In Situ Real-Time Monitoring for Aseptic Drilling: Lessons Learned from the Atacama Rover Astrobiology Drilling Studies Contamination Control Strategy and Implementation and Application to the Icebreaker Mars Life Detection Mission

Open camera or QR reader and scan code to access this article and other resources online. In Situ Real-Time Monitoring for Aseptic Drilling: Lessons Learned from the Atacama Rover Astrobiology Drilling Studies Contamination Control Strategy and Implementation and Application to the Icebreaker Mars Life Detection Mission Rosalba Bonaccorsi, 1,2 Brian Glass, 2 Mercedes Moreno-Paz, 3 Miriam Garcı´a-Villadangos, 3 Kimberley Warren-Rhodes, 1,2 Victor Parro, 3 Juan Manuel Manchado, 3 Mary Beth Wilhelm, 2 and Christopher P. McKay 2 Abstract In 2019, the Atacama Rover Astrobiology Drilling Studies (ARADS) project field-tested an autonomous rovermounted robotic drill prototype for a 6-Sol life detection mission to Mars (Icebreaker). ARADS drilled Marslike materials in the Atacama Desert (Chile), one of the most life-diminished regions on Earth, where mitigating contamination transfer into life-detection instruments becomes critical. Our Contamination Control Strategy and Implementation (CCSI) for the Sample Handling and Transfer System (SHTS) hardware (drill, scoop and funnels) included out-of-simulation protocol testing (out-of-sim) for hardware decontamination and verification during the 6-Sol simulation (in-sim). The most effective five-step decontamination combined safer-to-use sterilants (3%_hydrogen-peroxide-activated 5%_sodium-hypochlorite), and in situ real-time verification by adenosine triphosphate (ATP) and Signs of Life Detector (SOLID) Fluorescence Immunoassay characterization of hardware bioburden and airborne contaminants. The 20to 40-min protocol enabled a 4-log bioburden reduction down to <0.1 fmoles ATP detection limit (funnels and drill) to 0.2–0.7 fmoles (scoop) of total ATP. The (post-cleaning) hardware background was 0.3 to 1–2 attomoles ATP/cm 2 (cleanliness benchmark background values) equivalent to ca. 1–10 colony forming unit (CFU)/cm 2 . Further, 60–100% of the in-sim hardware background was £3–4 bacterial cells/cm 2 , the threshold limit for Class <7 aseptic operations. Across the six Sols, the flux of airborne contaminants to the drill sites was *5and*22 amoles ATP/(cm 2 $day), accounting for an unexpectedly high Fluorescence Intensity (FI) signal (FI: *6000) against aquatic cyanobacteria, but negligible anthropogenic contribution. The SOLID immunoassay also detected microorganisms from multiple habitats across the Atacama Desert (anoxic, alkaline/acidic microenvironments in halite fields, playas, and alluvial fans) in both airborne and post-cleaning hardware background. Finally, the hardware ATP background was 40–250 times lower than the ATP in cores. Similarly, the FI peaks (FI max ) against the microbial taxa and molecular biomarkers detected in the post-cleaned hardware (FI: *1500–1600) were 5–10 times lower than biomarkers in drilled sediments, excluding significant interference with putative biomarker found in cores. Similar protocols enable the acquisition of contamination-free materials for ultra-sensitive instruments analysis and the integrity of scientific results. Their application can augment our scientific knowledge of the distribution of cryptic life on Mars-like grounds and 1 SETI Institute, Mountain View, California, USA. 2 NASA Ames Research Center, Moffett Field, California, USA. 3 Department of Molecular Evolution, Centro de Astrobiologı ´a (INTA-CSIC), Madrid, Spain. Rosalba Bonaccorsi et al., 2023; Published by Mary Ann Liebert, Inc. This Open Access article is distributed under the terms of the Creative Commons License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. ASTROBIOLOGY Volume 23, Number 12, 2023 Mary Ann Liebert, Inc. DOI: 10.1089/ast.2022.0133 1303 support life-detection robotic and human-operated missions to Mars. Key Words: Atacama—Microbial reduction— Real-time ATP assay—Robotic drilling—Life detection—Mars’Icebreaker Sample Handling and Transfer System (SHTS)—Atacama Rover Astrobiology Drilling Studies (ARADS). Astrobiology 23, 1303–1336. 1. Introduction Unlike Earth, the martian surface appears lifeless, with the subsurface proposed as the best target for detecting life. The Atacama Rover Astrobiology Drilling Studies (ARADS) system is a prototype for a future life detection drill mission to Mars, namely the currently planned Icebreaker Mission (Dave ´et al., 2013; McKay et al., 2013; Glass et al., 2014; Mann, 2018). ARADS performed field trials in the hyper-arid region of the Atacama Desert in the Antofagasta Region from 2016 to 2019 (Fig. 1), culminating in a 6-Sol end-to-end mission simulation. Astrobiology robotic drilling missions on Mars will face broad issues both in forward planetary protection (PP) from terrestrial microbes surviving sterilization of the spacecraft’s hardware, which is the primary concern (e.g., NAS, 2020; NAS, 2019; NAS, 2018; NRC, 2006) and in the integrity of sample analysis. The Committee on Space Research (COSPAR) PP policy for landed missions in a Martian Special Region (Category IVc PP, probability contamination: <10 -4 ) is the most strict requiring hardware bioburden contamination control of forward contamination (COSPAR, 2020). Spores can remain long-term viable in space (e.g., Horneck et al., 2010, 2012; Hallsworth, 2021), be shed on the surface (Marshall and Mancinelli, 2011), and enter the underground, becoming metabolically active under increased water activity (Rummel et al., 2014; Smith et al., 2017). A secondary concern is mitigating the risk of terrestrial contaminants transfer (spores, amino acids) into sample analysis generating false positives—life appears to be detected when there is not—(Kminek et al., 2019), or protecting science data integrity. Human-operated contamination remediation is a primary step for robotic drilling mission planning, beginning with clean room operations during Spacecraft Assembly, Test, and Launch Operations (ATLO). The human involvement continues during robotic drilling mission mock-ups, as the hands-off mission technology readiness is not yet mature, requiring field engineers to supervise FIG. 1. (A) ARADS 2019 Study Sites. (B) Playa test site (Lat. 2406¢08.3†S to Long. 7008¢17.5†W) about 22 km from the field support headquarter, the Yungay Station. (C) Site location of ‘‘Green Parrot’’ (GP) Lat. 245¢37.62†S; Long. 6954¢20.86†W). ARADS, Atacama Rover Astrobiology Drilling Studies; GP, Green Parrot. 1304 BONACCORSI ET AL. decontamination (Glass et al., 2023, Astrobiology this issue). Thus, developing and field-testing human-tended cleaning regimes of payload prototypes addressing PP knowledge gaps is mandatory even for a robotic mission (e.g., NASA, 2011) and can be advanced by field trials such as ARADS. We developed our Contamination Control Strategy and Implementation (CCSI) over the 4-year project and integrated lessons learned from several astrobiology and technology fields drilling simulations, for example, the 2005 MARTE Project at the Rio Tinto Pen ˜a de Hierro, Spain (Glass et al., 2006; Bonaccorsi and Stoker, 2008; Miller et al., 2008; Parro et al., 2008; Stoker et al., 2008; Eigenbrode et al., 2009; Amador et al., 2014; Cockell et al., 2018). These tests provided opportunities to assess and mitigate longand short-term bioburden built-up at *6and 12 months and evaluate cleaning protocols for re-contaminated deep-cleaned hardware hardware after 1 day of use. This paper describes field-feasible and effective practices for contamination risk reduction during the ARADS mission and consists of seven sections. Section 1 introduces ARADS sample drilling handling and transfer hardware. Section 2 informs spacecraft hardware contamination control. Section 3 outlines anthropogenic, environmental, and cross-contamination. Section 4 describes (1) the general contamination remediation (during off-hours rover, instrument maintenance, etc.); (2) multi-step cleaning, microbial reduction, and verification of the drill, scoop, and funnels; (3) multi-step protocols implementation for in-sim operations; and (4) source-to-sink contamination of airborne, sedimentary, and anthropogenic biomass to the drill sites. Section 5 reports the 2019 ARADS results. Section 6 discusses false negatives in cleaning verification and environmental and cross-contamination issues unique to terrestrial trials or a life-detection drilling mission to Mars. Section 7 highlights innovative elements stemming from ARADS: (1) an effective simplified decontamination protocol, (2) a source-to-sink contamination approach, (3) a reassessment of contamination risks in the Atacama Desert, (4) the novel use of the immunoassay-based life detection tool (Signs of Life Detector [SOLID]) for identifying forward-contaminants on Mars, and (5) hardware design of the arm-mounted scoop analog of the Phoenix soil acquisition device to mitigate cross-contamination. 1.1. Introduction to ARADS The ARADS project was part of NASA’s Planetary Science and Technology Through Analog Research (PSTAR) Program. This project explores the science, instrumentation, and automation in performing a near surface (to 1 m depth) drill operation on Mars, using the hyper-arid Atacama (Cabrol et al., 2007; McKay et al., 2003; Navarro-Gonzalez et al., 2003; Cabrol et al., 2007) as an analog drill site (e.g., Glass et al., 2019, 2022; Moreno-Paz et al., 2019; Stoker et al., 2023). The final ARADS field campaign concluded in September 2019 with a 6-day simulated mission at a novel ‘‘landing site’’ the ‘‘Playa’’ (Fig. 1), selected for multiple ground-truth field studies (Glass et al., 2023, this issue). The ARADS CCSI constituted an integral part of the rover field deployment in both in-sim and out-of-sim operations. There are four major subsystems to the ARADS payload: (1) The NASA Ames’ self-navigating midsize rover (KREX-2) as the mechanical integration platform for the 1 m TRIDENT (The Regolith and Ice Drill for Exploration of New Terrains) drill and the other subsystems hosted by the Rover payload deck (Fig. 2A). (2) The Honeybee Robotics’ rotary-percussive Drill, whose string spiral coils can retrieve ca. 10–60 g of geological cuttings from every 25 cm of drilled subsurface (Fig. 4C). (3) The Sample Handling and Transfer System (SHTS) includes an arm-mounted scoop analog to the Phoenix Icy Soil Acquisition Device (ISAD) shown in Fig. 2D (Bonitz et al., 2008; Glass et al., 2023) and four funnels on the top of the rover payload deck, protecting the science payload underneath (Fig. 2A): three *12 cm diameter instrument-feeding funnels, and one 13 cm diameter one (aka ‘‘Big_Fun’’) to deliver overflow samples into a storage unit for post-field analysis (Fig. 2C). (4) The science instruments included (i) The Signs of Life Detector (SOLID) or prototype, built by the Spanish Centro de Astrobiologia (CAB) (Parro et al., 2008) using fluoresce microarray immunoassay detection of 450 biomolecular compounds (amino acids, nucleic acids, proteins, and polysaccharides) at 1–2 parts per billion [ppb (ng/mL)] and whole cells, that is, 10 4 –10 5 cells/mL (Rivas et al., 2008); (ii) the JPL’s Chemical Laptopformerly Microfluidic Life Analyzer (MILA) (Willis et al., 2015; Mora et al., 2020); (iii) a brassboard version of the Phoenix’s Wet Chemistry Laboratory (WCL) for salt chemistry analysis; and (iv) the NASA Goddard Space Flight Center’s Linear Ion Trap Mass Spectrometer, or LITMS for organic compounds detection and identification, which was field tested March 2019 (Castillo et al., 2023). Next, we describe the sample handling and transfer to these instruments. 1.2. Life journey of a drilled sample The ARADS’s fully automated rover/drilling/sampling/ analysis system aimed at driving to locations of scientific interest and acquiring geological material for geochemical and biological study. The sample capture occurs in three steps. The stainless-steel drill string, nested inside the system’s primary drill, accesses the depth of science interest and retrieves ca. 10–60 g of powdered cuttings. Before the sediment is brought to the surface, the arm-mounted scoop lowers to the ground to receive the drilled material (Fig. 2B). Once sufficient powdered sediment from the target depth enters the scoop, this rises above the rover deck (Fig. 2A) to deliver material into the instruments’ funnels (Fig. 2C). Each funnel drops the cuttings into 1 cm diameter holes aligned with the entry port of the analytical tools beneath the rover payload deck (not visible). Finally, the powder left in the scoop is dropped into the fourth, larger funnel (aka ‘‘Big_Fun’’) and transferred into a pre-cleaned borosilicate glass jar as in previous ARADS campaigns (Wilhelm et al., 2018). The funnel-jar unit latches onto the rover payload deck top (Fig. 2A). The rover instrument suite immediately analyzes the dropped samples. Alternatively, the instrument is detached, brought into the laboratory trailer, and analyzed immediately by the science team. Only one instrument team scientist can work at the dedicated sanitized lab bench to minimize contamination during sample analysis. ARADS CONTAMINATION CONTROL STRATEGY IMPLEMENTATION 1305 Finally, samples in jars are stored in a clean box for later sub-sampling and analysis of the archived cored materials on request by the science team. For this purpose, we operated on the clean bench top using sterilized stainless-steel spatulas in the dedicated science trailer (Fig. 9A). We excluded a few low-priority drilled intervals from the simulation to maintain mission fidelity. However, unlike on Mars, we did not discard these low-priority samples from the drill but manually collected them in labeled sterile Whirl-Pakbags for later use by the science team. 1.3. Relevance of contamination control Cleaning, microbial reduction, and verification were fundamental to ARADS. Unlike the case for human and robotic exploration of Mars Special Regions, the forward contamination of the terrestrial poly-extreme Atacama Desert poses no real concern. However, because the hyper-arid surface of this region is predominantly devoid of microbes, instruments to detect molecular indicators of putative life must operate under low detection thresholds (high sensitivity). Even minor bioburden contamination during sample acquisition and transfer into analysis could contaminate scientific results and confuse interpretations. Bioburden is the number of microorganisms on the non-sterile surface material. Sterilization refers to chemical or physical procedures that kill, deactivate, or remove all life-forms (i.e., MHRA, 2006). A terminally sterilized surface device is ‘‘sterile’’ if the theoretical probability of a viable microorganism being present is equal to or less than 1 ·10 -6 (Lambert et al., 2021). NASA protocols involve dry heat for microbial reduction (DHMR) and chemical methods when DHMR is not attainable. However, long-term absolute sterility is unachievable or is compromised after a few-minute exposure to a contaminating environment. Further, hardy microorganisms can resist chemical treatment in clean rooms (Horneck et al., 2012; Moissl-Eichinger et al., 2015; Spry et al., 2021) and even use organic compounds in cleaning reagents as an energy source (e.g., Mogul et al., 2018). Field trials in extreme environments lack cleanroom facilities, and continuous human-tended operations enhance the risk exposure of bioburden contamination to the hardware payload and sample analysis pathways. FIG. 2. SHTS. (A) Payload deck’s top hosting the three sample delivery funnels and the large funnel (background) for the outflow sample transfer to the glass jar (not in place); (B) lowered scoop receiving drilled sample; (C) scoop above payload deck delivering cuttings to the funnel feeding the SOLID instrument below (not visible). The glass jar is visible in the background. (D) Phoenix ISAD design similar to ARADS scoop (Photo credit: NASA). ISAD, Icy Soil Acquisition Device; SHTS, Sample Handling and Transfer System; SOLID, Signs of Life Detector. 1306 BONACCORSI ET AL. The risk is enhanced under a time-constrained schedule for decontamination. Each Sol, the ARADS simulation initially allocated only a 20-min operational window for the microbial reduction and verification cycle, making critical the devising of a field-feasible and time-effective decontamination protocol (See details in Section 4). 2. Contamination Control Practices for Spacecraft Hardware During the ARADS campaigns, we followed state-of-theart contamination control procedures typical of those to prepare for Mars landed missions. The modifications fit the requirements posed by the terrestrial field trial. Carosso (2005) offered comprehensive background information on spacecraft decontamination and verification techniques shared with recent missions targeting planetary samples. Examples include the 2007 Phoenix Mission and other missions to Mars such as the 2011 Mars Science Laboratory (MSL) (Mahaffy et al., 2004) and the 2020 missions centered on the analysis of molecular organic compounds, and the 2016 OSIRIS Rex to the carbonaceous near-Earth asteroid Bennu (Lauretta et al., 2017; Dworkin et al., 2018). Although these missions were expected to advance decontamination procedures for sampling and sample handling mechanisms, they have yet to progress beyond the 2007 Phoenix Lander experience due to cost caps and slow advances in material technology (Miller et al., 2008). The Phoenix Mission, operated on Mars from 2008 to 2009, offers state-of-the-art contamination control for spacecraft hardware analog to ARADS. The Lander’s robotic arm-mounted digging scoop or ISAD collected and delivered ice-bearing regolith samples to the Thermal Evolved Gas Analyzer (TEGA) to determine indicators of habitable conditions in Mars ice-cemented ground, requiring strict contamination control (10 ppb) on the digging scoop to protect the scientific results from Mars. Once assembled, the compartmentalized scoop surfaces were inaccessible for cleaning, challenging crosscontamination control between samples scooped from the martian soil. Unlike the 1975–1976 NASA’s Viking landers, which were heat treated in their entirety (e.g., Soffen and Snyder, 1976), the Phoenix Mission sampling chain (blades and scoop) maintained only low organic contamination Level 75 A/10 requirement, that is, 100 ng/cm 2 (Borson, 2005). Control and assessment of contamination on the Phoenix TEGA included vacuum bake-out/dry-heat sterilization, containment for transport and testing, and contamination characterization on an engineering mock-up flight model. During Assembly, Test, and Launch Operations (ATLO) flight hardware in an ISO 14644-1 Class 8 controlled environment, bioburden requirement is £10 CFU/m 3 (air samples), 2000 CFU/m 2 (surfaces), and £1 CFU/ 5-finger glove print (ECSS, 2008a, b). During aseptic operations (Class <7), hardware assembly occurs on laminar flow under a stricter NASA Standard assay (NSA)-based bioburden level of <1 CFU/m 3 (air), £300–400 CFU/m 2 (surfaces), and <1 CFU (per gloveprint). Assuming that the NSA can detect only 1% of the total viable and non-viable burden, 400 CFU/m 2 corresponds to the threshold for aseptic operations of £40,000 microbial cells/m 2 (or £3–4 cells/cm 2 ) achievable by ARADS. 3. Sources of Contamination Whether or not a foreign material is a ‘‘contaminant’’ depends on the objective of the planned analysis and only if the material interferes with the instruments’ measurement. Foreign cellular materials and organic biomarkers detected in ARADS cores and hardware are contaminants. ARADS contamination control (Section 4) targeted anthropogenic, environmental, and cross-contaminations while addressing their risk level before and after mitigation. (1) Anthropogenic contamination by human-tended automated drilling tests is unavoidable and a significant concern for biologically depleted terrestrial and martian environments. In Atacama’s biologically lean surface environment, airborne microorganisms are lesser concerning due to airflow dispersion. Conversely, anthropogenic microbiome contamination of soil and air via handling, clothing, and skin particles during daily operations can compromise results from the life-detection suite. (2) Environmental contamination is a unique problem in terrestrial trials, with three potential source entries into the sample path: (i) non-putative biological (viable spores, microbes, cellular remains); (ii) fossil molecular organics detectable by LITMS and SOLID; and (iii) airborne mineral dust and salts detectable by the wet chemistry instrument. There is a similar opportunity for airborne inorganic contamination on the martian surface, which is an unlikely source of biological contaminants. Airborne dust from the vast oligotrophic desert pavement to drill sites was initially not of great concern with bio barriers to protect sample-handling elements when not in use. Still, airborne biology resulted in an unexpected potential contaminant to ARADS (Section 6.2). (3) Cross-contamination could occur between different sedimentary strata drilled as powdered samples and mixed from the target depth; once in the scoop, a new sample could mix with sediment residue adhering to the inside from previously drilled intervals. 4. Methods of Contamination Control We describe by-design, operational, and contamination control techniques we tested in ARADS. 4.1. Contamination remediation by design and operational methods To remediate cross-contamination, we used stainless-steel or aluminum sampling hardware, avoided soft materials and polymers poorly conducive to field decontamination, and chose single-use pre-cleaned stainless steel funnels and clean-certified borosilicate glass jars. The sample acquisition and handling system design also enabled contaminant dilution by accepting a relatively large sample size, up to 60 g, compared with the expected residual crosscontaminants (Fig. 12). The general contamination remediation operations included protective shields usage and drill operators training guidelines that involved: (1) mindful walking in ARADS CONTAMINATION CONTROL STRATEGY IMPLEMENTATION 1307 approaching the rover downwind to avoid stirring dust from the playa pavement (Fig. 3J); (2) only two operators tending the drill in-sim (Fig. 3H); (3) use of facemasks and sanitized gloves (ethanol [Et-OH]-wiped) when approaching hardware at close contact with samples or working on the instruments outlets; (4) hand-sanitization before wearing gloves; (5) touch not critical equipment surfaces and tools or contact with sampling surfaces with un-sanitized or contaminated gloves; (6) avoid contact between the clean gloves and body’s surface, or any compromised surface; (7) use of clean/sterilized tools whenever possible; and (8) wear a protection suit when appropriate (Fig. 3F). We used protective media for proactive contamination control before, during, and between operations. For instance, before assembly on the rover deck, we transported double-wrapped cleaned funnels in sanitized plastic containers (Sterilite TM) wiped down with high-purity isopropyl alcohol (IPA). While the rover moved toward the drilling site, we protected the cleaned drill, scoop, and funnels with heavy-duty Ultra High Vacuum (UHV) Aluminum Foil (AllFoil, Inc.), which was removed only before use. Between each drilling cycle, we kept the bio-barrier in place until the drill was ready to operate (Fig. 4C), the scoop was about to accept drilled samples, and the funnels could receive and transfer samples for analysis. We maintained a contamination-free sample path to each life detection instrument. For instance, the SOLID’s inlet valve opened only before the sample’s release into the analytical tool (Parro et al., 2008). At the end of each Sol, the rover-mounted drill string and scoop were rinsed off with distilled water (DW), wrapped in Al foil, and left onsite overnight. The funnels were unclamped from the rover deck, wrapped in Al foil, and transported to the Yungay Station (Lat. 244¢53.05’’S, Long. 6955¢9.51’’W) for a new cleaning cycle before the successive Sol’s operations. Each day, the rover-mounted drill string was ‘‘touched up’’ and cleaned (done on previously deep-cleaned surfaces) before use. 4.2. Chemical cleaning and disinfection We tested the deep-cleaning effectiveness on two Icebreaker drill segments: A (drill tip, 0–25 cm) and C (40– 60 cm) in Test 2 (March) and segment B (20–40 cm) in Test 3 (September) together with the sample handling hardware elements (Test 4). We used chemical precision cleaning, a core component of both Quality Assurance for robotic landed missions to Mars and PP (e.g., Frick et al., 2014). In 2019, we did not use heat sterilization because surfaces could be easily re-contaminated in the field, and decontamination cycles would require additional time. Further, heat sterilization would not remove organic contaminants. Tables 1 and 2 summarize the chemical cleaning FIG. 3. (A) Disassembled drill string, scoop, and funnel prepared for field deep cleaning and testing; (B) preliminary step (distilled water rinse) for removing particulate material from the drill string; (C) ATP swab assay of one instrument’s funnel hardware; (D) Surface Swab Assay samples for SOLID co-analysis; (E) ATP molecule (left) and hand-held luminometer with liquid (left) and surface (right) sampling devices. ATP, adenosine triphosphate. 1308 BONACCORSI ET AL. and microbial reduction steps we applied; their specific effectiveness for ARADS is addressed in Section 5 and Figs. 6A–H and 7. 4.2.1. Cleaning and disinfecting agents. The precision cleaning involved (1) adenosine triphosphate (ATP)-free distilled water (DW) or liquid chromatography/mass spectrometry (LC/MS) grade water for removal of visible dirt (adhering clays and dust particles); (2) solvents with decreasing polarity (pure grade acetone, methanol, and Et-OH) to remove the aromatic and aliphatic molecular organics; and (3) microbial reducing agents, for example, 60–80% EtOH, high-purity IPA, 0.5–5% household bleach (NaClO; fresh batch, unopened bottle as direct sunlight and organic matter inactivate NaClO), and 3% hydrogen peroxide (H 2 O 2 ) for disrupting bacterial biofilms and to inactivate/kill microorganisms. NaClO and H 2 O 2 were applied individually or simultaneously. We used various dispensing methods and multiple applications to maximize the chemical exposure to the target hardware, for example, bleach-soaked (impregnated) polypropylene wipes wrapped around the string’s surface, with wraps molded around the drill’s spirals for 1 h. Afterward, the drill was wiped off, air-dried, and thoroughly rinsed with ca. 0.5–1 L of ATP-free DW. 4.2.2. Dispensing and wiping. We adapted techniques based on hardware topography; in some cases, we used professional spray bottles to mist, spray, and stream each chemical. Still, highly volatile solvents evaporate quickly from the target hardware, decreasing their cleaning effectiveness. In some cases (drill deep cleaning), we preferred stroke-wiping hardware with lint-free anhydrous solventimpregnated wipes to remove contaminants instead of direct spraying onto the hardware. The standard Quality Assurance for robotic landed missions to Mars requires periodic precision cleaning of exposed flight hardware surfaces with high-purity IPA-soaked lint-free swabs and wipes (e.g., Carosso, 2005). We impregnated wipes (by spraying sufficient reagents onto them) for the most efficient disinfection (Panousi et al., 2009). In other cases (e.g., funnel’s narrow outlets), we nebulized aqueous IPA several times to maximize exposure to the disinfecting agent. Finally, a last iteration with (IPA)- soaked low-lint Kimwipes (KIMTECH KimWipes Task Wipers) removed cellular remains from the surface to maximum disinfection. The wiping best practice we applied involved consistent pressure on every stroke, working from top to bottom (for the drill string) and ‘‘pull and lift’’ from the interior (cleanest) to the exterior (dirtiest) areas (for scoop and funnels). FIG. 3. (Continued) (F) Mitigation contamination during ARADS 2016 ground truth sampling of soil pits (Photo credit: NASA Ames); (G) containment set up at Yungay Basecamp before field-testing of the NASA Goddard’s Ion Trap Mass Spectrometer. (H–K) ARADS practices for contamination mitigation examples. See description in Section 4.1. (H) Reduced field personnel during rover ops; (J) minimized unnecessary walking on playa toward rovers; and monitored footsteps on ground sites of scientific interest, that is, post-flood desiccated biofilms (K). ARADS CONTAMINATION CONTROL STRATEGY IMPLEMENTATION 1309 When folded, the wipes provide up to eight clean, usable sides to prevent re-contamination across the cleaned surface. 4.2.3. Water rinse. A third step included thoroughly rinsing with a stream of ATP-free gas chromatograph–mass spectrometer (GC-MS) grade water for the complete removal of cleaning agents, any Kimwipes’ fiber residue, dead/ inactivated microbes, and cellular constituents. 4.3. Hardware contamination verification assays We used two independent contamination verification methods: the ATP luminometry assay for viable total bioburden and the Fluorescence Sandwich Microarray FIG. 4. (A) Out-of-sim bioburden contamination tests with the TRIDENT drill in working position; (B) Top: Icebreaker drill segment including the Drill bit (0–25 cm). Bottom: drill Segment 20–40 cm with a smoother surface finish than the Drill bit; (C) mounted drill after a 24-h exposure to the GP environment and Al foil bio-barrier protecting drill before field testing; (D) potential sources of environmental contamination from surface biology at the GP Site: Vascular plant biomass; and (E) field of cyanobacteria-colonized halite. TRIDENT, The Regolith and Ice Drill for Exploration of New Terrains. Table 1. Multi-Step Protocol, as Described in Table 2 These steps involve mechanical cleaning (1–5), solvent-based molecular organics removal (6–8), physical and chemical microbial reduction/sterilization (9–15), and preventive actions (16–17). Household bleach is 5.25% sodium hypochlorite (NaClO). DW =distilled/deionized water; Et-OH =ethanol; IPA =isopropyl alcohol; LC/MS =liquid chromatography/mass spectrometry; MetOH =methanol. 1310 BONACCORSI ET AL. Immunoassay (FSMI) for broader characterization of microbial taxa and their molecular and cellular contaminants. Both techniques, described next, involve a solvent swab assay of the post-cleaning exterior drill string segments and the interior surface of the scoop and funnel. Best practices include a consistent swabbing technique (horizontal, vertical, and diagonal), directional pressure, and time ca. 5 min (see e.g., Barengoltz, 2005). Swab samples require subsequent wet extraction for contaminants analysis. Depending on the surface roughness, we used two swabs to retrieve the ATP from the hardware surface and combined the resulting values or assayed smaller surfaces if covered by mineral dust. 4.3.1. ATP luminometry assay. We relied on the ATP Swab assay for the in situ real-time, pre-screening contamination risk-reduction and disinfection certification because all living organisms use ATP as the universal energy carrier (Lundin and Thore, 1975a, 1975b), and this molecule quickly degrades after cellular breakdown (HolmHansen and Booth, 1966). The assay is used for the bioburden monitoring of spacecraft hardware in cleanrooms (NASA Procedural Requirements 8020.12D; 2011, Venkateswaran et al., 2003). The Hygiena ATP assay system (Fig. 3E) includes a self-calibrating (25–28C) handheld Luminometer (EnSURE) with a dynamic range of 0.1–2000 femtomoles, operating with an ultrasensitive ATP surface swab (SuperSnap), whose limit of detection (LOD) is 0.1 ·10 -15 moles, or 0.1 fmoles ATP. The luminometer measures the light emitted by the firefly’s (Photinus pyralis) luciferin-luciferase (L-L) enzymatic reactions (pH 7.6-buffered) binding with the ATP released by living cells in the presence of oxygen and magnesium (Balkwill et al., 1988), proportionally to the ATP in the sample (Lundin and Thore, 1975a; McElroy and DeLuca, 1983). The photodetector converts the generated photons into Relative Luminosity Units, or RLUs (15-s readings), that can translate into ATP concentrations (1 RLU equivalent to 0.1 fmoles ATP) with dilutions of ATP standard (Na_ATP salt). ATP þDLuciferin þO2þLuciferase Mg2þ  >Oxyluciferin þAMP þPyrophosphate þCO2 þhv (560 nm) 4.3.2. Fluorescence sandwich microarray immunoassay. The Quality Assurance for robotic landed missions to Mars (Phoenix, MSL, and 2020) requires follow-up standard molecular cleaning verification by Solvent Swab Analysis (SWA) of flight hardware’s swabbed surfaces with a Freonbased solvent via Fourier transform infrared (FTIR) spectroscopy or GC-MS (Anderson et al., 2002; Mahaffy et al., 2004; Blakkolb et al., 2014). A subset of post-cleaned hardware (Sol 5 and Sol 6) analyzed for ATP (Table 6) was also co-analyzed repurposing the SWA for use by the SOLID’s FSMI targeting SOLIDrelevant biological contamination in place of contaminating compounds detectable by FTIR or GC-MS. We used Teflon Table 2. Contamination Control Practices Used for the Rover Astrobiology Drilling Studies Subsystems See Table 1 for an explanation of each protocol’s coded items. SHTS =Sample Handling and Transfer System. ARADS CONTAMINATION CONTROL STRATEGY IMPLEMENTATION 1311 analyzed with the LDChip200 immunosensor for broader characterization of the Sol 5 and Sol 6 hardware background residue (ATP synthase and cellular protein biomarkers). The full dataset and immunogens’ intensity peaks are provided in Supplementary Information Fig. S1, Table S2a and S2b. We could perform only one sample extraction and concentrate the volume to improve detection. In our experience, no differences in fluorescence values resulted when using the same extract twice. Error bars correspond to the standard deviations of three internal replicates of each antibody array we tested. 5.2.2.1. ATP synthase antibodies. ATP synthase subunit 1 (ASB and 20_prot ASF1 [ATP synthase subunit 2]) is a protein that forms the intracellular energy storage molecule ATP in the mitochondria. We detected ATP synthase in all hardware surfaces co-analyzed with the ATP Luminomety Assay (Table 6). Although the signal in the blank swab (negative control) was high, after subtracting the negative control, the immunoassay yielded near-zero FI signals for ATP synthase (20_prot ASF1) in the scoop and drill (September 20 assay). The only FI signal for the protein’s other subunit ASB (FI: 2347 –850) relates to Sample 1 (scoop, September 19) and subunit ASF1 in airborne samples 9 and 10, in agreement with values of ATP above 6.0 RLUs (Table 6). High RLU luminescence matches the increased fluorescence signal of ASB 2, whereas low RLU corresponds to negative detection (FI: 0) or trace signals (FI: 55–235). The two signals appear positively correlated (R=0.836) when including the two air samples, but the correlation drops when considering only hardware samples (N=5), and more analysis is necessary for a robust correlation. 5.2.2.2. Cellular and protein biomarker antibodies. The FI peaks (max values) against the microbial taxa and molecular biomarkers detected by the LDChip200 were higher than those against ATP synthase in the post-cleaned hardware (Table 7 and Fig. 8). Further, we compared each biomarker detected in the hardware with those identified in the air dust samples (Supplementary Table S1a, b). The comparison informs whether the background contaminant was from an airborne source or potential crosscontamination from drilled sediments residues transferred to the sample handling hardware and analysis. Fifty percent or less of taxa (trace residues) detected in the drill, scoop, and funnels were also in the dust. In comparison, more than 50% were detected only in the background residue transferred from drill to funnel to scoop (putative subsurface biome). Very low signals (FI <400–600) against Anabaena sp. are ubiquitous in post-cleaned hardware and dust samples. The highest FI-positive values in the swabbed hardware samples are against proteins Gram-negative bacteria ancestral beta-lactamase (GNCA), Gammaproteobacterial ancestral beta-lactamase (GPBCA), and NifS2 (Leptospirillum ferrooxidans). The peptidoglycan (polymer sugars and amino acids forming bacteria cell wall) is found only in the 19-Sept scoop (FI <800 and 8.4 –1.6 amoles ATP/cm 2 ) but not in the dust. Very weak signals against Bacillus subtilis and Streptomyces sp. are detected only in post-cleaned hardware from Sol 6 (September 20), whereas the peak intensity (<700) is against Planococcus sp., a Gram-positive psychrotolerant, halophilic bacterium. FIG. 8. Linear plot with negative correlation (R=-0.673) between post-cleaning surface ATP background (ATP luminescence) and Immunoassay-based microbial taxa background (Fluorescence max intensity) for Sol 5 and Sol 6 hardware. Note the weak signals (FI: <700), except for the drill bit’s positive signal (FI >1000), as detailed in Table 7. FI, fluorescence intensity. Table 7. Post-Cleaning Microbial Cells and Protein Biomarkers Detected by the LDChip_200 in the Hardware Background and Co-Analyzed for Adenosine Triphosphate Biomarker by Adenosine Triphosphate Luminometry (Thick-Bordered Columns) From left to right: total ATP swab Assay of in-sim cleaned HD; ATP per surface unit; maximum value of FI (peak) against microbial taxa (whole cell) and molecular biomarkers (proteins). ‘‘Presence in air samplers’’ indicates the fraction (%) of microbial taxa and protein biomarkers detected in one or both airborne samplers and the HD background. We calculated the percent of molecular biomarkers in air samplers from the antibodies raised against whole cells, proteins, biofilms, and sediment mixture cultures in the HD residue (Supplementary Table S2a and S2b) and those detected in the two air samplers (Table 9). GNCA =gram-negative bacteria ancestral beta lactamase. 1318 BONACCORSI ET AL. The surface swab of the post-cleaned hardware yielded relatively low background contaminants (Table 7), indicated by traces (<300) or weak (<1000) FI signals, except for a few moderate signals (1000–3000). The intensity signals from protein biomarkers contaminants (FI peak *1000–1600) are also higher than those against cellular materials (FI peak: *300–1000). A few high signals (>3000) pertained only to materials from the air samplers (Fig. 10 and Table 9). The results suggest that solvent-cleaning and microbialreduction steps effectively mitigated residual and active bioburden contamination in the ARADS hardware, with a few exceptions. The chip detected a moderate FI signal against Geobacter sp. (FI: 1008) and the Gammaproteobacterial lactamase protein (FI: 1472) in the post-cleaned drill bit (Sol 6, September 20). However, this highest background intensity signal corresponds to one of the lowest signals detected against ATP synthase (FI: <300), which correlates with the ATP biomarker (1.2 –0.6 amoles/cm 2 ). The ATP background contamination was equivalent to traces of viable bioburden of ca. £1 cell/cm 2 , or a total of *100–200 cells in the drill bit (0.6 –0.0 amoles/cm 2 ATP) based on *1–2 amoles ATP per bacterial cell (Okanojo et al., 2017). Further, a negative correlation exists between the FI peak against microbial cells (gray columns) and the ATP background signal (r=-0.673) as shown in Fig. 8. The relationship suggests that only a negligible fraction (if any) of the residual microbes detected by the LDChip200 may be viable (ATP activity) and that the FI relates to residual traces of dead or partially fragmented cells. There is no significant correlation between ATP background and FI peak against proteins (r=-0.196), whereas no correlation exists between the cells’ and the proteins’ FI peak (r=-0.017). Correlation data are not presented. 5.3. Airborne contamination Figure 9A–D and Table 8 illustrate monitoring setup for airborne biomass flux to collectors 1 and 2. 5.3.1. Anthropogenic and environmental airborne biomass. The ATP-based airborne biomass flux at Collector 1 ranged from 23.7 amoles to 47.4 amoles/cm 2 across the 6-Sol simulation. The deposition rates of anthropogenic ATP at basecamp, average of 217.8 fmoles/(m 2 $day), were 4.2 times higher than the rates of environmental ATP, which averaged 51.6 fmoles/(m 2 $day). The flux of airborne anthropogenic delivered over 5 days at basecamp (Coll. 1) was 110.7 amoles, averaging 22.1 amoles/ (cm 2 $day), or 221 fmoles/(m 2 $day), whereas Playa Collector 2 received a total amount of 46.4 amoles/cm 2 ATP over the same period exposure. After background subtraction (Bkgr: 20.6 amoles/cm 2 ), the total flux of airborne environmental biomass to the playa (Coll. 2) was 25.8 amoles/cm 2 , equivalent to 5.2 amoles/(cm 2 $day) or 51.6 fmoles ATP/(m 2 $day). Further, by assuming that basecamp Collector 1 would receive airborne dust in similar amounts to the Playa Collector 2, the theoretical net anthropogenic component could be calculated as the difference between the two airborne fluxes at the two collectors (Coll 1 Flux -Coll 2 flux), which is 166.2 fmoles ATP/(m 2 $day). Table 8 reports the calculated airborne flux (environmental biomass) based on ATP swab data and normalized FIG. 9. (A, C) Airborne collectors’ location and deployment. (A) Basecamp Collector 1 outside the Science trailer (dark red); (B) Collector 2, with basecamp in the background; (C) Top view of the 23 ·23 cm collecting surface and sub-sampling areas (A–H). Supplementary Tables S1a and b in the Supplementary Information report the surface area and ATP swab data used for flux calculations. (D) Monitored 6-Sol airborne contamination in collectors 1 and 2 as ATP attomoles/cm 2 . Estimated daily flux data are from Table 8. Square symbols and error bars indicate the average and STD (Column F, Table 8) of individual measurements (not shown here; see the gray-shaded Column J in the Supplementary Table S2 instead). The solid bars indicate the total amount of surface ATP measured over 5 days (Column D). STD, standard deviation. ARADS CONTAMINATION CONTROL STRATEGY IMPLEMENTATION 1319 for the entire collector’s surface (23 ·23 cm). For collectors 1 and 2 airborne flux calculations, see Supplementary Information for pre-deployment setups, post-sanitation ATP background determination (Supplementary Table S0a and b), and daily airborne biomass raw data (Supplementary Table S1a, b and Fig. S3). 5.3.2. Microbial taxa in airborne samples. The two collectors were located downwind from the Playa and received a higher flux of contaminants from the wind-scoured playa surface sediments (Fig. 13 [3, 4]). LDChip200 detected in situ bacteria, archaea, and molecular biomarkers at the two sites, with mostly very weak signals of the relative FI (FI <600–1000) across the two sites, but a few exceptions (Table 9 and Fig. 10). There are a few intensity signals (weak signals) in both Basecamp Collector 1 (sample #9) and the distal alluvial fan/Playa Collector 2 (Sample #8), although #9 has more positive ones. The 10 m track sample is a little more contaminated given the moderate signals (FI >1000) for antibodies against Dechloromonas,Desulfovibrio (Coll 1); the highest fluorescence peak (FI >6000) related to the PhaC1 (a protein from aquatic cyanobacteria, Anabaena sp., Nostoc sp.) and the Ferridoxin enzyme. The aquatic microflora was present in both collectors, but the cyanobacterium Xenococcus, characteristic of semi-arid hypolithic communities (the Atacama B sequence, WarrenRhodes et al., 2006), occurred only in the distal Collector 2. Overall, the positive immuno-reactions occurred with antibodies raised against: (1) Gram-negative acidophilic ironand sulfur-oxidizing bacteria (Acidithiobacillus,Desulfovibrio); (2) perchlorate-reducing Betaproteobacteria (Dechloromonas, Ideonella), Gammaproteobacteria (Shewanella,Pseudomonas, Gammaproteobacteria ancestral beta-lactamase); (3) Cyanobacteria: aquatic (Anabaena,Nostoc) and hypolithic (Xenococcus); (4) Gram-positive Actinobacteria (Streptomyces)and Firmicutes (Bacillus,Desulfosporosinus); (5) strictly anaerobic sulfateand metal-reducers bacteria (Desulfosporosinus,Shewanella); and (6) halophilic heterotrophic archaea and methanobacteria (Methanobacterium). The immunogram data in Table 9 show that the microbes exclusively present at Collector 2 are Gram-negative bacteria, with Gram-positive (Firmicutes, actinobacteria) and Archaea in the basecamp’s air sampler. The immunoassay data are not statistically relevant (only a collector for each area), whereas the immunograms match well the ATP data. 5.4. Microbial biomass of drilled and surface sediment Figure 11 shows the downcore distribution of ATP biomarker from three Playa drilling sites. The average ATP biomarker is 100–250 fmoles/g and appears to decrease with depth in Hole 5 (60 to <0.2 fmoles/g, below detection limit [DL]) and Hole 2 (from 350 to 5 fmoles/g). The distribution of ATP biomarkers in these cores appears to be from complex putative life (e.g., Warren-Rhodes et al., 2019) rather than from contaminants transferred during drilling via hardware, post-drilling sample manipulation (e.g., hand skin reading 200 fmoles), or airborne biomass-rich sediment nearby (Fig. 13), with these contaminants being prevented by our strict contamination control practices during operations. Table 8. Flux of Airborne Environmental (Collector 2) and Anthropogenic (Collector 1) Biomass From left to right: Column B: Swabbed surface area out of the total collector surface (529 cm 2 ); Column C: ATP swab values normalized to 529 cm 2 ; Column D: airborne biomass in attomoles ATP per surface unit (amol cm 2 ). Columns E and F: Summary statistics (min, max, Avg., and STD) for data relative to individual surfaces sub-areas A, B, C, D, etc. (not included, see Table S1 in the Supplementary Information). The last two rows report averaged ATP flux as femtomoles per square meter per day [fmoles/(m 2 $day)] and attomole/ (cm 2 $day). 1320 BONACCORSI ET AL. FIG. 10. Immunogram with the relative FI for the LDChip200 positive detection of microorganisms and polymeric biological markers at the basecamp and distal collectors. Note that some immunogens are present only in either one of the two collectors, whereas others occur in both ones. See Section 5.3.2. for a description and Table 9 for the list of antibody names. 1321 The unusually higher values of ATP in the upper-level samples (0–10, 40–50, and 10–20 cm) and the decreases of ATP with the borehole depth (50–60, 60–70, and 30–40 cm) might be due to naturally occurring metabolically active biomass present in these levels as the result of the latest ponding events. For comparison, 15.1–32.4 ng/g, or ppb, of total amino acids (Leu, Val, Ser, Ala, and Gly) were detected by the MILA instrument in three subsurface samples acquired in 2019 (values calculated from Mora et al., 2020; Supplementary Table S2). Finally, ‘‘Green Parrot’’ drilled dark brown clays (March 2019) yielded higher values (*2·10 3 to *7·10 4 cells/g) against the coarse-grained surface pavement (*10 2 –10 3 cells/g), with no recovery of microbial ATP in five out of FIG. 11. Depth distribution of microbial ATP in three holes expressed as fmoles of ATP biomarker per gram of sediment. Horizontal error is STD (Table S3 in the Supplementary Information). Table 9. Peak Numbers, Antibody Names, and Immunogens of Intensity Peaks for the Immunogram of Air Samples from Fig. 10 1322 BONACCORSI ET AL. seven trials (Supplementary Table S3). Supplementary Figure S2A–E reports protocol test results, and Supplementary Table S3 provides Summary statistics for the ARADS surface and drill samples dataset. 5.4.1. Surface distribution of biomass contaminant at Playa drill site. We measured the surface distribution of total, free, and microbial ATP in smectite clay-rich playa sediments as potential contaminant indicators to the ARADS sites (Fig. 12 and Supplementary Table S3 in the Supplementary Information). Samples included (1) a 3 mm layer of the desiccated microbial mat as positive control, that is, 4.7 –0.58 ·10 4 fmoles total ATP/g (N=9); (2) excavated clay from a nearby pit (3–5 cm-depth), that is, 7.2 –0.7 ·10 3 fmoles/g (N=10); and (3) never-ponded surface clay (0– 3 cm-depth), that is, 212 –268 fmoles/g (N=6). These values are equivalent to an estimated viable microbial biomass of 10 3 cells/g (0–3 cm-depth) and 10 5 cells/g (3–5 cm-depth) to very high levels (10 6 cells/g) in the 3 mm layer of the dried microbial mat (positive control samples) a 4-log or a 5-log higher microbial biomass than in drilled sediments or the majority of the Playa surface. This unexpectedly highly active surface biomass related to a thin crust of desiccated microbial mat (Fig. 3K) we sampled from small areas of the post-flooded Playa and was still preserved after a short-lived pond formed at its margin during an extreme flood event in 2017 (Personal communication; Kim Warren-Rhodes, 2019), 2 years before our campaign. In ephemeral ponds, spore-forming cyanobacteria can proliferate and leave post-flooded desiccated yet viable biomass on the surface. The desiccated, very brittle biofilm could be eroded, carried, and delivered by wind, increasing the contamination risk when entraining exposed cored material or the scoop and funnels during the sample transfer (Fig. 13). However, most surface biomass was 3to a 4-log lower than boreholes sediments or the high biomass patches. 5.5. Contamination transfer to drilled samples An integral component of the contamination plan included a source-to-sink assessment of potential contamination in the drilled samples (hardware-to-sample transfer, airborne biomass contamination on sampling hardware, etc.). Results from these tasks are infused into a conceptual model (Fig. 13) and integrated into the Discussion (Section 6). Figure 12 shows the hardware-to-sample contamination transfer calculated using post-cleaning hardware background data (Table 4), and by assuming an average drilled sediment (60 g) getting into contact with a 50 cm segment of the drill string (616 cm 2 ), 100% of the scoop (217 cm 2 ), and funnel (196 cm 2 ), and 100% contamination transfer (drill to scoop to funnel). 6. Discussion 6.1. Effectiveness of ARADS methods We developed integrated CCSI practices for ARADS to prevent and mitigate contamination of the SHTS and the acquired samples and learn about potential contamination pathways across the operating environments. Table 10 summarizes the mitigation tasks, key findings, and their applicability to field conditions. The ARADS practices are similar to those established to mitigate contamination during a landed mission to Mars (Spry et al., 2021). A very close relationship exists between microbial reduction and organics removal to protect science. Background contaminants in spacecraft hardware pose two risks to a life detection mission drilling into the martian ice-cement ground Table 10. Description and Field Application of Rover Astrobiology Drilling Studies Mitigation Practices (Items #1–6) and Critical Results PPE =personal protective equipment. ARADS CONTAMINATION CONTROL STRATEGY IMPLEMENTATION 1323 like Icebreaker (Heldmann et al., 2013; McKay et al., 2013) and the Mars Life Explorer (NAP, 2021). First, forward contamination from terrestrial microbes surviving spacecraft’s hardware sterilization is the primary concern (e.g., NRC, 2006; NAS, 2018, 2019, 2020). Spores can remain long-term viable in space, even without metabolic activity at the low temperatures of Mars’ surface (e.g., Horneck et al., 2010, 2012; Hallsworth, 2021), be shed on the surface during landing (Marshall and Mancinelli, 2011), and enter the underground, becoming metabolically active under increased water activity (Rummel et al., 2014; Orosei et al., 2018); this will increase the forward contamination risk should more habitable conditions exist underground or near the surface Category IV bioburden level limits are £30 +(2 ·10 5 ) spores from the landed spacecraft’s surface, mated, and encapsulated hardy bioburden surviving decontamination. Second, whole microbes or cellular fragments could release mission-relevant analytes such as amino acids—or any other target biomarker along the ‘‘ladder of life detection’’ (Westall et al., 2015; Neveu et al., 2018)—into the analytical suite. For instance, amino acids are a primary target of the Icebreaker life analyzers and other life detection missions. Bacillus spores contain free (l-glutamate and l-arginine) and bound amino acids, that is, mucopeptide and protein (Nelson and Kornberg, 1970; and references therein) releasable on fragmentation. If dislocated and transferred to analysis, only one B. subtilis spore in the space hardware could release more than 0.24 pg (or 10 -12 g) total AA. 6.1.1. False negatives in cleaning verification. We demonstrated that false negative detection of contaminants undermines microbial reduction and verification, resulting in more concern for ARADS than false positives, that is, the level of recovery of swabbed ATP decoupled from the actual bioburden background. False negatives can originate from many confounding factors. They are: (1) The hardware’s surface characteristics (discussed in Section 6.1.3). (2) The amount, distribution, and complexity of hardware bioburden (biofilms-organics layering) and mineral coatings (dust or oxidized). When nonvisible organics and the mineral film were present on the drill, the ATP swab assay yielded false negatives for two plausible reasons. (i) Mud particles and oily residues obscured the L-L ATP reaction, and (ii) molecular organic layers encasing the microbes protected them against detection (Fig. 6C, D) and microbial reduction. The exposed active bioburden could be detected and reduced only after acetone removes inorganic and organic polymers (oils, grease, lipids) before microbial reduction. This problem did not concern scoop and funnels, whose bioburden decreased linearly after treatments (Fig. 6G, H). Unlike the drill, no complex organic bio(films) build-up occurred during the long-term storage of scoop and funnels. The cleaning test results were consistent with the common wisdom that the cleaner a surface can get, the more successful microbial reduction and verification will be (e.g., Meszaros et al., 2005). Given what has been stated earlier, we suggest performing microbial reduction and verification on an acetone-cleaned drill. Disinfection-resistant biofilmorganics layering can be a common issue because the drill system in gears, bearings, and sliding surfaces of the drill tube connections makes routine use of standard lubricants, for example, 3-in-1 SAE 20 motor oil containing petroleum, LubriMatic White Lithium Grease, and WD-40 (Miller et al., 2008). (3) Cleaning agents’ residues on hardware. We first mitigated the false negative issue (a detection one) by conducting verification assays on solvent-rinsed hardware. Rinsing with water removes interfering cleaning and disinfecting agents and addresses one limitation posed to the ATP assay in clean rooms (see review by Shama and Malik, 2013). (4) Luminometry quenching effects by chromogenic pigments in samples decrease the light emission. Nanophase clays, FeOx, and salts can interfere with L-L reactions and chemiluminescence photodetectors. (5) Environmental field conditions. High diurnal temperatures (‡25–40C) and direct sunlight can trigger very high RLU values generating false positives (Bonaccorsi and Stoker, 2008). The use of chemiluminescence photodetectors over photomultiplier FIG. 12. ATP biomarkers (total, free, and microbial) in cored versus surface sediments and derived values of microbial biomass as cells/g sediment. Note the negligible contribution of the potential contamination transfer from hardware (Total ATP, shortest bars) to drill samples. 23.7 –2.1 amoles/g for DP and 63.4 –28 amoles/g (6-Sol Bkgr) Bkgr: post-cleaning hardware background. DP, deep cleaning. 1324 BONACCORSI ET AL. tubes can mitigate the latter issue. Further, windgenerated triboelectric effects could produce unusually high RLU values even at night, as we experienced during field measurements. We included positive and negative controls to address the assay performance. 6.1.2. Effectiveness of microbial reduction and verification. Countless cleaning protocols exist for laboratory, cleanroom applications or aseptic drilling of icy materials (e.g., Christner et al., 2005; Kuhn et al., 2014; Goordial et al., 2017; Kayani et al., 2018; Coelho et al., 2022). Yet, only a few have been designed for complex astrobiology life-detection technology trials. Miller et al. (2008) applied a multi-step protocol for aseptic drilling (Milli-Q water, 10% Lysol aqueous solution, 70% Et-OH, flame sterilization) (Table 3) and immunoassaybased deep biosphere characterization in the Rio Tinto region (Parro et al., 2008), consistently with other multi-reagent cleaning protocol certified by immunoassay-based Limulus Amebocyte Lysate (LAL) and GC-MS swab analyses (Eigenbrode et al., 2009). ARADS multi-reagent protocols combined immunoassay and ATP luminometry techniques to detect a broader range of viable and nonviable biological contaminants. Overall, the 6-h microbial reduction protocol enabled a 4-log drop of ATP background from unmitigated to aseptic conditions (1–2 amoles/cm 2 reference background within the aseptic threshold of £3–4 bacterial cells/cm 2 ); see Section 2. During the six Sols, aseptic drilling was possible by extending the protocol time (after Sol 2) across the timeconstrained drilling simulation flow (Fig. 7), and consistently with other multi-reagent cleaning protocols (Miller et al., 2008; Eigenbrode et al., 2009). Across the 6-day simulation, we tested NaClO-activated H 2 O 2 for more effective microbial reduction. After squirting 3% H 2 O 2 on a drill tip sprayed with NaClO, the resulting singlet oxygen ( 1 O 2 ) sterilant agent—decontaminated the hardware 30 times faster than bleach alone to levels achievable during the extensive 6-h deep cleaning in the lab (Table 4). The activation of H 2 O 2 by bleach is an easy and relatively safe, field-effective procedure for reducing bioburden after only a 2-min exposure, compared with the long exposure required by concentrated bleach alone (10–20¢), 10% bleach (3 h), 70% IPA (1 h), or 3% H 2 O 2 alone (>3 h). The redox reaction between the oxidizing hypochlorite ion (OCl - ) and the reducing H 2 O 2 agent forms water (H 2 O), chlorine ion (Cl - ), and 1 O 2 (Held et al., 1978; Greer, 2006) such as: H2O2(aq)þNaCIO (aq)!1O2(g)þNaCI (aq)þH2O(aq) The disinfection action was due to the excited state of 1 O 2 , which contains more energy than ground oxygen (Held et al., 1978; Wirz, 2009). 1 O 2 unfolds proteins (e.g., Del Maestro et al., 1980; Davies, 2003) and kills bacterial and eukaryotic cells (Dahl et al., 1987; Djimeli et al., 2014; Taewan et al., 2019) with microorganism inactivation proportional to 1 O 2 exposure. Low-temperature chemical sterilants such as 35% H 2 O 2 Gas Plasma fumigation, Ethylene Oxide (EtO), and Ozone (O 3 ) are all absolute biocides commonly used in clean rooms (e.g., Gordon et al., 2012). However, these techniques are inapplicable in the field because they are highly toxic and require complex equipment (e.g., containment chambers, deep vacuum draw zapping, gas monitoring systems) and several hours-exposure (Meszaros et al., 2005; Gordon et al., 2012). 6.1.2.1. Cleanliness verification and contamination knowledge. We used two methods for cleanliness verification, the ATP Assay, and the SOLID instrument microarray immunoassay technology, as a novel approach for contamination knowledge of the hardware background. We could not use the culture-based NSA, which enumerates (plate counting) dry heat-surviving aerobes endospores after a 72 h growth beyond the ARADS real-time timeline. In addition, the NSA overlooks non-cultivable cells, underestimating the total microbial burden, spore-forming strict anaerobes (Clostridium), and extremophilic archaea microorganisms of PP concern (e.g., Moissl-Eichinger et al., 2015). For these reasons, Bernardini and Venkateswaran (2016) co-analyzed ATP and NSA data from the MSL spacecraft. They proposed the ATP assay (per NASA-HDBK-6022, 2010) as the benchmark threshold for spacecraft surface cleanliness, that is, 2.3 ·10 -11 mmol ATP/25 cm 2 (or <0.9 fmole/cm 2 ). Several methods exist to detect and characterize hardware contaminants matrix-assisted laser desorption/ionization-time of flight, but complex setups limit their use in the field. Other real-time tools designed for field detection, such as Minion Oxford Nanopore for DNA and RNA sequencing (10 min) or the LAL assay for Gram-negative microorganisms, have been used (post-mission) in field trials (Eigenbrode et al., 2009; Cockell et al., 2018). However, the LAL assay does not detect only viable microbes and requires more extended sample processing and analysis time than the ATP Luminometry assay, that is, 15–60¢versus 15 s -5 min, respectively. Their application is beyond the scope of the quick disinfection validation requirement across the six Sols. In situ real-time and ultrasensitive ATP assay can detect 0.1 fmoles of ATP on hardware surface as an independent molecular proxy for active bioburden/microbial biomass, viable spores, or intact recently dead cells. Further, the ATP synthase FI signal was consistent with ATP’s post-cleaning hardware background levels. This independent verification corroborates the effectiveness of in-sim cleaning and microbial reduction. Further, the fluorescence signal against the protein biomarkers was higher than that against the whole cells. The two signals decoupled, likely because more fragmented cellular materials versus whole cells were present in the wind-carried dust (aeolian particles subject to photooxidation) and subsurface sediment. SOLID immunograms of whole cells, proteins, and other cellular compounds provided an effective verification tool to self-check against false positives contaminating science measurements. One limitation is that the immunoassay does not differentiate between living or dead cells (a forward contamination risk), but ATP assay co-analysis addresses this gap. Finally, GC-MS and high-performance liquid chromatography identify background contaminants in space hardware (contamination knowledge) with substantial time elapsed between sampling and analysis. For the ARADS hardware ARADS CONTAMINATION CONTROL STRATEGY IMPLEMENTATION 1325 contamination knowledge, we used SOLID immunograms; such an approach could reduce time and costs for the simultaneous assessment of forwarding and science contamination. 6.1.3. Role of hardware surface features or geometry. The space hardware design, geometry, surface topography, porosity, and roughness can affect contamination control effectiveness. For the same reason, these features can determine bioburden retention, dislodgement, transfer, or dispersal to the environment. Specifically, the stainless-steel arm-mounted Phoenix-like scoop interior was difficult to access due to its 90edges between the interior surfaces. The surface of the external scoop blade was more accessible but rougher than the scoop interior. As a result, the scoop’s interior and edge exhibited a post-cleaning threshold background higher than the funnels. Similarly, the TRIDENT drill bit’s finishing was threedimensional (3D) printed and etched to augment the recovery of drilled fines. This drill bit surface was rougher than the drill string (Fig. 2), resulting in high bioburden background (Fig. 6G, H and Tables 4 and 5). However, an augmented cleaning protocol reduced the background. Mitigating solutions could include current good manufacturing practices such as ‘‘coving’’ and polished surface finishing. Coving minimizes geometry-related contamination by replacing 90and 3D corners with curved surfaces and improves sterilization of clean rooms and hospital floors. Curved corners inside the scoop surfaces (inner walls) would prevent the accumulation of biological contaminants or drilled material (conducive to cross-contamination) and facilitate the pathway to cleaning and swabbing. An additional implementation could use state-of-the-art mirror-like polished surfaces or the drill and SHTS. Electropolishing ensures ultra cleaning of embedded contaminants in Titanium alloys for tissue engineering and surgical implants, which must be sterile before use. In addition, electropolished microscopically featureless surfaces reduce microbial adhesion and surface contamination (Bagno and Bello, 2004; Oshida, 2007; Shimakura, 2007; Tajima et al., 2008). 6.2. Contamination pathways We propose a conceptual model (Fig. 13) that includes the significant contamination pathways highlighted by data analysis and tests on ARADS materials (hardware, sediments, air samplers, etc.). Interestingly, the study of airborne materials delivered two surprising results: a higherthan-predicted environmental contamination (pathway 8–9 to 3–4); and a lower-than-expected anthropogenic one (pathway 1–2 to 3). The emerging picture is fascinatingly complex due to the intrinsic nature of the Atacama Desert. 6.2.1. Environmental sources of airborne contaminants. Wind-transported mineral dust can efficiently carry and disperse viable microbial life across the Atacama (Parro et al., 2011; and references therein). Whether this biomass constitutes an anthropogenic (human biome) or natural (environmental) contaminant, a cross-contaminant, or a ‘‘native’’ element is not always clear to us or understood (Azua-Bustos et al., 2019, 2022). Our tests revealed that airborne biomass represented an unusually high source of potential contamination, unexpected for the hyperarid core of the Atacama Desert. The high-intensity signal (>6000)—raised against aquatic microflora (Anabaena or Nostoc sp. cyanobacteria) in air samplers—points to a primary source of airborne biomass consistent with ATP data from the post-flooded Playa (10 6 cells/g sediment, ATP cell equivalent) and the airborne FIG. 13. Conceptual model illustrating source-to-sink contamination. Figure not to scale. Semi-quantitative cell estimates [cells/g, cells/cm 2 ,andcells/(cm$day 1 )] and numbers near each pathway. Basecamp [1] and anthropogenic contamination[2]industColl.1[3]andColl.2[4]. The thin dashed arrows indicate Rover [5] and cross-contamination pathways: (a) drill-to-subsurface [6] (a forward contamination analog); (a–d) drillto-sample transfer and analysis: (a) drill, (b) scoop, (c) funnel, (d) life detection instrument. Blue solid-lined arrows with ‘‘X’’ indicate unlikely contamination pathways, for example, desiccated postflood microbial mat [8] sourcing aquatic cyanobacteria [9] to sampling hardware (a–c) or subsurface sedimentary layers [6] and not flooded playa surface [7] delimited by the Playa shoreline (black dashed lines). Large green arrows indicate unexpected airborne biology, for example, Anabaena sp. [9] in collectors [3, 4]. Blue and red parallel dashed arrows represent heterogeneous (more distal?) sources of airborne biomass [10] to the collectors or sampling hardware. Coll., collector. 1326 BONACCORSI ET AL. biomass—5–22 amoles/(cm 2 $day) ATP, across the 6-Sol mission (Fig. 12). This suggests an unexpectedly high viable biomass lateral transfer from an unusual upwind source, that is, desiccated ponds within the nearby Playa a few tens of meters away from the drill site and 300–400 m from Basecamp Collector 1. Such an airborne sedimentary biomass represents a potential source of high contaminant (also very close to one of the drill sites) that could have entered robotic sampling and analysis if not mitigated. Ephemeral water-induced microflora population growth is uncommon and represents a novel phenomenon for the hyper-arid core of the Atacama Desert due to the recent onset (in 2015) of climatic change bringing yearly storms and floods to the Atacama region (Ferna ´ndez-Martı ´nez et al., 2019; K. Warren-Rhodes 2022, unpublished data). However, bursts of microflora growth can occur after extreme rain events in arid hot desert environments worldwide (e.g., Cire ´set al., 2017; and references therein; Bonaccorsi and McKay, 2022). The weaker-intensity signals (£1000) with LDChip point to secondary low-intensity sources other than the Playa. For instance, transport from the Atacama soil microenvironments containing strong acids (Quinn et al., 2005) could explain the detection of bacteria from acidic habitats (Parro et al., 2011; and references therein). Similarly, the detected archaea Methanobacterium could have been wind-carried from the same source of the aquatic cyanobacteria, the postflooded Playa. Methanogenesis can occur at the water-sediment interface, in anoxic micropores of fine-grained clay-rich sediments during floods. Gammaproteobacteria (Pseudomonas) and their associated biomarker are found in extreme desiccated environments (Parro et al., 2011), including Atacama playas and alluvial fan samples (Ferna ´ndez-Martı ´nez et al., 2019), subsurface hypersaline habitats (e.g., Salar Grande). The anaerobic Shewanella inhabits heavy metalcontaminated (iron, lead, and uranium) environments (mine activity). All these environments such as the GP Site (Fig. 4D, E) exist in closed or distal proximity to the ARADs’ drill sites and can provide biological contaminants of concern depending on their relative abundance and viability. The hyper-arid Atacama’s surface soil is mostly extremely low in viable microbes and associated biomolecular compounds (e.g., Cameron, 1969; Navarro-Gonzalez et al., 2003; Skelley et al., 2005; Warren-Rhodes et al., 2006; Ca ´ceres et al., 2007; Cockell et al., 2008) with biomass increasing with some depth (e.g., Ferna ´ndez-Martı ´nez et al., 2019). A few exceptions can include extensive surface source areas such as salars, microflora-colonized halite fields (e.g., Davila et al., 2010; Parro et al., 2011; Wierzchos et al., 2006, 2011; Davila and Schulze-Makuch, 2016), groundwater-fed oases with sparse vascular plants (Fig. 4D, E), or microflora blooms in response to sporadic and transient rain events (e.g., Schulze-Makuch et al., 2018; Uritskiy et al., 2018; 2019). Airborne dust from the sources cited earlier might carry contaminants to drilled samples or enter the analytical instrument, thus confusing the scientific results. A secondary but no less impactful contaminants are ‘‘fresh’’ and degraded fossilized biomolecules, for example, ATP, Phospholipid fatty acids (PLFA), DNA, lipopolysaccharides (LPS), which the hyper-arid surface soil of the Atacama Desert can preserve well (Lester et al., 2007; Bonaccorsi et al., 2010; Powers et al., 2018; Wilhelm et al., 2018). 6.2.2. Anthropogenic airborne contamination. We monitored the flux of airborne contaminants during the in-sim operations to evaluate the anthropogenic contamination risk (expected high) to the Playa environment and ARADS science. The biomass flux at the anthropogenic-dominated basecamp area was higher [22.1 amoles/(cm 2 $day) ATP] than that monitored by Collector 1, 250 m away from basecamp [5.2 amoles/(cm 2 $day) ATP; non-anthropogenic sources]. This flux of ATP-based biomass was higher than the average ATP background of the (post-cleaning) SHTS hardware, that is, 0.9 –1.1 to 3.6 –4.5 amoles/cm 2 (Table 5), implying that we effectively mitigated the delivery of airborne contaminants to the hardware. However, at Basecamp, LDChip200 detected microorganisms naturally occurring in the Atacama Desert (airborne from proximal and distal potential source areas). Only a few taxa detected with a weak signal (FI <1000) (B. subtilis spores; Streptomyces spp., spore, and mycelium) might be related to the human microbiome. However, B. subtilis and Streptomyces spp. are usual bacteria from soil (also common in the Atacama). B.subtilis spores can survive for up to 15 months in the hyper-arid core of the Atacama Desert with 15% survival (Dose et al., 2001). Streptomyces is a filamentous actinobacterium naturally present in the soil. This alkaliphilic and thermophilic genus is found in arid, hypersaline, or heavy metals-rich extreme habitats (Kampfer et al., 2014), including the hyper-arid Atacama (Santhanam et al., 2012; Goodfellow et al., 2017). Streptomyces can also be found in a healthy human microbiome (a rarer occurrence), where culture-based techniques have overlooked its presence due to low growth, but recent molecular-based studies highlighted its presence in the gastro-intestinal (Bolourian and Mojtahedi, 2018) and respiratory tracts (Huang et al., 2015; Herbrı ´ket al., 2020) and in the healthy skin (Gallo and Hooper, 2012). In particular, skin flakes are rich in somatic cells (5 ·10 8 cells shed every day or 2 ·10 7 cells every hour (Weschler, 1978)) and associated microbes. For instance, the hand microbiota varies *4·10 4 to 4.6 ·10 6 CFU/cm 2 (Leyden et al., 1987; Messager et al., 2004) and 1 ·10 7 culturable bacteria per cm 2 (Fredricks, 2001); yet these culture-based estimations are likely underestimates (Edmonds-Wilson et al., 2015; and references therein). On a cautionary note, Bacillus and Streptomyces taxa could come from the environment rather than human biome contaminants. However, we cannot discard this hypothesis either. An explanation for the weakness of the airborne anthropogenic signal could be inefficient sampling (undersampling) of airborne particles, that is, the swab was vertically placed to collect dust particles directly from the air. Another possibility is that the anthropogenic contaminants (skin, clothing, and droplets) shed by the human crew during operations were non-deposited, removed after deposition, or dispersed by airflow and wind. Further, the ARADS CONTAMINATION CONTROL STRATEGY IMPLEMENTATION 1327 signs of life detector-life detector chip during a Mars drilling simulation campaign in the hyperarid core of the Atacama Desert. Astrobiology 2023;23(12); doi: 10.1089/ast.2021 .0174 NAP. Planetary Science and Astrobiology Decadal Survey. Mars Life Explorer Mission Concept Study Report, 2021. NASA-HDBK-6022. Handbook for the Microbial Examination of Space Hardware. National Aeronautics and Space Administration: Washington, DC; 2010. National Aeronautics and Space Administration (NASA). Planetary Protection Provisions for Robotic Extraterrestrial Missions. NPR 8020.12D, National Aeronautics and Space Administration: Washington, DC; 2011. National Academies of Sciences, Engineering, and Medicine (NAS); Space Studies Board; Committee on the Review of Planetary Protection Policy Development Processes. Review and Assessment of Planetary Protection Policy Development Processes. The National Academies Press: Washington, DC; 2018; doi: 10.17226/25172 National Academies of Sciences, Engineering, and Medicine (NAS); Space Studies Board; Committee to Review the Report of the NASA Planetary Protection Independent Review Board. NASA Planetary Protection Independent Review Board (PPIRB) Report to NASA/SMD. 2019. National Academies of Sciences, Engineering, and Medicine (NAS); Space Studies Board; Committee to Review the Report of the NASA Planetary Protection Independent Review Board. Assessment of the Report of NASA’s Planetary Protection Independent Review Board. The National Academies Press: Washington, DC; 2020; doi: 10.17226/25773 National Research Council (NRC). Preventing the Forward Contamination of Mars. The National Academies Press: Washington, DC; 2006. doi: 10.17226/11381 Navarro-Gonzalez R, Rainey FA, Molina P, et al. Mars-like soils in the Atacama Desert, Chile, and the dry limit of microbial life. Science 2003;302:1018–1021. Nelson DL, Kornberg A. Biochemical studies of bacterial sporulation and germination. 18. Free amino acids in spores. J Biol Chem 1970;245(5):1128–1136. Nescerecka A, Juhna T, Hammes F. Behavior and stability of adenosine triphosphate (ATP) during chlorine disinfection. Water Res 2016;101:490–497; doi: 10.1016/j.watres.2016.05 .087 Neveu M, Hays LE, Voytek MA, et al. The ladder of life detection. Astrobiology 2018;18(11):1375–1402; doi: 10 .1089/ast.2017.1773 Okanojo M, Miyashita N, Tazaki A, et al. Attomol-level ATP bioluminometer for detecting single bacterium. Luminescence 2017;32:751–756; doi: 10.1002/bio.3246 Orosei R, Lauro SE, Pettinelli E, et al. Radar evidence of subglacial liquid water on Mars. Science 2018;361(6401): 490–493. Oshida Y. Surface science and technology—Titanium dental implant systems. J Soc Titanium Alloys Dentist 2007;5:52–53. Pandey S, Clarke J, Nema P, et al. Ladakh: Diverse, highaltitude extreme environments for off-Earth analogue and astrobiology. Int J Astrobiol 2019;19(1):1–21; doi: 10.1017/ S1473550419000119 Panousi MN, Williams GJ, Girdlestone S, et al. Evaluation of alcohol wipes used during aseptic manufacturing. Soc Appl Microbiol Lett Appl Microbiol 2009;48:648–651. Parnell J, Cullen D, Sims M, et al. Searching for life on Mars: Selection of molecular targets for ESA’s Aurora ExoMars mission. Astrobiology 2007;7:578–604. Parro V, Ferna ´ndez-Calvo P, Rodrı ´guez Manfredi JA, et al. SOLID2: An antibody array-based life detector instrument in a Mars Drilling Simulation Experiment (MARTE). Astrobiology 2008;8(5):987–999. Parro V, de Diego-Castilla G, Moreno-Paz M, et al. A microbial oasis in the hypersaline Atacama subsurface discovered by a life detector chip: Implications for the search for life on Mars. Astrobiology 2011;11(10):969–996. Parro V, Puente-Sa ´nchez F, Cabrol NA, et al. Microbiology and nitrogen cycle in the benthic sediments of a glacial oligotrophic deep Andean lake as analog of ancient martian lake-beds. Front Microbiol 2019:10:929; doi: 10.3389/fmicb.2019.00929 Patel A, Malinovska L, Saha S, et al. ATP as a biological hydrotrope. Science 2017;356:753–756. Powers LS, Smith HD, Kilungo AP, et al. In situ real-time quantification of microbial communities: Applications to cold and dry volcanic habitats. Glob Ecol Conserv 2018;16:e00458. Quinn R, Zent AP, Grunthaner F, et al. Detection and characterization of oxidizing acids in the Atacama Desert using the Mars Oxidation Instrument. Planet Space Sci 2005;53(13): 1376–1388; doi: 10.1016/j.pss.2005.07.004 Rivas LA, Garcı ´a-Villadangos M, Moreno-Paz M, et al. A 200antibody microarray biochip for environmental monitoring: Searching for universal microbial biomarkers through immunoprofiling. Anal Chem 2008;80;7970–7979. Rummel JD, Beaty DW, Jones MA, et al. A new analysis of Mars ‘‘special regions’’: Findings of the second MEPAG special regions science analysis group (SR-SAG2). Astrobiology 2014;14:887–968. Rutala WA, Weber DJ; the Healthcare Infection Control Practices Advisory Committee. Guideline for Disinfection and Sterilization in Healthcare Facilities. 2019; p. 163. Available from: https://www.cdc.gov/infectioncontrol/pdf/guidelines/ disinfection-guidelines-H.pdf [Last accessed: December 11, 2023]. Sanchez-Garcı ´a L, Ferna ´ndez-Martı ´nez MA, Moreno-Paz M, et al. Simulating Mars drilling mission for searching for life: Ground-truthing lipids and other complex microbial biomarkers in the iron-sulfur rich Rı ´o Tinto analog. Astrobiology 2020;20(9):1029–1047; doi: 10.1089/ast.2019.2101 Santhanam R, Okoro CK, Rong X, et al. Streptomyces atacamensis sp. nov., isolated from an extreme hyper-arid soil of the Atacama Desert, Chile. Int J Syst Evol Microbiol 2012;62: 2680. Schulze-Makuch D, Wagner D, Kounaves SP, et al. Transitory microbial habitat in the hyperarid Atacama Desert. Proc Natl Acad Sci U S A 2018;115(11):2670–2675; doi: 10.1073/ pnas.1714341115 Uritskiy G, Getsin S, Munn, A, et al. Response of extremophile microbiome to a rare rainfall reveals a two-step adaptation mechanism. bioRxiv 442525, 2018; doi: 10.1101/442525 Shama G, Malik DJ. The uses and abuses of rapid bioluminescence-based ATP assays. Int J Hyg Environ Health 2013;216:115–125. Shimakura M. Efficient polishing method of titanium. J Soc Titanium Alloys Dentist 2007;5:32–33. Siragusa SR, Dorsa WJ, Cutter CN, et al. Use of a newly developed rapid microbial ATP bioluminescence assay to detect microbial contamination on poultry carcasses. J Bioluminesc Chemiluminesc 1996;11(6):297–301. Skelley AM, Scherer JR, Aubery AD, et al. Development and evaluation of a microdevice for amino acid biomarker detection and analysis on Mars. Proc Natl Acad Sci U S A 2005;102:1041–1046. 1334 BONACCORSI ET AL. Smith SA, Benardini JN, Anderl D, et al. Identification and characterization of early mission phase microorganisms residing on the Mars Science Laboratory and assessment of their potential to survive Mars-like conditions. Astrobiology 2017;17(3):253–265; doi: 10.1089/ast.2015.1417 Soffen GA, Snyder CW. First Viking mission to Mars. Science 1976;193:759–766. Spry JA, Siegel B, Kminek G, et al. Planetary protection knowledge gaps and enabling science for human Mars missions. Bull Am Astron Soc 2021;53(4):205. Stanley PE. Extraction of adenosine triphosphate from microbial and somatic cells. Methods Enzymol 1986;133:14–22. Stoker CR, Cannon HN, Dunagan SE, et al. The 2005 MARTE Robotic Drilling Experiment in Rio Tinto Spain: Objectives, approach, and results of a simulated mission to search for life in the Martian subsurface. Astrobiology 2008;8(5):921–945. Stoker CR, Glass BJ, Glass, Stucky TR, et al. A mission simulating the search for life on Mars with automated drilling, sample handling, and life detection instruments performed in the hyperarid core of the Atacama Desert, Chile. Astrobiology 2023;23(12):1284–1302; doi: 10.1089/ast.2022-0055 Taewan K, Hyung-eun K, Jiyoon C, et al. Inactivation of Escherichia coli and MS2 coliphage via singlet oxygen generated by homogeneous photosensitization. Korean J Chem Eng 2019;36(11):1785–1790; doi: 10.1007/s11814019-0353-4 Tajima K, Hironaka M, Chen K, et al. Electropolishing of CP titanium and its alloys in an alcoholic solution-based electrolyte. Dent Mater J 2008;27(2):258–265. Uritskiy G, Getsin S, Munn A, et al. Halophilic microbial community compositional shift after a rare rainfall in the Atacama Desert. ISME J 2019;13(11):2737–2749; doi: 10 .1038/s41396-019-0468-y Venkateswaran K, Hattori N Duc MTL, et al. ATP as a biomarker of viable microorganisms in clean-room facilities. J Microbiol Methods 2003;52(3):367–377. Warren-Rhodes KA, Rhodes KL, Pointing SB, et al. Hypolithic cyanobacteria, dry limit of photosynthesis, and microbial ecology in the hyper-arid Atacama Desert. Microb Ecol 2006; 52:389–398. Warren-Rhodes KA, Lee KC, Archer SDJ, et al. Subsurface microbial habitats in an extreme desert Mars-analog environment. Front Microbiol 2019;10:69; doi: 10.3389/fmicb .2019.00069 Weschler CJ. Characterization techniques applied to indoor dust. Environ Sci Technol 1978;12(8):923; doi: 10.1021/ es60144a011 Westall F, Foucher F, Bost N, et al. Biosignatures on Mars: What, where, and how? Implications for the search for martian life. Astrobiology 2015;15:998–1029. Wierzchos J, Ascaso C, McKay CP. Endolithic cyanobacteria in halite rocks from the hyperaridcore of the Atacama Desert. Astrobiology 2006;6:415–422. Wierzchos J, Camara B, de los Rios A, et al. Microbial colonization of Ca-sulfate crusts in the hyperarid core of the Atacama Desert: Implications for the search for life on Mars. Geobiology 2011;9(1):44–60; doi: 10.1111/j.1472-4669.2010 .00254.x Wilhelm MB, Davila AF, Parenteau MN, et al. Constraints on the metabolic activity of microorganisms in Atacama surface soils inferred from refractory biomarkers: Implications for martian habitability and biomarker detection. Astrobiology 2018;10(7):955–966. Willis P, Creamer JS, Mora M. Implementation of microchip electrophoresis instrumentation for future spaceflight missions. Anal Bioanal Chem 2015;407:6939–6963. Willson D, Bonaccorsi R, Gold R, et al. Contamination Management of Sample Collection Devices and Ocean Analogs for Life Detection on Icy World Plume Fly-through Missions. Paper ID. B5.3-64-18. 42nd COSPAR Scientific Assembly, July 14–22, 2018, Pasadena, CA, 2018. Wirz KPJ. Photochemistry of Organic Compounds: From Concepts to Practice. John Wiley & Sons: Chichester, UK; 2009. Zwartkruis E, Bettray G, Wilke T. Rapid testing of microbial surface contamination on veal carcasses using ATP bioluminescence. Fleischwirtschaft 1999;79:101–103. Address correspondence to: Rosalba Bonaccorsi NASA Ames Research Center MS. 245-3 Moffett Field, CA 94305 USA E-mail: [email protected] Submitted 10 October 2022 Accepted 2 July 2023 Associate Editor: Jack Mustard Abbreviations Used 1 O 2 ¼singlet oxygen 3D ¼three-dimensional ARADS ¼Atacama Rover Astrobiology Drilling Studies ASB ¼ATP synthase subunit 1 ASF1 ¼ATP synthase subunit 2 ATLO ¼Assembly, Test, and Launch Operations ATP ¼adenosine triphosphate Avg. ¼average CCSI ¼Contamination Control Strategy and Implementation CFU ¼colony forming unit COSPAR ¼Committee on Space Research DHMR ¼dry heat for microbial reduction DP ¼deep-cleaning DW ¼distilled/deionized water Et-OH ¼ethanol FI ¼fluorescence intensity FSMI ¼Fluorescence Sandwich Microarray Immunoassay FTIR ¼Fourier transform infrared GC-MS ¼gas chromatograph–mass spectrometer GNCA ¼Gram-negative bacteria ancestral beta-lactamase GP ¼Green Parrot GPBCA ¼gammaproteobacterial ancestral beta-lactamase H 2 O 2 ¼hydrogen peroxide HD ¼hardware IPA ¼isopropyl alcohol ISAD ¼Icy Soil Acquisition Device LAL ¼Limulus Amebocyte Lysate LC/MS ¼liquid chromatography/mass spectrometry LITMS ¼Linear Ion Trap Mass Spectrometer L-L ¼luciferin-luciferase LOD ¼limit of detection LPS ¼lipopolysaccharides Met-OH ¼methanol ARADS CONTAMINATION CONTROL STRATEGY IMPLEMENTATION 1335 MILA ¼Microfluidic Life Analyzer MSL ¼Mars Science Laboratory ND ¼no detection NSA ¼NASA Standard assay PLFA ¼phospholipid fatty acids PP ¼Planetary Protection ppb ¼parts per billion PPE ¼personal protective equipment PSTAR ¼Planetary Science and Technology Through Analog Research RLU ¼relative luminosity units SHTS ¼Sample Handling and Transfer System SOLID ¼Signs of Life Detector STD ¼standard deviation SWA ¼swab analysis TEGA ¼Thermal Evolved Gas Analyzer TRIDENT ¼The Regolith and Ice Drill for Exploration of New Terrains UHV ¼ultra high vacuum WCL ¼Wet Chemistry Laboratory 1336 BONACCORSI ET AL.