home astro/eps 12 bio projects talks photos weather code links


My main research focus is on atmospheric gas and cloud composition, especially as tracers of atmospheric chemistry, climate, evolution, and origins.

I study planetary atmospheres using a wide range of techniques and facilities: the mass spectrometers on the Galileo Probe and Curiosity Mars Rover, sensors on spacecraft missions like Cassini, and remote sensing with ground-based and space telescopes like Keck and Hubble.


SAM on Curiosity measured accurate ratios of isotopes and molecules in Mars' atmosphere, producing some interesting surprises. A higher than expected argon/nitrogen ratio presents an intriguing puzzle when compared with Mars meteorite data: how could this gas ratio have changed so much over only a few million years? Oxygen has been observed by MSL (ChemCam and SAM) to vary over the martian year, but this too is completely unexpected. The atmosphere of Mars is more compositionally dynamic than we thought!


G480, G512

Collaborator Chris Go was the first to notice in early 2006 that White Oval BA had become Red Oval BA, and some people nicknamed the storm "Red Spot Jr." Our team used Hubble's finest-resolution camera (the ACS/HRC; upper left picture) to image the Red Oval, and we retrieved very precise velocity fields with the data. We also studied the variation of haze above the Oval, which did not change much between 1995 and 2008.

While studying Oval BA with Hubble and the Keck telescope, and comparing it with the Great Red Spot (GRS) and smaller ovals, we made several key discoveries. We described the circulation within these vortices that preserves them from decaying away. We found that large vortices like the GRS and Oval BA may have thin arcs around them in infrared images at 5 μm (lower left image), while smaller anticyclones are completely surrounded by complete rings in images at this wavelength. Finally, we found that dynamical models of the Oval BA and the GRS require that the vortices must extend down to the water cloud layer---and that the abundance of water in this layer is supersolar.

Most recently, we imaged the GRS with Hubble to help figure out whether its rate of shrinking really is accelerating, as suggested by amateur observations.


G400, G410, G411, G468

We made the first detection of the ammonia ice spectral feature at 10 μm in the thermal infrared. One possible explanation for the general lack of ammonia ice spectral signatures in Jupiter's atmosphere is that "soot" produced by stratospheric photochemistry drifts down and contaminates fresh ammonia ice particles, masking their spectral signatures. With Kostas Kalogerakis, we studied this effect in the lab and with models.

We submitted a "fresh clouds" model that estimates cloud densities based on the strength of updrafts delivering cloud-forming gases from below.

We have been collecting high-spectral resolution measurements of Jupiter's radiation at 5 μm wavelengths, where methane features can serve as "spectral dipsticks," a technique pioneered by Gordy Bjoraker to determine the depths of the otherwise hidden water cloud.


Advances in technology used by amateur astronomers led to the discovery three impacts on Jupiter in 2009 and 2010. The first of these, discovered by Anthony Wesley and confirmed by Chris Go, was large enough to leave a "scar" of high-altitude aerosols that appear bright in the methane-band images shown here.

The first observations in the sequence serendipitously became very the first science data taken by Hubble's Wide Field Camera 3, because the impact happened while Hubble was being checked out after its final servicing mission. At that time, I was a Visiting Scientist at STScI (Hubble headquarters).



Jupiter's clouds are ever changing: belts and zones shift colors, intriguing new features appear and disappear, the powerful east-west jets shift slightly in latitude and speed, and spectacularly large convective plumes erupt from the deep water cloud layer and reach all the way up to the stratosphere.

These events are interesting in their own right, and we work hard to describe the physics behind the events. But the rich history of time-domain Jupiter data is also of great value because it provides a whole new angle of attack (distinct from spatial-domain imaging data or from spectral-domain data) to understand processes such as heat transport, atmospheric structure and evolution, and the formation of clouds and hazes.

The images to the left show one of the changes in 2007: a drop in the cloud density close to the equator, darkening the equatorial zone. I proposed a technique (2007 AGU poster, below) for using this cloud change to measure the source of Jupiter's equatorial tropospheric haze.



With Máté Ádámkovics I researched a "morning drizzle" in Titan's atmosphere. This important part of the methane cycle on Titan---analogous to the hydrologic cycle on Earth---involved ground-based hyperspectral adaptive-optics imaging on some of the world's largest infrared telescopes.



Binary asteroids are the holy grail of the study of asteroid interiors, because measuring their orbits yields the mass of the system, which leads to a determination of the asteroids' density. Trojan asteroids, which orbit two jovian "months" ahead and behind Jupiter, are thought to be remnants of the same population of planetesimals that formed the outer planets. 617 Patroclus and 624 Hektor are the only known binary Trojans. Franck Marchis and I discovered Hektor's companion during an adaptive optics observing run at Keck. We have also been accumulating lightcurve data on many Trojan asteroids and other small solar system bodies.


G451, G464

One of my first projects upon arriving at Berkeley to work as a postdoc with Imke de Pater was to analyze near-infrared spectroscopic observations of Jupiter's ring and moons taken with NIRSPEC at Keck. We also attempted to measure thermal radiation emitted from elusive larger bodies in Jupiter's ring using the ultra-sensitive Spitzer Space Telescope, but unfortunately those data were awash with stray light from the much brighter Jupiter.

G320, G321, G323

I spent my graduate years in Michigan working with a tiny but priceless stream of some 7000 integers. These numbers from the Galileo Probe Mass Spectrometer were the only direct in situ record of Jupiter's atmospheric composition, between the levels of about 0.5 and 22 bar. From this data we measured nitrogren and noble gas isotopic ratios, cosmochemical constraints that are difficult or impossible to collect remotely.

We measured strange profiles of the cloud forming gases, and struggled to explain the "jovian desert" the probe had descended into. Unfortunately, the deep water abundance was not measured due to the meteorology of the probe entry site. Indirect measurements, now including vortex models, generally agree that the probe would have measured a supersolar water abundance, had it descended into a "normal" area of the planet.

We also measured surprisingly high levels of complex hydrocarbons... at levels as deep as 10 bar! These chemicals are thought to form in the rarified stratosphere and diffuse down into the troposphere, but it seems impossible that the high concentrations measured could have been stratospherically generated. The probe did not carry an imaging camera, so even if a hydrocarbon spray from a jellyblimp---savagely slain by a razorwing---had splattered the GPMS inlets, we would have had no sign that anything was amiss, other than elevated hydrocarbon levels.

Jellyblimps and swordtails taken from Our Universe.