Jovian Atmosphere: Near-Ultraviolet Absorption

Greenspan and Owen (1), in a stimulating discussion of the Jovian atmosphere, have reconsidered Stecher's (2) rocket ultraviolet spectroscopy of Jupiter. They conclude that, while the general reflectivity between 2000 and 3000 A can be matched by some 12 km-atm of H2 and a small surface reflectivity, there still exist two significant absorption features in this wavelength range. This residual absorption, shown by the dashed line connecting the error bars in Fig. 1, includes a feature centered at 2600 A and a general reflectivity decline shortward of 2400 A, with the possibility of very high absorption below 2100 A. The existence of the 2600-A absorption feature had in fact been suspected earlier. From near-infrared spectra, Zabriskie (3) had estimated the abundance of molecular hydrogen above the Jovian cloudtops as several kmatm, and the expected near-ultraviolet spectrum determined by Rayleigh scattering from this amount of hydrogen was calculated by Brandt (4). When this spectrum was compared with rocket observations, of Jupiter at 2700 A performed by Boggess and Dunkelman (5), Brandt noticed a significant deficit in the observed flux below the calculated values. I then attempted (6) to determine whether any likely constituents of the Jovian atmosphere might be responsible for this absorption. The observed or suspected major atmospheric constitutents-hydrogen, helium, neon, methane, ammonia, and water-can all be excluded. Of the most abundant interaction products of these gases, expected both on the basis of laboratory simulation experiments (7) and on the basis of quenched thermodynamic equilibrium calculations (8), none have appropriate absorption features in this region. Other simple molecules capable of absorption at these wavelengths-such as aldehydes and ketones-require water vapor as a precursor. But the water will be condensed out at lower levels and will be unavailable. For these reasons I concluded that the absorption does not occur in the gas phase above the clouds, but rather occurs within the clouds. Since the Jovian clouds are not likely