Chapter 2 presented an overview of the work of the SterLim and JAXA (Japan Aerospace Exploration Agency) teams, together with a review of the models and assumptions used therein. This chapter investigates some additional arguments regarding planetary protection requirements for a sample return mission from the martian moons. The organization of this chapter is designed to provide the committee’s answers to the questions posed in the statement of task (see the Preface).
The committee’s first task was as follows:
Review, in the context of current understanding of conditions relevant to inactivation of carbon-based life, recent theoretical, experimental, and modeling research on the environments and physical conditions encountered by Mars ejecta during the following processes:
- Excavation from the martian surface via crater-forming events;
- While in transit through cismartian space;
- During deposition on Phobos or Deimos; and
- After deposition on Phobos or Deimos.
In this context, the committee reviewed the work of the SterLim and JAXA teams. Subsequent sections of this chapter summarize the methodology of the two teams; the issues identified by the committee were discussed in Chapter 2.
Potential Microbial Density on the Martian Surface
The SterLim team assumed the same microbial density as soils from the Atacama Desert. The JAXA team used a similar number with a slight correction.
Finding: The committee finds that if life exists on Mars, its cell density and even its biochemical nature, is unknown. Therefore, the cell-density estimate employed by the SterLim and JAXA teams is, as is appro-
priate for a planetary protection calculation, a very conservative estimate based on current understanding of life as it exists in Mars-like extreme environments on Earth.
Mars Ejecta Formation and Transportation from the Martian Surface
SterLim models were based on the 2011 report by Melosh et al.1 The JAXA team used smoothed particle hydrodynamics (SPH) computations newly conducted for statistical analysis. The models of the SterLim and the JAXA teams predict significantly different amounts of mass transported to the martian moons. For Phobos, the SterLim model predicts 1.6 × 106 kg (comparable to the 1.1 × 106 kg from the 2011 report from Melosh et al.), while the JAXA model predicts 5 × 107 kg.
Finding: The committee cannot identify why there is a significant discrepancy in the amount of material transported to the martian moons as determined by the SterLim and JAXA teams. Nevertheless, these uncertainties represent, in some sense, the current state of the art.
Sterilization During Mars Ejecta Formation
The SterLim team did not use a specific model (i.e., the microbial survival rate was 100 percent). The JAXA team added a sterilization model during meteoroid impact according to a numerical simulation.
Finding: The consensus in the committee is that shock heating is a highly localized process. When trying to resolve this adequately in numerical simulations, very high spatial resolutions are required.
Finding: The committee was unable to define a survival rate based on the information available. However, the proposed survival rate of 10 percent is a reasonable estimate, albeit one lacking significant experimental evidence.
Sterilization by Aerodynamic Heating of Mars Ejecta
The SterLim team did not include such sterilization. The JAXA team conducted thermal analysis of Mars ejecta along various trajectories.
Finding: The committee finds that the JAXA team’s conclusion that particles smaller than 10 cm do not escape the martian atmosphere is not well supported, and that subsequent analyses relying on this limit be treated with care.
Finding: The committee supports the JAXA team’s conclusion that aerodynamic heating of ejecta during passage through the martian atmosphere does not cause any significant sterilization.
Sterilization During Hypervelocity Impact on Phobos/Deimos Surfaces
The SterLim team calculated a microbe survival rate ~ 0.1 for velocity <2 km/s, based on impact experiments. The JAXA team reworked the impact sterilization model using SPH and trajectory analysis, resulting in a survival rate of only 10−4.
Finding: The committee finds that the experimental hypervelocity impact data generated during the SterLim study are limited with respect to the large spectrum of possible impact conditions on the martian
1 H.J. Melosh, K.C. Howell, L. Chappaz, and M. Vaquero, 2011, “Material Transfer from the Surface of Mars to Phobos and Deimos,” 22nd AAS/AIAA Space Flight Mechanics Meeting, Univelt Inc., San Diego, Calif.
moons, could be biased, and are not conclusive. Given the small footprint of the data within the vast parameter space, extrapolations drawn from the experimental data currently available seemed ill-advised. The committee notes that SterLim’s impact data were then used to calibrate the exponential function used by the JAXA group to estimate and extrapolate the likely sterilization due to impact.
Distribution of Mars Ejecta Fragments by Impact, Recirculation, and Reimpact
The SterLim team assumed a homogeneous deposition by averaging the incoming flux. The JAXA team took account of crater formation by Mars ejecta with retention and scattering of Mars ejecta fragments.
Finding: The committee finds that the estimations of the two teams were based on the different and limited experimental data (the SterLim team used its own data, and the JAXA team used literature data). Therefore, a factor of uncertainty remains in the fraction deposited at the first impact.
Sterilization by Radiation on Phobos/Deimos Surfaces
The SterLim team used a sterilization model based on experimental data. The JAXA team used a similar model but averaged the microbial survival rate for its specific depth by integrating the survival rate in the depth direction and then dividing by the depth. This enables the JAXA team to account for the fact that the microbial density decreases toward the exposed surface (the averaged density for each depth is smaller than the local density at each depth, but the sampling operation gathers material along the full depth path).
Finding: The committee finds that the use of aluminum, rather than chemically inert surfaces, as a simulant environment for irradiation on Mars/Phobos is problematic. In addition, the samples were irradiated in a frozen state, whereas the surface temperatures on the surfaces of the martian moons is frequently above the freezing point of water.
Finding: The committee finds that diurnal temperature cycling is an extremely significant factor in determining the survival of martian organisms deposited on the surfaces of Phobos or Deimos. Desiccation is bactericidal to even the most radiation-resistant microbes in a matter of months.
Phobos/Deimos Surface Reformation by Natural Meteoroid Impacts
The SterLim team expected this effect to be low and did not take it into account in its calculations. The JAXA team calculated the continuous natural meteoroid impacts on martian moons. Neither team considered the effect of thermal fatigue, which is likely to occur at a rate orders of magnitude faster than fragmentation due to micrometeoroid impact. The fragmentation of surface material into smaller pieces could significantly enhance the rate at which any biological material present is degraded by exposure to the radiation.
Finding: The committee finds that the sterilizing effect of meteoroid impacts following deposition of martian material on the surface of Phobos and Deimos to be minor due to the low flux of impactors. However, the effects of thermal fatigue could significantly enhance the rate at which any biological matter present is exposed to and degraded by radiation.
The second task of the committee was as follows:
Recommend whether missions returning samples from Phobos and/or Deimos should be classified as “restricted” or “unrestricted” Earth return in the framework of the planetary protection policy maintained by the ICSU Committee on Space Research (COSPAR).
The SterLim team’s recommendation on restricted or unrestricted changed between the first and the second meeting of this committee. Initially, their recommendation was for an “unrestricted” mission. However, additional work—conducted between the committee’s two meetings—on individual cratering events at Mars (“discrete ejections”) in general and the Zunil impact crater in particular, resulted in a recommendation for restricted Earth return. The latter recommendation was based on the uncertainty as to whether an unidentified large (>10 km) young crater might exist on Mars that could have contributed significantly to the deposition of martian materials on Phobos. All such craters less than 20 Ma old appear to have been identified in a search conducted by Werner et al.2 The SterLim team concluded that unless such large craters can be shown to be ancient (at probabilities over 99.9999 percent), then Phobos cannot be said to be free from hypothetical martian life. They noted that if the Mars ejecta contains low levels of life, the chance of transferring hypothetical martian organisms to Deimos is below the 10−6 criterion.
Finding: The committee finds that it is highly unlikely that such a large, young crater exists and has somehow escaped detection.
The JAXA team felt that due to the large uncertainties in many key assumptions used in the models (most importantly, the initial bioload of the martian surface), they were unable to explicitly say whether the probability of sampling a live microbe was greater or less than 10−6. However, they made a comparative argument based on the amount of martian material that arrives on Earth naturally, in the form of martian meteorites, to the amount of martian material expected to be present on Phobos (or Deimos) that would be delivered to Earth by a sample return mission. They investigated another “chain of events,” that of ejecta transported from the martian surface to an interplanetary trajectory, and finally hitting Earth, after passing through Earth’s atmosphere. Their conclusion was that the natural flux of direct samples from Mars to Earth is orders of magnitude greater that the flux from robotic sample return. Thus, samples returned from the martian moons should be characterized as unrestricted, regardless of the uncertainties.
The JAXA team’s argument was based on the inventory of martian meteorites (a total of ~19.2 kg) that have been identified on Earth. This committee has revisited the same argument, to elaborate further this issue.
Ratio of Natural to Spacecraft Flux of Martian Material to Earth
In considering the planetary protection requirements for sample return missions from Phobos and Deimos, it is instructive to compare the natural flux of martian material to Earth relative to that of a robotic sample return mission. Assuming life exists in the surface and near-surface regions of Mars, the flux ratio is a proxy for the relative contribution of spacecraft-transported martian life-forms arriving on Earth compared to those arriving on Earth naturally. The JAXA team calculated that the maximum mixing ratio of martian material to Phobos regolith for the material coming from the Zunil impact was ~100 ppm.3 Therefore, a 100 g sample collected by a nominal Phobos sample return mission will return 10−5 kg of Mars-derived material to Earth. As shall be shown, this mass is negligible compared with the mass of unsterilized Mars material that has arrived on Earth over the last million
2 S.C. Werner, A. Ody, and F. Poulet, 2014, The source crater of the martian shergottite meteorites, Science 343(6177):1343-1346.
3 See Section 9 of in K. Fujita, K. Kurosawa, H. Genda, R. Hyodo, T. Mikouchi, S. Matsuyama, and the Phobos/Deimos Microbial Assessment Team, 2018, “Assessment of Microbial Contamination Probability for Sample Return from Martian Moons,” GNG-2018003, http://sites.nationalacademies.org/SSB/CurrentProjects/SSB_181917.
years since the Zunil impact. Indeed, the argument holds even if there is a comparatively young, large crater that has been overlooked so far.
Following the work of Gladman,4 a minimum of 15 Mars rocks with a mass of ~100 kg arrive annually.5 It is known that the preentry masses of Mars rocks may exceed 10 kg.6 Taking into account only the Zunil impact, the ejected mass is around 2 × 1010 kg.7 Gladman estimated that ~5 percent of this material eventually will hit Earth and that 0.5 percent hit in the first million years,8 whereas Mileikowsky et al. estimated that 6 percent hit over a period of 10 million years.9 The accretion rate of Mars rocks by Earth over the first 10 million years is approximately constant and negligible after that time. Thus, the mass from Zunil that has already hit in the last million years is 108 kg. This mass applies to Zunil alone, but within uncertainty (an order of magnitude) it also applies to the flux from the small number of Mars rock producing impacts in the last 10 million years.
Thus, about 108 kg of Mars (i.e., 0.05 percent of 2 × 1010 kg) have arrived in the last million years, compared with 10−5 kg in the proposed Phobos sample. Therefore, the ratio of spacecraft flux to natural flux over the last million years is 10−13.
The committee compared the relative influx of microbes from the Phobos sample to the natural influx following the series of vicissitudes outlined in the JAXA report (see Chapter 2). The processes that eject material from Mars into space are essentially the same. Rock-size (~kg) pieces are needed to survive transport to Earth and to embed themselves into the regolith of Phobos. Rocks reaching Phobos and Earth both need to be ejected at cosmic velocities. The ejection velocity for Earth is higher than for Phobos. Weakly shocked Mars rocks where microbes would have survived are known,10 so this effect only modestly reduces the flux to Earth relative to Phobos.
The committee considered sterilization of microbes within Mars rocks on the way to Earth with sterilization of rocks on the surface of Phobos. In both localities, the number of living microbes decreases exponentially with time:
N = N0 e-t/trad
where N0 is the initial number of live microbes, t is time of radiation exposure since leaving Mars, and trad is the time scale for microbial death.
The JAXA report discusses the Zunil crater on Mars with a rounded age of 1 million years. One in 104 buried microbes survived on Phobos, giving trad = 108,000 years. The most recent impact dominates the microbial load on Phobos. Importantly, any microbe now collected on Phobos has endured this sterilization.
- For Earth-bound ejecta, 1 in 108 microbes now arriving at Earth survived radiation, because trad is 54,000 years. The radiation times differ by a factor of 2 (with Phobos trad of 108,000 years) because Earth-
4 B. Gladman, 1997, Destination: Earth. Martian meteorite delivery, Icarus 130(2):228-246.
5 One obtains the total influx of meteorites to the Earth by taking a target of known surface area (for example licensed cars in North America) where impact of a meteorite will be reported and the extrapolating to the full area of the Earth. Then one multiplies by the fraction of Mars rocks ~1/20,000 to total meteorites observed in falls and collections from glacial ice and deserts where there is little sampling bias. Note that Mars rocks resemble Earth rocks. Their provenance is rarely recognized unless they are observed to fall, hit human-made objects, or are collected from deserts and glacial ice. Almost all Mars rocks material fall in the oceans or impact rural areas and are never detected. The tiny fraction of Mars rocks that end up in collections comes into researchers’ flux calculations only in determining the ratio of Mars rocks in total meteorites.
6 Gladman (1997) shows that atmospheric entry results in an ablative mass loss of some 40-to-60 percent. Thus any meteorite with a mass greater than approximately 5 kg had a preentry mass of 10 kg or greater. Ten of the 104 martian meteorites cataloged in “The Martian Meteorite Compendium” (see https://curator.jsc.nasa.gov/antmet/mmc/index.cfm) maintained by Astromaterials Research and Exploration Sciences curation facility at NASA’s Johnson Space Center have masses greater than 5 kg. For example, the Tissint and Zagami meteorites have masses of 12 and 18 kg, respectively.
7 As determined by integrating the curves in Figure 8-7 of the JAXA report above the escape velocity. See Figure 8-7 in Fujita et al., 2018, “Assessment of Microbial Contamination Probability for Sample Return from Martian Moons.”
8 B. Gladman, 1997, Destination: Earth. Martian meteorite delivery, Icarus 130(2):228-246.
9 See p. 399 of C. Mileikowsky, F.A. Cucinotta, J.W. Wilson, B. Gladman, G. Horneck, L. Lindegren, H.J. Melosh, H. Rickman, M. Valtonen, and J.Q. Zheng, 2000, Natural transfer of microbes in space; 1. From Mars to Earth and Earth to Mars, Icarus 145:391-427.
10 B.P. Weiss, J.L. Kirschvink, F.J. Baudenbacher, H. Vali, N.T. Peters, F.A. MacDonald, and J.P. Wikswo, 2000, A lower temperature transfer of ALH 84001 from Mars to Earth, Science 290:791-795.
bound martian ejecta receives radiation from all directions, whereas the Phobos sample is irradiated only from above.
- It is important to note that some martian ejecta (meteoroids) reached Earth soon (i.e., months) after the Zunil impact and, thus, essentially escaped radiation sterilization during their journey to Earth.
- The chance of a Mars rock (that eventually reaches Earth) reaching Earth is 1 in 10−7 per year and constant at small (greater than a few months and less than few million years) times.11
- Thus, about 0.5 percent of the ejecta (that arrived in the last 1 million years since the Zunil impact) arrived during the first trad and were unscathed by radiation.
- This fraction would dominate the total flux of live microbes. That is, the Phobos fraction depends on a negative exponential of time since the impact, while the Earth fraction of live microbes depends linearly on the decay time.
Next, Phobos organisms needed to survive impact on a solid surface at cosmic velocities. About one in 104 microbes are expected to survive. Somewhere between 10 and 100 percent of the microbes in a sample survive passage through Earth’s atmosphere. This survival rate comes from the numerical analysis undertaken by the JAXA team. With an atmospheric entry velocity of ~5 km/s, the JAXA team used the same atmospheric sterilization model that it used for Mars and the same sterilization criterion (i.e., heating to 773 K for 0.5 second) to show that meteorites >10 cm across suffer a survival rate of between 20 and 80 percent. This fraction is somewhat misleading, as parts of the Mars rock are strongly heated and sterilized, while other parts of the rock remain unheated and unscathed. Impact at a gentle velocity on Earth’s surface is benign. Assuming an overall survival rate of 100 percent for atmospheric entry and landing is, at face value, the conservative choice. However, for the purposes of this calculation, it is preferable to underestimate the number of microbes arriving via martian meteorites. This may appear paradoxical. But the goal is to calculate the ratio of microbes potentially contained in a sample collected on Phobos or Deimos to the natural flux of life-forms to Earth via martian meteorites. Overall, the conservative choice is to maximize the ratio by minimizing the denominator. Therefore, the conservative choice is to assume a 10 percent survival rate during atmospheric entry and landing.
Collecting terms and retaining only orders of magnitude, 10−4 of the microbes from Phobos survive impact on its surface. Of these, a fraction of 10−4 survived subsequent radiation, giving a net survival of some 10−8.
A fraction of 10−2 of microbes evaded radiation by quickly coming to Earth. Of these, a fraction of 10−1 reached Earth’s surface alive, giving a total survival of 10−3. This ratio of spacecraft to natural live relative microbial fluxes, 10−5, is multiplied by the mass flux ratio of 10−13, to give an overall ratio of 10−18.
Note that this ratio remains small (~10−12) even if the current sterilization factor of 10−8 is used for radiation rather than the overall rate of 10−2 for rocks not hitting Earth.
For comparison, the committee estimates the flux of live microbes to Earth during the last 50,000 years, when modern humans were present. The mass flux was similar to that in the first 50,000 years after Zunil, 108 kg of Mars, compared with 10−5 kg in the sample, for a ratio of 1013. Only 1 in 108 of the organisms in Mars rocks survived radiation in space, and only 1 in 10 survive entry to the surface of Earth, for a net survival ratio of 10−9. For the Phobos sample, 1 in 104 survived impact on Phobos, and of these 1 in 104 survive radiation exposure on Phobos, for a net ratio of 10−8. So, the ratio of surviving microbes in Mars rocks to those in the Phobos sample is the mass ratio 1013 times the relative survival ratio (10−9/10−8), or 1012. Even over the last 50 years, a factor of 109 more microbes have arrived in martian meteorites (assuming life actually exists on Mars) from Zunil than will arrive in the sample returned by a spacecraft. Uncertainty in factors, such as the survival probability during Earth entry, will not change the overall conclusion that the mass of naturally delivered martian rocks (and microbes) is many orders of magnitude greater than that delivered by a sample return mission to Mars’s moons.
11 B. Gladman, 1997, Destination: Earth. Martian meteorite delivery, Icarus 130(2):228-246.
Restricted Versus Unrestricted Earth Return
In determining whether samples returned from Phobos or Deimos should be classified as restricted or unrestricted Earth return, the committee considered the following factors:
- The work represented by the SterLim and JAXA teams can be considered as the state of the art, in regard to the modeling of the process of deposition of martian material on the surface of the martian moons. Nevertheless, significant deficiencies exist in understanding, and there remain experimental and computational challenges associated with the quantitative estimation of ejecta mass and temperature distributions. Even though issues still exist with the modeling work that was performed (see Chapter 2), the work is convincing in showing that there is significant sterilization introduced during the whole chain of events.
- The issue of desiccation on the surface of the martian moons for any present martian microbes was not considered. At temperatures above the freezing point of water, desiccation is bactericidal to even the most radiation-resistant microbes in a matter of months. (See “Sterilization by Radiation on Phobos/Deimos Surfaces,” in Chapter 2.)
- The relative influx of martian microbes from the Phobos/Deimos sample versus the natural influx of direct Mars-to-Earth transfer can be shown to be smaller by several orders of magnitude.
Recommendation: After considering the body of work conducted by the SterLim and JAXA teams, the effect of desiccation on the surfaces of the martian moons, and the relative flux of meteorite- to spacecraft-mediated transfer to Earth, the committee recommends that samples returned from the martian moons be designated unrestricted Earth return.
The third task of the committee was to elaborate
In what specific ways is classification of sample return from Deimos a different case than sample return from Phobos?
The different orbits and cross-sectional areas of Phobos and Deimos result in differences in the velocities associated with impacts of martian ejecta to their surfaces, and also in the total mass of martian material expected to be delivered to each moon. Both of these factors affect the total likelihood that microbes could survive delivery to the moons from Mars, and therefore raises the important question of whether Phobos and Deimos should be treated differently with respect to planetary protection requirements.
The JAXA study concluded that although more martian material was likely to be present on Phobos than on Deimos, more total organisms could theoretically survive transfer from Mars to Deimos. This conclusion was strongly dependent on the specific ejecta geometries and velocities associated with the Zunil impact modeling. In this scenario, martian material impacted Phobos’s surface at significantly higher velocities than it impacted Deimos. When coupled with the assumptions about hypervelocity impact sterilization rates as a function of velocity, these impact velocity differences were more significant in determining the probability that microbes could survive transfer than the total mass of delivered martian material.
The committee considered these results in debating the question of whether differences in planetary protection requirements for Phobos versus Deimos are appropriate. The committee uncovered significant uncertainties associated with the impact sterilization assumptions (see “Sterilization During Hypervelocity Impact on Phobos/Deimos Surfaces,” in Chapter 2) and noted that choosing a different hypervelocity impact sterilization rate would affect the results from the JAXA work. Specifically, if assumed impact sterilization rates were lowered (especially at high velocities), the effects of differences in impact velocities between Phobos and Deimos may no longer dominate total martian material as the factor that drives total number of microbes that could have survived transfer to Phobos and Deimos.
Recommendation: Given uncertainty associated with impact sterilization assumptions, the committee recommends that Phobos and Deimos should not currently be treated differently in their planetary protection requirements.
The fourth task was to identify
What relevant information for classification of sample return is available from published studies of martian meteorites on Earth?
An overview of the literature is included in Chapter 1 (see “Earth Inventory of Martian Meteorites”). The main result from studies of martian meteorites is that coherent, solid rocks may be ejected to Mars’s escape velocity (>5.03 km/s). The degree to which the rocks are affected by the impact-ejection process ranges from weakly shocked—represented by igneous clinoproxenites, the nakhlites (5-10 GPa)—to strongly shocked basaltic rocks—the so-called shergottites (~55 GPa).
The current inventory of martian meteorites in collections around the world is biased toward sampling of igneous crust of Amazonian age (i.e., the last 3 billion years of martian history). Despite observation of sedimentary rocks on Mars, no such rocks exist in the known meteorite inventory, suggesting a bias in the delivery process toward young, coherent rocks. However, as no microbe has been so far reported from a martian meteorite, this suggests that the martian bioload is small, possibly smaller than the value assumed by the SterLim and JAXA teams.
Finding: The committee finds that the study of martian meteorites provides important context for studies of Mars and its moons and limited information (e.g., mass and flux to Earth) of relevance to planetary protection considerations. The unambiguous detection of an indigenous martian organism in a meteorite would be of great scientific and societal significance.
The committee’s fifth task was to answer the question
What are the planetary protection consequences of taking a surface sample at depths of 0-2 cm versus taking a sample extending down to depths of 2-10 cm or deeper?
The most penetrating forms of ionizing radiation suffer little attenuation over the two depth ranges, and so its sterilizing power 2 to 10 cm below the surface is essentially the same as it is in the top 2 cm. However, the committee identified two factors that could cause microbial survival probabilities to be different in these two depth ranges, as follows:
- Ultraviolet radiation, and
- Diurnal temperature cycling.
Ultraviolet radiation would decrease microbe survival rates at the surface of Phobos or Deimos, but such radiation is attenuated within the top few millimeters of surface material. There are therefore unlikely to be significant differences in returned samples due to ultraviolet irradiation from ~0.5-2 cm versus 2-10 cm.
A larger difference for samples from 0-2 cm depth versus the 2-10 cm depth will be the different temperature variations these regions experience. The temperatures on the surface of Phobos and Deimos (see Table 1.1) range,
respectively, from ~150 K to 300 K and ~160 K to 270 K according to the time of day.12,13,14,15 This temperature cycling will add an additional sterilization factor in the upper ~0.3-1.0 cm of Phobos’s regolith, but will have little effect at depths below 2 cm because Phobos has a very low thermal conductivity and associated diurnal skin depth.16 The corresponding surface thermal conductivity of Deimos is assumed to be similar to that of Phobos.
Sterilization by thermal cycling was not considered by the JAXA study, yet the results still showed that acceptably low microbial loads were expected to be present in samples. Therefore, samples from shallower depths on Phobos or Deimos have a lower risk for microbial contamination that those at a greater depth due to sterilization by thermal cycling. However, this additional factor is not needed to give confidence that samples from 2-10 cm depth will be below the established planetary protection limits for expected microbial contamination.
Recommendation: The committee recommends that no differences need to be made in planetary protection requirements for samples collected on the martian moons from depths 0-2 cm versus samples from 2-10 cm.
Last, the committee’s sixth task was to
Suggest any other refinements in planetary protection requirements that that might be required to accommodate spacecraft missions to and samples returned from Phobos or Deimos.
The committee limits its response to this task to comments on three specific topics, uncertainty quantification, implications for Mars sample return missions, and the publication of the work of the SterLim and JAXA teams.
The work of the SterLim and JAXA teams provides prime examples of attempts to reach a specific conclusion about real-world activities based on combining the results from multiple numerical simulations and laboratory experiments. Each individual calculation or experiment is subject to various degrees of uncertainty such as the following:
- Parameter uncertainty—Numbers required as input to models are unknown or poorly constrained—for example, the microbial bioload on the martian surface.
- Parametric variability—The range of variables used in modeling may not match the circumstances being modeled—for example, the range of impactor velocities and angles of incidence.
- Structural uncertainty—The physics underlying the processes being modeled is not well understood—for example, the choice of the equation of state used in an impact simulation.
- Algorithmic uncertainty—Errors or approximations made in the implementation of a particular numerical model need to be understood—for example, the validity of the SPH approach to modeling impacts.
13 R.O. Kuzmin and E.V. Zabalueva, 2003, The temperature regime of the surface layer of the Phobos regolith in the region of the potential Fobos–Grunt Space Station landing site, Solar System Research 37:480.
14 D.K. Lynch, R.W. Russell, R.J. Rudy, S. Mazuk, C.C. Venturini, H.B. Hammel, M.V. Sykes, R.C. Puetter, and R.B. Perry, 2007, Infrared spectra of Deimos (1-13 µm) and Phobos (3-13 µm), The Astronomical Journal 134:4.
15 J.L. Bandfield, S. Piqueux, T.D. Glotch, K.A. Shirley, T.C. Duxbury, J.R. Hill, C.S. Edwards, J.J. Plaut, V.E. Hamilton, and P.R. Christensen, 2018, “Mars Odyssey THEMIS Observations of Phobos: New Spectral and Thermophysical Measurements,” 49th Lunar and Planetary Science Conference, Abstract #2643, https://www.hou.usra.edu/meetings/lpsc2018/pdf/2643.pdf.
16 R.O. Kuzmin and E.V. Zabalueva, 2003, The temperature regime of the surface layer of the Phobos regolith in the region of the potential Fobos–Grunt Space Station landing site, Solar System Research 37:480.
- Experimental uncertainty—The observation error and natural scattering of results inherent in any experimental measurement may not be sufficiently well understood absent a sufficient number of repeated measurements—for example, a limited number of impact velocities and angles of incidence explored in laboratory impact studies.
- Extrapolation and interpolation uncertainty—The data available from numerical simulations or laboratory studies is insufficient to explore the full parameter range of interest—for example, the use of the results of limited impact experiments to calibrate an analytic function subsequently used to explore a parameter space beyond that explored experimentally.
The quantitative study, characterization, and reduction of uncertainties of these various types is the province of the discipline of uncertainty quantification. Practitioners of uncertainty quantification seek to determine the likelihood of specific outcomes for a system given that specific aspects of it are unknown or only weakly constrained.17,18 Examples of such uncertainty quantification approaches can be found in other fields—for example, in the Predictive Science Academic Alliance Program (PSAAP) developed by the U.S. Department of Energy,19 and this area is the subject of a number of conferences and workshops.
Recommendation: The committee recommends that a significant effort be made by the planetary protection community to formally develop an uncertainty quantification protocol that can be used to estimate the cascading uncertainties that result from the integration of multiple computational models or other factors relevant to the quantitative aspects of planetary protection. Specific attention should be given to consideration of the significant uncertainties in the model inputs that exist because of limited available experimental or observational data.
Implications for Mars Sample Return
What, if any, implications do the results of this study and the work of the JAXA and SterLim teams have for the planetary protection aspects of returning samples directly from the surface of Mars? There are several reasons why Mars sample return (MSR) missions differ from those for collecting samples from Phobos and Deimos, including the following:
- The MSR missions will be dedicated to studying the history and evolution of life in four different environments—hydrothermal, sedimentary, subaerial, and rock hosted—according to the recent International MSR Objectives and Samples Team (iMOST) study.20 The collected samples will be selected according to specific criteria designed to maximize the chance of sampling evidence of extant or extinct life. Therefore, the starting point of the quantitative evaluations undertaken by the SterLim and JAXA teams in terms of the number of colony-forming units in the martian regolith (see “Potential Microbial Density on the Martian Surface,” in Chapter 2) will very likely be invalid when considering MSR.
- The various physical processes evoked in the microbial contamination assessment (excavation by impact, collision with Phobos, sterilization, etc.) do not have to be considered for the assessment of potential microbial density for MSR, which could increase drastically the potential microbial density in comparison to the Phobos and Deimos samples.
17 See, for example, R.L. Iman, and J.C. Helton, 1988, An investigation of uncertainty and sensitivity analysis techniques for computer models, Risk Analysis 8(1):71-90, doi:10.1111/j.1539-6924.1988.tb01155.x.
18 See, for example, W.E. Walker, P. Harremoës, J. Rotmans, J.P. van der Sluijs, M.B.A. van Asselt, P. Janssen, and M.P. Krayer von Krauss, 2003, Defining uncertainty: A conceptual basis for uncertainty management in model-based decision support, Integrated Assessment 4(1):5-17, doi:10.1076/iaij.188.8.131.5266.
19 See, for example, P.-H.T. Kamga, B. Li, M. McKerns, L.H. Nguyen, M. Ortiz, H. Owhadi, and T.J. Sullivan, 2014, Optimal uncertainty quantification with model uncertainty and legacy data, Journal of the Mechanics and Physics of Solids 72:1-19.
20 Beaty et al., 2018, “iMOST: Potential Science and Engineering Value of Samples Delivered to Earth by Mars Sample Return,” white paper, posted August, by MEPAG at https://mepag.jpl.nasa.gov/reports.cfm.
- The reasoning regarding natural flux does not apply directly to samples returned from the Mars surface. The material will be gently sampled and returned directly to Earth. The sample may well come from an environment that mechanically cannot become a Mars meteorite. The microbes may not be able to survive impact ejection and transport through space. Samples with current liquid water and recent ice seem especially fragile to natural transport to Earth.
Finding: The committee finds that the content of this report and, specifically, the recommendations in it do not apply to future sample return missions from Mars itself.
Publication of the Work of the SterLim and JAXA Teams
Planetary protection policies and the studies underlying them have a reputation in some circles as being based on faulty or outmoded ideas and approaches. The immense amount of work undertaken by the SterLim and JAXA teams makes it clear that these criticisms are, at least in this case, unfounded. Another criticism of planetary protection is a lack of transparency as to how particular conclusions and policies were reached. The planetary protection, astrobiology, and planetary science communities would greatly benefit from the publication of the work undertaken by the SterLim and JAXA teams.
Recommendation: The committee recommends that the SterLim and JAXA teams formally publish the details of and results from their studies or make them readily available in some publicly accessible form.
This page intentionally left blank.