Low Radio Frequency Observations from the Moon Enabled by NASA Landed Payload Missions

On the lunar dayside, photoelectrons are quasi-constantly emitted from the Moon's surface, and this electron flux acts to typically charge the dayside lunar surface a few volts positive. In arriving at an equilibrium surface potential, the surface will charge in order to balance the two primary currents: the outgoing photoelectron flux, J_p , against the incoming solar wind electron thermal flux, J_e . In nominal solar wind conditions, J_p > J_e and the surface charges are positive, trapping most of the photoelectrons (Farrell et al. 2013). The photoelectrons are related to surface charging, as well as the charging of exploration vehicles on the lunar surface. They also play a role in the mobility of lunar dust particles, which consequently varies based on the solar wind conditions. Therefore, the best possible understanding of the photoelectron sheath is important for future exploration of the lunar surface. There are several simulation code results published for the lunar photoelectron sheath. Figure 3 shows two plots from such simulations showing the significant increase in the electron density inside a height of 10 m. Note also that these two plots, which are simulations for the same interval (1998 May 1–3), do not have exactly the same results, which ROLSES is designed to resolve.

Zoom In Zoom Out Reset image size

The sheath will impact radio observations on the order of 100 kHz and attenuate radio waves at lower frequencies. Of course, variations in the solar wind have a significant effect on the photoelectron sheath, as shown in Figure 3. Assuming that these models are correct, then the good news is that an astrophysics-oriented radio observatory could make daytime observations down to less than 1 MHz without any impact from the electron sheath. On the other hand, any observatory making lower-frequency observations, such as a heliophysics solar radio burst imaging observatory, would likely have refraction and attenuation issues for observations at less than 100 kHz due to the photoelectron sheath. To better understand this situation, ROLSES will make measurements of the photoelectron sheath density scale height and determine the frequency at which the incoming radio waves will be attenuated/cutoff. ROLSES measures these values based on a thermal noise spectrum, possible wave activity at the electron plasma frequency (fpe), and absorption of the radio spectra of remote radio sources at frequencies at and below fpe. Figure 4 shows examples of the incoming radio waves that ROLSES will be detecting.

Zoom In Zoom Out Reset image size

Spacecraft in the interplanetary environment or orbiting planets may be struck by dust particles, which releases electrons and ions from the surface, affects the surface photoelectron environment, and creates signals that are detectable. ROLSES might detect dust impacting the NOVA-C lander in a similar way. The time resolution of 4 s does not permit detection of individual dust particles, but could detect dust "clouds," like those in Figure 5.

Zoom In Zoom Out Reset image size

There is no ground plane below the ROLSES antennas, so some detected radio waves will likely penetrate the lunar surface. They may be reflected at some depth, and ROLSES may be able to detect such reflection, assuming its duration is at least 4 s, providing some information about the structure below the lander. Previously, the Apollo 17 lunar Surface Electrical Properties instrument made such measurements at six frequencies, with a signal generator that sent radio waves down to a few kilometers into the Moon. Grimm (2018) described a recent analysis of the Apollo 17 results in detail, e.g., "Because no deep interfaces were detected, the thickness of the Taurus-Littrow volcanic fill must exceed 1.6 km and possibly 3 km."

ROLSES will provide continuous spectra of RFI from terrestrial transmitters (Figure 6) for the ∼14 day mission; information to confirm how well a nearside lunar surface-based radio observatory could observe and image solar radio bursts in the frequency range of 0.01 to ∼10 MHz for the first time. Its deployed STACER lengths are 2.5 m. Its frequency range of 10 kHz–30 MHz is divided into two bands: 10 kHz–1 MHz and 300 kHz–30 MHz. The LF band resolution is ∼1.76 kHz, and the high-frequency band resolution is ∼58.01 kHz. The low-band sensitivity is expected to be ∼3 μV Hz⁻¹, which will be further refined after the lander and system noise are fully defined.

Zoom In Zoom Out Reset image size

Like all lunar surface payloads, the ROLSES electronics need to survive the maximum lunar surface temperatures. To deal with this, each of the ROLSES components has a radiator plate attached to it to keep temperatures below ∼65°C. This will permit it to function for the complete 14 day NOVA-C mission. NOVA-C receives the data from ROLSES and sends it to the ground stations. Currently the NOVA-C mission is targeting landing on the Moon in Q4 of 2021. Intuitive Machines has chosen the landing site to be in Oceanus Procellarum near Vallis Schröteri (Schroter's Valley), in part because there should be many smooth-surface areas there. The location relative to the Earth–Moon line is 25°N, 50°W. The location is reasonable for the ROLSES payload, but not unique.

While the Schrödinger Basin (see Figure 1) is located on the lunar farside, it will not be completely radio-quiet near the low end of the LuSEE frequency band. In fact, at 100 kHz, the level of suppression of radio interference will vary significantly across the basin. As shown in Figure 7, the level of attenuation varies from a factor of approximately 25 dB at the southwest corner of the basin to near 55 dB at the northeast corner. Based on these levels of suppression, it is likely that RFI and AKR will be present at a sufficient level to be detected by LuSEE. Specifically, the AKR from Earth's dayside will be observed to be comparable to the flux density of the Galaxy at 100 kHz (∼10⁵ Jy), whereas the AKR from Earth's nightside will be ∼40 dB brighter than the Galaxy in the SW quadrant of Schrödinger (see Figure 2 of Zarka et al. 2012).

Zoom In Zoom Out Reset image size

However, the presence of radio interference in the Schrödinger Basin provides a method with which to test predictions of the radio environment. In addition to its primary science goals, LuSEE will provide a ground-truth measurement with which to compare the results of the numerical simulations of Bassett et al. (2020). Further work is underway to extend the results of Bassett et al. (2020) to include a tenuous lunar ionosphere and electron sheath, which will likely influence the propagation of waves at frequencies as low as 100 kHz. These simulations, in conjunction with LuSEE measurements, will provide the most comprehensive study of the radio environment of the farside, and specifically the Schrödinger Basin, to date.

In contrast to the radio interference at lower frequencies, LuSEE high-band measurements at frequencies above 10 MHz will be shielded from terrestrial (artificial) shortwave emissions, which are attenuated by a factor of at least 90 dB over the entire basin. This lack of contamination makes the Schrödinger Basin an excellent landing site for LuSEE to make precision observations of the radio sky, as detailed in the following section.

The science targets of the LuSEE suite include lunar surface plasma physics, the science of dust transport near the surface, and to serve as an LF radio pathfinder mission. The lunar surface plasma environment is relatively poorly understood, with a true paucity of measurements. However, any surface-landed asset (human or robotic) must function in the surface plasma environment. The surface electric potential that will control the transport of (charged) dust should be highly variable, depending on the upstream plasma parameters, solar UV irradiance, solar zenith angle, and local shadowing (Reasoner & Burke 1972) with predictions in the range of a few volts to ∼100 V (Poppe et al. 2012; Stubbs et al. 2014). LuSEE will make true measurements of the DC electric potential using a current-biased double-probe technique, which is appropriate for electric field measurements in low-density plasmas (e.g., Mozer et al. 1983). The Moon is thought to sustain a tenuous "exo-ionosphere" supported by photoionization, electron impact ionization, and charge exchange. Radio occultation measurements have suggested rather large plasma densities 500–1000 cm⁻³ with scale heights of tens of kilometers, while in situ measurements suggest smaller surface plasma density. The THEMIS-ARTEMIS mission measured electron plasma waves, which suggested an ionospheric density of ∼0.1–0.3 cm⁻³ (Halekas et al. 2018). LuSEE is designed to make sensitive measurements of electron plasma waves/noise using a technique called "quasi-thermal noise spectroscopy" (e.g., Meyer-Vernet & Perche 1989) and can measure local electron density and its variability, and in some cases the electron temperature, on the lunar surface. At higher frequencies, up to ∼20 MHz, LuSEE will make sensitive radio frequency measurements of interplanetary radio bursts, Jovian and terrestrial natural radio emission, and the galactic synchrotron spectrum. LuSEE will provide best-yet spectral measurements of the quiet radio sky from the lander surface.

The LuSEE experiment is derived from flight spare and engineering model hardware from the NASA Parker Solar Probe "FIELDS" instrument (hereafter PSP/FIELDS; Bale et al. 2016), the NASA STEREO WAVES (S/WAVES; Bougeret et al. 2008; Bale et al. 2008), the NASA Van Allen Probes Electric Field and Waves (EFW) experiment (Wygant et al. 2013), and the ESA Solar Orbiter Radio and Plasma Waves (RPW) suite (Maksimovic et al. 2020).

The LuSEE system consists of a "Main Electronics Package," which is comprised of three digital signal processing engines, a low-voltage power converter, and data processing units with interfaces to the spacecraft command and data-handling system. These units are all flight spares, or nearly build-to-print, from the PSP/FIELDS instrument on-orbit and have demonstrated excellent scientific capability. The Radio Frequency Spectrometer (RFS) subsystem is a dual channel, baseband digital radio receiver that operates in two frequency regimes (10 kHz–2 MHz and 1.3–19.2 MHz), and is described by Pulupa et al. (2017). The RFS uses polyphase filter techniques to reject out-of-band noise sources and is highly configurable. Early results from the RFS on PSP (Pulupa et al. 2020) show superb sensitivity and capability, resolving the galactic spectrum down to a few hundred kilohertz (and a few nanovolts). The PSP/FIELDS RFS is arguably the most capable LF radio instrument ever put into orbit, and LuSEE inherits that capability. The LuSEE digital fields board (DFB) measures DC- and AC-coupled voltage inputs from the antennas and search coil magnetometer (Malaspina et al. 2016) and produces time-series measurements to 150 k-samples/s and spectral and cross-spectral matrices to 75 kHz. DFB is highly configurable and uses a triggered "burst mode" to capture impulsive signals. The LuSEE preamp-DFB system has capabilities to measure DC voltage perturbations to +/- 100 V, producing direct measurements of the lunar surface potential. The LuSEE time domain sampler (TDS) is a snapshot waveform sampler that measures five analog channels (and one digital) at speeds up to 2 M-samples/s and some accumulated statistics associated with these waveforms (Bale et al. 2016). The TDS on PSP/FIELDS has produced spectacular results on the distribution of interplanetary dust (Page et al. 2020; Malaspina et al. 2020; Szalay et al. 2020) and the physics of spacecraft/dust interactions (Bale et al. 2020). A central processing unit, the data controller board, and low-noise power converter round out the complement in the electronics package.

LuSEE will use three voltage probes and two boom-mounted magnetometers as sensors. Figure 8 shows the LuSEE sensors mounted on a strawman concept lander. The V1 and V2 antennas are flight spare units from the STEREO/WAVES instrument (Bougeret et al. 2008). The V1/V2 antennas are 6 m long, ∼1'' diameter BeCu STACER antennas (Bale et al. 2008); STACERs are cold-rolled metal springs that deploy under their own spring force (no motor) and have been used on hundreds of space projects, with no known failures. The V1/V2 antennas will be deployed out away from the spacecraft and at an angle to the horizontal so that they sag in lunar gravity to form a dipole, which has been modeled mechanically. The V1/V2 dipole will be the primary measurement for the RFS/radio frequencies. The V3 antenna is a flight spare unit from the EFW experiment on the Van Allen Probes mission (Wygant et al. 2013) and also uses STACER technology. However, V3 holds a small spherical voltage probe that is current-biased and will be used to measure the DC voltage between V3 (at 6+m above the surface) and the V1/V2 pair, yielding a measurement of the vertical electric field. Magnetic field measurements on LuSEE will be made from a three-axis fluxgate magnetometer (MAG), derived from PSP and MAVEN heritage and a three-axis search coil magnetometer (SCM) flight spare from the Radio and Plasma Waves instrument (Maksimovic et al. 2020) on the ESA Solar Orbiter mission. The MAG sensor measures magnetic fields with <1 nT resolution at cadences to 293 Samples/s, while the SCM has no DC sensitivity, but otherwise an excellent response to tens of kilohertz.

Zoom In Zoom Out Reset image size

Beginning with the distinctive radio-quiet characteristics of the farside described in Section 2 as a requirement, an astronomical array at frequencies 0.1–40 MHz is driven by sensitivity and spatial resolution. The sensitivity is determined by the number and type of antennas used in the array, and the resolution is governed by the maximum baseline. For FARSIDE, we use 100 m tip-to-tip thin-wire antennas. It was determined that 128x2 dipole antennas will provide the needed sensitivity to span the science cases described in Section 4.2, especially probing the space plasma weather and magnetospheres in exoplanetary systems. With an array diameter of 10 km, FARSIDE achieves a resolution of 10° FWHM at 200 kHz and 10' at 15 MHz.

Sensitivity specifications for FARSIDE are summarized in Table 1. The array sensitivity, or minimum detectable flux density, in Table 1 depends upon the frequency bandwidth, the integration time, the system temperature, and the effective area of the array. The effective area is determined by the dipole length and the antenna impedance. This was modeled using NEC4.2 simulations that take into account that the dipole wires rest directly on the regolith. The system temperature depends upon the sky and regolith temperatures, and the front-end amplifier. See Burns et al. (2019a) for complete details of the modeling and simulations.

The FARSIDE instrument front end uses the electrically short antennas to achieve sky background noise-limited observations. On the lunar farside, the lunar highlands are thick, have low conductivity, and vary slowly with depth, thus removing the need for a ground plane. The net impact of this multipath effect is to introduce a direction dependent component to the array synthesized beam. FARSIDE will use an orbital beacon for calibration and to map both the antenna beam pattern and the array synthesized beam to fully capture this effect. The 100 m antennas are embedded within the tether and will be placed directly on the lunar surface. The array leverages existing designs from high heritage front-end amplifiers, fiber optics, and the correlator system based upon ground-based observatories such as the Owens Valley Radio Observatory Long Wavelength Array (Anderson et al. 2018).

The concept design calls for operations during the lunar night as well as the lunar day using proven low-temperature electronics and batteries. Recent tests indicate that core constituents such as resistors and capacitors, forming the basis of the electronic components of the FARSIDE receiver nodes, are capable and survivable down to temperatures <−180 C (Ashtijou et al. 2020).

The 128 antenna-node pairs are distributed along four spirals, as shown in the Figure 9 insert. This configuration with four JPL Axle rovers (Figure 10) would allow for the full deployment of the array in one lunar day. A tether connects the nodes to each other and to the central base station for data transmission and power. For each spiral arm, the rover would carry a set of antenna nodes from the base station, unspooling the tether, before returning to the base along a different path. We assume that the landing site and rover path are relatively flat, free of obstacles with scale sizes ≳1 m.

Zoom In Zoom Out Reset image size

The base station includes solar panels and batteries to provide power for the receivers and the signal processing using an FX correlator (e.g., Kocz et al. 2014), as well as communication of the data back to Earth, possibly via NASA's lunar Gateway. The rovers draw power along the tether from the base station. The rover will either be operated from Earth (with ≈ 5 s latency) or teleoperated via astronauts aboard the lunar Gateway (with ≈ 0.5 s latency; Burns et al. 2019b). After cross-correlation via the FX correlator, visibility data will be conveyed to Earth for further analysis. This is estimated to be about 130 GB of data per 24 hr period. The goal is to process 10,000 square degrees of the sky every 60 s over 1400 frequency channels. This would then produce dynamic spectra at the location of each bursting object including Type II/III solar emissions and bursts in any of the several thousand nearby exoplanetary system counterparts as well as CME-driven auroral emissions from nearby terrestrial exoplanets.

Recently, Hegedus et al. (2020) utilized a combination of the Common Astronomy Software Applications package (McMullin et al. 2007), lunar surface maps from Lunar Reconnaissance Orbiter (LRO) Lunar Orbiter Laser Altimeter (LOLA; Barker et al. 2016), and Spice (Acton 1996) to define a simulated array on the Moon's surface. This approach properly aligns the reference frames from the lunar surface to the sky for the purpose of observing LF emissions. Here, the software pipeline has been adapted for the simulation of FARSIDE observations. The 10 km, four-armed spiral pattern of FARSIDE on top of an elevation map from LOLA data is shown in panel (A) of Figure 11. The array consists of 256 antennas, 128 pairs of X and Y polarizations that lie nearly side by side, each of which are 100 m in length. The instantaneous u-v coverage (projected baselines) of the array for an observation at its zenith is shown panel (B) of Figure 11.

Zoom In Zoom Out Reset image size

One of the scientific targets of FARSIDE is imaging solar Type II and III radio bursts. The simulation pipeline for FARSIDE has been applied to this problem, using a model-inspired Type II burst as input to the array in order to assess its reconstructive capabilities. This model Type II burst is shown in panel (C) of Figure 11. This model emission was obtained by taking a subset of data from an MHD recreation of the radio-loud, Earth-directed CME observed on 2005 May 13 (Manchester et al. 2014). A subset of the data was identified via applying parameter-based thresholds on the simulation data for each time step in the simulation. The first threshold used for these data was that the regions included must have an increase in local entropy of at least a factor of four compared to the pre-shocked data. This entropy threshold generally captured the location of the shock within the simulation. The other threshold included regions with a local de Hoffmann–Teller speed of at least 2250 km s⁻¹. This is the speed in the frame where the upstream plasma bulk velocity and magnetic field are parallel, and therefore there is no upstream convective (V × B) electric field. This de Hoffmann–Teller speed was identified by Pulupa et al. (2010) to be most highly correlated with in situ observations of Langmuir waves from accelerated electrons at 1 au coinciding with shocks, agreeing well with the predictions of fast Fermi acceleration theory.

Panel (C) shows the projected sky distribution after applying the above parameter-based thresholds, illustrating only the regions that have upstream plasma frequencies 6–6.1 MHz. This map is then slightly blurred by applying a Gaussian kernel with an FWHM of 007. The resulting "truth image" is then normalized to a brightness of 10⁶ Jy and fed into the simulated FARSIDE array. The simulated response of the FARSIDE antenna on the lunar surface assumes A_eff/T_sys = 0.065 at 6 MHz, consistent with the array parameters in Table 1. Appropriate thermal noise was added to create a more realistic response of FARSIDE to the model Type II emission. These data were then imaged and "cleaned" (i.e., deconvolved) using the widefield imaging software WSClean (Offringa et al. 2014), with the resulting image shown in panel (D) of Figure 11. This simulation pipeline confirms that FARSIDE would provide unprecedented imaging capabilities for LF solar radio bursts, and could help explain how energetic particles are being accelerated at CME-driven shocks.

During the solar illumination of the lunar day, FARSIDE will observe solar radio bursts, which are one of the primary remote signatures of electron acceleration in the inner heliosphere. FARSIDE extends the spectroscopic observations of LuSEE and ROLSES by producing the first radio images of Type II and Type III radio bursts at ≳ 2R_⊙ (Figure 11). Type II bursts originate from suprathermal electrons (E > 100 eV) produced at shocks. These shocks are generally produced by CMEs as they expand into the heliosphere with Mach numbers >1. Emission from a Type II burst drops slowly in frequency as the shock moves away from the Sun into lower-density regions at speeds of 400–2000 km s⁻¹. Type III bursts are generated by fast (2–20 keV) electrons from magnetic reconnection, typically due to solar flares. As the fast electrons escape at a significant fraction of the speed of light into the heliosphere along open magnetic field lines, they produce emission that drops rapidly in frequency. As discussed in Section 4.2.3, FARSIDE will also attempt to detect intense type II and type III emissions from other stars in the solar neighborhood.

Acceleration at shocks. Observations of CMEs near Earth suggest that electron acceleration generally occurs where the shock normal is perpendicular to the magnetic field (Bale et al. 1999), similar to acceleration at planetary bow shocks and other astrophysical sites. This geometry may be unusual in the corona, where the magnetic field is largely radial. There, the shock at the front of a CME generally has a quasi-parallel geometry. Acceleration along the flanks of the CME, where the magnetic field-shock normal is quasi-perpendicular, would seem to be a more likely location for the electron acceleration and Type II emission. FARSIDE has sufficient resolution (Figure 11) to localize these acceleration site(s), yielding the preferred geometry of the shock normal relative to the magnetic field direction for radio emission around CMEs.

Electron and ion acceleration. Observations at 1–14 MHz made with the Wind spacecraft showed that complex Type III-L bursts are highly correlated with CMEs and intense (proton) solar energetic particle events observed at 1 au (Cane et al. 2002; MacDowall et al. 2003). While the association between Type III-L bursts, proton solar energetic particle events, and CMEs is now secure, the electron acceleration mechanism remains poorly understood. Two competing sites for the acceleration have been suggested: at shocks in front of the CME or in reconnection regions behind the CME. For typical limb CMEs, the angular separation of the leading edge of the shock and the hypothesized reconnection region behind the CME is approximately 15 when the CME shock is 3–4 R_⊙ from the Sun.

CME interactions and solar energetic particle intensity. Unusually intense radio emission can occur when successive CMEs leave the Sun within 24 hr, as if CME interaction produces enhanced particle acceleration (Gopalswamy et al. 2001, 2002). Statistically associated with intense solar energetic particle events (Gopalswamy et al. 2004), this enhanced emission could result from more efficient acceleration due to changes in field topology, enhanced turbulence, or direct interaction of the CMEs. Lack of radio imaging makes it difficult to determine the nature of the interaction. Images with ∼2° resolution would give Type II locations and permit identification of the causal mechanism and the relation to intense solar energetic particles.

FARSIDE provides the requisite spatial resolution, as well as exquisite signal-to-noise and imaging capability (due to a large number of unique baselines) spanning the 1–14 MHz band observed by the Wind spacecraft and extending much wider at both lower and higher frequencies. It will be a uniquely powerful tool for understanding the complex interplay of CMEs, solar energetic particles, and associated shocks and particle acceleration.

In 1955, the planet Jupiter was serendipitously detected to be a source of bright, highly variable radio emission at decametric wavelengths (Burke & Franklin 1955). This radio emission was manifested as intense, highly polarized bursts, so bright as to often outshine the Sun at these low frequencies (<40 MHz). Bursts were only detectable over specific ranges of rotational phase of the planet, with certain flavors also found to be strongly modulated by the orbit of the volcanic moon Io (Bigg 1964). This discovery revolutionized our understanding of planetary magnetospheric physics, providing the first direct confirmation of the presence, strength, and extent of the Jovian magnetosphere. In particular, the radio bursts, generally attributed to electron cyclotron maser emission, were found to be produced at the electron cyclotron frequency, ν_ce ≈ 2.8B MHz, where B is the magnetic field strength (in Gauss) at the source of the burst, and thus enabled direct determination of magnetic field strength.

All of the magnetized planets in our solar system, including Earth, have since been found to produce similar bright coherent radio emission at low frequencies, predominantly originating in high magnetic latitudes and powered by auroral processes. It is notable that only Jupiter, with a maximum polar magnetic field strength of 14 G, produces radio emission that can penetrate through the Earth's ionosphere. All other planets (and Ganymede) have magnetic fields of <2 G, requiring a space-based observatory for detection. For example, the magnetospheric radio emissions of Saturn, Neptune, and Uranus were initially detected in situ by Voyager during fly-bys (Zarka 1998).

These auroral processes are driven by: (a) magnetic reconnection between the planetary magnetic field and the magnetic field carried by the solar wind (e.g., Earth, Saturn, Neptune, and Uranus), (b) the departure from co-rotation with a plasma sheet residing in the planetary magnetosphere (e.g., Jovian main auroral oval), or (c) interaction between the planetary magnetic field and orbiting moons (e.g., Jupiter-Io current system; Zarka 1998 and references therein). As well as providing diagnostic information on the presence, strength, and extent of planetary magnetospheres, the detected radio bursts are the only means to accurately determine the rotation period of the planetary interior for the gas giants.

FARSIDE will be the first space-based radio telescope with sufficient sensitivity for the detection of radio emission from all of the solar system's magnetized planets from cis-lunar space. Monitoring the impact of the variable solar wind on the auroral radio flux density at the planets, including during interplanetary shocks from CMEs, is one application. Long-term monitoring of the Kronian radio emission period, modulated by the solar wind, will refine measurement of the true planetary rotation period (Zarka et al. 2007). The magnetospheric radio emissions from Neptune and Uranus have not been observed at radio wavelengths since Voyager.

In addition, Saturn and Uranus were found by Voyager to produce radio emission associated with atmospheric lightning (Zarka et al. 2004). This radio emission can potentially inform us about atmospheric dynamics and composition, when compared to optical imaging of cloud data (Zarka et al. 2012). The bursts are expected to have duration of 30–300 ms according to Zarka et al. (2004). The rms noise is 100 mJy in 60 s at 200 kHz. As one example, 300 ms bursts will be attenuated by a factor of 200x in a 60 s image, and thus the 1σ sensitivity to such bursts is 20 Jy. This is sufficient for frequent detection of the spectral energy distribution and detection of the brightest electrostatic discharge bursts.

Finally, FARSIDE would offer the unique possibility of searching for radio emission from large bodies beyond Neptune out to hundreds of astronomical units. This includes, but is not limited to, the putative Planet 9 (e.g., Batygin & Brown 2016). The expected flux density of Planet 9 is estimated by scaling from outer solar system planets assuming the radio emission is proportional to Poynting flux on the magnetospheric cross section. A semimajor axis of 600 au and radius of 3 R_⊕ is assumed (Batygin et al. 2019). Planet 9 can be interacting with the local interstellar medium (ISM) or the heliotail. For the heliotail, the B field and velocity are extrapolated from the inner solar wind. For the ISM, it is assumed that the magnetic field and density are consistent with measurements of the local ISM from Voyager 1 and 2 (Izmodenov & Alexashov 2020) and that the velocity of Planet 9 relative to the ISM is dominated by the motion of the Sun through the local ISM (∼25 km s⁻¹). A magnetic to radio efficiency of 2 × 10⁻³ is assumed in both cases (Zarka et al. 2018). A flux density of 10 mJy is estimated for the heliotail scenario and 3 Jy for the ISM interaction, if a significant planetary magnetosphere is present, although the uncertainty is very large.

As described above, the radio emission from the Sun and planets provides key insights into the bulk motion of plasma in the solar corona and interplanetary medium and a means for direct measurement of planetary magnetospheres. Extending to other stellar and planetary systems in the solar neighborhood will provide essential data on the impact of energetic particles and CMEs on planetary atmospheres and habitability, together with the role of planetary magnetic fields as a buffer against such activity. The FARSIDE array, in particular, will be uniquely capable as an instrument for this science due to its ability to extend by a factor of a hundred times lower in frequency with equivalent or better sensitivity than current or planned ground-based arrays, enabling the routine detection of stellar CMEs, solar energetic particle events, and the magnetospheres of extrasolar terrestrial planets for the first time (Figure 12).

Zoom In Zoom Out Reset image size

Detection of stellar CMEs and solar energetic particle s. As discussed in Section 4.2.1, solar CMEs and solar energetic particle events can be accompanied by radio bursts at low frequencies, particularly Type II bursts, as well as a subset of Type III bursts (complex Type III-L bursts). The emission is produced at the fundamental and first harmonic of the plasma frequency and provides a diagnostic of the density and velocity (a few 100 to >1000 km s⁻¹) near the shock front, while the flux density of the burst depends sensitively on the properties of the shock and solar wind (Cairns et al. 2003). Ground-based radio astronomy can only trace such events to a heliocentric distance of a few solar radii, whereas space-based radio antennas can trace the propagation of shocks out to the Earth and beyond, which is particularly relevant for characterizing geoeffective CMEs and solar energetic particle events. The detection of equivalent interplanetary Type II and III events from stars other than the Sun is one of the goals of the FARSIDE array.

FARSIDE would detect the equivalent of the brightest Type II and Type III bursts out to 10 pc at frequencies below a few megahertz. By imaging >10,000 deg² every 60 s, it would monitor a sample of solar-type stars simultaneously, searching for large CMEs and associated solar energetic particle events. For the αCen system, with two solar-type stars and a late M dwarf, it would probe down to the equivalent of 10⁻¹⁵ W m⁻² Hz⁻¹ at 1 au, a luminosity at which solar radio bursts are frequently detected at the lowest frequencies available to FARSIDE (Krupar & Szabo 2018). The nearby young active solar-type star, εEridani (spectral type K2), is another priority target. For the case of M dwarfs, FARSIDE would be able to detect Type II bursts formed at the distance where super-Alfvénic shocks should be possible for M dwarfs (Villadsen & Hallinan 2019) and directly investigate whether the observed relationship observed between solar flares and CMEs (Aarnio et al. 2011; Osten & Wolk 2015) extends to M dwarfs. If it does, FARSIDE should also detect a very high rate of radio bursts from this population.

Detection of the terrestrial planet magnetospheres. All of the magnetized planets in our solar system, including Earth, produce bright coherent radio emission at low frequencies, predominantly originating in high magnetic latitudes and powered by auroral processes (Section 4.2.2; Zarka 1998). Detection of similar radio emission from candidate habitable planets is the only known method to directly detect the presence and strength of their magnetospheres. The temporal variability of this radio emission can also be used to determine the rotation periods and orbital inclinations of these planets. As described in Section 4.2.2, the radio emission is emitted at the electron cyclotron frequency (Section 4.2.2).

Extending to the exoplanet domain requires a very large collecting area at low frequencies, with the first detections likely to be ground-based. Indeed, in a recent breakthrough, radio emission has been detected from a possible free-floating planetary-mass object (12.7 ± 1 M_jup; Kao et al. 2018). This is the first detection of its kind and confirmed a magnetic field much higher than expected, >200 times stronger than Jupiter's, reinforcing the need for empirical data. However, the detection of the magnetic fields of candidate habitable planets will almost certainly require a space-based array if the magnetic fields are within an order of magnitude in strength of Earth's magnetic field. Detection of the magnetic field of planets orbiting in the habitable zone of M dwarfs is particularly key, as such planets may require a significantly stronger magnetosphere than Earth to sustain an atmosphere (Lammer et al. 2007).

The detection of radio emission from planets orbiting nearby stars is very sensitive to the stellar wind conditions imposed by its parent star and can serve as an indirect diagnostic of its velocity and density. It is possible to estimate the expected radio power from such planets, based on scaling laws known to apply to radio emission from the solar system planets (Zarka 2007 and references therein). These scaling relations are not only descriptive but predictive, with the luminosity of both Uranus and Neptune predicted before the Voyager 2 encounters and found to be in excellent agreement with the measurements (Desch & Kaiser 1984). Extrapolations to terrestrial exoplanets in the habitable zone of M dwarfs, and thus embedded within a dense stellar wind environment, predict radio luminosities that are orders of magnitude higher than that of the Earth (Farrell et al. 1999; Zarka et al. 2001; Grießmeier et al. 2007; Grießmeier 2015; Vidotto et al. 2019). During enhanced solar wind conditions, the Earth's radio emission can increase in luminosity by orders of magnitude (Figure 13), and a similar effect is expected for exoplanets. Therefore, as is the case for detecting Type II bursts associated with CMEs, it is essential to have the capability to monitor large numbers of planets simultaneously to detect periods where exoplanets greatly increase in luminosity.

Zoom In Zoom Out Reset image size

As noted in Section 4.1, FARSIDE will image 10,000 deg² of sky visible from the lunar farside every 60 s, in 1400 channels spanning 100 kHz–40 MHz. The sensitivity at the lower end of the band, enabled by the unique environment of the lunar farside, is particularly well suited to the search for radio exoplanets. In each FARSIDE image, there will be ∼2000 stellar/planetary systems within 25 pc. At these low frequencies, the spatial resolution is limited to a few degrees due to scattering in the interstellar medium and interplanetary medium. However, with one stellar system within 25 pc per ∼5 square degrees, identification of the source of detected emission should be straightforward, particularly as FARSIDE is most likely to detect exoplanets within 10 pc (Figure 13). An advantage at these low frequencies, is that the galaxy is optically thick at distances <100 pc, and thus long integrations are possible without classical confusion noise. Over the course of a 2 yr observing program, FARSIDE will have accumulated an average of 4000 hr of observing time on each of 8000 stellar planetary systems within 25 pc. Radio exoplanets will be identified by two means:

• 1.  
  Deep 2500 hr integrations will be used to place limits on the average radio power from radio exoplanets.
• 2.  
  Individual 2.5 hr integrations will be searched for heightened emission, expected during CME interaction.

Figure 13 highlights that FARSIDE would detect the radio emission from a population of Earth-sized, and super-Earth-sized planets orbiting nearby M dwarfs, including a number of candidate habitable planets, providing the first measurements of terrestrial planet magnetospheres outside our solar system. The radio emission can be distinguished from that of the star by rotational modulation, as well as circular polarization, which is largely absent for interplanetary solar radio bursts (Zarka et al. 2007; Hallinan et al. 2013). Detection of magnetospheres, if present, would identify the most promising targets for follow-up searches for biosignatures, as well as providing a framework for comparative analysis of exoplanet magnetospheres and atmospheric composition.

ROLSES will attempt to probe the subsurface layers beneath its lander. FARSIDE has the potential to sound the mega-regolith and its transition to bedrock expected at ∼2 km below the surface (Hartmann 1973). The Lunar Radar Sounder on board the KAGUYA (SELENE) spacecraft has provided sounding observations of the lunar highlands (Ono et al. 2010) and found potential scatterers hundreds of meters below the subsurface. However, the results are inconclusive due to surface roughness. FARSIDE, by virtue of being on the surface, would not be affected by roughness. Data from a calibration beacon in orbit could be synthesized to identify deep scatterers and the transition to bedrock at kilometer depths by virtue of the low frequencies, which are significantly more penetrating. Deep subsurface sounding can also be performed passively using Jovian bursts from 150 kHz to 20 MHz (e.g., Romero-Wolf et al. 2015). The array covers a 10 km × 10 km area on the lunar highlands, which could provide a three-dimensional image of the highland subsurface structure.