1.6 x 10~13 molcm~2 above the 0.2mbar level
Bezard et al. (1999)
6 x 10~8 at p < 5 mbar
Feuchtgruber et al. (1997)
a Fraction of H2 with eqm ortho/para.
a Fraction of H2 with eqm ortho/para.
ratio determinations (Conrath et al., 1998), and with the presence of significant levels of tropospheric and stratospheric CO, whose v.m.r. has been estimated from millimeter observations to be 2.2 x 10~6 in the upper stratosphere, decreasing to just under 1 x 10~6 in the lower stratosphere/upper troposphere (Hesman et al., 2007). Such a CO profile requires both an internal and external source of CO with the lower stratosphere/upper troposphere abundance requiring an O/H value of several hundred (Lodders and Fegley, 1994). The source of upper stratospheric CO is unclear. The material is unlikely to come from Neptune's moons and rings, but it might possibly come from water arriving in the form of meteorites and interplanetary grains. However, the flux of water required is significantly higher than has been detected in Neptune's stratosphere from ISO observations (Feuchtgruber et al., 1997), and Lellouch et al. (2005) suggest that the CO may instead have arrived from a cometary impact in the last few hundred years. This suggestion is supported by the observation that when Comet Shoemaker-Levy 9 struck Jupiter's atmosphere in 1994, most of the comet's water was converted to CO by shock chemistry.
If a significant fraction of nitrogen in Neptune's atmosphere does exist mostly in the form of N2 and not ammonia, then this may explain why the He/H2 ratio derived from Voyager far-IR measurements apparently exceeds the solar value, an observation which is almost impossible to explain theoretically. However, an N2 mole fraction of only 0.3%, which is equivalent to an N/H value of ^55 x the solar value (i.e., similar to the C/H ratio), reduces the derived He/H2 ratio to solar, which is much more plausible and is also consistent with the Uranian estimate. Once transported to the stratosphere, nitrogen atoms may be produced by the dissociation of N2 molecules by galactic cosmic ray (GCR) impacts. An alternative source of nitrogen atoms may be from atoms escaping from neighboring Triton, which are then captured by Neptune.
Clearly N2 is very much a disequilibrium species in the cold reducing atmosphere of Neptune and thus as fast as it may be uplifted, a certain fraction per second will convert to NH3. Why then do we still not see much ammonia in Neptune's atmosphere? It is likely that ammonia formed from N2 at pressures less than ~40 bar will react with H2S (which may be more abundant in the atmospheres of Neptune and
Uranus by a factor of S/N > 5) to form NH4SH. Alternatively, NH3 formed at pressures less than —8 bar should freeze out to form ammonia crystals. It is interesting to note that the Voyager 2 radio occultation experiment estimated the ammonia v.m.r. at —130 K and 6 bar to be 6 x 10-7, which is close to the s.v.p. of ammonia under those conditions, suggesting the presence of ammonia crystals.
Of the other possible tropospheric species—H20, AsH3, GeH4, and PH3—only PH3 has any thermal-IR features in regions of the spectrum where the extremely cold Neptune atmosphere emits in any strength. To date no phosphine has been detected, although ground-based millimeter studies suggest that the possibility of supersaturation is very small (Encrenaz et al., 1996). Some formation models of Neptune suggest that the bulk H20/H2 fraction must be several hundred times the solar ratio and some models find that just such a large fraction is necessary to account for the detected high abundance of CO in Neptune's upper troposphere/upper stratosphere (Lodders and Fegley, 1994) outlined earlier, although this does not fit well with the SCIP model of Owen and Encrenaz (2003), described in Chapter 2.
The abundance of methane in the stratosphere is estimated to be 3.5 x 10-4 (Baines and Hammel, 1994), but may be as high as 1.7 x 10-3. This is much greater than the saturated vapor pressure of methane at the temperature of Neptune's tropopause determined by Voyager, which is (1-3) x 10-5. The implication of this may be that methane is transported to the stratosphere not just as a gas, but also as ice crystals from the troposphere, which subsequently sublimate at the higher stratospheric temperatures. It has been proposed that the high internal heat flux of Neptune drives moist convection that produces convection cells vigorous enough to punch their way up through the tropopause and into the stratosphere. There is good observational evidence of localized, high, thick methane clouds and other transient storms, especially at midlatitudes, which are considered to be sufficiently vigorous to lift methane ice crystals to the stratosphere. The shadows cast by these clouds on the underlying main 3.8 bar cloud deck were occasionally observed during the Voyager 2 flyby (Figure 4.25) and were used to determine that their cloud tops were near the 1 bar level, consistent with them being composed of methane crystals. An alternative explanation for the high abundance of methane in Neptune's stratosphere comes from ground-based observations of Neptune's South Polar region in 2006, which showed the temperature near the South Pole to be greatly enhanced at the 100mbar level (Orton et al., 2007a). If this hotspot is due to localized solar heating, as Orton et al. (2007a) suggest, and not due to adiabatic heating from downwelling, then the "cold trap'' to methane is lifted in the polar regions, allowing methane to diffuse easily up into the stratosphere with a v.m.r. as high as 1%, before being transported latitudinally and mixed with less methane-enriched air.
Once in the stratosphere, methane is photolyzed at high altitudes to form hydrocarbons. The only hydrocarbons that have been detected so far are ethane and acetylene with v.m.r.s of —1 x 10-6 and 6 x 10-8, respectively (Bezard and Romani, 1991; Orton et al., 1992) and also CH3 (Bezard et al., 1999). Emission from stratospheric methane and ethane can be seen in ground-based observations, and this is seen to be enhanced polewards of 70°S (Hammel et al., 2007; Orton et al., 2007a). The enhancements are too large to be explained by increased methane and
ethane abundance, and instead are probably due to a temperature increase in the stratosphere of ~3K. In addition, disk-averaged observations of the emission from stratospheric methane and ethane have been used to monitor a steady rise in emission since 1985 (Hammel and Lockwood, 2007; Hammel et al., 2006; Marten et al., 2005).
Both stratospheric water (~3 x 10~9 at p < 0.6 mbar) and C02 (6 x 10~8 at p < 5 mbar) have been detected by ISO/SWS, indicating an external source of oxygen, probably via the continuing capture of interplanetary dust (Feuchtgruber et al., 1997).
To first order, the composition and pressure levels of the major cloud layers are similar to those of Uranus. Cloud layers of CH4, H2S or NH3, NH4SH, and H20
(ice and aqueous) are expected at about 2 bar, 8 bar, 50 bar, and greater than 50 bar, respectively. The H2O cloud in particular is likely to be extremely massive and heavily precipitating since the deep O/H ratio may be several hundred times greater than the solar ratio. However, similar to Uranus, only two observable cloud decks have been confirmed (Figure 4.13): a very thin cloud near the 1.5 bar level (r ~ 0.1), and an optically thick cloud beneath with a cloud top at ^3.8 bar, detected using observations of the hydrogen 4-0 and 3-0 quadrupole lines and 681.8 nm methane line (Baines and Hammel, 1994; Baines et al., 1995a). The observed abundance of methane rapidly reduces with height at the 1.5 bar level indicating that this is indeed the methane cloud. However, the optical depth is again very much less than that predicted by ECCM models. It is thought that particle growth in this cloud may be very rapid and thus that "raindrops" quickly form, which drop back through to below the condensation level thus keeping the visible and near-IR optical depth of this cloud low. The deeper thick cloud is most likely to be the top of the expected H2S cloud, although it has an unexpectedly strong absorption at red wavelengths, which contributes to Neptune's blue color. However, forming an H2S cloud at bar requires ammonia to be severely depleted at this level. While some ammonia can be absorbed in forming aqueous ammonia and NH4SH cloud decks, the amount that can be trapped by these mechanisms is insufficient to explain a pure H2S cloud at bar unless the S/N ratio is significantly supersolar, as was alluded to earlier. Atreya et al. (2006) suggest that the ammonia may instead be absorbed in a water-ammonia ionic ocean at very deep levels, first mooted by Hubbard (1984) to explain aspects of Neptune's and Uranus' magnetic fields. Clearly Neptune's main cloud deck is more red-absorbing than Uranus' since Neptune appears significantly bluer than Uranus (Baines and Bergstrahl, 1986; Baines et al., 1995a). The blueness of the deep cloud is puzzling since the coloration is not experimentally observed in H2S and NH3 ices. Irradiated frosts do have a dip in reflectance of 0.6 ^m, but the albedo recovers at longer wavelengths whereas the observed clouds are dark for A > 0.7 ^m. An alternative interpretation is that the 3.8 bar cloud is partially transparent (Sromovsky et al., 2001c) and thus that the properties may have more to do with the wavelength dependence of scattering efficiency.
The general background appearance of Neptune is that the midlatitudes are slightly darker than the equator and poles (as can be seen in Figure 1.6). This variation is due to variable reflectance from the deeper cloud and could be due to variations in the scattering efficiencies of the particles in this cloud, variations in the cloud top height, or perhaps due to variations in the tropospheric methane abundance. However, superimposed on this background structure are small convec-tive clouds (presumably composed of methane), which were observed to be common during the Voyager 2 flyby in 1989 and widely occur at many latitudes. Subsequently, convective activity appears to be increasing, and such clouds are observed to be most common at midlatitudes from 25-60°S and 20-40°N (due to Neptune's obliquity, it is not possible currently to view latitudes polewards of ^40°N from the ground). Sromovsky et al. (2001c, d) noted a clear increase in midlatitude cloud activity from HST observations between 1996 and 1998 and Max et al. (2003) used ground-based observations to show that cloud activity was continuing to increase and estimated the cloud tops to be just below the tropopause. Gibbard et al. (2003) analyzed ground-based spectral observations between 2 ^m and 2.3 ^m to estimate that northern midlatitude clouds had cloud tops in the lower stratosphere, while those at southern mid-latitudes had cloud tops near the tropopause at between 0.1 bar and 0.14 bar. In addition, clouds were also seen near 70°S with cloud tops between 0.27 bar and 0.17 bar. Gibbard et al. (2003) used these observations to suggest that upwelling was occurring at southern midlatitudes, and suggested that northern midlatitude clouds were due to a subsidence of stratospheric haze material. The small clouds near the South Pole were inferred to be isolated convective events in a region of general subsidence.
Higher up in the stratosphere methane is photolyzed into hydrocarbons, which diffuse down through the atmosphere and freeze out as haze layers as the temperature decreases towards its minimum of 50 K at the tropopause. Using photochemical models and estimated eddy-mixing coefficients to calculate the hydrocarbon v.m.r. profiles consistent with measurements, it is predicted that C4H2 condenses at p > 2 mbar, C2H2 condenses at p > 6 mbar, and C2H6 condenses at p > 10 mbar (Baines et al., 1995a). Neptunian hydrocarbon hazes appear to be more abundant and optically thicker than their counterparts in the Uranian atmosphere, and indeed absorption of sunlight by these aerosols may explain why the lower stratosphere of Neptune is some 40 K warmer than Uranus. It is estimated that 6% to 14% of incident solar UV and visible flux is absorbed by stratospheric hazes. However, other sources of heating are possible such as tidal heating by Triton, or the breaking of vertically propagating gravity waves generated in the more vigorous Neptunian troposphere. The mean particle size of stratospheric hazes is estimated to be 0.2 ^m, and the average visible optical depth (619 nm) is estimated to be 0.025 (Baines et al., 1995a; Pryor et al., 1992). Particle formation requires the presence of foreign condensation nuclei or ions. At lower stratospheric altitudes these may be provided by hazes "drizzling" down from above, but at the highest altitudes the likely sources of condensation nuclei, such as ions generated by meteoritic impacts or UV and cosmic-ray ionization, are limited. Hence, it is possible that hydrocarbon vapors may become significantly supersaturated before condensation starts. Indeed it has been postulated that haze formation may be episodic with partial pressures slowly rising to levels greatly in excess of the s.v.p., triggering the onset of condensation and the rapid formation of haze particles, which reduce the supersaturation level to 1.0 and then fall down though the atmosphere.
The disk-averaged albedo of Neptune has been monitored since 1950 and distinct trends are seen. Sromovsky et al. (2003) considered observations from 1970 onwards and found a good correlation between observed trends and a lagged seasonal model where disk-averaged albedo was greatest near the solstices. However, Lockwood and Jerzykiewicz (2006) found this model to be inconsistent with a longer time series of data and Hammel and Lockwood (2007) note that the brightness is better correlated with subsolar latitude.
A particularly curious feature of Neptune's stratospheric aerosols was that for a long time their reflectance appeared to be correlated with solar activity (Baines, 1997b; Baines et al., 1995a; Lockwood et al., 1991), whereas no such correlation was seen in Uranus' stratosphere, which is otherwise so similar to Neptune's. It was proposed that over time the hydrocarbon aerosol particles of both Uranus and Neptune are "tanned" by radiation from the Sun, which makes them more absorbing, and the rate of "tanning" depended on solar activity. On Neptune, the higher optical depth of stratospheric hazes means they contribute more to disk-integrated reflectivity, and the particles sizes are large enough (0.2 ^m) that tanned particles fall reasonably quickly through the atmosphere to be replaced by fresh white particles, which themselves slowly tan. Hence, a fair degree of correlation might be expected between solar activity and disk-averaged reflectivity. On Uranus, however, the haze particle size is estimated to be only of the order of 0.1 ^m and thus these particles remain for much longer in the stratosphere before settling. If haze particle residence time is a sufficiently large fraction of the solar cycle period of —11 years then there might not be expected to be such a strong tracking between mean aerosol brightness and reflectivity. In addition, the smaller optical depth of stratospheric hazes in Uranus' atmosphere would mean variations in stratospheric haze reflectivity would contribute a smaller fraction of total disk reflectance. Although an elegant and appealing theory, Lockwood et al. (1991) noted that anti-correlation was becoming harder to observe as the disk-averaged albedo of Neptune continued to increase in the early 1990s. Subsequently, Lockwood and Jerzykiewicz (2006) note that anti-correlation started to break down in 1990 and is no longer visible. Whether the "tanning" theory is fundamentally flawed, or whether the effect is now being masked by the increased reflectance from Neptune's tropospheric clouds is unknown.
Was this article helpful?