Hi

<7.60 x 10-9

Cassini far-IR

Fouchet et a/. (2004a)

The ammonia profile is expected to have two "knees" since the abundance is first expected to be depleted at the level of the ammonium hydrosulfide cloud layer at around 1 bar to 2 bar and then to remain fixed until the formation of an ammonia ice cloud at 700 mbar to 500 mbar. Such a profile is consistent with ground-based microwave observations of the disk-averaged spectrum, where the deep ammonia abundance was estimated to be roughly 3x the solar value. However, like water, ammonia is observed to be severely depleted in 5 ^m hotspot areas, both from measurements of the near-IR and 5 ^m spectrum and also from the in situ observations of the Ga/i/eo entry probe. However, one surprising result of the probe analysis was that although the ammonia abundance was found to be severely depleted above about 2 bar, it increased rapidly with depth to a maximum value of over 5x the solar value at 19 bar. de Pater et al. (2001) have reanalyzed their ground-based disk-averaged microwave spectra in terms of this new estimated ammonia profile (assuming it to apply globally) and have found that it is also consistent with their spectra. If the estimated 5 ^m hotspot ammonia profile is really representative of the globally averaged ammonia profile, then some means of globally depleting the abundance of ammonia in the upper troposphere of the Jupiter is required. de Pater et al. (2001) suggest that more ammonia might dissolve in the aqueous ammonia cloud than is currently expected (laboratory data do not exist at Jovian temperatures and so must be extrapolated from room temperature measurements), or alternatively that the ammonia molecules combine with more than one hydrogen sulfide molecule to produce products such as ammonium sulfide (NH4)2S (where each H2S molecule combines with two NH3 molecules). Another possibility is that more ammonia may adsorb onto the solid NH4SH particles than is currently estimated. At higher altitudes, analysis of mid-IR spectra show that the ammonia abundance in the upper troposphere decreases much more rapidly with height than would be expected from saturation alone. This is due to the horizontal averaging effect, described in Section 4.3.3, and also because ammonia is photolyzed at roughly 100mbar and vertical mixing brings this ammonia-depleted air to lower altitudes. At these altitudes (—400mbar), mid-IR observations by Cassini (Achterberg et al., 2006; Irwin et al., 2004) have revealed a clear latitudinal variation in ammonia abundance, which matches well the observed belt/zone structure with moist air rising in the zones and descending in the belts. Achterberg et al. (2006) also found increased levels of ammonia over the Great Red Spot, indicating rapid upwelling. This belt/zone difference is also apparent in ground-based microwave images (de Pater, 1986; de Pater et al., 2001), shown in Figure 6.21.

Although phosphine does not condense in the Jovian atmosphere, its abundance is found to decrease rapidly above the 1 bar level due to photodissociation near the tropopause and vertical mixing. Its "deep" abundance is estimated from Cassini CIRS observations to be 2-3 times the solar value (Irwin et al., 2004), which is consistent with the observation that the enrichment of almost every other element appears to be —3-5 x the solar value. Of course, since phosphine is a disequilibrium species, the deep bulk P/H ratio may actually be higher. The abundance of phosphine from Cassini CIRS observations also appears to vary with latitude in the manner just described for ammonia and there is also, as for ammonia, an indication of increased abundance over the Great Red Spot (Irwin et al., 2004). In addition to phosphine, the other main disequilibrium species observed are germane (GeH4) and arsine (AsH3). The detected tropospheric (5-8 bar) mole fractions are approximately 6.1 x 10~10 and 1.9 x 10~10, respectively, and indicate rapid vertical uplifting. Unfortunately, it has not been possible with existing measurements to determine how the abundances of germane and arsine vary with height or latitude. Similarly, a higher than expected level of carbon monoxide has also been detected in the 5 bar to 8 bar region with a mole fraction of approximately 1 x 10~9, which requires rapid upwelling from the deep interior. However, the abundance of CO in the stratosphere has been found to be higher than that found in the troposphere, rising to perhaps 4 x 10 ~9 just above the tropopause (Bezard et al., 2002). This observation requires an external source of CO in addition to the internal source and, like Noll et al. (1997), Bezard et al. (2002) conclude that this may be formed by shock chemistry from the infall of kilometer-size to subkilometer-size Jupiter family comets. Other gases that appear to be introduced to Jupiter's atmosphere by cometary impacts include HCN and CO2, which were detected after the impact of Comet Shoemaker-Levy 9 in 1994. Observation of how the abundance of such gases has changed subsequently can reveal much about the circulation of Jupiter's atmosphere, which we will return to in Chapter 5.

Another indicator of vertical and meridional flow in the atmospheres of the giant planets is the observed para-H2 fraction, fp. The latitudinal variation of temperature and fp at roughly the 300 mbar level has been calculated from Voyager IRIS measurements by Conrath et al. (1998) for all four giant planets. For Jupiter, the upper tropospheric temperatures are found to be cool above zones and warm above belts while fp is found to be low and high, respectively, supporting the view of uplift in the zones and subsidence in the belts. At higher altitudes, in the stratosphere, Fouchet et al. (2003) used disk-averaged Infrared Space Observatory Short Wavelength Spectrometer (ISO/SWS) observations to determine the value offp at 1 mbar and 10 mbar and found that it is close to the value at the tropopause and does not vary with pressure. Fouchet et al. (2003) propose that since ortho/para conversion is mainly accomplished through the catalytic effects of aerosols, and since the Jovian stratosphere is observed to be relatively clear of aerosols, the stratospheric fp value is frozen at its tropopause value.

In the stratosphere the photolysis products of methane, ethane, and acetylene, are observed with peak abundances occurring towards the lower part of the main photolysis region between 1 ^bar and 0.1 mbar. Smog-like haze particles are probably also produced at these altitudes by further complex chemical reactions. Both the hydrocarbons and the hazes spread vertically to other pressure levels through eddy mixing. The mole fractions of ethane and acetylene are estimated in this lower region (1-10mbar) to be 4 x 10~6 and 3 x 10~8, respectively (Nixon et al., 2007). Hydrocarbon abundances decrease with increasing pressure due to vertical mixing and conversion back to methane. Nixon et al. (2007) find that at 5 mbar, the abundance of acetylene is highest at 20°N and then decreases towards both poles by a factor of ~4, while at lower altitudes (200 mbar level) its abundance changes little with latitude. In contrast, the abundance of ethane shows no north-south asymmetry, and instead of decreasing towards the poles its abundance is seen to increase by a factor of ~2. Nixon et al. (2007) argue that these observations are due to a combination of the facts that the UV production rate of these molecules decreases towards the poles and that the chemical lifetime of acetylene in the Jovian atmosphere is much shorter than that of ethane. It is argued that meridional circulation transports air from low latitudes and high altitudes, where hydrocarbons are produced, towards the poles and down to lower altitudes faster than the hydrocarbons can decay, in the case of ethane, but not fast enough in the case of acetylene.

Water vapor has been observed in Jupiter's stratosphere, as was outlined in Section 4.3.4, and is believed to have an external source.

Clouds and hazes

If we follow a parcel of Jovian air, with the assumed "deep" composition, traveling up through the atmosphere with its temperature decreasing adiabatically, three main cloud decks are calculated to form by an ECCM. Atreya et al. (1999) calculate the following cloud layers: (1) an aqueous-ammonia cloud blending into a water ice cloud at higher levels based at approximately 7 bar with a maximum column density of ^ 1,000 kg m (2) a solid ammonium hydrosulfide cloud based at 2.5 bar with a maximum column density of 22kgm~2; and (3) an ammonia ice cloud based at 0.8 bar with a maximum column density of 12kgm~2. While the cloud bases calculated by an ECCM are fairly reliable, the cloud densities are likely to greatly exceed the actual mass density of the condensed clouds since they neglect the precipitation (and thus re-evaporation) of condensed aerosols and also horizontal mixing with nearby dry air.

In reality, the Jovian cloud structure in the troposphere appears to be much more complicated than this, and a post- Voyager review was made by West et al. (1986) and a post-Cassini review made by West et al. (2004). One immediately obvious problem is that the clouds of Jupiter appear colored, while pure water, ammonium hydro-sulfide, and ammonia condensates produced in the laboratory are all pure white. Hence, the yellow-ochre appearance of Jupiter is intriguing. The origin of the colors, or chromophores, observed in the Jovian atmosphere has always been a source of

0 500 1000 1500 2000

Forward scattering (counts)

0 500 1000 1500 2000

Forward scattering (counts)

Figure 4.9. Relative cloud profile of Jupiter deduced from the Galileo probe nephelometer experiment cloud results. From Ragent et al. (1998).

much speculation. The red appearance of the Great Red Spot (GRS) is often ascribed to triclinic phosphorous P4(s) and candidates for the various yellow, red, and brown colors seen elsewhere include allotropes of sulfur or hydrocarbon "smog" particles produced by photolysis in the stratosphere, such as tholins (Cruikshank et al., 2005). However, Atreya and Wong (2005) point out that since H2S condenses well below the UV penetration depth, the existence of free sulfur at the cloud tops is unlikely. It should also be remembered that the pictures of Jupiter, and the other giant planets, that we have all become accustomed to are heavily enhanced and colour-stretched. The "true" appearance of the giant planets is much blander, and indeed it has been argued that the apparent redness of the GRS has more to do with how the human eye perceives color in different lighting conditions than with a real intrinsic redness (Young, 1985)!

For a long time, the main cloud deck responsible for observed albedo contrasts in the visible and near-IR was believed to be the ammonia ice cloud with a base at about 0.8 bar. The first spectral indication of this was made by Brooke et al. (1998), who found that the 3 ^m part of the ISO/SWS disk-averaged spectrum was well-modeled with the inclusion of ammonia ice particles. However, Irwin et al. (2001) pointed out that this would introduce features in other parts of the infrared spectrum that are not observed. Observations with Cassini CIRS (Wong et al., 2004a) in the mid-IR (10 ^m) report a more wide-area detection of ammonia ice absorption in the North Tropical Zone at latitudes between 22°N and 25°N, but only if the particles are assumed to be non-spherical. However, why such an absorption feature would be generally visible at 10 ^m and not at any other wavelength is again not clear.

Wide-area visible and near-IR measurements by the Galileo orbiter of the height of the main cloud deck reveal a structure which is significantly different from a simple ECCM calculation. Estimates from Galileo SSI observations (Banfield et al., 1998; Simon-Miller et al., 2001) place the lower cloud deck just above the 1 bar level, while Galileo NIMS observations (Irwin and Dyudina 2002; Irwin et al., 2001) place the main cloud just beneath the 1 bar level. Irwin et al. (2005) showed that part of this discrepancy was due to the SSI analysis assuming a discrete thin lower cloud, while the NIMS analysis assumed an extended cloud. Analysis of mid-IR (7.2 ^m) CIRS Jupiter observations by Matcheva et al. (2005), which also assumed an extended cloud distribution, also placed the main cloud somewhere in the 1.1 bar to 0.9 bar range. Hence, the consensus does appear to be that the main cloud deck of Jupiter is somewhere close to the 1 bar level, which is inconsistent with it being composed of ammonia ice unless the N/H ratio is significantly supersolar at these altitudes, which as we saw in the last section it does not seem to be. Instead, it has been suggested by a number of authors that this cloud might perhaps be the top of the NH4SH cloud, or that the ammonia ice particles are contaminated, perhaps by haze material, which increases their sublimation temperature.

Comparison of visible or near-IR images of Jupiter with those made at 5 ^m show a clear anti-correlation with regions that are bright in the visible and near-IR appearing dark at 5 ^m, while regions that are dark in the visible and near-IR appear bright at 5 ^m. This observation is consistent with visibly bright regions being cloudier, which thus reflect sunlight better, but blocks the escape of thermal radiation from deeper levels. Furthermore, the close anti-correlation between visible/near-IR and 5 ^m brightness suggests either that a single cloud deck is responsible for the opacity at both visible/near-IR and 5 ^m wavelengths, or that the opacity of overlying different cloud layers is closely correlated. However, analysis of the limb darkening of Galileo NIMS 5 ^m observations (Roos-Serote and Irwin, 2006) concludes that the main cloud deck responsible for the variations in opacity seen at 5 ^m must exist at pressures of less than 2 bar, suggesting that the same cloud deck is responsible for the observed opacity variations at both 5 ^m and visible/near-IR wavelengths.

The scenario where the clouds are made predominantly of contaminated ammonia ice particles would be consistent with the observed absence of ammonia ice spectral features (discounting the ISO detection at 3 ^m of Brooke et al, 1998) throughout the infrared spectrum of Jupiter (Atreya et al., 2005; Irwin et al., 2005; Kalogerakis et al., 2008), except in small, localized regions of rapid uplift (Figure 4.10, see color section), which condense pure spectrally identifiable ammonia clouds (SIACs) at high altitude (Baines et al., 2002), seen by Galileo NIMS to cover less than 1% of Jupiter. It is thus suggested that pure ammonia ice particles are only condensed in regions of rapid uplift, but that they are then quickly coated with haze materials, which mask their identity. In the stratosphere, the primary constituent of the haze is expected to be polycyclic aromatic hydrocarbons (PAHs) (Atreya and Wong, 2005), which slowly settle down through the atmosphere and then probably mix with the hydrazine haze expected to form in the upper troposphere through the photolysis of ammonia. Such a material is expected to have a grayish color and would provide an ideal coating material for the ammonia ice particles. As well as being seen in Galileo NIMS observations (Baines et al., 2002), the appearance, evolution, and disappearance of an SIAC was observed by the New Horizons spacecraft during its flyby of Jupiter in February 2007 (Reuter et al., 2007). A small white cloud with an identifiable ammonia ice signature was seen to appear at 35°S, and then expand, broaden, and disappear within the space of five Jupiter days.

At deeper levels, there is now increasing evidence for the detection of a deep water cloud at roughly 5 bar. This has been detected both by Galileo NIMS 5 ^m measurements (Nixon et al., 2001), and by visible/near-IR observations of Jovian thunderstorm clouds, which have bases in excess of 4 bar (Figure 4.11, see color section) and have clearly detected lightning activity (Banfield et al., 1998; Dyudina and Ingersoll, 2002; Gierasch et al., 2000; Little et al., 1999) shown in Figure 4.12. In addition, the possible spectral absorption of water ice has now been observed in Voyager far-IR spectra (Simon-Miller et al., 2000), which suggests that water ice particles may on occasion be lifted up to pressures less than 1 bar. Figure 4.13 shows a summary of the estimated cloud structure of Jupiter and of the other giant planets.

At higher altitudes, above the radiative-convective boundary but below the tropopause, the "modified ammonia" cloud appears to blend into the haze layers formed possibly by the dissociation products of ammonia and phosphine, and also methane haze products settling from the stratosphere. The upper tropospheric haze layers are seen mostly over the GRS and the northern edge of the Equatorial Zone (EZ) and appear much more zonally spread out than the convective clouds seen in the

18 16 14 12 10

Wrsl I nngit.irif

Figure 4.12. Galileo SSI images of a convective storm (left panel) and the associated lightning (right panels) in Jupiter's atmosphere. The left-hand image shows the dayside view of a storm cloud while the right-hand images show a close-up of lightning strikes from the storm some 2 hours later when the feature had rotated around to the nightside. The two lightning images were taken about 4 minutes apart. The dayside image was recorded at 727 nm, while the night images were recorded with the "red" filter to improve the throughput. Courtesy of NASA.

18 16 14 12 10

Wrsl I nngit.irif

Figure 4.12. Galileo SSI images of a convective storm (left panel) and the associated lightning (right panels) in Jupiter's atmosphere. The left-hand image shows the dayside view of a storm cloud while the right-hand images show a close-up of lightning strikes from the storm some 2 hours later when the feature had rotated around to the nightside. The two lightning images were taken about 4 minutes apart. The dayside image was recorded at 727 nm, while the night images were recorded with the "red" filter to improve the throughput. Courtesy of NASA.

troposphere (Figure 4.14, see color section). The main altitude of ammonia photodissociation occurs in the 30mbar to 300mbar region, and peaks at the 80mbar pressure level for phosphine. The main photochemical product of ammonia is likely to be hydrazine, which should form ice particles that slowly settle through the atmosphere and are pyrolyzed at deeper levels. Similarly, diphosphine particles may also be present in the upper tropospheric hazes. In addition, red phosphorus P4(s) may also be produced (as was mentioned in Section 4.3.2).

Observations of stratospheric hazes have been made both from ground-based methane band measurements (West, 1979a, b; West and Tomasko, 1980) and more recently from Galileo (Rages et al., 1999). Such observations are described more fully in the context of Saturn's stratospheric hazes (Section 4.4.2). These hazes may be produced from the higher mass photolysis products of methane. However, the stratospheric aerosols of Jupiter (and Saturn) in the polar regions are found to be highly UV-absorbing and very different from those seen at other latitudes. In these regions it would appear that haze production results from a different mechanism possibly by charged particles from the solar wind and magnetosphere which travel down the magnetic field lines to strike air molecules in the upper atmosphere causing ionization.

J up i 1er

Neptune

J up i 1er

Was this article helpful?

0 0

Post a comment