Introduction
False positives during life-detection experiments on the Martian surface will be influenced by at least the following five conditions: (1) planetary protection (PP) protocols prior to launch to reduce launched bioburdens (Frick et al., Reference Frick, Mogul, Stabekis, Conley and Ehrenfreund2014), (2) the biocidal effects of interplanetary space during transit (Moores and Schuerger, Reference Moores and Schuerger2020), (3) biocidal conditions on the Martian surface after landing (Bischof et al., Reference Bischof, Moores, Ordinaria and Schuerger2026; Rummel et al., Reference Rummel, Beaty, Jones, Bakermans, Barlow, Boston, Chevrier, Clark, de Vera, Gough, Hallsworth, Head, Hipkin, Kieft, McEwen, Mellon, Mikucki, Nicholson, Omelon, Peterson, Roden, Lollar, Tanaka, Viola and Wray2014), (4) the dispersal of viable spores from spacecraft surfaces by aeolian processes (Fenton et al., Reference Fenton, Marshall, Schuerger, Smith and Kelley2024, Reference Fenton, Schuerger, Marshall, Kelley, Smith and Smith2026) and (5) the probability of recovering viable spacecraft microbiota dispersed on to the local terrain. Although recent studies into the lethality of interplanetary space (Moores and Schuerger, Reference Moores and Schuerger2020) and the Martian surface after landing (Bischof et al., Reference Bischof, Moores, Ordinaria and Schuerger2026) have suggested extremely low survival rates for common spacecraft microbiota, it remains plausible that some viable spores or cells might be dispersed onto the terrain immediately adjacent or downwind of a landed spacecraft. Of all of the lethal factors on Mars, UVC irradiation remains the most dominant parameter (Bischof et al., Reference Bischof, Moores, Ordinaria and Schuerger2026).
Here we examine the probabilities of dispersing viable spores or cells (henceforth spores) from spacecraft surfaces on to the Martian terrain and subsequently picking up the viable spores during a sampling event later in the mission. Spore dispersal can be affected by mechanical processes during removal (e.g., crushing or killing of spores during sand-impact events; see Fenton et al., Reference Fenton, Marshall, Schuerger, Smith and Kelley2024, Reference Fenton, Schuerger, Marshall, Kelley, Smith and Smith2026), dispersal (e.g., biocidal effects of UVC irradiation; see above) and duration of spores exposed on the surface terrain after dispersal (e.g., UVC, biotoxic soils, volatile oxidants, etc.; see Bischof et al., Reference Bischof, Moores, Ordinaria and Schuerger2026; Rummel et al., Reference Rummel, Beaty, Jones, Bakermans, Barlow, Boston, Chevrier, Clark, de Vera, Gough, Hallsworth, Head, Hipkin, Kieft, McEwen, Mellon, Mikucki, Nicholson, Omelon, Peterson, Roden, Lollar, Tanaka, Viola and Wray2014). Here we assume that all spores remain viable during the time from spacecraft removal to subsequent sampling. The primary goal of the current study was to develop a 1st-order probability model for spore dispersal during the initial entry, descent and landing (EDL) profile of a Martian spacecraft under no-wind or wind-driven scenarios.
Materials and methods
Two spore dispersal models were developed. One dispersal scenario was modeled for a non-mobile lander in a no-wind situation at landing that used active descent engines for landing. Dispersal in the first scenario might be due to vibrational effects of the spacecraft, electrostatic repulsion from the spacecraft, direct effects of descent engine exhaust on exposed bioburdens, or spore removal via saltation (see Fenton et al., Reference Fenton, Marshall, Schuerger, Smith and Kelley2024, Reference Fenton, Schuerger, Marshall, Kelley, Smith and Smith2026). The second dispersal scenario was for a rover landing system that created a vertical spore plume that reached 100 m in height and was carried downwind in a unidirectional air flow with winds between 1 and 10 m−2. Overall, we did not assume any specific bioburden on the upper decks of the spacecraft (i.e., neither in diversity or abundance) but assume that at least 100 spores are dispersed per landing event. However, standard planetary protection protocols for Category IVb Mars spacecraft (Frick et al., Reference Frick, Mogul, Stabekis, Conley and Ehrenfreund2014) limit the total surface launched bioburdens to <300 spores/m2 (Frick et al., Reference Frick, Mogul, Stabekis, Conley and Ehrenfreund2014). Furthermore, we assume that regolith sampling devices will be deployed later in the missions and scoop samples from 2.01 cm2 areas of regolith or rocks (i.e., equivalent to the area sampled by the Curiosity Rover drill) in the dispersal cones. And finally, it is assumed that the dispersed spores will remain viable until the sampling events.
Spore dispersal in a stationary lander
The released spores are assumed to spread out radially from the spacecraft, coming to rest in a ring with minimum and maximum radii that are defined below. The code assumes that 100 spores (N) are distributed radially from the lander without a preference for direction or distance. The concentration of spores, and thus, the probability of recovering a single spore decreases with distance because each incremental distance increases in area of the circle the further it is from the lander. We do not assume that any individual spore is more or less likely to land closer to or further away from the source, and thus, the fall-off in concentration–and therefore the probability of contamination of a surface location–is purely geometric as per:
$P\left(r\right)=A_{\textit{SAMPLE}}{N \over 2\pi r(R_{MAX}-R_{MIN})}$
In Equation (1), P is the probability that a sampling event that excavates an area ASAMPLE obtains a single spore from N spores released. The minimum dispersal radius, RMIN, is assumed to be 1 m for simplicity, roughly consistent with many landed spacecraft on Mars. The maximum dispersal radius, RMAX, comes from simulations of saltation on Mars which suggest that typical grains take paths that extend from 20 to 120 m (Almeida et al., Reference Almeida, Parteli, Andrade and Herrmann2008). Higher probabilities arise from smaller areas; therefore, to be conservative in our calculations and to simplify this order-of-magnitude discussion, we selected 10 m and 100 m for our maximum radii in the two models. Finally, we assume 2.01 cm2 as ASAMPLE, equivalent to the area excavated by the 1.6 cm diameter Curiosity drill (Abbey et al., Reference Abbey, Anderson, Beegle, Hurowitz, Williford, Peters, Morookian, Collins, Feldman, Kinnett, Jandura, Limonadi, Logan, McClosky, Melko, Okon, Robinson, Roumeliotis, Seybold, Singer and Warner2019; Okon, Reference Okon2010).
Spore dispersal in landing scenarios with rovers
In the second scenario, the act of landing is assumed to create a plume of material which includes spores mixed with larger sand/dust particles from Martian regolith. Moores et al, (Reference Moores, Schieber, Kling, Haberle, Moore, Anderson, Katz, Yavrouian, Malin, Olson, Rafkin, Lemmon, Sullivan, Comeaux and Vasavada2016) examined the landing plume of the Curiosity Rover and suggested that the particles could range up to an order of ≤500 µm in diameter and could be lofted up to 100 m above the lander before being swept downwind. Using the dispersal model of Moores et al. (Reference Moores, Schieber, Kling, Haberle, Moore, Anderson, Katz, Yavrouian, Malin, Olson, Rafkin, Lemmon, Sullivan, Comeaux and Vasavada2016) – and the same number of released particles and sampling area as in section, Spore dispersal in a stationary lander – it becomes possible to determine the probability of detecting a single spore within the fall-out cone of the plume if the windspeed is known. Previous studies on Mars (e.g., Holstein-Rathlou et al., Reference Holstein-Rathlou, Gunnlaugsson, Merrison, Bean, Cantor, Davis, Davy, Drake, Ellehoj, Goetz, Hviid, Lange, Larsen, Lemmon, Madsen, Malin, Moores, Nornberg, Smith, Tamppari and Taylor2010; Newmann et al., Reference Newman, Gomez-Elvira, Marin, Navarro, Torres, Richadson, Battalio, Guzewich, Sullivan, de la Torre, Vasavada and Bridges2017) indicate that the most probable average wind speed is 4 m s−1, however, winds can also be as high as 16 m s−1, the highest speed reported at the Phoenix landing site (Holstein-Rathlou et al., Reference Holstein-Rathlou, Gunnlaugsson, Merrison, Bean, Cantor, Davis, Davy, Drake, Ellehoj, Goetz, Hviid, Lange, Larsen, Lemmon, Madsen, Malin, Moores, Nornberg, Smith, Tamppari and Taylor2010), or greater. Based on winds at the Curiosity landing site, a range of 1 m s−1 to 10 m s−1 was selected to capture all winds that have an occurrence frequency greater than 10% compared to a peak wind flux (Newmann et al., Reference Newman, Gomez-Elvira, Marin, Navarro, Torres, Richadson, Battalio, Guzewich, Sullivan, de la Torre, Vasavada and Bridges2017).
MATLAB coding (version R2024, MathWorks, Natick, MA, USA) was used to generate data given in Figures 1 and 2. The data in both figures are based on the assumptions given above.
Negative [−] log10 probabilities of recovering one spore in a 2.01 cm2 sample area for a dispersal pattern with zero wind in a 10 m radius (A) and 100 m radius (B) circles when 100 spores are dislodged by all mechanisms at landing in one event (see text).
Negative [−] log10 probabilities of recovering one spore in a 2.01 cm2 sample for a dispersal pattern of a landing plume that reaches 100 m in height and is then dispersed downwind in a 1 m/s wind (A) or a 10 m/s wind (B and C). The red box in Figure 2C indicates the region depicted in Figure 2B.
Results
Three key assumptions for interpreting the data in Figures 1 and 2 are (i) that 100 viable spores are dispersed from spacecraft surfaces and are uniformly distributed in the sample areas being described, (ii) the spores remain viable during transit and while waiting for the sampling event and (iii) the sample area is constrained to the area predicted for the Curiosity rover drill bit (i.e., 2.01 cm2). Thus, the following probabilities (P) represent the chances that a single viable spore is recovered in an individual sample collected within the described sampling cones.
In Figure 1A, with a 10 m radius sample area, the maximum P of approx. 3.5 × 10−4 or 0.035% is adjacent to the spacecraft and decreases to ≤ 3.5 × 10−5 or 0.0035% at 10 m. For a 100-m radius sampling circle (Figure 1B), the P of sampling one viable spore is approx. 3.0 ×10−5 or 0.0030% right next to the spacecraft and decreases to P ≤ 3.2 × 10−7 or 0.000032% at the edge of the sampling area. The probabilities in Figure 1 fall-off quickly as sampling moves away from the landing site.
In the case of wind-blown plumes (Figure 2), probabilities are essentially zero when sampling orthogonal to the wind vector. In addition, the P levels for detection along the wind vector diminish quickly as sampling becomes more distant from the landing site. In a 1 m/s wind (Figure 2A), P ≤ 7.2 × 10−6 at 10 m, ≤ 3.2 × 10−7 at 100 m, ≤ 1.0 × 10−7 at 200 m and ≤ 3.0 × 10−10 at 900 m. If the wind is stronger, for example at 10 m/s (Figure 2B), the region of high probability is constrained (i.e., narrower dispersal cone) but longer as a greater proportion of particles are swept further downwind before settling out of the atmosphere. Thus, the P in a 10 m/s wind at 100 m downwind is ≤ 1.1 × 10−6, ≤ 3.5 × 10−7 at 200 m, ≤ 8.1 × 10−8 at 500 m, ≤ 2.8 × 10−8 at 1000 m.
The downwind cones in a 1.0 m/s wind reaches only 900 m long by 200 m wide before all spores are deposited onto the terrain. In contrast, in a 10 m/s wind the dispersal cone might reach as much as 7000 m downwind with the same width (Figure 2C). Furthermore, the primary settling out of spores in a 10 m/s wind does not occur immediately adjacent to the landed spacecraft (Figure 2B, slight offset of the start of the plume) but begins at approx. 20 m downwind as the spores begin at the level of the deck and are swept downwind before they can settle.
If the sample areas are larger than the 2.01 cm2 used here (i.e., the area of the Curiosity drill bit), the probabilities will increase as the sample areas increase according to the following equation.
$P-new\ sample\ area = [P-predicted\ here] * \rm [area\ of\ new\ sampler / 2.01\ cm^2]$
For example, the Curiosity sample scoop on the Sample Acquisition/Sample Processing and Handling (SA/SPaH) system (Jandura, Reference Jandura2010) has a cross-sectional area of 31.5 cm2 (4.5 cm × 7 cm; Jet Propulsion Lab, 2012) which would increase P for recovering one viable spore by 15.7 times higher for any scenario chosen in Figures 1 and 2. However, the increase in P for the SA/SPAH scoop assumes that 100% of the regolith is delivered to a life-detection experiment. If the amount delivered to a life-detection experiment is actually much less, the P value would decrease proportionally. Caution should be used in applying the dispersal probabilities presented here until a more thorough effort can be developed with all parameters that will affect the survival of spores on future landers or rovers carrying a life-detection payload (e.g., sample scoop areal cross section, volume delivered to a life-detection payload, effects of saltation on spore survival (Fenton et al., Reference Fenton, Schuerger, Marshall, Kelley, Smith and Smith2026) and the lethality of the Martian environment on dispersed spores (Bischof et al., Reference Bischof, Moores, Ordinaria and Schuerger2026).
Discussion
Two spore dispersal scenarios for landed Mars spacecraft were considered. First, stationary landers that have 2-m robotic arms for collecting regolith samples were considered for no-wind scenarios. Second, mobile rovers with 2-m robotic arms for sampling were considered for two wind scenarios of 1 and 10 m/s. The winds were considered to be unidirectional and not fluctuating. We consider here the probabilities of recovering one viable spore–out of 100 dispersed spores–in sampling circles (no wind) or cones (with wind) from plumes of spores dislodged during EDL events. If new modeling indicates that dispersed spore loads are either lower or higher than the 100 modeled here, the P values presented herein would change linearly based on the new assumptions of how many spores are dislodged at landing. We only briefly consider below chronic spore dispersal scenarios throughout the mission.
The greatest risks for collecting regolith contaminated with viable spores from Earth are during stationary and mobile spacecraft missions immediately after vehicles have landed and within 10 m of the vehicle. If 2-meter robotic arms are quickly deployed within the first few sols, the risks may be as high as 1 part in a thousand (P ≤ 1 × 10−3) if all spores are dispersed no further than inside a circle with a radius of 10 m (Figure 1A). The low P ≤ 1 × 10−3 given above for a stationary lander or rover is close to the bioburden transport model of P ≤ 4.2 × 10−3 for contamination around the Mars 2020 rover (Chen et al., Reference Chen, Ly, Mikellides, Bernard and Cooper2023).
This first scenario is applicable for stationary landers, or rovers not yet on traverses. Furthermore, even for rovers that are engaged in traverses, spacecraft vibrations, electrostatic charging of surfaces, sand saltation and dust-devil encounters–for example–may lift spores and disperse them to the immediate vicinity of the moving spacecraft. Current sampling arms do not extend beyond approx. 2 m from the landers or rovers, and thus, the 10-m outer boundary in Figure 1A indicates an upper bound on the risks of sampling at least one viable spore immediately around a lander or rover on Mars.
In all rover scenarios, the greatest risks of sampling viable spores are if the traverses start and then remain within the spore dispersal circles (Figure 1) or cones (Figures 2A and 2B). In a no-wind scenario, the P for sampling one viable spore quickly approaches ≤1 × 10−6 after only a few 10’s of meters distal from the landing site (Figure 1B). As wind increases, the dispersal cones will elongate and narrow. In contrast, any sampling that is orthogonal to a downwind dispersal cone will have essentially a P = 0 for collecting one viable dispersed spore, except in cases in which new spores are dislodged during the actual sampling event. And finally, even if rover traverses remain entirely within the spore dispersal cones (e.g., Figure 2), the P values rapidly decrease and approach ≤1 × 10−7 after a few hundred meters.
The spore dispersal model assumptions discussed here are very liberal (i.e., benign). For example, we assume that all 100 viable spores retain their viabilities through EDL, dispersal, residence times in the local terrain and then the collection of regolith samples. If the spore dispersal models here are combined with a recently published Mars Microbial Survival (MMS) model (Bischof et al., Reference Bischof, Moores, Ordinaria and Schuerger2026), more realistic scenarios can be developed that suggest that orders of magnitude lower survival rates are likely than the data presented in Figures 1 and 2. For example, the MMS model predicts that greater than 12 orders of magnitude of biocidal activity is likely on the upper decks of landed spacecraft–or on unshielded areas of the regolith surface–within 1 sol on Mars under clear sky conditions. The biocidal factors combined in the MMS model include the effects of UVC irradiation (190 to 280 nm), temperature, low-pressure and the biocidal nature of Mars regolith. However, several studies on the biocidal nature of the Martian environment suggest as many as 17 to 20 biocidal conditions persist on Mars (e.g., Rummel et al., Reference Rummel, Beaty, Jones, Bakermans, Barlow, Boston, Chevrier, Clark, de Vera, Gough, Hallsworth, Head, Hipkin, Kieft, McEwen, Mellon, Mikucki, Nicholson, Omelon, Peterson, Roden, Lollar, Tanaka, Viola and Wray2014; Schuerger et al., Reference Schuerger, Ulrich, Berry and Nicholson2013; Stoker et al., Reference Stoker, Zent, Catling, Douglas, Marshall, Archer, Clark, Kounaves, Lemmon, Quinn, Renno, Smith and Young2010). If additional biocidal factors are added to the MMS model, then the P values for recovering one viable spore in the scenarios discussed herein would be even lower.
One scenario that warrants more study is the effect of sand saltation on scouring the lower 2–3 dm of spacecraft landing struts, wheels and other hardware on the undersides of landed vehicles. For example, several studies (Fenton et al., Reference Fenton, Marshall, Schuerger, Smith and Kelley2024, Reference Fenton, Schuerger, Marshall, Kelley, Smith and Smith2026; Marshall and Mancinelli, Reference Marshall and Mancinelli2011) have demonstrated that saltating sand grains on Mars can effectively sand blast the lower structures of landers resulting in the removal–and inactivation–of spores from spacecraft surfaces. Recently, Fenton et al. (Reference Fenton, Schuerger, Marshall, Kelley, Smith and Smith2026) demonstrated that spores of Bacillus subtilis on aluminum coupons can be crushed (i.e., and thus presumably killed) by saltating sand particles under conditions applicable to the surface of Mars. Saltation activities on Mars are examples of chronic long-term spore removal and inactivation processes that might occur throughout a mission. The results for Fenton et al. (Reference Fenton, Marshall, Schuerger, Smith and Kelley2024, Reference Fenton, Schuerger, Marshall, Kelley, Smith and Smith2026) and Marshal and Mancinelli (Reference Marshall and Mancinelli2011) are consistent with other studies on microbial survival of spores encumbered within saltating sand on Mars (Bak et al., Reference Bak, Larsen, Moeller, Nissen, Jensen, Nornberg, Jensen and Finster2017, Reference Bak, Larsen, Jensen, Nornberg, Moeller and Finster2019). Viable spores that survive saltation would distribute additional spores into the local terrains immediately around landers and would fall within the predictions of Figure 1A; that is, the contamination is likely to settle out quickly after removal because the removal process is close to the surface. However, more work is required to develop quantitative models on the precise levels of biocidal activity from saltating particles on spacecraft bioburdens on Mars.
In summary, if we assume that 100 spores are dislodged and dispersed away from landers during EDL activities, that all spores remain viable during the dispersal and sampling processes, then the P predictions for recovering one viable spore are highest for stationary landers or rovers within a few meters of the lander. When downwind dispersal is considered, the highest P values predicted are for terrains adjacent to the spacecraft but decrease rapidly downwind. The P values of recovering one viable spore are lowest in the high-wind scenario beyond a few hundreds to a few thousands of meters in the dispersal cones and approach ≤10−7 to 10−8 beyond 500 m downwind of the landing site. Beyond 5000 m, the P values for recovering one viable spore in a 10 m/s wind approach ≤10−9. Results suggest that the risks of false positives in life-detection experiments are very low–even under optimistic spore dispersal and survival scenarios and if larger sampling scoop sizes are considered–and may approach zero if biocidal factors on the Martian surface are combined with the dispersal P values described herein.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1473550426100433.
Data availability statement
All experimental data are provided here and in the Supplemental sections.
Author contributions
The manuscript was conceived, written and edited by ACS and JEM. In addition, JEM developed the numerical models described in Equations (1 & 2).
Funding statement
ACS was funded by a NASA Planetary Protection grant (#80NSSC22K0212), and JEM was funded by a Natural Sciences and Engineering Research Council of Canada grant (#RGPIN-2025-06299).
Competing interests
The authors declare no conflicts of interest in the current paper.
Open access
