LIBS and the Mars Curiosity Rover

NASA’s Curiosity rover landed inside the 3.7-billion-year-old Gale Crater on Mars on August 6, 2012, and it has been obtaining data about the planet’s rocks and soils with its ChemCam instrument ever since. We recently spoke with Roger Wiens of the Los Alamos National Laboratory, the Principal Investigator of the ChemCam instrument, about the instrument’s laser-induced breakdown spectroscopy (LIBS) capabilities.

What are the primary goals of the ChemCam LIBS system on the Mars Science Laboratory rover “Curiosity”? Since the rover landed on Mars in August 2012, has the LIBS system been successful in obtaining the data needed to achieve those goals?

ChemCam is designed to do rapid chemical and visual reconnaissance of rocks and soils around the rover with a broader reach than can be done with the rover’s arm instruments (up to 7 m in all directions versus 2 m in the forward direction for the arm). In the first 2 years on Mars ChemCam has returned 160,000 LIBS spectra on 4500 locations, along with 2600 high resolution images; these are all available to the public. More than 20 papers are in press or in the process of publication, supporting the Curiosity rover’s findings on the habitable conditions in Gale Crater’s past.

What are the system’s limitations in terms of determining the composition of rocks and soils on Mars?

The Curiosity rover has only eight geological calibration targets for LIBS on board. The geology at Gale crater has been surprisingly diverse, and so these eight targets don’t begin to cover the compositional phase space. The accuracy is thus more limited than desired; however, the precision has been remarkably useful in distinguishing among different compositions along the rover’s traverse.

Before the ChemCam deployed to Mars, were terrestrial rock and soil samples used to test the system? If so, what modifications or improvements to the ChemCam LIBS system resulted from those tests for its use on Mars?

We mostly analyzed powdered rock standards before deployment. We used these to have homogeneous targets at the scale of the focused beam (≤0.5 mm). One of the parameters that we developed when we started analyzing real rocks on Mars was the number of laser shots per observation point. We needed to use enough to get below any dust or weathering layer. The dust is essentially all removed by the fifth shot, and we have found very few weathering layers, so we now use 30 shots per point, which gives us 25 dust-free spectra to use for statistical analysis. We also worked out the number of observation points to use per rock. If the rock is very coarse-grained and heterogeneous, we have used up to 25 points, which still leaves some variability. On fine-grained rocks we can use just five points, checking these five against each other for homogeneity.

How are specific Mars sites or rocks selected for LIBS analysis?

A geology “science theme group” is staffed each day with several scientists, and it is their job to select the targets. The group comes together from around North America and Europe via phone lines and a chat room. They survey images of the terrain around the rover and using a software tool they select targets by clicking on a location on the image. The MSLICE software [Mars Science Laboratory Interface, the result of a collaborative effort between NASA’s Ames Research Center and Jet Propulsion Laboratory] provides the target’s coordinates and distance from the rover. The theme group decides how many laser observation points to use and how many context images to take for each target, depending on the complexity of the target and the time available. Once the targets are selected, uplink engineers prepare the command sequences and deliver them to Jet Propulsion Laboratory. The sequences are combined with other rover activity sequences, which are uplinked together to the rover. The different instrument teams coordinate their activities with each other and with the rover drivers and the arm mobility personnel. We always work to optimize the activities, trying to use all of the time and energy available on the rover each day. On Fridays we program the rover to work over the weekend.

What is the range of distances used so far for the laser shots? How does distance affect the quality of the data?

Our nominal distance limit is 7 m (23 ft). We’ve gone just a little farther than that on Mars. From ChemCam’s position on the mast it is about 2 m down to the ground, so that is generally our shortest distance. The system is quite stable to about 5.5 m and then the accuracy drops off. At 7 m we can determine in a broad sense the type of material we have hit, but we don’t obtain the elemental composition to good accuracy.

Besides the near-distance LIBS observations, which are the main objective of the instrument, we have found a number of other scientifically interesting applications. We have used ChemCam’s spectrometers passively for observations of objects several kilometers away, taking reflectance spectra of dunes and high-resolution images of Mount Sharp. One of our team members points the instrument up at the daytime sky to monitor the water and oxygen column abundances of the Mars atmosphere. This week we took the first spectra and images of stars in preparation for the close encounter of comet Siding Spring with Mars in October. 

The 2020 Mars mission will also include a LIBS system in the mission’s SuperCam instrument, for which you have been named the Principal Investigator. What are the main differences between the Curiosity Rover’s ChemCam system and the SuperCam system? What additional capabilities will the SuperCam system have?

The big changes are the addition of a stand-off Raman spectroscopy system and an infrared spectrometer. Both of these will provide mineralogy information and will identify organic materials if present. Determining the elemental chemistry and the molecular structure is a powerful combination. And by doing LIBS first, the plasma shock wave cleans the surface of dust, providing an unobscured target for all of the techniques. We expect the LIBS system to be nearly the same on SuperCam except we will significantly increase the number of on-board calibration standards.

How do you think the results obtained with the ChemCam system will influence the design of the SuperCam system?

ChemCam has given us a good feel for the number of targets we can expect to interrogate in a day, and it has given us a feel for the distance range for most targets (mostly within 3.5 m of the rover). It has strongly validated our need for a high-resolution context imager to reveal fine-scale features on our target. On SuperCam that imager will be upgraded from black and white to color.