Detectability of Chlorofluorocarbons in the Atmospheres of Habitable M-dwarf Planets

Thousands of exoplanets have so far been discovered from space-based telescopes, such as Kepler and TESS, as well as from ground-based observatories. Detection methods can constrain the orbital position and bulk properties of such planets, but follow-up observations of planetary spectra in transit, emitted, or reflected light can provide information about the presence and composition of a planet's atmosphere. These methods have already been demonstrated for the spectral characterization of gas giant atmospheres (e.g., Charbonneau et al. 2002; Benneke et al. 2019; Tsiaras et al. 2019), while detecting and characterizing the atmospheres of smaller, Earth-sized planets remains an ongoing priority for exoplanet science.

One of the astrobiological motivations for the spectral characterization of planetary atmospheres is the possibility of detecting evidence of life on an exoplanet. This has inspired the search for "biosignatures," which refer to remotely detectable spectral features that could indicate evidence of life on an exoplanet. The idea of searching for spectral biosignatures has received significant attention with regard to identifying plausible biosignatures, assessing their detectability limits, and developing strategies for conducting a search for them in tandem with the broader goals of the astrophysics community (e.g., Seager et al. 2012; Grenfell 2017; Kaltenegger 2017; Catling et al. 2018; Fujii et al. 2018; Meadows et al. 2018; Schwieterman et al. 2018; Walker et al. 2018; Lammer et al. 2019; O'Malley-James & Kaltenegger 2019).

Biosignatures refer generally to any remotely detectable evidence of life, while "technosignatures" (Tarter 2007) specifically describe observational evidence of technology that could be detected through astronomical means. Technosignatures are a logical continuation of the search for biosignatures, both of which draw upon the history of life and technology on Earth as examples of planetary evolution (NASA Technosignatures Workshop Participants 2018). The science of identifying and classifying technosignatures, and developing cost-efficient methods to search for them, remains in a state of infancy compared to biosignature science (Wright 2019; Haqq-Misra et al. 2020; Lingam & Loeb 2021). Nevertheless, several possible classes of technosignatures have already been identified that include waste heat (Dyson 1960; Carrigan 2009; Wright et al. 2014; Kuhn & Berdyugina 2015), artificial illumination (Schneider et al. 2010; Loeb & Turner 2012; Kipping & Teachey 2016; Tabor & Loeb 2021), artificial atmospheric constituents (Owen 1980; Campbell 2006; Schneider et al. 2010; Lin et al. 2014; Stevens et al. 2016; Kopparapu et al. 2021), artificial surface constituents (Lingam & Loeb 2017), stellar pollution (Shklovskii & Sagan 1966; Whitmire & Wright 1980; Stevens et al. 2016), nonterrestrial artifacts (Bracewell 1960; Freitas Jr & Valdes 1980; Rose & Wright 2004; Haqq-Misra & Kopparapu 2012), and megastructures (Dyson 1960; Arnold 2005; Forgan 2013; Wright et al. 2016).

The modern era of SETI (Wright et al. 2018; i.e., the search for technosignatures) began with the realization that humanity could use existing technology to detect a similar level of technology over interstellar distances (Cocconi & Morrison 1959; i.e., powerful, deliberately directed radio signals). A major milestone in SETI would be achieved when present-day detection technologies become sensitive enough to detect humanity's ongoing and passive technosignatures at such distances. The full Square Kilometer Array, for instance, is thought to be sensitive enough to detect humanity's typical radar emission at distances of a few parsecs (Loeb & Zaldarriaga 2007); however, future projections in which Earth's radio leakage decreases significantly would be much more difficult to detect (Forgan & Nichol 2011).

Industrial pollution represents a class of atmospheric constituents on Earth that could conceivably be technosignatures if observed in the spectra of an exoplanet. One example is nitrogen dioxide (NO₂), which has large sources on Earth from combustion that are greater than nonanthropogenic sources. A study by Kopparapu et al. (2021) showed that the absorption features of NO₂ in the 0.2–0.7 μm range could be detectable with the Large Ultraviolet Optical IR Surveyor (LUVOIR; LUVOIR Team 2019). Kopparapu et al. (2021) found that a 15 m LUVOIR-like telescope could detect Earth-like levels of NO₂ for a planet around a Sun-like star with ∼400 hr of observation, while planets orbiting K-dwarf stars would require even less time due to the reduction in loss of NO₂ from photolysis in such systems. The detection of elevated NO₂ levels in the atmosphere of an exoplanet could be consistent with ongoing industrial processes on the surface, although any such observations would need to be evaluated against possible nontechnological explanations before concluding that the NO₂ must be a technosignature. In this regard, such a search for technosignatures shares the same ambiguity or tentative nature as many or most searches for biosignatures.

In this study, we examine halocarbons—molecules that contain carbon and halogen atoms—as another class of pollutants that could serve as technosignatures. We specifically focus on chlorofluorocarbons (CFCs), which are only produced in significant quantities on Earth from industrial uses as refrigerants, blowing agents, and cleaning agents. CFCs are potent greenhouse agents with long atmospheric residence times. The only sink for most CFCs is photolysis by ultraviolet radiation in the stratosphere, which releases chlorine atoms that cause the depletion of stratospheric ozone on Earth (Seinfeld & Pandis 2006). The Montreal Protocol of 1987 placed limits on the production of certain CFCs in order to prevent further damage to the ozone layer (Velders et al. 2007). These provisions have been successful to the extent that ozone-depleting compounds are much less present in the stratosphere; however, the recovery of the ozone layer to pre-1980s levels appears to be slow and shows large uncertainties in both measurements and model projections (e.g., Eyring et al. 2010; Chipperfield et al. 2017).

Observing CFCs in the atmosphere of an exoplanet would be compelling evidence of a technosignature. The accumulation of CFCs on an exoplanet could be the result of ongoing industrial processes (Owen 1980; Campbell 2006; Schneider et al. 2010), particularly for a planet on which ozone loss is not a concern. CFCs could also be useful in artificially increasing the greenhouse effect on a planet, which could have applications in terraforming a planet to increase its suitability for life (Marinova et al. 2005; Dicaire et al. 2013). The detectability of CFCs for a planet orbiting a white dwarf host star was examined by Lin et al. (2014), who focused specifically on CCl₃F (known as CFC-11) and CF₄ (known as CFC-14) because they both show strong absorption features in the IR. Lin et al. (2014) estimated that CFCs at abundances 10 times greater than present-day Earth could be detected in a white dwarf system by the James Webb Space Telescope (JWST) with ∼1.7 days of observing time.

Here we calculate detectability limits of CFCs for an Earth-sized planet orbiting an M-dwarf star. Such systems are likely targets for characterization by upcoming space missions such as the JWST or mission concepts such as LUVOIR, the Habitable Exoplanet (HabEx; Gaudi et al. 2018), Origins (Meixner et al. 2019), or the Large Interferometer for Exoplanets (LIFE; LIFE collaboration et al. 2021; Quanz et al. 2019), in addition to large ground-based observatories such as Extremely Large Telescopes (ELTs; Quanz et al. 2015; Snellen et al. 2015). We focus on the detectability of CFC-11 (CCl₃F) and CFC-12 (CCl₂F₂), which are two of the most abundant CFCs in Earth's atmosphere with elevated levels that have persisted despite the Montreal Protocol. Our model simulations are constrained to planets within the habitable zone of the host star, which represents the circumstellar region where a terrestrial planet could sustain surface liquid water (Kasting et al. 1993; Kopparapu et al. 2013). Planets within the habitable zone of low-mass stars are expected to fall into synchronous rotation, so that one side of the planet experiences perpetual day with the other side in perpetual night. We use a three-dimensional general circulation model (GCM) to calculate the equilibrium climate state for such synchronously rotating habitable planets at Earth-like and elevated abundances of CFCs. We then use the steady-state output from these GCM simulations to calculate synthetic IR spectra to show that absorption features of these CFCs could be detectable with missions such as the JWST and Origins.

The climate simulations in this study are conducted with the model called resolving orbital and climate keys of Earth and extraterrestrial environments with dynamics (ROCKE-3D; Way et al. 2017). ROCKE-3D is a GCM that has been developed by the NASA Goddard Institute for Space Studies for the study of planetary habitability. ROCKE-3D has been used to understand the climate of ancient Venus (Way et al. 2016; Way and Del Genio 2020), to explore possible habitable climates for specific exoplanets (Kane et al. 2018; Del Genio et al. 2019; Fauchez et al. 2020), and to constrain the dependences of general habitability limits on planetary properties (Fujii et al. 2017; Way & Georgakarakos 2017; Checlair et al. 2019; Colose et al. 2019; Olson et al. 2020; Salazar et al. 2020). Our configuration of ROCKE-3D assumes aquaplanet (i.e., ocean-covered) conditions with a 1 bar atmosphere composed of N₂ and H₂O with 400 ppm CO₂. We used a "slab" ocean with a fixed 50 m depth and a present-day Earth q-flux parameterization of oceanic energy transport. The use of a slab ocean significantly reduces computational time, although the use of a dynamic ocean can cause differences in surface temperature by a few degrees (Colose et al. 2021). The model includes 40 vertical layers and a 4° × 5° horizontal resolution. We assume a planet with zero obliquity that has the mass and gravitational acceleration of Earth. A more thorough discussion of the model configuration, including details of the radiative transfer methods, is provided by Colose et al. (2021), while a broader technical description of ROCKE-3D is given by Way et al. (2017).

Here we perform ROCKE-3D simulations of an Earth-sized planet in synchronous rotation around a 3300 K and TRAPPIST-1 host stars with increased abundances of atmospheric CFC-11 and CFC-12. These species are included in ROCKE-3D as part of the Earth system model. They have strong absorption bands in the mid-IR part of the spectrum (Figure 1); however, this part of the IR spectrum is dominated by several other greenhouse gases such as CO₂, H₂O, and CH₄ (Figure 2) that could make a detection of CFCs difficult with upcoming observatories. In Section 3 we discuss the detectability of CFC-11 and CFC-12 in detail. We focus on these two particular systems because they are characteristic of systems likely to actually be studied by missions such as the JWST and Origins, in which CFCs could actually be detectable.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Our model experiments consist of four simulations conducted with different atmospheric abundances of CFC-11 and CFC-12. Both of these are potent greenhouse gases with an average lifetime of ∼55 yr for CFC-11 and ∼140 yr for CFC-12 in the Earth-Sun system (Seinfeld & Pandis 2006). The 0× case contains no CFCs and provides a reference control case for comparison. The 1× case includes 0.225 ppb CFC-11 and 0.515 ppb CFC-12, which are the present-day abundances of these CFCs on Earth. ¹¹ The subsequent 2× and 5× cases contain CFC-11 and CFC-12 abundances that have been increased by 2 and 5 times, respectively. These elevated CFC abundances are realistic projections, shown in Figure 3, that could have occurred on Earth if the Montreal Protocol had been ineffective (see Young et al. 2021).

Zoom In Zoom Out Reset image size

The CFC abundance is fixed in the model, with no sinks or sources due to chemistry. On Earth, the sources of CFC-11 and CFC-12 are industrial sites at the surface, while the only sinks occur when the CFCs rise into the stratosphere and are photolyzed at wavelengths between 185 and 210 nm. This results in the relatively long atmospheric lifetimes of CFC-11 and CFC-12 on Earth. Planets orbiting M dwarfs tend to receive even less incident shortwave radiation on average, so the lifetime of CFCs may be even longer. This study presents a set of steady-state calculations with fixed CFC abundances that are intended to constrain the order of magnitude of detectability, but further studies with a coupled chemistry-climate GCM would provide more robust constraints on plausible upper limits for the abundance and lifetime of CFCs.

We consider two planet-star system configurations, both of which are motivated by previous studies of planetary habitability using GCMs. The first configuration uses a climate configuration for TRAPIST-1e that was defined and studied in previous model intercomparisions (Fauchez et al. 2020, 2021). The second configuration uses a climate configuration for a planet in the habitable zone of a 3300 K host star that has been examined in previous studies (Kopparapu et al. 2017; Colose et al. 2021).

2.1. TRAPPIST-1 Host Star

2.2. 3300 K Host Star

Our GCM simulations give steady-state solutions for atmospheres with fixed abundances of CFC-11 and CFC-12, which can be used to identify the strength of potentially observable spectral features. We use the calculated outgoing IR flux values, cloud opacity, and greenhouse-gas mixing ratio values from our GCM simulations as input to the Planetary Spectrum Generator (PSG; Villanueva et al. 2018) to calculate synthetic spectra and assess limits of detectability. PSG is an online radiative transfer software that can be used to compute planetary spectra (atmospheres and surfaces) for various objects of the solar system and beyond. It includes a wide range of wavelengths (UV, visible, near-IR, IR, far-IR, THz, submillimeter, and radio) from any observatory, orbiter, or lander and also includes a noise calculator. In this work, we use a PSG add-on called Global Exoplanet Spectra (GlobES), which allows the user to ingest data from a variety of GCMs to accurately synthesize planetary spectra that are then computed with PSG. GlobES has been used for climate studies of TOI-700 day—the first habitable-zone terrestrial size planet discovered with TESS (Suissa et al. 2020).

We consider the GCM calculations in Section 2 for planets around TRAPPIST-1 and a 3300 K host star. For our synthetic spectra calculations, we use updated parameters for TRAPPIST-1 and TRAPPIST-1e from Agol et al. (2021). (Although our GCM simulations use older parameters from Grimm et al. 2018, this discrepancy does not have any quantifiable impact on our results.) The reason for focusing on M dwarfs is that near-term atmospheric characterization telescope missions focus on planets around M dwarfs, and their instruments will operate in the IR part of the electromagnetic spectrum. Likewise, as mentioned in earlier sections, CFC-11 and CFC-12 have dominant absorption in the IR, so the confluence of instruments operating in the IR region and the corresponding strong absorption of CFCs makes M-dwarf planets ideally suited for studying CFC detectability. Here we focus on generating the transit spectra of planets around M dwarfs, and we then estimate the observing time needed to detect the dominant CFC features with the JWST and Origins.

Figure 4 shows transit spectra for TRAPPIST-1e and an Earth-like planet around 3300 K star, both with 0×, 1×, 2×, and 5× Earth-level abundances of CFC-11 and CFC-12. The strongest absorption features are between the 8 and 12 μm range. We focus our detectability estimates within this region. As expected, higher amounts of CFCs increase the transit depth, which is represented as the planet-to-star contrast ratio in parts per million (ppm) on the y-axis. The depth of the 8–14 μm features for TRAPPIST-1e is higher than that of a planet around a 3300 K star because the stellar host is correspondingly smaller (TRAPPIST-1 is just slightly larger than Jupiter), which would enable a relatively larger surface area of the star to be transited by the planet and would therefore increase the transit depth. This suggests that CFC absorption features might be easier to detect around late M dwarfs than the earlier ones.

Zoom In Zoom Out Reset image size

To estimate the signal-to-noise ratio (S/N) for detecting these CFC features, we kept the TRAPPIST-1e planet at a star-planet separation of 0.029 3 au and at a distance of 12.4 pc. We used the TRAPPIST-1e planet parameters from Grimm et al. 2018 for our GCM simulations as done in THAI (Fauchez et al. 2020). For the planet around a 3300 K star, we kept the planet at 0.1 au, corresponding to the habitable zone for this star, and at a distance of 5 pc from Earth. Any S/N calculations can be qualitatively scaled to other distances based on our results for this star type. We have calculated the S/N by subtracting the 0× case with no CFCs from the 1×, 2×, and 5× cases and dividing the result with the noise, all within a wavelength range of the instrument under consideration. This method evaluates the S/N with respect to the 0× level of the observed spectra.

Figure 5 (left) shows the S/N values for 1× (blue), 2× (black), and 5× (red) CFC versus in-transit observation time in hours (i.e., no overheads) with the JWST Mid-Infrared Instrument low-resolution spectrometer (MIRI-LRS) using a resolution R = 50. Two levels of noise floors are shown as well: 10 ppm (dashed) and 50 ppm (dotted). We note that these noise floors are applied to multiple coadded observations, and the 50 ppm floor is therefore expected to be very conservative. With a 10 ppm noise floor, concentrations at the current Earth level (1×, dashed blue) can be detectable with an S/N ∼ 3 in ∼100 hr. While this may not be a robust detection, this could hint at a possible feature for further observations. However, the features of a 2× CFC level, which might have been Earth's current CFC level without the Montreal protocol, would be detectable with an S/N ∼ 5 with the same amount of time. We would be able to detect the features of a planet with a 5× CFC level (dashed red) with an S/N ∼ 10 in 100 hr of JWST MIRI-LRS time on TRAPPIST-1e. However, if the noise floor is set at 50 ppm, then even the 5× CFC features would not be detectable with a reliable S/N, regardless of the JWST MIRI-LRS observation time. While upper limits can be placed, these limits will not be robust to make any meaningful analysis.

Zoom In Zoom Out Reset image size

Figure 5 (right) shows similar S/N calculations for CFC on an Earth-like planet around a 3300 K host star using Origins. The system is located at 5 pc from Earth, which is intended to provide an optimal scenario for calculating CFC detection limits for Origins. There are about 60 stars within 5 pc, most of which are M dwarfs with about half in binaries, all of which are on the Breakthrough Listen target list (Isaacson et al. 2017). The spectral features are muted (see Figure 4, right) owing to a larger star, and the S/N values are lower than for TRAPPIST-1e for the present Earth-level case. However, higher CFC concentrations generate much higher S/N values. For the 2× and 5× CFC cases, an S/N ∼ 5 can be achieved with ∼600 hr of Origins observing time.

Note that the estimated integration times only take into account in-transit times. The real observation times including overhead would likely be two or three times longer. Moreover, no instrument noise systematics were considered in our JWST simulations, which only include pure white noise (i.e., noise that decreases with 1/$\sqrt{{photons}}$). The real noise will decrease at a slower rate and will therefore increase the required number of transits to achieve 3 or 5σ. Furthermore, for long integration times over a large aperture like that of the JWST, instrument systematics are expected to lead to a noise floor that is at a level of noise that cannot be reduced by adding more observations, as shown in Figure 5 (Fauchez et al. 2019).

As mentioned above, ∼100 hr of JWST MIRI-LRS observing time are needed to detect present Earth-level CFCs on TRAPPIST-1e with a noise floor of 10 ppm. Almost every transit in the JWST mission lifetime would need to be observed to reach ∼100 hr of in-transit time. There are ∼100 transits observable between 2022 June and 2028 July; this is technically longer than the nominal mission lifetime of 5.5 yr of the JWST, but NASA now estimates that JWST "should have enough propellant to allow support of science operations in orbit for significantly more than a 10 yr science lifetime." ¹² TRAPPIST-1e is likely going to remain the best target for such observations in the foreseeable future, so such an investment would be an expensive but potentially highly rewarding program.

Here, we have only calculated whether the effects of CFCs might be detectable at all and what its form would be, and not whether they can be unambiguously retrieved from a real spectrum given realistic ambiguities of the host planet's atmospheric structure and composition, and host star's intrinsic spectrum. We leave an analysis of these complexities to a future study.

In summary, the absorption features of CFC-11 and CFC-12 could potentially be detectable by upcoming missions such as the JWST, depending on the noise floor levels. Present or past Earth-like abundances of CFCs could be detected with observing times of ∼100–300 hr at an S/N ≳3–5. Large observing programs have been conducted previously, such as ∼400 hr for the Hubble Ultra Deep Field (Beckwith et al. 2006) or ∼900 hr for the CANDLES galaxy evolution survey (Grogin et al. 2011), so this requirement remains plausible. Moreover, such large observing programs are smaller than the estimates for biosignatures in a modern Earth-like atmosphere on TRAPPIST-1e (Fauchez et al. 2019; Lustig-Yaeger et al. 2019), i.e., ∼600 hr for O₃ detection at 9.6 μm or ∼800 hr for O₂ at 6.4 μm, with CH₄ and H₂O being undetectable in this scenario. An interesting point here is that the time needed to detect some present-Earth biosignature gases (∼600 hr) is longer than the observing time needed to detect present-Earth abundances of CFCs (∼300 hr) with the JWST. Furthermore, any attempt at characterizing spectral technosignatures would be conducted in tandem with a more general effort to characterize a planet's atmosphere and identify any potential biosignatures. Calculations such as those presented in this paper are useful in determining observability thresholds for detecting particular technosignatures, such as CFCs, which can aid in the development of observing strategies as well as motivate the design of new technology for future missions.

Halocarbons such as CFC-11 and CFC-12 are industrial compounds on Earth and thus could be evidence of extraterrestrial technology if observed in an exoplanet atmosphere. In the event of such an observation in exoplanet spectra, any absorption features claimed to be technosignatures would need to be examined against nontechnological and nonbiological alternatives. On Earth, there are no known abiotic or nonanthropogenic sources of CFC-11 or CFC-12 (Seinfeld & Pandis 2006), likely because such molecules are thermodynamically challenging to produce; however, this does not necessarily mean that CFC-11 or CFC-12 could not be generated on an exoplanet through nontechnological means. For example, other halocarbon species can be emitted to the atmosphere by phytoplankton, although these tend to be short lived (Lim et al. 2017). This is an instance of a more general problem in biosignature science, as identifying false positives that occur on exoplanets but not on Earth is a challenging task. Nevertheless, advancing the science of technosignatures will require evaluating such false positives for technosignatures such as CFCs, just as biosignature science continues to evaluate false-positive spectral signatures for habitability.

The calculations presented in this study indicate that CFCs could potentially be detectable by upcoming missions and future mission concepts given noise floor constraints, even at present-day Earth levels. The detection of absorption features within the 8–14 μm region, and subsequent identification of CFCs as the source of these features, will remain challenging, and future work will be needed to improve constraints on the detectability of CFCs for specific targets as observed by the JWST, Origins, or other missions. This study focused on the detectability of two M-dwarf systems that could plausibly be observed by the JWST or Origins, but future work could examine the possible detectability of CFCs in the atmospheres of habitable planets around solar-type stars with other mission concepts, such as LIFE.

We note that our study has focused on the concept of "detectability" for CFCs, and the actual detections of absorption features within the 8–14 μm region would not themselves be sufficient evidence to conclude the presence of CFCs in an exoplanet atmosphere. For an atmospheric composition like present-day Earth, CFCs may show some unique spectral features that help distinguish it from other gases (Figure 2), but other molecules such as CH₄, NH₃, CO, H₂S, and PH₃ also have absorption features that overlap with those of CFC-11 and CFC-12. Resolving any ambiguity regarding the identity of absorbing species will require further observations at other wavelengths that resolve additional absorption features. Theoretical modeling can also help to provide constraints on possible false-positive (and false-negative) scenarios for detecting CFCs in the atmospheres of exoplanets.

One limitation of this modeling study is that CFC abundances were assumed to be fixed and scaled without consideration of atmospheric chemistry. The use of a coupled chemistry-climate model would allow for a more self-consistent prediction of CFC abundances that could be sustained in the atmosphere, which would depend on the atmospheric composition as well as on the stellar spectrum. Our GCM cases also assumed a uniform distribution of CFCs across the atmosphere, which assumes that localized sources or sinks of CFC production are well mixed over time. However, scenarios in which strong, localized, and continuous sources and sinks of CFCs dominate other sources and sinks on the planet would require further study with a coupled climate-chemistry model to more accurately constrain the detectability of CFCs.

It is worth asking whether it is even reasonable to consider the detectability of planets with CFC abundances much greater than those on Earth today. Governments on Earth have banned the use of CFCs that could deplete ozone, while the threat of exacerbating climate change keeps our civilization from allowing long-lived radiatively active CFCs to accumulate in significant quantities. Such risks could also motivate extraterrestrial civilizations to minimize use of CFCs, although this is strongly dependent on the details of the climate needs of this species, the atmospheric composition of the planet, and many assumptions about the long-term aims and coordination of the species. For instance, the species might be incapable of preventing the buildup (because it is insufficiently organized or motivated to), indifferent to the buildup (because it has no important effects on the planet or because they are unaware of those effects), or causing the buildup deliberately (because it wishes to warm the planet, perhaps).

This study focused on planets in M-dwarf systems because such systems are likely to be characterized by upcoming missions, and the steady-state photochemistry on such planets will differ due to lower reaction rates for many atmospheric constituents. For example, planets orbiting K- and M-dwarf stars can accumulate O₃ more easily (e.g., Segura et al. 2010; Arney 2019), which could allow for CFCs to accumulate to higher levels than on Earth before causing environmental problems. Such scenarios should be considered in future work and should also be expanded to include a broad range of halocarbons beyond CFC-11 and CFC-12 only. Similarly, this study focused on the detectability limits of upcoming space missions, but future high-resolution ground-based spectroscopy facilities such as Extremely Large Telescopes (e.g., Cavarroc et al. 2006; Snellen et al. 2013; Kuhn & Berdyugina 2015; Lovis et al. 2017; Birkby 2018) may also be able to detect the presence of CFCs in exoplanet atmospheres.

We do not know the extent to which the specific CFCs produced on Earth would be prevalent elsewhere, even for extraterrestrial civilizations with similar industrial processes. The family of halocarbons is large, and much more work is needed to assess the detectability of a broader range of industrial molecules. This effort would be a step toward constructing a library of technosignatures for planning and interpreting future observations. But even such a library may be unable to identify industrial molecules that are chemically possible but have never been generated on Earth. Efforts to explore other possible industrial molecules, as well as their spectral signatures, could also help to constrain the use of halocarbons in general as technosignatures.

The CFCs are a notable example of a technosignature on Earth, and the detection of CFCs on a planet like TRAPPIST-1e would be difficult to explain through any biological or geologic features we know of today. Our civilization continues along a path of growth in both population and energy consumption, while we are only beginning to understand the extent to which our technology could be detectable at astronomical distances. Continued exploration of how the past, present, and future of civilization will affect Earth's detectability remains an important objective for understanding the prevalence of biosignatures and technosignatures in our galaxy.

In this study, we have shown that with the launch of the JWST, humanity may be very close to an important milestone in SETI: one where we are capable of detecting from nearby stars not just powerful, deliberate, transient, and highly directional transmissions like our own (such as the Arecibo Message), but consistent, passive technosignatures of the same strength as our own. Note that this is not a symmetric situation: the detectability of CFCs in an Earth-like planet's atmosphere is strongly dependent on the radius and spectrum of the host star, and the TRAPPIST-1 system in particular is extremely favorable in this regard. Even if Earth were seen to transit from that system, the Sun's large radius and Earth's orbital distance mean that the JWST would not detect CFCs around Earth from the distance of TRAPPIST-1.

In the next few decades, at least two of Earth's passive technosignatures, radio emissions and atmospheric pollution, would be detectable by our own technology around the nearest stars. It is possible that other plausible atmospheric technosignatures will prove even more detectable once their signal strengths have been calculated. We conclude that atmospheric technosignatures are at least as promising as communicative, radio technosignatures in this regard, especially since they can be searched for concurrently with biosignatures.

The authors thank Tom Greene and Knicole Colon for their suggestions regarding JWST instrument and noise floor specifics. J.H.M., A.F., J.T.W., and M.L. gratefully acknowledge support from the NASA Exobiology program under grant 80NSSC20K0622. R.K. and T.F. acknowledge support from NASA Goddard's Sellers Exoplanet Environments Collaboration (SEEC), which is supported by NASA's Planetary Science Division's Research Program. This work was facilitated through the use of advanced computational, storage, and networking infrastructure provided by the Hyak supercomputer system at the University of Washington. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of their employers or NASA.

Software: The ROCKE-3D source code is freely available at https://simplex.giss.nasa.gov/gcm/ROCKE-3D/. The Planetary Spectrum Generator is available online at https://psg.gsfc.nasa.gov/. The NCAR Command Language (Brown et al. 2012) and CET Perceptually Uniform Color Maps (Kovesi 2015) were used in post-processing.