Mars methane detection and variability at Gale crater

See allHide authors and affiliations

Science  23 Jan 2015:
Vol. 347, Issue 6220, pp. 415-417
DOI: 10.1126/science.1261713

Of water and methane on Mars

The Curiosity rover has been collecting data for the past 2 years, since its delivery to Mars (see the Perspective by Zahnle). Many studies now suggest that many millions of years ago, Mars was warmer and wetter than it is today. But those conditions required an atmosphere that seems to have vanished. Using the Curiosity rover, Mahaffy et al. measured the ratio of deuterium to hydrogen in clays that were formed 3.0 to 3.7 billion years ago. Hydrogen escapes more readily than deuterium, so this ratio offers a snapshot measure of the ancient atmosphere that can help constrain when and how it disappeared. Most methane on Earth has a biological origin, so planetary scientists have keenly pursued its detection in the martian atmosphere as well. Now, Webster et al. have precisely confirmed the presence of methane in the martian atmosphere with the instruments aboard the Curiosity rover at Gale crater.

Science, this issue p. 412, p. 415; see also p. 370


Reports of plumes or patches of methane in the martian atmosphere that vary over monthly time scales have defied explanation to date. From in situ measurements made over a 20-month period by the tunable laser spectrometer of the Sample Analysis at Mars instrument suite on Curiosity at Gale crater, we report detection of background levels of atmospheric methane of mean value 0.69 ± 0.25 parts per billion by volume (ppbv) at the 95% confidence interval (CI). This abundance is lower than model estimates of ultraviolet degradation of accreted interplanetary dust particles or carbonaceous chondrite material. Additionally, in four sequential measurements spanning a 60-sol period (where 1 sol is a martian day), we observed elevated levels of methane of 7.2 ± 2.1 ppbv (95% CI), implying that Mars is episodically producing methane from an additional unknown source.

Because Earth’s atmospheric methane is predominantly biologically produced (1), determining the abundance and variability of methane in the current martian atmosphere is critical to assessing the contribution from a variety of potential sources or reservoirs that may be biological [such as methanogens (1, 2)] or abiotic (1). These latter processes include geological production such as serpentinization of olivine (3), ultraviolet (UV) degradation of meteoritically delivered organics (46), production by impacts of comets (7), release from subsurface clathrates (8) or regolith-adsorbed gas (9, 10), erosion of basalt with methane inclusions (11), or geothermal production (12). Several detections of Mars methane have been published. Ground-based observations from the Canada-France-Hawaii Telescope in 1999 found a global average value of 10 ± 3 parts per billion by volume (ppbv) (2), and those using the NASA Infrared Telescope Facility (IRTF) telescope in 2003 reported (13) methane release in plumes from discrete sources in Terra Sabae, Nili Fossae, and Syrtis Major that showed seasonal changes with a summertime maximum of ~45 ppbv near the equator. The planetary Fourier spectrometer (PFS) on the Mars Express (MEX) spacecraft reported detection in 2004 (14) with an updated global average abundance of 15 ± 5 ppbv (15), with indications of discrete localized sources and a summertime maximum of 45 ppbv in the north polar region. From the thermal emission spectrometer (TES) of Mars Global Surveyor, methane abundances from 5 to 60 ppbv were deduced (16) as intermittently present over locations where favorable geological conditions such as residual geothermal activity (Tharsis and Elysium) and strong hydration (Arabia Terrae) might be expected. Using data from NASA IRTF acquired in February 2006, Krasnopolsky (17) reported a detection of 10 ppbv at mid-latitudes (42° to 7°N) over Valles Marineris but an upper limit of 3 ppbv outside that region. In December 2009, Krasnopolsky obtained an upper limit of 8 ppbv and noted that data from both observations agreed with those of Mumma et al. (13). More recent ground-based observations report methane mixing ratios that have diminished considerably since 2004–2006 to a 2σ upper limit of 5 ppbv (1719), suggesting a very short lifetime for atmospheric CH4 and contradicting the MEX claim that methane persisted from 2004–2010. At Curiosity’s Gale crater landing site (4.5°S, 137°E), published maps of PFS data (15) show an increase from ~15 ppbv in fall to ~30 ppbv in winter, whereas the TES trend (16) is opposite: ~30 ppbv in fall and ~5 ppbv in winter.

Observational evidence for methane on Mars has been questioned in the published literature (2022) because photochemical models are unable to reconcile the observed amounts with their reported spatial gradients and temporal changes over months compared with the expected ~300-year methane lifetime. Contradictions were noted between the locations of maxima reported from ground-based observations and maps inferred by PFS and TES from Mars orbit. The plume results (13) were questioned (22) on the basis of a possible misinterpretation from methane lines whose positions coincided with those of terrestrial isotopic 13CH4 lines. Krasnopolsky (7) argued that cometary and volcanic contributions were not sufficient to explain high methane abundances—noting for the latter possibility the lack of current volcanism or hot spots in thermal imaging (23)—and the extremely low upper limit for Mars SO2 that in Earth’s volcanic emissions is orders of magnitude more abundant than CH4, as predicted for Mars (24). Model calculations including expected atmospheric transport and circulation (20, 25) are, to date, all unable to reproduce the spatial and temporal characteristics of the observed high-concentration methane plumes, whether resulting from possible clathrate release (8) or surface adsorption by or desorption from the regolith (9) or for UV degradation of surface organics (46), despite the introduction of a variety of putative loss mechanisms (2628).

The tunable laser spectrometer (TLS) of the Sample Analysis at Mars (SAM) (29) instrument suite on the Curiosity rover has a spectral resolution (0.0002 cm−1) that offers unambiguous identification of methane in a distinct fingerprint spectral pattern of three well-resolved adjacent 12CH4 lines in the 3.3-μm band (30). The in situ technique of tunable laser absorption in a closed sample cell is simple, noninvasive, and sensitive. TLS is a two-channel tunable laser spectrometer that uses both direct and second harmonic detection of infrared (IR) laser light. One channel uses a near-IR tunable diode laser at 2.78 μm that has yielded robust data on carbon, oxygen, and hydrogen isotopic ratios on Mars (31). The second channel uses an interband cascade tunable laser at 3.27 μm for methane detection alone, scanning across seven rotational lines that include the R(3) triplet used in this study. This laser makes 81 passes of a 20-cm-long sample cell of the Herriott design fitted with high-vacuum microvalves that allow evacuation with a turbomolecular pump for empty cell scans or filled to Mars ambient pressure (~7 mbar) for full cell runs. Our methane determination is made by differencing the measured methane abundances in our sample cell when filled with Mars atmosphere from measurements of the same cell evacuated, as detailed in the supplementary materials (32).

From our first six observations spanning a 234-sol period (1 sol = 1 Mars day = 24 hours 37.3 min), we previously reported (33) a mean value of 0.18 ± 0.67 ppbv that was not precise enough to claim detection of Mars methane but instead set an upper limit of 1.3 ppbv [95% confidence interval (CI)] that was significantly lower than those reported (5 ppbv, 95% CI) from recent ground-based observations (1719).

We have now reprocessed our entire data set [with a small modification explained in (32)]. Our data set now extends the measurement period over 605 sols, including 11 direct-ingest measurements and two recent measurements using a methane enrichment experiment run on sols 573 and 684. In this latter procedure, the atmospheric methane is effectively enriched 23 ± 1 times by flowing the ingested gas slowly over a carbon dioxide scrubber material. Before running on Mars, the instrument script was optimized using the test-bed SAM suite (32). Results from the complete data set are given in Table 1 and plotted in Fig. 1. We partition our data points of Fig. 1 into three groups for independent analysis: (i) the low-methane direct-ingest results of sols 79, 81, 106, 292, 313, and 684; (ii) the low-methane enrichment results for sols of 573 and 684; and (iii) the four sequential high-methane runs of sols 466, 474, 504, and 526, as there is no statistically significant variation within each grouping. Mean values for these grouped data sets (final three rows, Table 1) form the basis for our analysis and conclusions. We note the good agreement between the direct and enriched experiments that were run back-to-back on sol 684. The daytime run of sol 306 is not included in group (iii) because it was not part of the high methane sequence, nor is it included in the low-methane group (i) because it is clearly higher than the background average.

Table 1 Curiosity TLS-SAM methane measurements at Gale crater (4.5°S, 137.4°E) over a 20-month period.

Ls, solar longitude. CI values are ±2 SEM for individual results and are explained in the supplementary materials (32) for the grouped results in the last three rows.

View this table:
Fig. 1 The TLS-SAM methane measurements versus martian sol.

Plotted values are from Table 1, with error bars representing ±1 SEM. Those with larger error bars are the direct-ingest results, and the two with smaller error bars labeled “EN” are the values retrieved from the methane enrichment runs. All measurements were made from nighttime ingest, except the two marked “D” that were ingested during the day and analyzed at night. The last direct-ingest value (plotted near sol 700) occurred during the same sol 684 as the last enrichment run but is offset to a higher sol value to separate the points for visual clarity. The shaded boxes show the occurrence and duration of the SAM evolved gas analysis runs for Rocknest (RK), John Klein (JK), Cumberland (CB), and combustion during which gases evolved from rock samples were introduced into SAM and a portion was fed into the TLS sample cell for analysis.

Our low-methane enrichment experiments produce a mean value for atmospheric methane of 0.69 ± 0.25 ppbv (95% CI), as described in (32). The direct-ingest (nonenrichment) group yields a mean methane value of 0.89 ± 1.96 ppbv (95% CI), agreeing with the higher-precision enrichment value within error. For the high-methane abundance seen in direct-ingest group, we measure a mean value of 7.19 ± 2.06 ppbv (95% CI) for the four sols 466, 474, 504, and 526. In the supplementary materials (32), we provide arguments to rule out the possibility of terrestrial contamination and therefore conclude that the enrichment result and the high-methane result independently produce detection of methane at two levels of abundance. Although TLS samples only the very lowest part (~1 m) of the Mars atmosphere in the Gale crater region, the atmospheric mixing time of a few months suggests that our measured value of 0.69 ± 0.25 ppbv (95% CI) is probably representative of the mean background level for Mars atmospheric methane abundance, which is only expected to vary substantially and seasonally over the winter poles (20).

The principal sources of organics delivered exogenously to Mars are isotropically accreted interplanetary dust particles (IDPs) and low-mass carbonaceous chondrites containing up to 10% organics by weight (1, 46). Recent observations by SAM on Curiosity have detected the presence of chlorobenzene and simple chlorinated alkanes (34) in a drilled martian mudstone in Gale crater. Laboratory studies of meteoritic materials have shown that UV irradiation of organic molecules can produce methane either directly (5) or through secondary photochemical reaction (35) and that certain molecules can form a photoresistant layer leading to methane over extended time periods (6). Constrained by laboratory production rates, models have assessed the rate and size of infall of meteoritic material such as IDPs, carbonaceous chondrites, and other sources of organic carbon to the martian surface that might reproduce methane observations under Mars-like UV conditions. The UV/CH4 model of isotropically accreted IDP organics of Schuerger et al. (5) predicts that the UV-induced production of methane is carbon-limited and over geological time can produce a globally averaged methane abundance of 2.2 ppbv methane for a 20% conversion rate of organic carbon to methane. No significant diurnal or seasonal changes are predicted by this model, which cannot explain the variability of methane over relatively short time scales observed in earlier studies (13, 15). Even with consideration of single large bolide impacts or multiple airburst events, the models struggle to emplace sufficient carbon over the large surface areas of the plume observations and, more importantly, cannot supply methane fast enough to create plumes over the observed time frame (5).

Our background methane abundance reported here of ~0.7 ppbv from the low-methane enrichment is significantly lower than the 2.2 ppbv obtained from the Schuerger UV/CH4 model estimate (5) described above, despite the fact that measured surface UV levels from Curiosity’s Rover Environmental Monitoring Station (REMS) instrument (32, 36) agree with the model values (5). This implies that either the quantity of delivered carbon or its conversion efficiency to methane is one-third of the model’s estimates or that an indigenous source may be having an effect. It is also likely that the fresh analog material used by Schuerger et al. (5) is not completely representative of the bulk of material (UV-processed IDPs) being delivered to Mars.

As detailed in (32), our high-methane result of 7.19 ± 2.06 ppbv (95% CI) shows no significant quantitative correlation with relative humidity, atmospheric pressure (carbon dioxide abundance), ground or air temperature, inlet pointing, or radiation levels measured by other Curiosity instruments: the REMS (36), the Chemistry and Camera complex [ChemCam (37)], and the Radiation Assessment Detector (38). The REMS observations suggest a plausible anticorrelation with water abundance as well as air and ground temperatures, and both REMS and the Curiosity mast camera (39) show a possible anticorrelation with atmospheric opacity (32), but our first enrichment measurement on sol 573 spoils this. However, all methane measurements (including sol 573) support an anticorrelation of methane abundance with column measurements of oxygen abundance and water vapor as measured by the ChemCam instrument (see fig. S9), the latter contrasting with the weak positive correlation observed by the Mars Express PFS (8, 40). However, the lack of O2 and H2O data for the solar longitude range Ls = 160 to 220 spoils this comparison, and we must await future measurements to assess this fully.

Concerning the possibility of spatially variable methane abundance, although the high-methane measurements were observed within 200 to 300 m of each other (fig. S10), the rover had not traveled far (~1 km) since the lower value of sol 466, and the high methane disappeared after traveling only a further ~1 km away. Typical ground winds of ~7 m/s (25) would cover that distance in only 2 min, and given the rotation of diurnal wind, it is impossible to isolate one location from another. This suggests a short-duration event that is either local and weak or more distant and stronger. The persistence of the high-methane values over 60 sols and their sudden drop 47 sols later is not consistent with a well-mixed event but rather with a local production or venting that, once terminated, disperses quickly. Most of our data are taken at night, when prevailing winds are likely from the south. The marginally higher daytime values suggested in sols 306 and 526 indicate a source to the rover’s north, because prevailing daytime winds would advect toward the rover location. The change in rover location is therefore unimportant as this is a temporal, not locational, variation. With a concern that Curiosity transit over varying surface materials [identified by the alpha particle x-ray spectrometer (41) measurements] could be associated with the high-methane observations, we studied rover stand time and local terrain composition (32) and ruled out such potential contributions. Although we cannot rule out possible clathrate release (8) or surface adsorption into the regolith with subsequent release (9), both these mechanisms do not support the local, short–time scale variation that we observe.

Our measurements of a background methane abundance of ~0.7 ppbv can be reconciled with photochemical models that include an exogenous source such as UV degradation of organics (5) because model results likely represent upper limits with extensive UV processing in space before delivery to Mars. Like the earlier plume measurements, our higher transient methane amounts of ~7 ppbv require an additional source of methane, in our case suggesting advection to the rover location from a local unidentified source. If that source were a recent bolide striking Gale crater and producing 1% methane, we estimate that it would have to be several meters in size and leave a crater of tens of meters in diameter, but no new impact craters have been observed within Gale crater from Mars orbit time-series imaging (42) since landing. Our measurements spanning a full Mars year indicate that trace quantities of methane are being generated on Mars by more than one mechanism or a combination of proposed mechanisms—including methanogenesis either today or released from past reservoirs, or both.

Supplementary Materials

Materials and Methods

Supplementary Text

Figs. S1 to S13

Tables S1 and S2

References (4345)

The MSL Science Team Author List

References and Notes

  1. See supplementary materials on Science Online.
  2. Acknowledgments: The research described here was carried out in part at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with NASA. Data described in the paper are further described in the supplementary materials and have been submitted to NASA’s Planetary Data System (PDS) under an arrangement with the Mars Science Laboratory (MSL) project. Funding is acknowledged for J.M.-T. and M.-P.Z. from the Spanish Ministry of Economy and Competiveness, J.C.B. from the UK Space Agency, and R.G. from the Canadian Space Agency.
View Abstract

Stay Connected to Science

Navigate This Article