overlying semitransparent tropospheric haze region.
Just as in Jupiter's atmosphere, molecules such as ammonia and phosphine are photolyzed near the tropopause, probably contributing to the production of upper tropospheric haze. Likewise, in the stratosphere the photolysis of methane leads to detectable levels of ethane, acetylene, and other hydrocarbon products. These products diffuse downwards, and heavier hydrocarbons may possibly condense near the tropopause, where the temperatures are lowest. The highest concentrations of these smog products might thus be expected at the subsolar latitude, which varies during the period of Saturn's orbit from 26.7°N to 26.7°S (Beebe, 1997). During the Voyager flybys, Saturn's year was just entering northern spring and thus this simple model would predict that the haze would be thickest over the equator. However, measurements by Voyager in the UV (Smith et al., 1982) found the stratospheric haze to be thickest over the North Pole, moderately thick in northern mid-equatorial latitudes and almost absent at southern midlatitudes (the South Pole was in darkness at the time). This north/south asymmetry was also observed in ground-based observations in the near-IR methane absorption bands (West et al., 1982) and would thus seem at odds with the simple photolyzed methane haze model.
Ground-based observations of the reflection spectrum of Saturn have now been made from near-infrared to visible wavelengths for almost 30 years (Karkoschka and Tomasko, 1992; Ortiz et al., 1993, 1995, 1996; West et al., 1982) and have been extended into the ultraviolet by HST (Karkoschka and Tomasko, 1993, 2005; Perez-Hoyos et al., 2005). The vertical and horizontal haze structure may be inferred from these observations as follows. At visible continuum wavelengths, reflection from clouds and hazes at all levels is seen, while in the near-IR methane absorption bands, only light reflected from the upper haze layers may be seen which appear bright against a dark background. In the UV, Rayleigh scattering from the air molecules becomes important and thus, as for the near-IR methane absorption bands, only light reflected from the upper atmosphere is seen. However, since Rayleigh scattering from the air molecules is conservative, high-altitude hazes appear dark against a bright background. Hence, images of giant planets at ultraviolet and methane absorption wavelengths should, to a first approximation, be complementary and thus any differences that are present may be used to infer particle size and absorption properties.
The results of these studies suggest that there are two distinctly different types of hazes in Saturn's atmosphere (West et al., 1983): a stratospheric and a tropospheric haze layer. The thin stratospheric haze layer is comprised of small particles (0.1—0.2 ^m), which are highly absorbing in the UV (Karkoschka and Tomasko, 1993, 2005; Munoz et al., 2004; Ortiz et al., 1996; Perez-Hoyos et al., 2005), while the tropospheric haze layer is made up of larger particles (1-2 ^m), which are optically thicker (Karkoschka and Tomasko, 1993, 2005). In some models, such as Karkoschka and Tomasko (1993), the tropospheric haze spreads between the tropopause and the expected ECCM base of the ammonia cloud at 1.8 bar, while in most recent studies, there is a clear gap between the bottom of the tropospheric haze at the radiative-convective boundary at 400 mbar to 600 mbar and the lower condensation clouds (Perez-Hoyos and Sanchez-Lavega, 2006a). This region is thought to be cleared by eddy-mixing processes and convection.
The tropospheric haze is probably associated with the ammonia cloud, but if it is, its spectral features are not clear. Kim et al. (2006) have tentatively identified a feature at 2.96 ^m, which they attribute to ammonia ice particles below 390 mbar to 460 mbar. However, Kerola et al. (1997) used lower resolution 3 ^m observations to exclude the possibility of an ammonia haze unless it lies beneath the 700 mbar level. The expected ammonia ice absorption features at 9.4 ^m and 26 ^m have not been observed by CIRS.
The tropospheric haze appears to be thickest over the Equatorial Zone (EZ) and in general seems correlated with the belt/zone structure. Karkoschka and Tomasko (2005) find that there are significant seasonal variations in the size of troposperic haze particles, with the particles being largest in summer and smallest in winter. Hence, aerosol sizes and their scattering phase functions (Section 6.5) inferred at a particular season should not be taken to be representative of Saturn's atmosphere at other seasons. Perez-Hoyos et al. (2005) find that the tropospheric haze extends from 75 mbar to 400 mbar with an optical thickness (at 814 nm) that varies from 20-40 at the equator to only 5 at the pole.
Stratospheric haze is found to be most abundant in polar regions and is highly UV-absorbing (Karkoschka and Tomasko, 2005) with its opacity appearing to be directly correlated with the level of solar irradiation (Perez-Hoyos et al., 2005). Stratospheric aerosols are thus thought to be produced directly from gas by auroral processes in the polar regions, as is the case for Jupiter. At the equator, stratospheric aerosols may derive directly from the photochemical products of methane, or they may perhaps arise through the "overshooting" of tropospheric particles. This latter interpretation is perhaps supported by the observation of Ortiz et al. (1993, 1995, 1996) that the brightness of the EZ in near-IR methane absorption bands increased dramatically from 1991 to 1992 and appeared to be decaying in 1993, suggesting an increase in the optical depth and/or height of the tropospheric haze. This period was just after the "Equatorial Disturbance" (or Great White Spot) of 1990. The 1994
Equatorial Disturbance was shown earlier in this book in Figure 1.5 and such disturbances are discussed more fully in Chapter 5. At continuum wavelengths a north/south asymmetry was observed with the southern hemisphere appearing darker at longer wavelengths, suggesting smaller particle sizes in the tropospheric haze. These latitudes had just emerged from the shadow of the rings, which may have had an effect. The distribution of clouds and hazes observed by HST in 1998 is shown in Figure 4.18 (see color section).
The nature of tropospheric haze particles is puzzling. Simplistically, one would expect these to be composed predominantly of ammonia. However, as is the case on Jupiter (except in small regions of localized vigorous upwelling), there has been no unambiguous spectroscopic identification of pure ammonia ice on Saturn. One suggestion that has been made is that, as for Jupiter, the ammonia crystals perhaps become coated with stratospheric haze material settling down from above. Alternatively, it may be that the thermal history of tropospheric particles hides their identity, or perhaps the photochemical products of ammonia and phosphine produced near the tropopause combine with ammonia ice in some way to produce a hybrid particle. However, it may simply be that the tropospheric haze layer on Saturn is not composed or based on ammonia at all since current haze models have a clear gap between the bottom of the haze layer and the top of the ammonia cloud (Perez-Hoyos and Sanchez Lavega, 2006a).
The presence of a relatively optically thick tropospheric haze layer increases the absorption of sunlight just below Saturn's tropopause and is probably responsible for the local maximum in temperature, or temperature "knee" observed in the 150mbar to 300mbar region by Voyager (Hanel et al., 1981, 1982). This is consistent with the scattering model analysis of Perez-Hoyos and Sanchez-Lavega (2006b), who conclude that the penetration level of sunlight in the 0.25-1.0 ^m range is at —250mbar. Cassini has made more precise observations of the latitudinal and vertical extent of this knee (Fletcher et al., 2007b) and has found it to be higher and less vertically extended over the equator than other latitudes, suggestive of upwelling.
Was this article helpful?