Shedding Light on Desert Dwellers

Tim Hallatt, Sarah Millholland · 2025

Read Shedding Light on Desert Dwellers by Tim Hallatt, Sarah Millholland (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-22893.txt.

Introduction

A sparse zone of close-in planets, called the sub-Jovian desert, has puzzled planet hunters. Tim Hallatt and Sarah Millholland test whether wrecked hot Jupiters can fill that zone. Desert planets are dense and fairly heavy, and their stars look like hot Jupiter hosts. The pair first show the host stars seem older, as if they descend from hot Jupiter hosts.

Tides can peel a giant planet down to a dense core, but only if the flow wastes angular momentum. That lossy flow can stock the whole desert width. It also predicts slow orbital decay for one known desert planet, plus faster star spin. A brief glow of extra infrared light may mark such events.

Catalog record

CREATORTim Hallatt, Sarah Millholland
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE94 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-22893

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Shedding Light on Desert Dwellers Tim Hallatt, Sarah Millholland Abstract The “sub-Jovian desert” ( 2 ≲ R p ≲ 10 ​ R ⊕ 2{\lesssim}R_{\rm p}{\lesssim}10\ R_{\oplus} , periods ≲ \lesssim 3 days) is sparsely populated but no longer empty. Recent surveys have revealed that planets residing in the desert are dense ( ρ ≳ 1 {\rho}{\gtrsim}1 g/cm 3 ), massive ( ∼ \sim 10 − {-} 50 M ⊕ M_{\oplus} ), and orbit metal-rich stars that are indistinguishable from those hosting hot Jupiters. However, their origins remain mysterious. In this work we adopt and test the hypothesis that tidal destruction of hot Jupiters can populate the sub-Jovian desert with stripped remnant planets. We first show that stars hosting desert dwellers exhibit Galactic kinematics indicative of an older population descended from those hosting hot Jupiters. We highlight that tidally-driven Roche lobe overflow (RLO) can indeed populate the desert with planets similar to those observed, but only if angular momentum transfer during RLO is inefficient (“lossy” RLO). The entire width of the sub-Jovian desert can be backfilled with the remnants of hot Jupiters that possessed their empirically inferred spread in entropy. In this picture, current desert dwellers such as LTT 9779b should be tidally decaying at an observationally testable rate of ∼ 0.5 {\sim}0.5 ms/yr. Our theory also predicts that desert dweller host stars may rotate up to an order of magnitude more rapidly than field stars; rotation period differences may persist ∼ {\sim} Gyr after RLO. Lossy RLO may also manifest as a burst of IR excess that could outshine the host star for up to ∼ 10 3 {\sim}10^{3} yr. If these predictions are confirmed by observations, our theory indicates that desert dwellers can be leveraged to study the interiors of giant planets in exquisite detail. I Introduction Exoplanets exhibit a remarkable diversity of sizes, spanning gas-poor super-Earths ( 1{-}1.7\ R_{\oplus} ), and sub-Neptunes ( 1.7{-}4\ R_{\oplus} ), to gas-rich sub-Saturns ( 4{-}8\ R_{\oplus} ) and giant planets ( {\gtrsim}10\ R_{\oplus} ) ( Howard et al. 2010 ; Howard et al. 2012 ; Fressin et al. 2013 ; Petigura et al. 2013 ; Petigura et al. 2018 , e.g.) . Inside orbital periods {\lesssim} 3 days however, the planet population bifurcates: planets with sizes intermediate between super-Earths and gas giants become exceedingly rare. Early analysis of Kepler transiting planet data found this dearth of hot, intermediate-sized planets to be so pronounced so as to dub it the “sub-Jupiter desert” ( Szabó & Kiss 2011 ) . Follow-up work confirmed that this sub-Jovian desert was indeed remarkably empty, especially in light of the favorable conditions to observe such planets in the desert if they existed ( Beaugé & Nesvorný 2013 ; Mazeh et al. 2016 ; Lundkvist et al. 2016 , e.g.) . Thanks to the TESS mission however ( Ricker et al. 2015 ) , several “desert dweller” planets have recently been discovered that lie squarely in the regime that was previously thought uninhabited ( Vissapragada & Behmard 2025 ; Doyle et al. 2025 , e.g.) . The desert is no longer empty. Initial characterization of the desert delineated a triangular region in mass/radius versus orbital period, opening inward from a locus at {\sim} 5 {-} 10 days; these boundaries separated gas giants above from super-Earths below by a region virtually devoid of planets ( Mazeh et al. 2016 ) . Theoretical work suggests that the desert’s bottom boundary can be carved out by photoevaporative mass loss catalyzed by stellar high energy emission ( Owen & Jackson 2012 , e.g.) , as low/intermediate mass planets readily lose their atmospheres at such tight orbital distances to become denuded rocky cores comprising the lower desert edge ( Owen & Lai 2018 ; Hallatt & Lee 2022 ) . The desert’s upper boundary however is unlikely to be sculpted by evaporation as massive planets resist significant mass loss ( Owen & Lai 2018 ; Ionov et al. 2018 ; Vissapragada et al. 2022 , though see Kurokawa & Nakamoto 2014 for a differing view, and Thorngren et al. 2023 for consideration of exceptionally low density Saturns which may evaporate in the presence of extra internal heating) . Instead, Owen & Lai 2018 point to high eccentricity migration as a means to sculpt the desert’s upper edge; in this picture, the boundary reflects the tidal disruption limit where planets may safely park after their orbits circularize. Alternatively, both upper and lower desert boundaries may be set by the tidal disruption barrier which scales differently with planet mass for large and small planets, respectively ( Matsakos & Königl 2016 ) . In-situ formation of hot Jupiters could also help to create the desert’s upper edge ( Bailey & Batygin 2018 ) . Recent work has brought the desert and its denizens into sharper focus. Bias-correcting the desert boundaries places the outer locus at {\sim} 3.2 days ( Castro-González et al. 2024 , roughly corresponding to an irradiation flux {\sim}550{\times} that of the Earth; Magliano et al. 2024 ) and suggests differing formation pathways between bona fide desert dwellers ( {<}3.2 days) and those in the neighboring regime beyond the desert edge ( Bourrier et al. 2023 , the “ridge” and “savannah”;) . In particular, Vissapragada & Behmard 2025 showed that desert dweller host stars are preferentially metal-rich and have a metallicity distribution that is indistinguishable from that of hot Jupiter host stars (see also Dong et al. 2018 , who first pointed out a preference for metal-rich host stars among hot Neptunes) . Desert dwellers might also share the same elevated stellar multiplicity rates as hot Jupiters, unlike longer period planets ( Eeles-Nolle & Armstrong 2025 ) . Vissapragada & Behmard 2025 highlight that shared host stellar properties between desert dwellers and hot Jupiters is consistent with a “top-down” picture in which desert dwellers are the descendants of hot Jupiters (albeit with a possible difference in host star ages; see Section II ). In this picture, hot Jupiters could be destroyed either during their high eccentricity migration or over longer timescales due to tidal inspiral. Doyle et al. 2025 furthermore find that desert dwellers boast high densities consistent with the absence of gaseous atmospheres, in stark contrast with planets beyond the desert edge that frequently host sizable gas envelopes. Massive, dense planets devoid of gas are also suggestive of the exposed cores of giant planets ( Armstrong et al. 2020 ) . In this paper, we explore whether tidal destruction of hot Jupiters can populate the sub-Jovian desert with gas-poor remnant planets similar to those observed. We focus on the tidal inspiral channel, rather than destruction during high eccentricity migration. The tidal inspiral picture is supported by at least three empirical results. First, it is empirically known that hot Jupiters orbit stars that are kinematically young, strong evidence that they do indeed decay to destruction while their host stars are on the main sequence ( Hamer & Schlaufman 2019 , see also Miyazaki & Masuda 2023 ) . Second, Sun-like stars hosting hot Jupiters are observed to spin more rapidly than those with more distant giant planets or lower mass planets ( Tejada Arevalo et al. 2021 ) , as expected under tidal orbital decay.

Read the complete text