1. WHAT Are We Looking For?

Right up front, we need to decide upon what we, as astrobiologists, are actually looking for on Mars. This may seem like an unnecessary step, but how can we solve a problem if we cannot even properly phrase the question (Cleland & Chyba, 2002; McKay 2004)? To perform this step, we have to ask a couple of very fundamental questions:

These turn out to be two huge questions. To discuss them in proper depth would take volumes – well beyond the scope of this chapter. However, to search for life, we do need to know what it is we are looking for. After all, scientists and engineers cannot develop the necessary tools and protocols without a clearly defined end goal in mind (Chyba & Hand, 2005). In order to properly engineer these tools and procedures, we will need to begin by making a few simplifying assumptions. In general, scientists do not like to make assumptions; we like absolutes. However, any program manager will tell you that we simply do not have the time, money, and resources to address every possible variable, in every conceivable permutation. Of course, that also presumes – another assumption – that we actually know what all of the variables are. We do not.

Now, the more assumptions we make, the easier our task will be. Conversely, as these assumptions accumulate, we will quickly find ourselves steering closer to a preconceived earth-centric solution (Crawford et al. 2001; Kounaves et al. 2002). So, this approach may seem counter intuitive to the scientific method. However, if we shape our assumptions into a hypothesis, we can then set about the task of testing this hypothesis in the laboratory (Earth analogs) and in the field (Mars). This is precisely what we need to do (Cleland & Chyba, 2002; Nickles 2010). Otherwise, we will continue to blindly look for every possible thing in every possible way; an impossible task. An old proverb tells about three blind men who set out to explore and describe an elephant. Depending upon how and where they approached this animal, they came up with three very conflicting results. This is the path we are currently on. We need to jointly decide on which characteristics are most significant about this thing we call life, and where and how we should go about looking for them. Only then can we reach consensus and acceptance of any results.

But, is life a "thing", like an elephant, or is it a "process" (McKay 2004)? Here, we may be starting to delve into semantics and philosophy (Cleland & Chyba, 2002). But, is that necessarily a bad thing? The beauty of this discipline we call "astrobiology" is that we are free to use any and all tools and methods to solve our underlying question; "Are there other examples of life in the universe?" To that end, we can use biology, geology, chemistry, physics, and even natural philosophy.

The task of defining life is very similar to how physicists have come to describe light. From the beginning, light has been a difficult phenomenon to pigeonhole. In many ways, light behaves like a wave (e.g., diffraction). In other important aspects, it is more like a particle (e.g., photovoltaics). So, which is it? We now know it is actually neither; it behaves like both. In our limited frame of reference, no single concept could adequately address all of the known characteristics of light. Therefore, we developed a model for light that encompassed both concepts. When it suits our needs, we treat light as if it were a wave. At other times, we treat it as if it were a particle. In actuality, as we define a wave and a particle, these are very conflicting concepts. Yet, this combined model of light now allows us to better understand how light actually works in nature. Granted, we still cannot precisely define it. However, by making a few fundamental assumptions we can now apply this hypothetical model to any situation we might imagine (Hawking & Mlodinow, 2010).

As previously stated, astrobiologists can use any discipline they need to accomplish their goals. This leaves us with a rather impressive array of tools to choose from. In that light, the term "astrobiology" might seem like a very narrow term to define what we actually do. Perhaps "exobiophysics" might be more applicable. At any rate, with so many diverse tools to choose from, we need to find an underlying concept that unifies these disciplines and our efforts.

If we look closely at each of these scientific disciplines, it appears that energy is the common denominator. Each of these disciplines has selected a set of observable physical characteristic and assigned this group a title: chemistry, biology, geology, and even philosophy. However, at the base all of this differentiation is energy. So, if we can utilize this unifying concept, we can dispose of the superfluous categories and get at the core solution to our two fundamental questions. In general, most researchers agree that terrestrial life "as we know it" requires three fundamental things (Beegle et al. 2007; Chyba & Hand, 2005; Cleland & Chyba, 2002, Kounaves et al. 2002; Lammer et al. 2009; McKay 2004; Mottl et al. 2007; NRC 2006):

Photons – quantum packages of light energy – from the sun directly provide the energy used by many of the terrestrial entities we refer to as "living". But there are many other sources of energy available (Chyba & Hand, 2005). So, we should not limit ourselves to a particular energy source. Instead, we need to learn how living organisms take and use this energy, regardless of the source, to produce the long list of characteristic that most lifeforms seem to share (Horneck 2000; Kounaves et al. 2002; McKay 2004).

Lets follow this energy perspective to see what physics tells us about life, and what characteristics a given site should possess in order to make it potentially suitable for life to form and flourish – to be habitable (Conrad 2009; Kounaves et al. 2007). For example, silicone-based life has often been discussed as a potential alternative to the carbon-based life that is the norm on Earth (Levin 1968; McKay 2004). Being in the same column of the periodic table, silicone (Si) shares many of the favorable properties of carbon. However, it is not quite as good of a "building block of life" as carbon is. Primarily, silicone cannot form the multiple bonds that carbon uses to make such a wide array of polymeric organic chains. So, since many forms of carbon are very abundant in the known universe, why should nature settle for second best? The efficiency of physics drives nature to use carbon. It is the best available option with the lowest energy expenditure. So, even though silicone-based life may be theoretically possible, carbon is definitely the best material to use to build a living organism. The polar physics of a water molecule is what makes liquid water such an effective solvent. On Earth, it allows the aqueous mechanisms of biochemistry to take place very efficiently. As we all know, every individual component of a living organism is essentially "abiotic". However, when these non-living constituents come together in the proper combination, in an aqueous solution, we get a collection of interdependent chemical reactions that produce the observable characteristics we jointly refer to as "life". Granted, there are other solvents out there that are almost as good as water. Ammonia (NH₃) is a polar molecule that can be found in its liquid phase under more limited environmental conditions than water. Liquid NH₃ is almost as good of a solvent as H₂O (Chyba & Hand, 2005, Mottl et al. 2007). Almost. Now, hydrogen is the most abundant detectable element in the known universe, and oxygen is the most abundant element on Earth. Oxygen is third only to helium (He) in universe (Mottl et al. 2007). Astronomers use absorption bands in the near-infrared (NIR) portion of the electromagnetic spectrum (1.5, 2.0, and 3.0 μm) to detect water and hydrated minerals (Mottl et al. 2007). In one phase or another, water seems to be ubiquitous in the universe (Lammer et al. 2009). Granted, the dominant phase of this water seems to be in its solid form – ice (Deming & Eicken, 2004; Mottl et al. 2007). That is certainly true in our solar system. However, given the proper ambient conditions, this water will assume a liquid state. So, once again, it would be simple efficient physics for nature to settle upon the best, most readily available solvent to form and support living organisms – water.

2. WHERE Do We Look?

So, it would appear from the discussion in the previous section that physics is the common fundamental concept we need to explain why the basic components of life behave the way they do. It is through simple natural efficiency that carbon, water, and those quantum packets of energy come together into that unique manifestation we call life. So, lets "chase the physics of life" – energy. But, how do we do that? On Earth, we see the interchange of energy taking place in living systems through the process of reduction and oxidation (redox) reactions. Electrons (e-) are passed back and forth between a wide variety of molecules to drive the biochemical reactions of living organisms. Now, many of these same reactions can – and do – take place abiotically (NRC 2002). However, a special set of organic molecules (enzymes) can intervene in these reactions and help them take place much more efficiently. So, if we could somehow physically track these enzymatic redox molecules in situ, we could see where the organically catalyze processes that are associated with living organisms are taking place. A visual accounting of these molecules should provide us with the very "biosignatures" – indicators of past or current life – we need to find.

Simply finding a potential biosignature would not unequivocally identify the presence of life. This would only be the beginning of the process. For example, recall what happened in 1996 with the "Mars meteorite". The ALH84001 research team reported on a set of four potential biosignatures. However, this has only managed to spark more controversy over the last 14 years (Ascaso & Wiezchos, 2002; Banfield et al. 2001; Lang et al. 2002; McKay 2004; Thomas-Keprta et al. 1997). So, as we set out on our manned mission to Mars, we need to look at the "big-picture". Once we remotely locate a potential biosignature – like an enzymatically directed redox reactions – we will then need to bring all of our astrobiological tools to bear.

Ever since the Mariner 9 orbiter visited Mars in 1972, we have received a constant influx of photographs providing geomorphological evidence that Mars was once a very wet place (Amann et al. 1995; Bada, et al. 2005; Baker 1981, 2006; Banfield et al. 2001; Beegle et al. 2007; Christensen 2006; Chyba & Hand, 2005; ESA 2010; Horneck 2000; Kerr 2000; Lang et al. 2002; Malin & Edgett, 2000; Squyres & Knoll, 2005). These initial observations also went on to inspire the Viking orbiter and lander missions. Anyone can simply cast on eye upon these photographs to see features that resemble terrestrial riverbeds and channels, valleys, alluvial fans, deltas, lakebeds, and even ocean shorelines. In fact, these features virtually cover the entire surface of Mars. Their very presence suggests a periodic set of events that have involved large quantities of water and ice throughout the entire 4.5 billion year history of Mars (Leuko et al. 2010). There are even indications that some of these aqueous events may still be occurring today (Baker 2006).

All of this aqueous activity has also provided many minerals and deposits that are typically associated with liquid water (Bada, et al. 2005; Sephton 2010). For example, ice-layers of the cryrolithisphere, 1 to 2 km thick have been found at the Martian equator, and up to 5 to 6 km deep at the poles (Baker 2006). High-resolution imagery has shown extensive planet-wide sedimentary layering (Bell 2002). Similar terrestrial deposits have proven to be a valuable record of long-term geologic and climatic variations. They are also a vast storehouse of biominerals and other significant biosignatures.

Even going back as far as the two Viking landers (1976), we have been receiving data describing the surface mineralogy of Mars. The basic regolith is apparently composed of fine grained basaltic sands and the same lava alteration products that are found on Earth (Soderblom et al. 2004). The highly oxidized regolith (Crawford et al. 2003) also contains large quantities of water-bearing phyllosilicates and iron-rich smectite clays, carbonates, iron oxides, and quantities of sulfate, chloride, bicarbonate, magnesium, sodium potassium, and calcium salts (Amaral & Frais, 2007; Baker 1981; Banfield et al. 2001; Bish & Vaniman, 2008; Crawford et al. 2008; ESA 2010). In fact, the discovery of so many sulfate salt deposits (e.g., jarosite) by the Opportunity rover has further supported the strong geomorphic evidence that large volumes of liquid water once covered the Meridiani plain (Chyba & Hand, 2005). This general mineralogical composition also appears to be relatively common on Mars. Samples collected by Pathfinder were found to closely resemble regolithic material collected several thousand kilometers away at the two Viking sites (Banfield et al. 2001). Although no organic material has yet to be found, one thing is certain. We are going to have to dig deeper and use tools with higher detection sensitivities if we ever hope to detect evidence of even the abiotic organic molecules that must be there (Bada, et al. 2005).

It has long been believed that the strong oxidizing nature of the Martian regolith destroys any organic material that should be present. However, no one could say with any confidence what this oxidant was. This may all change as a result of findings from the Phoenix Scout Mission. On 25 May 2008, Phoenix landed on the northern arctic plains (~68° N latitude) of Mars (Ming et al. 2009; Houtkooper & Schulze-Makuch, 2010). For 150 sols (Martian days) it collected data regarding the geology and climate of subpolar Mars on the polygonal terrain. This terrain was apparently created by subsurface ice and permafrost (Smith et al. 2009). Phoenix also applied an array of analysis instruments to investigate the prospects for past and present habitability in the Vastitas Borealis region. For example, the Microscopy, Electrochemistry and Conductivity Analyzer (MECA) contained a Wet Chemistry Laboratory (WCL) (Hecht et al. 2009a, b). The WCL was designed to detect and identify any soluble salts in the Martian regolith. Up to this point, only the elemental composition of the regolith had been studied. However, elemental analysis tells us nothing about the solution chemistry of this material. This data is essential to the understanding of how elements and ions might be made biologically activity. It is also tells us about the thermo-physical properties of any potential aqueous solutions (Hecht et al. 2009b).

The MECA added 25 cm³ of a dilute leaching solution to approximately 1 cm³ of regolith in four single-use cells or "beakers". The WCL then detected any ions that went into solution with a set of 26 ion-selective electrodes (ISEs) mounted into the walls of each beaker (Catling et al. 2009; Hecht et al. 2009b). The WCL indicated the presence of ~10 mM of dissolved salts, including 0.4 to 0.6% (by mass) of the negative ion (anion) perchlorate (Hanley et al 2009a; Hecht et al. 2009b; Houtkooper & Schulze-Makuch, 2010; Kounaves et al. 2010; Ming et al. 2009). The presence of perchlorates in the Martian regolith was entirely unexpected. The cation (positive ions) population in solution was dominated by magnesium (Mg²⁺) and sodium (Na⁺) (Hecht et al. 2009b; Renno et al. 2009). Perchlorate (ClO₄-) is formed from one chlorine atom and four oxygens. It is typically found in the form of salts derived from perchloric acid (HClO₄) (ITRC 2005). On Mars, these ClO₄− anions are probably associated with Mg²+ and Na+ cations in the form of Na(ClO₄)₂ and Mg(ClO₄) salts.

Why are astrobiologists so interested in Martian mineralogy? Well, microbial activity on Earth is very closely linked to the geochemical cycles that drive our planet (NASA 2008, Newman & Banfield, 2009). In fact, many of our mineralogical features were directly formed by microbes (Sephton 2010). Others were formed more indirectly as a result of the changes these organisms made to their biogeochemical environments due to their diverse metabolisms (Banfield et al. 2001). Either way, the results are geological features that can serve as potential biosignatures. This should be true even if the extraterrestrial life that caused them on Mars is very different from that on Earth. To that end, astrobiologists study the mechanisms and distributions of microbial transformations throughout our geologic history (Banfield et al. 2005). This approach drives a discipline that is at the core of much of our work – geomicrobiology. If we can learn how terrestrial microorganisms reshape their environments, we can apply this knowledge to any extraterrestrial arena (Banfield et al. 2005).

Mineralogy is often closely related to water (NASA 2008). Therefore any aqueous feature could be a potential site for the detection of past or present life. Clay minerals (phyllosilicates) are hydrated minerals that typically form though an interaction between rocks and water (Leveille & Konhauser, 2007; Sephton 2010). So, the presence of these minerals is a good indicator that the associated area was once exposed to substantial amounts of water for an extended period of time. Liquid water is one of the three fundamental components we have assumed are essential for habitability. On Earth, we have repeatedly seen how microbes play a significant role in determining the properties and behavior of clay minerals. Our work involving the geomicrobiological interactions taking place in soils, desert rock varnish (DiGregorio 2001; Dorn 2007; Kuhlman et al. 2004), cryptoendoliths (Nickles 2010), and evaporates (Crawford 2008) clearly shows that these processes have a major role in creating biofilms and layered microbial ecologies. Phyllosilicates are widely distributed on Mars.

We have long known of the existence of clay minerals on Mars. Early remote infrared (IR) analyses and powder X-Ray fluorescence performed by the Viking landers showed that the surface dust and regolith were composed of various clay minerals (e.g. iron-rich smectites) and their weathering products (Bish & Vaniman, 2008; Leveille & Konhauser, 2007, Poulet et al. 2005). More recently, clay minerals have been identified by orbiter-based spectroscopic analyses, as well as in situ observations and chemical analyses performed by the Mars Exploration Rovers (MERs) (Leveille & Konhauser, 2007). These observations and analyses clearly suggest that the aqueous degradation products of basalt (e.g., smectites) are a major portion of the geology of the Martian highlands (Bishop et al. 2008; Leveille & Konhauser, 2007; Sephton 2010). The Observatoire pour la Minéralogie, l'Eau, les Glaces et l’Activité (OMEGA) instrument onboard the Mars Express has used visible-near IR (NIR) hyperspectral reflectance imagery to detect other clay minerals like montmorillonite, nontronite, and chamosite from orbit (Bishop et al. 2008; Leveille & Konhauser, 2007; Poulet et al. 2007).

High concentrations of organic matter in terrestrial clay-rich deposits suggest a microbial role in the formation of these minerals (Leveille & Konhauser, 2007; Sephton 2010). The effects of this intimate interaction between microbes and clay minerals have been frequently observed. The presence of phyllosilicates can have a significant impact on the metabolism and survival mechanisms of various organisms (Morra et al. 1998). In fact, the direct application of small particle clays (e.g., smectite and montmorillonite) has been shown to be a stimulus for biofilms formation (Alimova et al. 2009). Additionally, microorganisms can adhere to the surface of clays particles in windblown (aeolian) dust. This interaction apparently helps protect some bacteria from extreme desiccation and the effects of ultraviolet (UV) radiation – including damage to DNA (Alimova et al. 2009). This relationship could be very valuable to an organism caught up in a Martian sand storm or dust devil.

Even with all of the evidence of past liquid water and ice activity, the surface of Mars now appears to be a very dry place. Any organism surviving there would need to develop "desiccation tolerance." Desiccation tolerance, or anabiosis, is the ability of an organism to be dried to equilibrium with the air, and then returned to normal biological viability at a later time when it is rehydrated (Alpert 2005). These organisms are also called "xerophiles". Xerophiles are extremophiles that can live and grow when there are very low levels of biologically available water (water activity (aw)). They can typically survive when the aw is less than 0.8 (80% relative humidity). To do this, many terrestrial microbes are also able to suspend all metabolic activity, often forming dormant spores. They return to their vegetative state when conditions finally 15 improve (Alpert 2006). They do this because their metabolic enzymes and biochemical processes cannot function properly without the presence of adequate liquid water. On Earth, most microbes are composed of 70 to 90% water (Deming & Eicken, 2004). Once their water content drops below 65 to 70%, the normal metabolic processes of even xerophilic organisms can no longer function.

Bacterial spores have been documented surviving for several hundred thousand to several million years in amber and even in brine inclusions in salt crystals (Vreeland et al. 2000). They tend to survive longer at low temperatures (permafrost or ice) and away from the harmful effects of UV radiation (e.g., below the surface or within rocks). Many microorganisms are also capable of creating and sustaining unique local environments on mineral surfaces (Alimova et al. 2009; Nickles 2010). For example, researches have found that small particle clays particles (e.g. montmorillonite) increase the survival rates of some organisms when exposed to desiccating conditions (Morra et al. 1998). Just how and why this interaction with clays helps organisms thrive and survive is currently unknown. However, these very same clays also make up a large portion of Martian mineralogy. Human explorers on Mars will need to analyze these ubiquitous clay formations for potential biosignatures and signs of habitability. Table 1 provides a partial list of other sites and conditions on Mars that astrobiologists should explore. The following paragraphs address a few of the more promising sites in a little more detail.

Liquid Water: Liquid water and water ice are commonly believed to be unstable on the Martian surface (Sephton 2010). The low average atmospheric pressure implies that any exposed liquid water should swiftly boil away to evaporation (Hanley et al 2009b; NASA 1999). Due the high specific heat of water, this rapid evaporation would draw off so much heat from the remaining liquid, it would freeze solid. This water ice should then sublimate to vapor into the thin Martian atmosphere. But, science teams from Viking through Phoenix (Smith et al. 2009) have seen evidence of ice frosts forming, not sublimating, on the Martian surface. The Phoenix lander also provided other evidence for the presence of water on Mars. For example, water ice was seen falling from the clouds at night and low altitude fogs were observed in the early morning (Houtkooper & Schulze-Makuch, 2010; Smith et al. 2009). More significantly, liquid saline-water (brine) may have been observed. A set of small "globules" were imaged on the leg struts on the Phoenix lander (Renno et al. 2009). Their status was tracked over a period of 36 sols using the camera on the robot arm (RA). These droplets may have been composed of brine mixed with regolith that were splashed onto the struts as the lander touched down. The droplets expanded in size over time as they apparently absorbed water from the atmosphere. They also moved down the struts under the influence of gravity and merged with other globules. Finally, the high-resolution cameras of the Mars Global Surveyor (MGS) Mars Orbiter Camera (MOC) recently revealed small gullies on Martian hill slopes (Baker 2006; Christensen 2006; Crawford & Newman, 2008; Hanley et al 2009b; Hecht 2002, 2009b; Kerr 2000; Malin & Edgett, 2000; NRC 2002). These features are similar to terrestrial hill slope gullies in the higher latitudes. The ones on Mars are so shallow that surface winds and dust storms should quickly erase them. Therefore, their presence clearly suggests recent periglacial or spring-like water seepages.

Now, we have been told that surface water has long been impossible on Mars, yet these observations suggest otherwise. How do we address such conflicting data? Well, the key phase

in the previous paragraph was "average atmospheric pressure". In order to compute a statistical average, you must have a range of values; some lower than the average, some higher. Looking at a phase diagram for pure water (Figure 1), we see that the triple point of water is at 6.11 millibars (at 0.01^o C). The thermodynamic triple point of a phase diagram is where the lines dividing the liquid, solid, and gaseous phases of the material converge (NRC 2006). This means that all three phases can exist simultaneously. So, it would only take a small change in temperature and/or pressure to result in liquid water. The average surface pressure on Mars is approximately 6 millibars (NASA 2007). Looking at Figure 1, we see that this value is just below the triple point for pure water. However, the actual air pressures recorded by the Phoenix lander were always higher than the triple point pressure (Renno et al. 2009; Smith et al. 2009). Viking recorded air temperatures only as high -17^o C. However, on a sunlit summer day, the soil temperatures were found to be as high as +27^oC. This was at a time when the recorded air pressures ranged from 6.7 to 10.0 millibars (Hecht 2002). Granted, these were only for short periods of time. After sunset, the temperatures at the Viking sites typically dropped to below -50^oC. However, according to Figure 1, there were many times when pure liquid water could have been stable on the surface of Mars. It’s all a mater of simple physics.

Orbital and surface imagery and analyses indicate the presences of water ice at the polar caps, at lower latitudes in permafrost, and even in the pore spaces of rocks and regolith (Christensen 2006). So, as we said in the previous paragraph, a small change of atmospheric pressure or temperature could easily result in liquid water forming within the pores of rocks or percolating downward to fill deep subsurface aquifers (Nickles 2010). Based upon the actual surface pressures recorded on Mars, the boiling point of pure Martian water should be about 2 to 7^oC above the melting point. However, if we add some of the salts known to exist in the Martian regolith to this surface water, this small temperature band for stable liquid water should be significantly expanded (Catling et al. 2009; Hanley et al 2009a, b; Hecht et al. 2009b; Houtkooper & Schulze-Makuch, 2010; Kounaves et al. 2010; Renno et al. 2009). The magnesium and sodium perchlorate salts (Na(ClO₄)₂, Mg(ClO₄)) apparently detected in Martian regolith by the Phoenix WCL may be the very material needed to produce this surface brine (Hanley et al 2009a). These salts, when placed in an aqueous solution, form a "eutectic system". A eutectic system is produced when two or more chemical compounds are mixed to form a single material that solidifies at a temperature lower than any of its individual constituents. For example, sodium chloride (NaCl) when dissolved in pure water forms a eutectic system. Instead the water freezing solid at 0^o C (standard terrestrial sea level pressure), the resulting brine (~23% by mass NaCl) would have a eutectic point of approximately -21^o C. Reducing the eutectic point produces a phenomenon known as "freezing point depression". This process also results in a "boiling point elevation". Together, they produce a wider potential liquid range for the aqueous solution. The same thing would happen if a perchlorate salt were mixed with pure water. A eutectic mixture of Mg(ClO₄) and water would have a eutectic temperature as low as -70^o C (Houtkooper & Schulze-Makuch, 2010; Kounaves et al. 2010).

Physics dictates that air pressure is a function of altitude. An increase in altitude means that the air column above is thinner and the resulting mass of the constituent gases will not push downward as much (compared to sea level). Conversely, a decrease in altitude means the air column is thicker and therefore pushes down with more force per unit area. So, to find the most favorable atmospheric conditions for liquid water on the surface of Mars, we should decrease our altitude as much as possible. To do this, we would need to enter the many valleys, chasms, and craters on Mars. For example, the Argyre Basin, Elysium Planitia, and Hellas Basin would be excellent places to look. In fact, a quick glance at Mars indicates that as much as 25 to 35% of its surface is composed of low lying places. This is especially true for the more moderate northern hemisphere.

For our first human astrobiological excursion on the surface, we would suggest a visit to the 2100 km wide Hellas Basin. Hellas is a low lying area nearly 9 km deep. It was probably formed by an ancient asteroid impact. The lower elevation means that the air pressure at the floor of this basin should be approximately twice the computed global average (~12 to 15 millibars). High resolution imagery of Hellas shows steep layered walls covered with scattered pockets of snow and ice. Some of these patches are protected against potential sublimation by layers of dust and sediment (Christensen 2006). This description is very reminiscent of the environment found in the McMurdo Dry Valleys of Antarctica (Kounaves et al. 2010; Mahaney & Dohm, 2010; Nickles 2010; Onofri et al. 2008). The Dry Valleys form a cold (sub-zero temperatures), arid (~0% humidity), and rocky polar desert. There is little snow derived moisture. What little snow that does fall in this area is quickly sublimated and is not biologically available (Nickles 2010). This area also experiences daily severe freeze-thaw cycles, high solar UV radiation (seasonal "ozone-hole"), strong desiccating winds, and elevated salt concentrations (Nickles 2010). However, within the pore spaces inside of some of the sun-heated sedimentary rocks, a stable microniche containing liquid water does exist; and where liquid water exists, even for short periods of time (Kounaves et al. 2007), life can develop and survive. These conditions are very similar to the zone of potential liquid water that should exist within the Hellas Basin. If so, this could be the astrobiological habitable zone (HZ) we have been looking for (Chyba & Hand, 2005; Des Marias & Walter, 1999; Lammer et al. 2009; Mottl et al. 2007). This could be the true "Goldilocks’ Zone" for Martian habitability.

3. Ice – Permafrost, Polar Caps, and Glaciers

Martian Water Ice: Orbital imagery (Bell 2002; NRC 2006) has indicated the presence of ice-rich materials in the mid-latitudes. This includes the lobate debris aprons (LDA), polygonal fields of permafrost (Phoenix landing site), and evidence for current glacial landforms and processes (Christensen 2006). Therefore, there should be large quantities of shallow subsurface water ice scattered about the planet (Bell 2002; Christner et al. 2005). And, we have already shown that the temperature and pressure conditions currently exist on Mars for this ice to periodically shift into its liquid state.

Permafrost: Permafrost typically refers to regions of ground ice (Baker 1981). On Earth, similar polygonal land features as those imaged on Mars are the result of near-surface permafrost zones (Baker 2006; Christner et al. 2005; Ming et al. 2009; Smith et al. 2009). So, this Martian permafrost should hold vast volumes of frozen water all over the planet. Daily and seasonal variations of temperature and pressure should result in the periodic availability of stable liquid water in the pore spaces of rocks and regolith. Therefore, these permafrost regions would be good places to study further for evidence of past and present indications of life.

Polar Caps: The north and south poles of Mars have been studied by astronomers on Earth for a couple centuries. The northern cap (Planum Boreum) is over 1200 km wide and up to 3 km thick in places. It appears to be composed primarily of water ice. The southern cap (Planum Australe) varies in thickness throughout the year. It is thickest during the winter when large quantities of frozen CO2 (dry ice) from Mars’ atmosphere are deposited there (Baker 1981). Planum Boreum is the larger of the two poles. Mars’ current axis of rotation with respect to its closest approach to the sun (perihelion) favors the more moderate conditions responsible for the development of the northern cap. Orbital observations indicate that the poles are covered by frozen CO₂ (dry ice), bright water ice (which never melts), and countless deposits of layered and terraced ice and aeolian dust (Christensen 2006; NRC 2006). Some of these layers appear to be 30 meters or more thick (Baker 1981; NASA 1999). If so, they could preserve a record of thousands of years of Martian climatology, geology, and perhaps even biology.

Glaciers: Glaciers are large sheets of water ice characterized by motion due to basal sliding. This motion occurs as a result of the base of the glacier being lubricated by a layer of melt water. On Earth, this basal melt water is produced by frictional and geothermal heating, as well as pressure-induced melting – as result of the thick mass of overlying ice. This is simple physics. For example, we commonly see pressure-induced melting with an ice skater. The mass of the skater is concentrated on the surface of the ice by the narrow blade. This pressure (force per unit area) results in a thin layer of water that lubricates the gliding of the blade. There is clearly enough ice at either Martian pole to produce the pressures needed for subglacial melting, resulting in yet another potential source of liquid water on Mars.

On Mars, glaciated landscapes are some of the most distinctive landform features documented by high-resolution imagery (Baker 1981, 2006). This being said, the actual presence of true Martian glaciers has yet to be proven. However there is plenty of supporting evidence for widespread ice-rich materials and rock covered glaciers in numerous areas, including Hellas (Christensen 2006).

Life in Ice: Other than ice being made of water, why should we care about such large deposits of ice on Mars? Recent research indicates that a significant number of viable microbes are being found within and under even the deepest and coldest terrestrial ice formations (Junge et al. 2004; Lang et al. 2002; Mader et al. 2006; Price 2000). Bacterial communities in small scale microclimates have been found within porous rocks under glaciers and within the surface ice itself (Bhatia et al. 2006; Skidmore et al. 2000). Some of these organisms are responsible for glacially related methanogenic activity (Bhatia et al. 2006). So, if similar organisms were to reside in the ample Martian ice formations, they could be one of the sources of the mysterious methane recently discovered in the atmosphere (Yung 2009).

Looking within the polar ice, terrestrial researchers have found a habitable brine-filled network of pore space and veins between the ice crystals (Mader et al. 2006; Price 2000). They have been observed even with ice as cold as -20°C (Junge et al. 2004).

This web of interlacing channels apparently supplies the microbes they contain with the water and nutrients required for survival (Price 2000). Many of these organisms appear to survive in this frigid brine by attaching themselves to the surface of small particles of clay sediment. How and why this works is unclear. The polar caps of Mars are believed to be far older than those on Earth. So, if there are similar organisms associated with this water ice, they would have to be very old. Microorganisms have reportedly survived in terrestrial glacial ice for hundreds of thousands of years, and for millions of years in permafrost (Christner et al. 2005; Mottl et al. 2007). So, it is possible that dormant or even viable microbes could be found in the Martian polar caps. If a similar network of pores and passageways should exist within this cap ice, then these brine-filled veins would be the true canals of Mars.

Rock Varnish: Rock varnish (Figure 2A), also known as desert varnish, is a paper-thin rock coating that is ubiquitous in the arid deserts of Earth. These varnishes are apparently formed when ferromanganese compounds are oxidized and deposited into a layered clay matrix. It is not known if this process is abiotic, biotic, or some interdependent combination. Varnish also contains oxides from other trace elements as well as varying levels of silica (DiGregorio 2001; Kuhlman et al. 2004; Spilde et al. 2008). Intriguingly, all of the Martian landers and rovers, going as far back as Viking, have imaged rocks with similar dark shiny coatings.

Rock varnish has long been linked to microbes (Northup et al. 2010). No one can say for certain whether they actually participate in the formation and maintenance of this material, or if they are merely residing in a habitable niche (DiGregorio 2001; Spilde et al. 2008). Organisms that oxidize and reduce manganese (Mn) are not uncommon. They are regarded as an active leg of the Mn redox cycle (Figure 3). According to Kuhlman et al. (2004), as many as 10⁷ to 10⁸ cells can be found in a single gram of powdered varnish. It is not certain if these organisms are viable, dormant, or long dead. Some of these organisms have been shown to be resistant to UV-24 C (100 – 280 nm) exposure (Northup et al. 2010). Additionally, both Mn and iron (Fe) are excellent UV shields for any organisms trying to survive in an otherwise desiccating environment (DiGregorio 2001; Kuhlman et al. 2004). This strongly suggests that since habitats like rock varnish can provide protective niches for terrestrial extremophiles, they may serve as habitable sites on Mars, too (Kuhlman et al. 2004; Northup et al. 2010).

Any microbe that could take advantage of the potential microniches within these endolithic pores could greatly increase their chances of survival under extreme conditions – no matter where they might be found (Horneck 2000; Mottl et al. 2007; Nickles 2010). Based upon all of our discussions up to this point, it is entirely likely that a similar cryptoendolithic community could very well exist within the sedimentary walls of the Hellas Basin. In the Antarctic, these microbial communities give the deposits of sedimentary rocks a bright orange and yellow coloration with small black mottling (Figure 4). So, if these microorganisms are in the Hellas Basin, it would be an easy matter for our human explorers to simply look and see.

Evaporites: There are other terrestrial locations where extremophiles have found protected niches within the surface layers of rocks (Crawford et al. 2008) formed by evaporation concentrations in arid environments (ITRC 2005). Evaporites (e.g., jarosite, halite, or gypsum) are sedimentary minerals where large microbial communities are also macroscopically visible. These communities are composed of one or two horizontal colored surface bands up to 4 cm deep. The colors typically range from tan to pink to green within a white salt matrix (Horneck 2000).

Since there is apparently a long history of water and sedimentary layering on Mars, it is not surprising to see that halite and sulfate evaporites have occurred there, too (Leuko et al. 2010). As with the nanostratographic layers of rock varnish (Figures 2B and C) and cryptoendolithic communities (Figure 4B), evaporites have shown their ability to attenuate UV radiation while transmitting photosynthetic wavelengths of light (400 – 700 nm) (Amaral & Frais 2007; Crawford et al. 2008). Therefore these lithic materials could form a habitable oasis for life upon the surface of Mars (Horneck 2000).

Subsurface: Due to the wide diversity of microbial metabolisms found on Earth, some organisms (chemolithoautotrophs) have demonstrated the ability to colonize subsurface sites devoid of sunlight (Horneck 2000, Mancinelli 2000). These same sites can also provide protection from the ionizing effects of unfiltered UV radiation and any soil oxidants. On Mars, similar organisms could reside in subsurface liquid water deposits within volcanic lava tubes. The surface morphology of Mars is rich with signs of its volcanic past. The MGS has imaged features that could very well match this description (Banfield et al. 2001).

On Earth we are constantly finding new forms of deep-dwelling microbial life. Granted, much of this life is metabolizing very slowly, but they are nonetheless viable (Kerr 2002). In fact, slow metabolisms would be in keeping with the nutritional limits and environmental constraints found within such a marginally habitable environment. The question is how deep would we have to go to find similar organisms on Mars? A potential microbial community may have been found deep (~1500 m) below eastern Washington in the subsurface aquifers of the Columbia River Basalts (Anderson et al. 1998; Mancinelli 2000; Kerr 2002; Stevens 1997). These organisms appear to use the CO2 and H2 produced by the chemical interaction of water and basalt (serpentination (Yung et al. 2010)) as their sources of carbon and energy. Although there is still considerable debate regarding their metabolism and community composition (Anderson et al. 1998), these organisms are nonetheless surviving by some means in these deep subsurface basaltic lavas. Of additional interest, some of these organisms are said to resemble the controversial nanobacterial fossils that were reportedly found in the ALH84001 Mars meteorite (Gibson et al. 1997; Thomas-Keprta et al, 1997).

The concept of nanobacteria (NB) first came to our attention in 1996 when a group of scientist suggested that some of the mineralized formations inside of the ALH84001 meteorite might be extremely small fossilized Martian organisms (Cisar et al. 2000; Vali et al. 2001; Gibson et al. 1997). Since then, many researchers have said that NBs may also be the smallest cell-walled living organisms on Earth. The have been reportedly found in human and cow blood, bile, tissue culture cells, wastewater, the atmosphere, sandstone formations, and even meteorites (Urbano & Urbano, 2007; NRC 1999). These tiny structures have also been called nanobes, nanoforms, and calcifying nanoparticles (CNP) (Çiftçioglu et al. 2006; Vali et al. 2001). They have been observed with the transmission and scanning electron microscope (TEM and SEM) as 20 to 500 nm diameter coccoid and rod-shaped particles, apparently having cell walls (Çiftçioglu et al. 2006; Cisar et al. 2000; Gibson et al. 1997; Kounaves et al. 2010; NRC 1999; Thomas-Keprta et al. 1998). They are also reported to be 0.2-μm filterable (Martel & Young, 2007; Vali et al. 2001). NBs typically have a hydroxyapatite and protein coating formed from the soluble calcium and phosphorus compounds in their environment (Urbano & Urbano, 2007; Vali et al. 2001).

Researchers have found that given the right environmental conditions, NBs rapidly aggregate in culture. When observed, this rapid aggregation makes them appear like microbes, even apparently undergoing cell division (Çiftçioglu et al. 2006). The big question is, are they alive? A large body of literature indicates they may be alive and associated with a broad array of diseases (Çiftçioglu et al. 2006; Martel & Young, 2007). Some researchers have gone as far as to say NBs are an overlooked primitive branch to the terrestrial tree of life (Çiftçioglu et al. 2006). Many of these claims seem to be based upon NB morphology and the fact that they seem to multiply rapidly in cultures (e.g., fetal bovine serum (FBS)) (Cisar 28 et al. 2000). Nevertheless, there is a growing belief that NBs are simply too small to be alive. However, it has long been known that some bacteria can survive periods of nutrient starvation by a significant decrease in cell size (Velimirov 2001). But all of this might be overruled by a growing pool of research indicating that the formation of these nanoparticles can be readily explained by abiotic means (Cisar et al. 2000; Martel & Young, 2007).

In response to the initial 1996 claim of potential NBs in ALH84001, a panel was convened (1998) by the US National Academy of Sciences to determine the lower limits of the size ranges theoretically possible for a living organism (NRC 1999). The panel concluded that these nanoparticles are simply too small to be alive, as we understand life. They believe that a particle 50 to 200 nm in diameter cannot hold all of the components considered essential to sustain life. A living, growing, cell must be large enough to hold the DNA, RNA, and enzymes required for the replication transcription and translation of a minimum set of essential proteins for the basic operation of a living cell. This means that if a cell is to accommodate the 250 to 450 genes considered essential for a viable cell, it must have a diameter in the range of 250 to 300 nm, and an overall volume between 0.014 and 0.06 μm3. The number of ribosomes needed for this genome expression is believed to be the major constraint on cell size (NRC 1999; Velimirov 2001). Even a single ribosome with its membranes and wall should require a sphere no smaller than 50 nm in diameter (NRC 1999). A fully functional E. coli cell possesses approximately 30 to 50,000 ribosomes. So, the general consensus is nanobacteria are too small to be alive, as we know it. However, should we expect these same arguments regarding minimal cell size to apply to some exotic extraterrestrial lifeform?

Carl Sagan, said that anyone who makes an extraordinary claim must be able to produce extraordinary evidence to support it (Sagan 1980). This has not been done for the case of nanobacteria. Although the images are compelling, we need a lot more than biomorphic mineralization processes to say that something is alive or not. Before we can assert that what we have found flies in the face of the established laws of physics, we better have some pretty solid data to back up our claims. Without having detailed data on composition, structure, and biochemistry, we cannot reasonably say that something is alive (Vali et al. 2001). All of this controversy exists over particles that have been found on our own planet – within our own bodies – and analyzed with a full battery of our state-of-the-art instruments and protocols. Yet, we still do not have consensus as to whether NBs are alive or not.

3. HOW Do We Look?

So, all a Mars explorer will have to do is go to one of these many sites we have discussed and look for evidence of biologically assisted redox reactions. Is it really that simple? Well, yes and no. The concept of "chasing the physics of life" is basic, but not simple. To be successful, we will need to be able to look for the right things, in the right way, in the right places. Many of the tools and protocols we will need are still in their infancy.

The designers of Viking landers (1976) may have had the right idea. Viking possessed the only remote astrobiological sensor package we have ever sent to another planet. They had three biophysics-related components in their biology package as well as access to a gas chromatograph and mass spectrometer (GC/MS) (Bada, et al. 2005; Beegle et al. 2007; Benner et al. 2000; Levin 2010). Together, they were designed to actively look for evidence of microbial life (Cleland & Chyba 2002). Ultimately, these biology experiments were designed to detect the physics of life – microbial metabolism. But, without knowing enough about the environmental conditions on Mars, and the concept and capabilities of terrestrial extremophiles (Kerr 2002; Lang et al. 2002; Levin 2010), this attempt was too limited. However, based upon what we have said up to now, it is entirely conceivable that these science packages could have achieved their goal.

The two Viking Landers soft-landed on the surface of Mars in 1976 using descent rockets that might have sterilized the very ground upon which they sat. Viking 1 landed in the Chryse Planitia (~22.4^o N latitude) and Viking 2 was approximately 4000 miles away in the Utopia Planitia (~48^o N latitude) (Levin 2010). Each lander carried an array of experiments designed to look for life similar to that which is found on Earth. There were a total of four biology-related experiments on Viking (Beegle et al. 2007). These were the (1) Gas Exchange experiment (GEX); (2) Labeled Release experiment (LR); (3) Pyrolytic Release experiment (PR); and (4) Gas Chromatograph / Mass Spectrometer (GC/MS). The GC/MS was not originally intended to serve as a key component of the astrobiology package. However, it became an essential element of the life detection process when the science team needed a way to cross-check the conflicting data from the GEX, LR, and PR (Bhatia et al. 2006; Cleland & Chyba, 2002).

The most promising results came from the LR (DiGregorio 2001; Levin 2010). The LR was designed to detect the uptake of radioactively tagged (¹⁴C) liquid nutrients by any potential microorganisms in the Martian regolith. Radioactive ¹⁴CO₂, or some other ¹⁴C-based gas, was expected to be byproduct of any microbial metabolism. The LR dropped this liquid nutrient solution onto a regolith sample in a small test cell (Levin 2007a). A radiation detector monitored the headspace over the dosed soil. The evolution of any ¹⁴C-labeled gases should provide direct evidence of metabolism for any extant life (Levin 2010). The LR apparently operated flawlessly (Levin 2010). A burst of radioactive gas was immediately released with the nutrient broth was added to the regolith. According to the protocol, any positive result required a control experiment to determine if this response was generated by some abiotic entity in the soil. To do this, a duplicate sample was inserted into a fresh cell and heated for three hours at 160ºC (Levin 1979, 2007a). Once sterilized, the sample was allowed to cool and then tested by the same means that resulted in the original positive response. When this was performed on the Viking landers, there was very little LR response. Therefore, this implied that the response was not due to some inorganic chemical reaction.

However, there were confusing results from the GEX and PR that seemed to conflict with the LR experiment. The Viking science team decided to use the GC/MS to try and resolve the issue. If the positive LR results were indeed due to microbes, then the samples should contain organic material. So, another regolith sample was heated to release any organic volatiles. These volatiles were separated using a gas chromatograph and then analyzed with the mass spectrometer. No organic material was identified. So, if there were no organics, there could be no life. Therefore, the Viking team announced that they had found no life on Mars.

Now, not all organic material is necessarily biological in origin. So, even if there were no indigenous Martian organics, large quantities of organic material have been raining down on the surface from comets, meteorites, comets, and interplanetary dust for the last 4.5 billion years (Atreya et al. 2006; Bada et al. 2005; Beegle et al. 2007; Bhatia et al. 2006). Unless the GC/MS was non-functional, something should have been found. In its absence, the Viking team proposed that some aggressive oxidants (H₂O₂?), or superoxidants, perhaps produced by the interaction of the unfiltered UV radiation with the surface mineralogy, might have resulted in the consumption of all of the expected organic material (Bada et al. 2005; Beaty et al 2005; Beegle et al. 2007). Actually, the GC/MS on each of the Viking landers did detect a couple of organic compounds. Chloromethane and dichloromethane were detected. However, these chlorine compounds were thought to have been contaminants left over from the sterilizing fluids used to clean the spacecraft prior to departing Earth.

The reported LR results were apparently consistent with the presence of living organisms. This was determined by numerous soil tests on Earth prior to the launch of Viking (Levin 2007a). However, with no organics, these results were dismissed (Levin 2009; NRC 2006). This conclusion has been hotly debated for the last 34 years. But until recently, there was no new evidence to significantly contest this decision. Furthermore, this finding of "no organics" and "no life" has driven the design and mission for every lander and rover to visit Mars since Viking. If this decision was wrong, then we may be heading in the wrong direction. The recent discovery of perchlorates in the Martian regolith may allow up to correct our course.

Very little is known about the behavior of perchlorate salts in low-temperature aqueous solutions. We do know that they are very soluble and do not readily precipitate (Hanley et al 2009a; Hecht et al. 2009b). Just based upon thermodynamics, perchlorates form extremely strong oxidizing agents. However, kinetically, they are very inert. The coordination of the four oxygen atoms in the ClO₄- ion forms an energetically stable tetrahedral geometry around the central chlorine atom. This makes the anion generally unreactive (Brown & Gu, 2006; Catling et al. 2009). Therefore, under the ambient conditions typically found on the surface of Mars, perchlorates are stable and should not destroy organic material. This stability probably explains why Phoenix apparently found such a large concentration of it in the Martian regolith (Catling et al. 2009).

Most perchlorate compounds only become a strong oxidant when heated (Ming et al. 2009). So, how does this apply to Mars? As we have already said, Mars is significantly colder that Earth. Therefore, organics and perchlorates should be able to intermingle in the regolith with absolutely no destructive oxidation occurring (Hecht et al. 2009a). In fact, perchlorates are so stable, they do not even oxidize the organics they are found with on Earth. Terrestrial microbes not only safely coexist with perchlorates, some actually used it as an energy source in their metabolism (Brown & Gu, 2006; Hecht et al. 2009a; Hecht et al. 2009b; ITRC 2005; Newman & Banfield, 2009). However, the samples collected by Viking were analyzed using a pyrolytic GC/MS to release the volatiles. Phoenix also heated its soil and ice samples in the Thermal and Evolved-Gas Analyzer (TEGA) (Hecht et al. 2009a). Therefore, the very processes used to look for organics on Mars may have catalyzed their destruction (Ming et al. 2009). Heating these samples should activate the oxidative properties of most perchlorate salts, leading to the rapid – almost explosive – combustion of any organic material. So, the only way any organics could be detected is if their concentration was adequate to survive the perchlorates. Due to the high concentration of perchlorates that have accumulated in the Martian regolith, this would take more organic material than should be present due to natural biotic and abiotic means.

So, could the Viking landers actually have detected organics 34 years ago, only to be disregarded? And if there were organics present, should the positive results of the LR now be reinterpreted as definite evidence of extant Martian organisms metabolizing in the regolith? To test this idea, Navarro-González et al. (2010) added 1% (by weight) Mg(ClO₄) to soil taken from the Atacama Desert in Chile. This hyperarid soil is used by researchers as a terrestrial analog for Martian regolith (Catling et al. 2009, Houtkooper & Schulze-Makuch, 2010). The Atacama soil is known to contain very low levels of organic compounds. The sample was heated in a similar manner as the Viking GC/MS. Virtually all of the organics in the sample were immediately destroyed. This also resulted in the release of the same chlorinated methane compounds as were detected by Viking, and dismissed as contaminants.

Instead of looking for life, NASA is now expending their astrobiological efforts to discover quantifiable metrics to better define the limits of terrestrial life (Conrad 2009). They will then extrapolate these limits to design and deploy remote orbiters, landers, and rovers to scout out potentially habitable extraterrestrial sites. These sites may exist as widespread oases wherever the correct set of conditions coalesce to permit life to form and thrive. We can greatly reduce the mind boggling scope of this herculean task by identifying, in advance, where these unique oases might be found. However, there is a huge caveat to this approach. By definition, life must have a habitable place in which to form and thrive. But, just because we identify a site as being potentially habitable does not mean that anything has ever lived there (Lammer et al. 2009).

In this vein, the MERs (Spirit and Opportunity) and the Phoenix lander were not designed to look for life. They were given an array of instruments to seek out potential areas that are, or once were, habitable. Based upon our list of minimum requirements for a living organism, Viking, Phoenix, and the MERs have clearly shown us that at one time in Mars’ history it had all of the building blocks essential for the formation and maintenance of life (Conrad 2009; Crawford et al. 2008; Horneck 2000; Kounaves et al. 2002). This ongoing mission to discover habitable sites on Mars will continue when the Mars Science Laboratory (MSL) rover eventually rolls upon the surface of Mars (Beegle et al. 2007).

Table 2 lists many of the current and developing tools astrobiologists may use to explore the sites we have discussed (Table 1) for indications of extant or extinct life. The actual tool or tools we will use depends upon what we are looking for and where these biosignatures may be found. The Viking and ALH84001 controversies may have inspired a broader interest in astrobiology, but their debated results have also caused many to be very skeptical of what astrobiologists report. So, whatever we decide to take with us to Mars must include a comprehensive array of analytical instruments designed to perform advanced in situ and laboratory analyses established with strict controls and counter checks (Beegle et al. 2007). The following paragraphs address some of the tools from Table 2 in a little more detail.

Drilling: The need for subsurface access to Mars is apparent, given its potentially biologically harsh surface conditions (Sephton 2010). The presence of intense UV radiation and oxidants in the regolith has possibly resulted in a sterile surface layer. But, does this sterile layer actually exist? If so, how deep would we have to go to escape its effects? The European Space Agency (ESA) plans to equip their upcoming ExoMars rover with a 2-meter drill in an attempt to get under this sterile layer (ESA 2010). This drill will also collect samples at 25 cm intervals to map the subsurface chemistry and measure the depth of the oxidizing layer and UV penetration (Beegle et al. 2007; Christner et al. 2005; Crawford et al. 2008; ESA 2010).

A major problem associate with drilling, and all sampling, is the risk of contaminating Mars with organisms we brought with us from Earth (NRC 2006). This is called "forward contamination". Besides the damage this could do to any extant Martian life, our organisms could also lead to false positive indications for organic material and life on Mars (Beaty et al. 2005; Christner et al. 2005). If organic molecules and other biosignatures are on Mars, they may be very scarce. The background noise generated by our forward contamination could easily blanket out the sensitivity of any sampling devices we may employ (Eigenbrode et al. 2009). In fact, all of the landers and rovers we have sent to Mars over the last 34 years may have already done irreparable damage. After all, we know that viable microorganisms in spacecraft do survive in space for prolonged periods of time (Horneck et al. 2010). So, there is little doubt that they can, and have, already survived multiple trips to Mars (Chyba & Hand, 2005; Crawford 2003, 2005).

To protect Mars from this source of contamination, as well as the integrity of future science missions, we need to establish robust anticontamination procedures (Crawford 2005; NRC 2002). Aseptic sampling techniques can be extremely tedious and time consuming, even on Earth. They can also be far from effective (Eigenbrode et al. 2009). Ice and rock drilling and coring procedures are inherently dirty processes. A great deal of effort goes into trying to preventing cross-contamination between layers while maintaining the integrity of the samples (Levin 1968; Mancinelli 2000; NRC 2002). A new method for drilling and coring has been proposed that utilizes an electrically heated dill head (See Table 2 and Figure 5) that can melt its way through the Martian surface (Mancinelli 2000). If successful, this self-sterilizing approach could solve many of the contamination issues associated with subsurface sampling.

However, all of these efforts may turn out to be for naught. We have learned a major lessen while performing our terrestrial geomicrobiological field studies, as well as conducting manned expeditions to the moon. No matter what we do, when humans land on the surface of Mars the deposition of microbes from Earth onto the Martian surface (forward contamination) and the exposure of the crew to Mars' surface material (back contamination) may be inevitable (Mancinelli 2000).

Redox: The need for all known living things to have an external source of energy suggests an on-site method for the detection of extraterrestrial biosignatures. By setting out to "chase the physics", we have developed a technique based upon fundamental thermodynamics properties and our basic assumptions regarding how life obtains energy from its environment (Crawford et al. 2001). In terrestrial biological systems, this energy is tapped by organisms via stepwise electron (e-) transport through enzymatically directed oxidation–reduction (redox) reactions along an electron transport chain (ETC) (Crawford et al. 2001, 2002; Lang et al. 2002). An ETC is central to the catalysis of redox reactions involving geobiochemically abundant species such as Mn and Fe. The biological manipulation of these materials profoundly changes their forms in the environment (Newman & Banfield, 2009). By focusing our efforts upon potentially redox-active sources on Mars, we can conduct rudimentary mineralogical biosignature detection and analysis (Banfield et al. 2001, 2005). In order to tap into the chemical free-energy flowing between redox agents, living organism have developed metabolisms that allows them to consume and store this energy in a controlled manner (Lang et al. 2002). Terrestrial microbes couple the production of adenosine triphosphate (ATP) to the enzymatic transport of electrons from a vast array of inorganic and organic substrates. Granted, any approach based upon this concept would only detect the biosignatures of currently viable organisms. However, we could extend this process to include dormant extremophiles (e.g., spores) if the collected samples were also given trace amounts of water so that these anabiotic entities could once again germinate and grow (Crawford et al. 2002).

There are a wide range of redox agents that are readily available on Earth. Microbes utilize them by shuttling electrons between redox pairs (acceptors and donors) with various redox potentials. For example, NH₄+, hydrocarbon molecules, reduced and oxidized metals, atmospheric gases (CO₂, CH₄, H₂, H₂S, CO, and O₂), and water are only a few of the available options. These materials are found in environments characterized by redox transitions (e.g., in sediments or at the interface between basaltic rocks) (Banfield et al. 2001). They can also be regenerated for repeated use through redox cycling (e.g., Figures 3 and 6) and photosynthesis (Lang et al. 2002). As we have already discussed, many of these redox agents are widely available on Mars.

Iron (Fe) is one of the most abundant redox-active components we know of in the Martian lithosphere (Banfield et al. 2001). Oxidized or ferric iron (Fe₃+) is what gives Mars its characteristic reddish hue. Fe₃+ is used as an e- acceptor by terrestrial microbe operating under anoxic conditions (Zhang et al. 2009). These same conditions can be found on the surface of Mars. There is also the possibility of a microbial iron-cycle (Figure 6) developing. Extant organisms could utilize the Fe₃+ present in Martian dust and sediments as a readily available e- acceptor. Other organisms could then take this reduced iron (Fe₂+) and oxidize it again, producing a fresh pool of Fe₃+. These changes in Fe redox states can also be linked to microbial carbon and energy flows as well as to the behavior of various inorganic compounds in soils and sediments (Banfield et al. 2001, Nickles 2010). Without a significant ozone (O₃) layer (Crawford et al. 2003; NASA 2007), Mars experiences a significantly higher UV radiation flux (UVB: 280-315 nm and UVC: 100-280 nm) than anywhere on Earth. Iron-bearing minerals can act as an UV screen to protect microorganisms (Amaral & Frais, 2007). So, given the right habitable conditions, Fe redox biogeochemistry could certainly sustain life on Mars while producing a cache of mineralogical biosignatures for us to detect (Crawford et al. 2002; Lang et al. 2002; Tratnyek et al. 2001).

Respiratory or "redox dyes" are a convenient method for directly observing redox chemistry in terrestrial soils. Theses dyes can be used to detect this process by "short-circuiting" the movement of electrons between the e- donor and acceptors (Crawford et al. 2001). We are currently using this concept to develop a redox biosignature detector (RBD) as an in situ method for analyzing Martian rock and regolith samples. This method that can be easily performed by remote probes or by our human explorers on the surface of Mars.

We use redox dyes to act as chemical probes to obtain fundamental insights into biogeochemical processes. Environmental scientists routinely used these dyes to study the redox properties of various natural systems. For example, indigo tetrasulfonate (I4S) can be reduced in anaerobic sediments. The results can be easily observed. The oxidized form of I4S is blue. When I4S is reduced by electrons in the sediments it turns a contrasting yellow (Tratnyek et al. 2001). Other researchers have used 5-cyano-2,3-ditolyl tetrazolium chloride (CTC). The oxidized form of CTC is nearly colorless and non-fluorescent. Once reduced, respiring bacteria can be directly located, observed, and enumerated using an epifluorescent microscope (Crawford et al. 2002).

The current design for our redox biosignature detector (RBD-2), detects potential biosignatures by observing the effects of applied redox dyes to our samples (Crawford et al. 2008; Lang et al. 2002). RBD-2 detects respiratory e- transport using the artificial e- acceptor (reduced) 2,3-dichlorophenol indophenol (DCIP) and the tetrazolium dye 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide inner salt (XTT) (Crawford et al. 2002). Living organisms in pure cultures and in soils samples readily reduced DCIP and XTT. These reductions have been visually observed. The results can also be quantified using a spectrophotometer. We have supplemented the basic approach by extracting, separating, and identifying a few of the responsible redox-active biochemicals (e.g., porphyrins or quinones) from microorganisms in regolith or powdered rock samples.

Variations of our RBD concept were verified, tested, and further developed using terrestrial soils. For example, one set of experiments utilized extracts taken from a nutrient-poor Idaho red sand and an Antarctica soil collected from the extreme environment of Lake Boney Basin. The red Idaho sandy soil was found to contain ~3x10⁴ cultivable cells per gram. As a control we also demonstrated that uninoculated and sterilized soil samples do not reduce these dyes (Figure 7). The ability of soil microbial communities to reduce artificial e- acceptors was also directly measured (Crawford et al. 2002). Based upon the initial success with our current redox dye RBD-2 concept, we plan to further develop the process in our work with rock varnish and Antarctic cryptoendolithic communities (Nickles 2010).

This general concept of detecting biologically mediated redox reactions forms the heart of our multistep procedure. Our basic RBD can also extract and separate redox signature chemicals and then analyze them using a standard set of equipment which can be readily miniaturized. Our ultimate goal is to integrate the RBD concept onto a ruggedized "laboratory-on-a-chip".

Our initial design (RBD-1) did not utilize redox dyes (Figure 8). The primary components of RBD-1 included a supercritical fluid extraction (SFE) module, a separation module, an amperometric detection module (no redox dyes), and an electrospray ionization tandem mass-spectrometer (ES-MS) (Crawford et al. 2001). The amperometric electrochemical detector (ECD) was used to detect the redox active compounds. An ECD is basically a simple device that can be easily miniaturized and packaged onto a lab-on-a-chip. ECDs are also highly selective for redox-active components and can perform their nondestructive analyses over a wide dynamic range (Lang et al. 2002). The potential range for typical e- donor and acceptor pairs vary from 11 to 21 V. Our initial ECD used a dual electrode series. The first electrode (reducing potential) was set at 20.5 V. The second electrode, set at 11.0V (oxidizing potential), was the detector.

The current RBD-2 concept replaces or supplements the ECD with redox dyes (Figure 9). As with RBD-1, we can also extract suspected signature compounds from the sample using SFE with supercritical CO₂ (SCCO₂). This step is most effective when the cells in the samples are completely disrupted. Twenty minutes of ultrasonic-aided mechanical cell disruption is used. Longer times do not result in better high performance liquid chromatograph (HPLC) responses. After disruption, the sample is subjected to hydrolysis to separate the redox compounds from any larger organic molecules to which they might be attached (Crawford et al. 2002, Lang et al. 2002).

SCCO₂ was chosen as the primary extraction agent because it is a powerful, tunable solvent that can extract both polar and nonpolar molecules from the samples. Since the extraction media is CO₂, it can be easily removed by releasing the pressure and venting it from the extraction residue (Crawford et al. 2001). Additionally, CO₂ is readily available on Mars. It is the primary constituent of the thin Martian atmosphere (~95%) and is a major component of the polar caps (dry ice) (Crawford et al. 2002, 2008; Lang et al. 2002; NASA 2007). Therefore, our manned mission to Mars could forgo the need bring terrestrial organic solvents. This would save on overall mission weight and complexity. This would also reduce the risk of an organic spill resulting in the serious long-term contamination of the surface of Mars (Crawford 2005; Eigenbrode et al. 2009).

With an instrument package such as RBD-2, the collected soil (pulverized to a dust) is used to inoculate sterile microtiter plates containing a wide variety of media containing the most likely combinations of electron donors and acceptors that might be naturally available to Martian microbes (e.g., CH₄or CO as donors, and Fe₃+ or even perchlorate (an anaerobic e- acceptor recently detected in Martian regolith) as acceptors). A positive response (microbial growth) would be indicated by a contrasting visual change in color (Figure 7). These dramatic color changes can be observed visually, photographed, and quantified using spectrophotometry. For example, the absorbance of inoculated wells containing the oxidized DCIP is measured at 600 nm. Reduction of the initially colorless XTT increases absorbance at 465 nm due to the orange color of the reduced material (Crawford et al. 2002). These data could be used to characterize the thermodynamic and kinetic properties of a potential Martian microbial ecosystem (Lang et al. 2002; Tratnyek et al. 2001).

If RBD-2 receives a "positive" visual indication for a given media inoculated with Martian soil, a portion of the sample could then be diverted for further analysis (dashed blue arrow in Figure 9). The responsible redox compounds could then be extracted by the SCCO2 SFE, separated by the HPLC, and injected into the ES–MS for possible identification. The core structures of these Martian redox-mediating molecules might resemble the porphyrins, quinones, flavins, and nicotinamides which perform a similar function in terrestrial organisms. If so, the ES-MS might be able to identify terrestrial compounds that are similar to these Martian signature redox species. But is it reasonable to assume that similar redox-active molecules could support a similar set of metabolic processes in an extraterrestrial life form? In the case of Mars, that might be a reasonable assumption. Both planets are believed to have experienced a similar set of environmental conditions during their first billion years or so of existence. So, since they must both follow the same set of natural physical laws, similar circumstances could very well result in a similar set of initial organisms utilizing a similar set of biochemical processes (Kounaves et al. 2002). Granted, the actual molecular structures for these compounds could vary in some details from those seen on Earth. But, similar redox molecules performing a similar biochemical role should possess similar core structures. If so, these structures should be recognizable by their characteristic fragmentation patterns (Crawford et al. 2001, 2002).

The primary limitation of the RBD-2 design is the sensitivity of the redox dyes. If there are viable organisms at all in our samples, they should be able to grow in an appropriate medium to a concentration adequate to elicit a positive redox response (color change). However, there might not be enough cells in a small sample to trigger the extant life identification portion of the process. On Earth we assume we can only detected about 1% of the actual living cells present in biologically poor, sandy soil samples (typically ~3x10⁴ cultivable cells per gram). If that limitation holds for Mars, then the detection limit of the RBD would be ~3x10⁶ cells/g. This problem could lead to controversy over the initial findings. A solution to this dilemma is to also integrate the ECD from RBD-1 into the RBD-2 design. The ECD has extremely low detection limits, especially when coupled to an HPLC or capillary electrophoresis (CE) for separation. The reported detection limit for CE amperometry is 10^-19 mol (Lang et al. 2002). However including the ECD would increase the complexity of the RBD-2 lab-on-a-chip design. Our experiments show that given an adequate sample size, the components on RBD-2 should be able to identify the responsible redox components, even if the overall cell density is inherently low. Therefore, this sensitivity problem could also be solved by analyzing a larger volume of sample material (microbial growth experiments) and with a continuous sample flow process (biosignature analysis).

RBD-2 can be used in a variety of situations to analyze samples in the field. It should be field test first using samples collected from locations such as the Dry Valleys of the Antarctic or the Atacama Desert. After validation, it could then be deployed to Mars. The RBD-2 design could encounter site-specific challenges in the extreme Martian environment. This is a major lesson learned from the controversy surrounding Viking’s biology results, and should be considered when developing an appropriate set of abiotic controls. The conditions on the Martian surface are severe (Crawford et al. 2008; NASA 2007). The regolith is acidic, arid, and exposed to high energy, oxidant-generating UV (Atreya et al. 2006; Benner et al. 2000; Crawford et al. 2003; ESA 1999; Horneck et al. 2010; Onofri et al. 2008; Wu 2007). So, organic compounds such as DCIP or XTT might be rapidly destroyed by the very samples we hope to analyze. Even though we probably can avoid this problem by taking our samples from subsurface material, this does illustrate the need to use more than one type of redox dye (Figure 9).

A wide variety of redox dyes would be incorporated into a microarray. Each microarray well would contain a different permutation of dye and redox pair combinations in an aqueous phase. There would be three series of wells. The first series would be inoculated with raw powdered sample material. If cells utilizing redox related metabolism are present in this material they should eventually produce a positive response in some or all of the wells. The second series of wells would contain an appropriate biocide to sterilize the added sample material. We have used 2% cresol or azide for this purpose in the past. These would serve as the abiotic controls. Finally, a third series of wells would not be inoculated at all. These wells would serve as a basis of comparison for determining where a positive or negative response occurred in the two inoculated series. A spectral scanner, or even a simple color camera, would be used to record the three series of wells. These data would be relayed to the human operators for further analysis. Coupled with the additional analyses for specific electron transport molecules (as related to their terrestrial counterparts), the use of RBD-2 could provide very convincing evidence of e- transport within the biological redox range. Therefore, RBD-2 would be a useful tool for detecting signs of life in extraterrestrial locations.

Once a potential redox biosignature has been detected it will still be necessary to conduct further analysis. Therefore, RBD-2 could easily be made part of a multiple instrument payload designed to examine a wide array of potential indicators of life. For example, integrating DNA or lipid analysis components into the overall modular design of the RBD-2 lab-on-a-chip should be a relatively easy addition. Granted, including these elements into our search could be consider a very Earth-centric approach. However, once a potential biosignature has been located, these options could provide useful supplemental information regarding aspects of a potentially new life form (Crawford et al. 2001; Lang et al. 2002).

Flyers and Penetrators: How could a small team of human explorers possibly scout out all of the potential sites listed in Table 1? Each site would require a preliminary survey prior to being considered for more exhaustive analyses. If this survey indicated that the site was potentially habitable, the team could then conduct a more detailed investigation. But even performing an initial survey on so many sites could take years. One approach for streamlining this process is to utilize a battery of inexpensive, ruggedized dagger-penetrator microprobes (Figure 3) deployed from autonomous airborne platforms (airplanes or airships). A network of simple microprobes, containing a basic set of instruments (Table 3), could be used to remotely scout out the more promising sites identified by all the previous orbiters, landers, and rovers missions.

These dagger-shaped penetrator microprobes would be small enough and inexpensive enough for numerous deployments over the entire planet. The unique blade-shape of the dagger-penetrator was selected to allow for the maximum possible penetration per unit mass, with the lowest possible impact dynamics imparted upon the payload. Depending upon the deployment altitude, these penetrators should encounter the Martian surface at approximately 100 to 200 fps with impact decelerations ranging from 20,000 to 50,000 G’s in the handle/hilt and blade subassemblies (NASA 1999). The actual physical properties of the encountered Martian surface (e.g., regolith, permafrost, glacial ice, etc.) could be extrapolated by how far the blade subassembly penetrates the surface. Depths of 0.5 to a few meters should be possible.

Much like the Deep Space 2 probes deployed with the ill-fated Mars Polar Lander in 1999 (NASA 1999), these dagger-penetrators are designed to separate upon impact. When the base of the "hilt" encounters the surface, the blade would separate, leaving the hilt/handle subassembly above the surface. Table 3 and Figure 10 provide details regarding the proposed configuration and payload for this microprobe. The hilt and handle would contain some basic surface meteorological sensors, datalogger, power source, and the datalink antenna and electronics. The dagger-blade-penetrator would be attached to the hilt subassembly via an umbilicus that would relay data and power. The blade would contain a simple set of ruggedized instruments to report on the presence of water, oxidants, redox biosignatures, and/or living cells able to reduce respiratory dyes (RBD-2S). The surface and subsurface data would be compiled and datalogged for periodic databurst transmissions to an orbiter. The orbiter would then relay these data back to Earth and to the human explorer teams on the Martian surface. These teams would assess the data collected from the planet-wide dagger-penetrator microprobe network to efficiently coordinate their more detailed life detection efforts.

4. WHEN Do We Look?

Currently, we only have a fundamental grasp of how living organisms function on our home planet. Even with that limited data set, we are constantly learning new limits and possibilities to life. We need to take our fledgling terrestrial knowledge and use it to develop clues about WHAT, HOW, and WHERE to look for signs of extraterrestrial life (Nickles 2010). As we do, we must be careful not to be like the man who lost his car keys in the dark and decided to search for them under the streetlight, because it was easier and safer. What he needed to do was determine where the keys could possibly be, and then safely equip his self to look in those places. Sure it could be harder, and might incur an increased element of risk, but it is his only hope of success. In a similar vein, if we decide that it is difficult and risky to access the sites where we believe the signs of past and present extraterrestrial life might be, then we need to know that ahead of time. Only then can we design the right tools and methods for the job. We cannot pick a site just because it is the safest place to land with the lowest technical risk. This is especially true of a manned mission. Only by combining the proper balance of hands-on and remote operations on the surface of Mars can we be sure that if life is there, we will find it.

There is always the chance that what we find will be so foreign that we may not even recognize it as life. This alien organism could conduct unknown metabolic processes in ways that have never been conceived, in environments we never even imagined (Levin 1968). If so, this would be an incredible challenge for our science, engineering, and philosophy. However, this would also be exactly what we ultimately want to find; a whole now approach to life. This find could teach us so much about who we are and where we came from. For, if we found life on Mars that was identical to life as we know it on Earth, what have we learned? Additionally, there will always be the nagging suspicion that what we found was no more that the remnants of forward contamination from one of our previous missions (NRC 2006).

This chapter has briefly addressed many of the Who’s, What’s, Where’s, and How’s regarding the human search for life on Mars. The remaining question is When? When do we send people to complete this task? As with all science and engineering endeavors, we need to use a "build-up" approach. We need to take what we learn as we go along to develop and deploy a progressively more sophisticated array of remote tools. This is the only way of ensuring the widest possible chance for success with the inevitable manned visit to Mars. As the Viking and ALH84001 experiences taught us, a positive result may be achievable, but there is not way to obtain a definite negative. There would always be the question of whether we used the right tool or even looked in the right place (Beaty et al. 2005). So, if we do eventually get a positive result with our advanced tools, then we will need to be ready to deploy the most powerful analytical tool we currently have available to study this life-altering discovery. We will need to deploy the human mind to Mars.

Acknowledgements: The authors would like to thank Dr. Pamela Conrad and the Jet Propulsion Laboratory (JPL) for the donation of two sandstone samples from their 2005 expedition to the McMurdo Dry Valleys in Antarctica. We would also like to thank the Environmental Biotechnology Institute (EBI) and the NASA Idaho Space Grant Consortium (ISGC) for their generous funding of our research.