1. Introduction

One of the greatest challenges in science today is to find living organisms that originate outside of Earth. A prime place to begin this quest is the neighboring planet, Mars, which exhibits the least extremes of temperature and pressure of all the other rocky planets. There is indirect evidence that indicates substantial life existed on Earth as far as 3.85 billion years ago (Mojzsis, 1996; Schidlowski, 1988), with speculation that life appeared on this planet as early as 4.2 billion years ago (Joseph 2010; Nemchin et al., 2008; O'Neil et al., 2008). Mars had a wet, relatively warm environment comparable to that of early Earth for at least 500 million years (Bibring, 2005; McKay and Stoker, 1989; McKay, 2004). If life originated on Earth, could it also have originated on Mars under similar conditions?

Today, the surface of Mars seems like an unlikely candidate to harbor life. It is bitterly cold and dry, and is continually bombarded with lethal doses of ultraviolet radiation (Ronto, 2003; Moreau and Muller, 2003). Moreover, the atmosphere is flooded with deadly superoxide ions that have diffused into the ground, likely sterilizing the top couple of meters of Martian soil (Yen, 2000). If life did originate on Mars, any surviving remnants would long ago have needed to retreat to a relatively warm, wet and safe environment in the planet’s interior.

At present, the most likely locations for extant life are in inaccessible regions deep underground. However, it may be possible that small communities of life exist near the surface in niche environments. At 145 million square kilometers, the surface area of Mars is huge - about the same area as the combined land surfaces on Earth. Finding these niche environments in such a large area is a difficult, time-consuming task. Robots can make this task easier by automatically searching large areas of the Martian surface, looking for these small, localized sites in the enormous environment. Mobile surface robots can search regions previously identified as potentially habitable using satellite or aerial vehicle remote sensing.

2. Niche Environments on Mars

In 2006, the Mars Exploration Program Advisory Group produced a report from the Mars Special Regions Advisory Group (SRSAG; Beaty, 2006). In this report, they place environmental limits on conditions necessary for life to reproduce. Areas where life from Earth could survive or Martian life forms might exist and reproduce were designated Special Regions. A Special Region is "... a region within which terrestrial organisms are likely to propagate, or a region which is interpreted to have a high potential for the existence of extant Martian life forms. Given current understanding, this applies to regions where liquid water is present or may occur. " To determine what constituted a Special Region, they sought life forms that exist under the worst of Earthly conditions. For example, in polar regions microbial life forms fail to show any reproductive behavior when the temperatures are less than -15?C, leading SRSAG to set a lower limit of -20°C as a Special Region boundary.

Water is a crucial element for all known life and various measures can be used to describe the availability of water to biological systems. The Mars Special Regions Advisory Group used water activity (aw), which integrates biology and geology definitions. Pure water has an activity of 1.0 and decreases as solutes concentrate in water and as the relative humidity becomes lower. On Earth, the lowest water activity allowing microbial reproduction is 0.62, leading SRSAG to set a lower boundary of 0.50 water activity. The SRSAG identified no special regions on Mars. However, certain gullies and pasted-on mantle regions were listed as uncertain and are to be treated as special until proven otherwise. These areas are controversial in that satellites have recorded images showing anomalies and erosion features that could be interpreted as being produced by very recent water flows (Malin & Edget, 2000).

3. Extremophiles

Life on Earth has been found to be adaptable to a wide range of seemingly extreme environments, many of which overlap environmental conditions thought to exist on Mars. For example, the Atacama desert in Chile may be the driest place on Earth, with some of its weather stations having never recorded any indication of rain. However, microbial life is found even in this seemingly impossible location, protected in the interior of halite (sodium chloride) crusts deposited in the crevices of rocks. In this case, the moisture needed for life to continue in this extreme environment is captured from the sparse fog that sweeps the area from time to time (Vitek, et al., 2010). Tests on the Mars-like soil samples from the driest regions of the Atacama showed active decomposition of organic molecules by abiotic processes. This suggests that all organic traces of any ancient surface life or more recent organisms that make their way to Mars’ surface could rapidly disappear.

Organisms have been found deep in ocean rift valleys, thriving near thermal vents under pressures of 1000 atm and temperatures of 150°C. These organisms exist entirely independent of sunlight and oxygen, with their energy coming from minerals and hydrogen sulfide (Ruby, et al., 1981). Life has also been found in ice buried deep within glaciers. Bacteria trapped in Antarctic ice have been cultured from deposits at least 20,000 years old (Christner et al., 2000). Due to their location, these bacteria are thought to have endured repeated episodes of desiccation, solar radiation, freezing and re-thawing processes. This is similar to what may happen to Martian microbes living in marginal habitats.

Methane producing microbes have been found deep in South African gold mines, 4 – 5 Km underground (Moser et al., 2005). These exist in an environment devoid of oxygen, under high pH and high temperature conditions. Moreover, these species get their energy from chemical processes that are independent of sunlight. High temperature is thought to be the limiting factor of how deep into the Earth’s interior life can exist. Since Mars is thought to have a cooler interior than Earth, life may well be able to exist considerably deeper than 5 Km below its surface.

Life on Earth has proven to be quite resilient in many hostile settings. For example, 250 million year old bacteria have been grown after being extracted from salt crystals (Vreeland et al., 2000). It is possible that Martian organisms could also survive hundreds of millions of years or longer in inhospitable environments, only to be revived on the brief occasions when conditions become more favorable.

4. Strategies for Finding Life

Mars today may not be able to support the majority of organisms found on Earth, but there are probably regions where species that are specially adapted to these extreme environments can survive. These regions may be difficult to access and may be widely separated geographically. How do we go about finding evidence of life in such a situation? Robots can perform key functions such as performing automated searches over large regions of land. Robots can also mechanically collect samples, manipulate them to look for biological activity and operate instruments to actually detect signs of life or life itself.

While it is unclear what Martian life might be like, we begin by assuming it has some similarities with life on Earth. Earth organisms are characterized by two distinct processes, genetics (for reproduction) and metabolism (the conversion of external energy into life-sustaining processes). One way to search for niche environments that might support underground life is to look for the waste products of metabolism in the near-surface Martian atmosphere. Such gases would slowly diffuse up through the soil until they reach the surface, where they will dissipate very quickly into the Martian atmosphere. Several candidate gases have been proposed, including methane, ammonia, hydrogen sulfide, carbon monoxide, nitrous oxide, sulfur dioxide and formaldehyde (Anderson et al., 2009).

Methane may be the most promising of these proposed gases. Because methane-producing bacteria show complete independence from surface conditions here on Earth, they are good candidates for Martian organisms (McKay, 1997; Stevens and McKinley, 1995). Moreover, if conditions in the subsurface of Mars are similar to those in the subsurface of Earth, these conditions would favor methanogens over iron reducers and other microorganisms. The Mars Express satellite and Earth-based instruments have detected small amounts of methane in distinct regions of the red planet’s atmosphere in the same regions where high water vapor concentrations are found (Mumma et al. 2009), thus identifying search locations where one can expect a candidate biogenic gas to be found along with water, a key ingredient for life. Since methane is estimated to have a very short lifetime in the Martian atmosphere (Lefèvre, 2009), there must be an underground source. It is unclear what abiotic processes could be producing methane in Mars’ interior, leading to speculation that it might be methane-producing microbes.

5. Instrumentation for Finding Life

To date, the only mission to Mars with experiments specifically designed to find signs of life were the Viking I and II missions of 1976 (Klein, 1976). Each Lander had a biology experiment with three subsystems: a Gas Exchange Experiment (GEx), a Pyrolytic Release Experiment (PR), and a Labeled Release Experiment (LR). The LR experiment was designed to detect radioisotope-labeled carbon dioxide that is produced through metabolism by microorganisms. The latter experiments showed positive results, but these outcomes seemed to be contradicted by an analysis that showed no signs of organic compounds in the soils used in the tests. A consensus was eventually reached that there were chemical agents in the soil producing the positive LR results, rather than the presence of life (Klein, 1978). This general consensus has been disputed by a few respected scientists, who claim, among other things, that the instrumentation on the Viking missions was not sensitive enough to detect low levels of organic compounds (Levin, 2010). While the debate continues as to the meaning of the Viking experiments, it is safe to say that the claims that life may have been found in these experiments is not universally accepted.

Today, NASA is investigating several strategies for detecting life on Mars (see NASA ASTID awards). The techniques vary in what they assume about Martian life, with the easiest to perform tests assuming it will be most like that found on Earth. A relatively easy test is to measure the isotope ratios of carbon molecules in atmospheric gases. For reasons that are unknown, life on Earth is enriched in light carbon compared to molecules of non-biological origin. If this is true on Mars, then a sensitive instrument such as a cavity ring down spectrometer may be able to detect gases high in light carbon ratios at specific locations on the surface of Mars.

In 1996, David McKay and fellow researchers astounded the world by publishing an article suggesting that a Martian meteorite found in Antarctica in 1984 contained evidence indicating the presence of past microbial life on Mars (McKay, 1996). While this claim has been greeted with a good deal of skepticism, a persistent claim is that magnetite nanocrystals found in the meteorite could only be of biological origin (Thomas-Keprta, 2009). Another surface technique might be to use ferromagnetic (FMR) and electron paramagnetic resonance (EPR) spectrometers to look for these iron biominerals, which would be able to survive for long periods on the Martian surface.

Other techniques require an actual sample of soil, ice or water to examine for life. Many types of spectroscopy have been proposed – absorption, Raman, time-of-flight mass, and desorption. These techniques are usually used to look for complex carbon molecules that might be indicative of life. Another method, fluorescence, is often employed to detect biomolecules in the laboratory. Typically, a fluorescent chemical is attached to a binding protein that is designed to affix itself to a very specific biomolecule. If molecules of interest such as amino acids or carbohydrates are present, the soil sample can be made to fluoresce. If we assume that Martian life must reproduce using DNA, RNA or similar genetic molecules, then polymerase chain reaction (PCR) techniques can be used to amplify and detect these molecules.

Perhaps the most general technique to detect living organisms assumes only that they must reproduce and use energy to survive. This method requires samples be placed into two identical chambers. One chamber is then sterilized by heating it. Water is added to both chambers and electrochemical sensors detect any differences between the two chambers that might indicate living organisms are present.

6. History of Mars Robots and Future Directions

The use of robotic rovers seems crucial to the automated search for life on Mars. Mars rovers have been steadily growing in size, capabilities and complexity (see Fig. 1). The rover Sojourner only explored a very local area near its Lander on the NASA Mars Pathfinder technology demonstration mission. Sojourner was followed by the Mars Exploration Rover (MER) mission, which employed twin rovers Spirit and Opportunity to perform robotic field geology in search of signs of past water activity at separate landing sites. Opportunity still operates today on what has resulted in a mission duration greater than 26 times longer than the nominally planned 90-day mission. Robotic rovers deployed to Mars thus far have not been equipped with the instrumentation to shed light on the question of life on the planet. This is changing as the evolution of robotic explorers responds to conclusive evidence of past water activity on the Martian surface produced by the Mars Exploration Rover mission and supporting evidence provided by Mars orbiter missions.

The next NASA Mars rover, the Mars Science Laboratory (MSL), is scheduled for launch in 2011 and is larger and more science-capable than Spirit and Opportunity. The MSL rover, dubbed Curiosity, is equipped with a comprehensive science payload to investigate the past or present potential of Mars to support microbial life by assessing the favorability of environmental conditions.

7. Current Plans for using Robots to Search For Life

Robots instrumented to conduct field astrobiology are expected to follow Curiosity’s mission, with launch currently planned for the latter part of this decade on the ESA-NASA ExoMars Program surface mission. This mission would deliver two rovers to Mars to search for signs of past or present life – the ESA ExoMars Rover and a proposed NASA Mars Astrobiology Explorer-Cacher (MAX-C) rover (MEPAG 2009). The ExoMars Rover would be equipped with exobiology and geochemistry instruments, as well as subsurface structure sensing and subsurface drilling/sampling capability. The proposed MAX-C rover would be equipped to cache samples containing possible evidence of past life and/or prebiotic chemistry as a precursor to a later sample return mission. Cooperative use of these rovers to search for life signs using complementary science payloads is a topic of current deliberation (MEPAG 2010).

8. What Types Of Mechanisms Might Be On Robots?

Spirit and Opportunity are equipped with a geology instrument suite and robotic arm for placement of a subset of instruments so that they are in contact with or in proximity to rocks or soil to be measured. Curiosity will be equipped with a robotic arm and tools to gather samples of rocks or soil, then crush and distribute them to deck-mounted instrument test chambers. The evolving line of exploration rovers maintains configurations consisting of wheeled mobility systems and robotic arms that facilitate instrument placement, measurement preparation, and sample collection and handling. Future exploration rovers for Mars will likely be similar, with adaptations in instrumentation and tooling for the science mission at hand. The instrumentation and mechanisms deployed will likely include drills of various types, interchangeable end-effectors, and smaller, retrievable marsupial robots deployed for specialized tasks. The marsupial robots are specialized devices that are tethered to a larger robot and can perform dangerous tasks such as climbing down the side of a steep crater wall (Kubota, 2004, Spenneberg, 2008, Nesnas, 2008).

Further generations of rovers will have the capability to access currently hard to reach terrain areas such as cliff faces and caves, potentially requiring various forms of mobility other than wheeled locomotion (Boston 2005, Kennedy 2006, Huntsberger 2007). Going beyond exploration, utility rovers will join the campaign to perform site characterization and work on precursor missions leading to human presence. They can also perform collaborative or cooperative work, service tasks, scouting with astronauts present, and continued pursuit of mission objectives after human departure.

On each subsequent mission, Mars rovers have gotten bigger and been endowed with greater capabilities than the previous generation. These robots tend to be specialized, with capabilities to perform specific tasks at specific locations. At unique locations, such as the sides of cliffs, distinctive robot capabilities are required that need to be specifically built into the robot design. Because these vehicles are designed for specific locations, they are not meant for general-purpose missions. There are many interesting terrains on Mars that are worth investigating, all requiring different robot capabilities. This means a degree of luck is required in order to chance upon a hard-to-find environment containing life. On the other end of the scale are large generalized robots that have many different capabilities. These can explore a variety of different terrains, but will not be optimized for any particular tasks. Because they must carry instrumentation and mechanisms for many tasks, these rovers will be bigger, heavier and use more power than the smaller specialized rovers. The advantage is that they can explore over large, varied areas of Martian terrain over long periods of time. The complexity of such rovers, however, will make them more likely to have at least some systems fail over time.

An intermediate approach is to have large robots carry smaller, specialized robots such as the marsupial devices described above (Howard, 2004). These marsupials could also be used to create reconfigurable robots where the two-wheeled devices could be combined to pick up and position different instrumentation or mechanical devices such as drills. They could detach after the particular task is complete, and reconfigure in different combinations for the next task.

Two factors impact what strategies can be used: the power requirements of robots and the their expected lifetime. While solar arrays have proven sufficient to support the Sprit and Opportunity rovers, the need for sunlight limits when and where the rovers can operate. Moreover, solar cells have a limit on the amount of power they can supply for a given size, making some power hungry applications such as drilling difficult. Alternative power sources such as radioisotope thermal generators (RTGs) or Stirling radioisotope generators (SRGs) will probably be necessary to fuel long-term astrobiology missions. Another factor impacting long duration missions is the reliability of robot hardware and robustness of rovers to non-mission critical failures. Prior to human presence, the inability to repair robots will limit their lifetimes. This dictates the use of mechanically simpler mechanisms and avoidance of high-risk tasks such as climbing down steep gullies.

9. Possible Robotic Search Strategies

The strategy for using robots to find life is to look for niche environments where life might exist, collect samples from these environments and then test the samples using various measurement strategies. Unfortunately, the niche environments we are looking for may be difficult to access and the measurements may be difficult to perform. There is therefore a tradeoff between doing what is easy to do and doing the far more difficult and risky procedures that are most likely to actually locate and detect Martian lifeforms. For example, we would really like to obtain samples from the interior of Mars by drilling down one to two Km into the crust at a couple dozen locations. However, the size of the drilling equipment and the complexity of the operation preclude doing this far into the foreseeable future.

Although deep core samples may be out of reach, NASA is working on lightweight percussive type drills that can obtain samples from five or more meters underground (see Satish, 2005, for example). With such limited capabilities there is a need to optimize search locations by looking for specialized geological features that might contain evidence of life near the surface. One way of doing this is to look in the atmosphere for signs of molecules emitted from microbes during metabolism and search for niche environments by measuring the near-surface atmosphere for excess concentrations of possible biogenic gases. LIDAR is a laser technology that can measure atmospheric gases over a broad range. More precisely localized measurements can be taken with a rover equipped with an open path spectrometer, which consists of a laser diode suite and light detector (Anderson, 2009). The rover can scan the area between it and a second rover or a platform on which is mounted a specialized mirror called a 360° retroreflector. By driving around the retroreflector in a spiral pattern and periodically taking measurements, the source of any gas emissions from the ground can be localized (Tunstel, 2007). The concept is illustrated in Fig. 2, which shows a two rover measurement approach. If high a concentration of water vapor or a biogas is detected, then the rover(s) will implement a search pattern to localize the source of the emission. Once the site of the emission is localized, other instruments such as ground penetrating radar (GPR) or a percussive drill can be used to perform further investigations. This approach could be carried out using the ExoMars and MAX-C rovers as a form of 2-rover cooperative search for life signs.

One area that may harbor near surface life is in the northern hemisphere in the vicinity of the landing site of the recent Mars Phoenix Lander mission. Experiments from the Phoenix mission showed signs that the ice/soil contains oxidizing perchlorates (Hecht, et al., 2009). It is theorized that the perchlorates may be produced by microorganisms that could survive near the surface in the far northern hemisphere. Although no direct signs of life were found in the simple Phoenix experiments, this remains an attractive place for further investigations. Investigations into Martian ice are attractive for other reasons, also. If the ice is flowing, it may be bringing up microorganisms from friendlier environments deeper in the surface. On Earth, long dormant bacteria has been cultured from Antarctic ice (Christner, et al., 2000), leading one to believe it is possible for any Martian life to remain in stasis for long periods of time in the ice. Even if the ice is not flowing, it may still contain the preserved remains of ancient life. To investigate the possibility of life buried in the ice, a robotic drill can be used. The drill could be a percussive/rotary or it could be a heated tip drill that melts through the ice. In either case, samples could be collected and tested using the instrumentation previously described.

It is also possible that microbes could be brought to the surface by flowing water. There is growing evidence of recent surface water flow from gullies within the walls of impact craters and other geologic formations (Malin, 2000). There must certainly be mechanisms for water distribution within Mars, even if these mechanisms are not fully understood. If water does diffuse flow toward the surface from deeper within the planet’s interior, it could well be carrying extant life forms with it. One way to tell might be to collect samples from the flow sites and test for organic molecules. Although we would ideally like to collect water as it is flowing, this seems to be a very rare event that would require enormous luck to be able to do. Still, a robot could place devices to collect and preserve water in crater gullies where previous flows occurred to see if samples could be obtained in this way.

An alternative explanation for the water gullies on the sides of craters involves Mars’ obliquity cycle, the way the tilt of Mars’ axis changes with time (Kargel, 2004). According to this theory, when Mars is tilted most on its side the poles of the planet heat up, releasing carbon dioxide. This raises the atmospheric pressure and temperature of the planet to the point where liquid water can exist on the surface. The ice that has accumulated during the rest of the obliquity cycle then melts, resulting in running water on the planet. The obliquity cycle is thought to be a major mechanism for water redistribution throughout the planet, possibly distributing life over a wide range of the planet. If this is the case, then wherever near-surface ice is found may be a prime location to search for extant life.

10. Final Thoughts

Based on what we know today, if life does exist on Mars it is likely to be harbored in remote, difficult to find locations. Robots can help automate the search, but there are many tradeoffs to consider in designing a search strategy. One point should be kept in mind while we proceed on our quest to find extraterrestrial life. The way we go about searching for life is rapidly evolving, as we learn more about the nature of life and the wide range of environments it can exist in. Knowledge of the history, hydrology and geology of Mars is increasing rapidly, with each new mission to the planet filling in gaps in our knowledge. For example, it is conceivable we may find active geothermal features on Mars, which would be obvious places to search for life. The discovery of deep caves or recent lava tubes would also provide prime locations for further investigation. The point is that better understanding of non- biological aspects of Mars will focus our search for life much more precisely. Finally, we need to keep in mind that even with the advances in knowledge that will surely be coming with each new mission, the search for life will require patience, persistence and a great deal of luck to be successful.