Issue |
A&A
Volume 693, January 2025
|
|
---|---|---|
Article Number | A244 | |
Number of page(s) | 13 | |
Section | Planets, planetary systems, and small bodies | |
DOI | https://doi.org/10.1051/0004-6361/202449942 | |
Published online | 22 January 2025 |
Sources of hydrogen in the primordial atmosphere of Venus
1
State Key Laboratory of Ore Deposit Geochemistry, Institute of Geochemistry, Chinese Academy of Sciences,
Guiyang
550081,
China
2
Research Center for Planetary Science, College of Earth and Planetary Sciences, Chengdu University of Technology,
Chengdu
610059,
China
3
Institute of Physics/AGP, University of Graz,
Graz,
Austria
4
CAS Center for Excellence in Comparative Planetology,
Hefei
230026,
China
5
College of Earth and Planetary Sciences, University of Chinese Academy of Sciences,
Beijing,
PR China
★ Corresponding authors; zhouyou06@cdut.edu.cn; liuyun@vip.gyig.ac.cn
Received:
11
March
2024
Accepted:
16
December
2024
Context. Understanding the hydrogen content in Venus’ primordial atmosphere is crucial for comprehending the hydrodynamic escape process that shaped its atmospheric evolution. The hydrogen originated from two main sources: molecular hydrogen (H2) from the solar nebula and water vapor (H2O) from geological degassing. The precise proportions of these sources remain uncertain, leading to different hypotheses about Venus’ atmospheric history. However, a systematic exploration of the parameter space regarding the proportions of these sources has not yet been conducted.
Aims. This study aims to constrain the hydrogen content and its sources in Venus’ primordial atmosphere by conducting extensive numerical simulations of early atmospheric escape scenarios.
Methods. We developed an improved energy-limited hydrodynamic escape model, integrated with a 1D radiative-convective equilibrium atmospheric model, to simulate the early atmospheric escape on Venus. Using isotopic data of Ne and Ar from the current Venusian atmosphere, we constrained the contributions of nebula-derived and degassing-derived hydrogen. Our simulations have explored over 500 000 scenarios, varying the initial H2 and H2O compositions and considering different solar extreme ultraviolet (EUV) irradiation conditions.
Results. Our results, based on the isotopic ratios of 20Ne/22Ne, 36Ar/38Ar, and 20Ne/36Ar observed in Venus’ atmosphere, indicate that the primordial atmospheric water content was limited to less than 0.01 ocean equivalents of H2 (0.0004 wt%) and less than 1.4 ocean equivalents of H2O. This suggests that if Venus ever had a primary hydrogen-rich atmosphere, it was mostly lost before forming its secondary, H2O-rich atmosphere. Furthermore, our method can be applied to constrain the primordial atmospheric compositions of other terrestrial planets, providing insights into their evolutionary histories.
Key words: astroparticle physics / planets and satellites: atmospheres / planets and satellites: physical evolution / planets and satellites: surfaces
© The Authors 2025
Open Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
This article is published in open access under the Subscribe to Open model. Subscribe to A&A to support open access publication.
1 Introduction
The evolution of early atmospheres on terrestrial planets is crucial to planetary science, as it directly influences the planet’s habitability and shapes its long-term atmospheric evolution. In the early stages of planetary evolution, a planet’s atmosphere would primarily consist of lighter elements such as hydrogen and helium, along with compounds originating from the solar nebula. Over time, hydrodynamic escape processes, driven by the young Sun’s intense extreme ultraviolet (EUV) radiation, would lead to the loss of these primordial atmospheres (Sekiya et al. 1980; Zahnle & Kasting 1986; Tian 2009; Lammer et al. 2014; Guo 2019; Zahnle & Kasting 2023). Hydrodynamic escape not only removes hydrogen, but also facilitates the loss of heavier atmospheric components, fundamentally altering the atmospheric composition and affecting the planet’s climate and surface conditions (Charnoz et al. 2023; Tian & Heng 2024).
Hydrogen is especially crucial in hydrodynamic escape, as it is the primary driving force of this process. Hydrogenrich species, particularly H2 and H2O, are central to the hydrodynamic escape process. The relative abundance of these species in the early atmosphere affects not only the escape rate of hydrogen but also the fractionation of other isotopes. Sekiya et al. (1980) first proposed that the early EUV-driven atmospheric escape could lead to mass fractionation of atmospheric noble gases. Research has demonstrated that isotopic ratios such as D/H,20 Ne/22 Ne, and 36Ar/38Ar can provide valuable constraints on the early escape dynamics and the initial atmospheric composition (Sekiya et al. 1980; Kasting & Pollack 1983; Gill- mann et al. 2009; Odert et al. 2018; Lammer et al. 2020; Zahnle & Kasting 2023). Hydrogen originates from nebula-derived H2, constituting the primary atmosphere, and from H2O contributing to the secondary atmosphere through magma ocean outgassing or impact degassing (collectively referred to as ’degassing’ hereafter). The photodissociation of H2O under EUV radiation breaks water molecules into hydrogen and oxygen, with hydrogen escaping into space (e.g. Guo 2019).
Among all terrestrial planets, Venus offers critical insights for understanding the role of hydrogen in atmospheric evolution, as it is believed to have experienced significant water loss due to hydrodynamic escape. This is particularly evident in the D/H ratio observed in Venus’ atmosphere (Kasting & Pollack 1983). Turbet et al. (2021) employed simulations focusing on cloud processes, challenging the hypothesis of a liquid water ocean forming on the Venusian surface. Therefore, a water-rich Venusian atmosphere might have provided enough degassing-source hydrogen for effective hydrodynamic escape. The predominant role of hydrogen in this escape process makes Venus an ideal subject for studying the dynamics of early atmospheric escape and its impact on planetary conditions.
Previous studies have extensively investigated hydrodynamic escape in Venus’ early atmospheric evolution. Gillmann et al. (2009) and earlier works (Sekiya et al. 1980; Kasting & Pollack 1983) highlighted that hydrodynamic escape, driven by the young Sun’s EUV radiation, could explain observed isotopic ratios such as 20Ne/22Ne and 36Ar/38Ar in Venus’ atmosphere, assuming H2O as the sole hydrogen source leading to significant water loss. Some studies (Gillmann et al. 2009; Lebrun et al. 2013; Zahnle & Kasting 2023) suggest that geological degassing played a significant role in supplying H2O for hydrodynamic escape. Subsequent studies by Odert et al. (2018) and Lammer et al. (2020) incorporated both nebula-derived H2 and outgassed H2O as hydrogen sources, suggesting that Venus’ secondary atmosphere could have initially contained between 0.004 and 3.5 wt% of nebula-derived H2 (equivalent to 1.3−1.1×103 oceans of hydrogen). Zahnle & Kasting (2023) modelled Venus’ atmospheric evolution assuming H2O as the exclusive hydrogen source, considering effects on residual atmospheric oxygen and surface oxidation, and constrained their model using isotopic systems including 20Ne/22Ne, 36Ar/38Ar,20 Ne/36 Ar, 15N/14N, D/H, and oxygen isotopic fractionation. They concluded that the Ne/Ar ratio in the early Venusian atmosphere, compared to solar abundance, would have been highly fractionated during the initial stages of EUV-driven hydrodynamic escape.
Different studies have employed varying EUV evolution models, affecting their conclusions. Gillmann et al. (2009) and others included CO2 in atmospheric escape models but used different EUV flux histories. Zahnle & Kasting (2023) relied on the EUV model developed by Claire et al. (2012). In contrast, Odert et al. (2018) and Lammer et al. (2020) employed EUV laws based on Tu et al. (2015), which proposed three distinct evolutionary paths for the young Sun’s EUV flux, supported by a larger dataset of exoplanet host stars. Based on this hypothesis, Odert et al. (2018) modelled a series of escape scenarios of H2-H2O hybrid atmospheres and estimated that Venus’ secondary atmosphere could have initially contained at least 0.03 wt% nebula-derived H2. Based on the atmospheric escape model in Odert et al. (2018), Lammer et al. (2020) incorporated giant impact events, which may have significantly eroded the primordial atmosphere. Their results showed that nebula-derived H2 could have contributed as much as 0.004 wt%, equivalent to about 1.3 oceans of hydrogen, during the early stages of Venus’ atmospheric evolution. In contrast, Zahnle & Kasting (2023) assumed that Venus’ secondary atmosphere contained no residual nebula-derived H2.
The amount of residual nebula-derived H2 in Venus’ secondary atmosphere is crucial for understanding its redox state and the potential for surface oxidation. However, systematic study of the precise H2-H2O composition of Venus’ secondary atmosphere has not yet been explored. The relative abundance of H2 and H2O can significantly influence the oxidation state of surface degassing products (Charnoz et al. 2023; Tian & Heng 2024). The degassing products of the magma ocean have been widely discussed (e.g. Elkins-Tanton 2008). It should be noted that in addition to magma ocean outgassing, impact degassing by volatile-rich impactors can also deliver substantial amounts of H2O into the atmosphere, as considered in earlier works (Lange & Ahrens 1982; Lange et al. 1985). Numerical models estimate that approximately 1021 kg of H2O could be released into the secondary atmosphere through the impact degassing process (Abe & Matsui 1986; Matsui &Abe 1986a,b).
In this study, we build upon the framework established by Odert et al. (2018) to further explore the possible range of H2O content in Venus’ early atmosphere during hydrodynamic escape. In addition to the isotopic ratios of 20Ne/22Ne and 36Ar/38Ar, we introduce observational data for 20Ne/36Ar as additional constraints on our model. For scenarios with little nebula-derived H2 remaining in the atmosphere, we further couple the energy-limited EUV-driven escape model of Odert et al. (2018) with a one-dimensional (1D) atmospheric model (Marcq et al. 2017). This model allows us to calculate temporal variations in the base-of-flow height as the escape progresses. We also investigate the impact of different initial escape times (t0) on the overall escape dynamics.
Our results suggest that Venus’s early atmosphere was primarily shaped by degassing-derived hydrogen (i.e., H2O). This work provides new constraints on the formation and evolution of Venus’s secondary atmosphere and its composition, offering broader insights for other terrestrial planets.
2 Method
2.1 Basic assumption and parameters
Previous studies have provided a comprehensive understanding of the dissipation and isotopic fractionation of noble gases in Earth’s early atmosphere. Sekiya et al. (1980) made a significant contribution by being the first to investigate the dissipation of the primordial H2–He atmosphere, demonstrating that noble gases such as Ne, Kr, and Xe were significantly depleted due to hydrodynamic escape driven by strong solar radiation and winds. Sekiya et al. (1981) extended this analysis with more refined modelling, offering new insights into the escape rates of heavy noble gases under strong solar influences. Sasaki & Nakazawa (1988) introduced a new perspective on the distinctive isotopic fractionation pattern of terrestrial Xe, highlighting the contrasting behaviors of Kr and Ar, which remained largely isotopically unfractionated. Pepin (1991) proposed a broader model that not only linked the isotopic distributions of noble gases to atmospheric loss processes but also incorporated planetary outgassing, thereby extending these insights to Venus and Mars.
We assumed that the early Venusian atmosphere is composed of the residual nebula gas (H2, Ne, and Ar) captured from the protoplanetary disk (i.e. the primary atmosphere), combined with the H2O–CO2 steam released by the degassing process (i.e. the secondary atmosphere). The same assumption has been applied in prior works (Odert et al. 2018; Lammer et al. 2020). We hypothesized that during the escape H2 and H2O in the upper atmosphere will undergo photodissociation due to extreme ultraviolet (EUV) irradiation, leading to the formation of H and O atoms. Therefore, for H2 and H2O, we consider them participating in escape in the form of H and O atoms. In other words, in our simulation of the escape scenarios, the atmospheric particles involved in escape include H, O, CO2, Ne, and Ar.
Guo (2019) modelled the photodissociation process on an Earth-like planet. They suggested that 1 µbar is a typical value for a H2O-rich atmosphere to be dominated by H atoms (i.e. via the major production of H2O by photodissociation) under a solar irradiation flux 100 times larger than the present. Benedikt et al. (2020) applied the 1 µbar level as an effective radius where the base of escape flow locates and where H atoms replace H2O to become the dominant species. For scenarios with initial nebular hydrogen less than 210 moles cm−2 (equivalent to 0.00004 wt%, 0.013 oceans), we coupled a 1D radiative-convective atmosphere model (Marcq et al. 2017) to the escape model to calculate the evolution of the atmospheric structure. The purpose of coupling this atmosphere model is to calculate the 1 µbar height (representing the effective radius of escape) and updates it at each time step during the escape. For atmospheres with more initial hydrogen, where the partial pressure of nebular hydrogen exceeds 50 bar, we solely employed an energy-limited hydrodynamic escape model to calculate the escape of all atmospheric compositions. Throughout all the calculations, the total atmospheric mass-loss rate is controlled by the flux of EUV received by Venus and the height of escape flow, r0 . Our parameter settings are shown in Table 1. The atmosphere tends to become inflated as a result of absorption of solar EUV; thus, our assumption of β = rEUV/r0 = 1 leads to a underestimation of the mass-loss rate of EUV-driven atmospheric escape.
Physical parameters.
2.2 Atmospheric structure-escape coupling model
We simulated the atmospheric hydrodynamic escape with an energy-limited model, which has been widely applied to atmospheric escape simulations on different types of planets (Watson et al. 1981; Erkaev et al. 2007; Odert et al. 2018; Benedikt et al. 2020; Lammer et al. 2020). Given that the particle numbers of Ne and Ar are three orders of magnitude smaller than H, we adopted a similar approach to Odert et al. (2018), wherein the escape of Ne and Ar does not impact the loss rates of H, O, and CO2.
For each atmospheric component, the model calculates an escape flux F (i.e. the number of escaping particles per unit area per unit time). This escape flux acts on a spherical shell located at a height, r0, which represents the base of escape flow, yielding the number of particles escaping per unit time. Odert et al. (2018) defined a range of r0 and selected specific values within this range as constant escape height. Benedikt et al. (2020) chose the 1 µbar height as the base of escape flow r0 and calculated the loss of H2 O–CO2 steam atmosphere for Moon and Mars-sized embryos. Lammer et al. (2020) coupled a 1D upper atmospheric model (Erkaev et al. 2015) and selected the height of photospheric radius as the base of escaping flow. Due to the computational complexity of this 1D upper atmosphere model, they also assumed a constant r0 . However, as escape progresses and partial pressures of atmospheric components decrease, the atmosphere tends to become optically thinner (Salvador et al. 2023) and, thus, the escape height, r0 , tends to decrease. The reduction in escape height further leads to lower efficiency of atmospheric mass-loss rate during hydrodynamic escape.
Therefore, for cases with initial hydrogen lower than 210 moles cm−2, in which the degassing-source hydrogen becomes a predominant component, we coupled an atmospheric escape model with a radiative-convective atmosphere model (Marcq et al. 2017). It takes the planetary surface temperature and partial pressure of atmospheric species (H2O and CO2) as inputs, applying the DISORT solver in the four-stream approximation to solve the radiative transfer equation, finally producing a pressure-altitude profile. As this radiative-convective atmosphere model focusses on the lower atmosphere, it takes hydrostatic equilibrium assumption. The ideal gas law is applied for CO2 . For cases with massive water vapor where H2O reaches supercritical phase, H2O would no longer be taken as an ideal gas. Instead, the steam tables for H2O in Haar (1984) have been applied for more accurate calculation. Here, we take the surface temperature of proto-Venus as 1500K and the altitude of 1 µbar as the location of r0 . Thus, we incorporated the radiative- convective atmosphere model to update the escape height r0 at each time step.
Furthermore, we derived an analytical form for the energylimited hydrodynamic escape model following the method of Odert et al. (2018), which is based on Zahnle & Kasting (1986) and Zahnle et al. (1990). We take:
(1)
(2)
(3)
(4)
where f is the mixing ratio of of a given species relative to H. V0 = GMVenus/r0 is the Venusian gravitational potential at r0, g the gravitational acceleration at r0 (all in cgs units); m is the mass of each species and the subscript H, O, k represents the major light species hydrogen, the major heavy species oxygen, and the minor heavy species k (i.e. CO2, Ne and Ar). Then, FEUV = LEUV/4πd2 is the stellar EUV flux, kB is the Boltzmann constant, and T is the upper atmosphere temperature (in K).
By combining Eqs. (4)–(6) from Odert et al. (2018), we can obtain the analytical expressions for the fractionation factors xO and :
(5)
(6)
where are the fractionation factors of oxygen and carbon dioxide, respectively, and correspond to the velocity ratio of oxygen or carbon dioxide to hydrogen. Then, we can obtain the escape flux FH,O,k (in cm−2 s−1 sterad−1) of hydrogen, oxygen, and the heavy minor species, k, at the height of r0 through:
(7)
(8)
Since the fk corresponding to Ne and Ar are magnitudes smaller than fO and , Eq. (7) is simplified as:
(9)
Thus we can calculate the number-loss rate dNH,O,k of each species at r0 at each timestep, dt:
(10)
We primarily used the energy-limited method to calculate the atmospheric escape process. The core formula Eq. (9) is essentially the same as Eq. (9) in Gillmann et al. (2009) and Eq. (24) in Zahnle & Kasting (2023). The analytical forms of xO and allow us to calculate the escape process faster. We assumed an initial Venus atmosphere composed of nebular- source H2 and degassing-source H2O–CO2 steam. As is shown in Eq. (10), integrating these escape fluxes over time will give us the atmospheric reservoir at any given moment. Meanwhile, there is a series of crossover mass mc , representing the mass below which the heavy species can be dragged to escape by H and O. As the escape progresses, the corresponding crossover mass mc for each heavy component decreases. According to Hunten et al. (1987), it can be written as:
(11)
where XO,k = NH/(NH + NO,k). N represents the total number of particles in the atmosphere. When mc,i ≤ mi, the component stops escaping. This allows us to obtain the isotope fractionation of Ne and Ar and the retention of CO2 during hydrodynamic escape.
Since the lighter particles can be dragged away by hydrogen more efficient than the heavier ones, 20Ne/22Ne and 36Ar/38Ar will continue to decrease during the escape until the crossover mass becomes lower than the mass of the particle itself.
Our criteria for valid results are as follows: the ratios 20Ne/22Ne, 36Ar/38Ar, and 20Ne/36Ar must fall within the range of present Venusian atmosphere values – 11.8 ± 0.7, 5.5 ± 0.6, and 0.25 ± 0.1, respectively (von Zahn et al. 1983; Donahue & Pollack 1983). The recently published study by Zahnle & Kasting (2023) also used these observational values to constrain their model results. Additionally, the crossover masses of Ne and Ar should be lower than their own respective masses, meaning that 20Ne, 22Ne, 36Ar, and 38Ar will not be affected by the dragging process during the latter stages of escape.
2.3 Solar EUV emission
As is shown in Eq. (9), the EUV from the young Sun is a key factor in driving atmospheric escape in young solar system. To explore the potential evolution of these emissions, we adopted the framework of Tu et al. (2015), which analyzed three distinct scenarios based on the evolution of stellar EUV luminosity (LEUV) for the 10th, 50th, and 90th percentiles of the rotation rate distribution, subsequently referred to as slow, moderate, and fast rotators, respectively (Eq. (12)).
In comparison, Claire et al. (2012) reconstructed the Sun’s EUV history using observations from a smaller number of nearby Sun-like stars of different ages, drawing on data from Ribas et al. (2005). The approach in Claire et al. (2012) provides a valuable reference for understanding solar EUV evolution, while the study by Tu et al. (2015) builds on this by combining observations of a larger number of solar-like stars with rotational evolution models, as illustrated in Figure 2 of their paper. These models indicate that initial stellar rotation rates lead to a wide range of possible rotational paths and associated EUV emissions at ages below 2–3 Gyr, which is consistent with X-ray observations of young Sun-like stars showing similar variability. For older stars like the Sun, these evolutionary paths converge, implying some uncertainty regarding the specific trajectory of solar evolution. Thus, the approach by Tu et al. (2015) allows for a broader exploration of possible EUV emission scenarios for young Sun-like stars.
According to Tu et al. (2015), the stellar EUV luminosity (LEUV) evolution for the three scenarios can be given by:
(12)
where t represents the stellar age in Myr and tsat indicates the saturation time.
The parameter ‘saturation time’ refers to the initial phase of stellar evolution during which the EUV luminosity remains at a high, nearly constant value. During this phase, the time variable, t, in the EUV luminosity equations is effectively set to the saturation duration, after which a power-law decline begins.
![]() |
Fig. 1 Distribution of matched cases in the entire case range is shown. Matched cases (i.e. red circles) represent the escape scenarios that can produce present Venusian atmospheric 20Ne/22Ne, 36Ar/38Ar and 20Ne/36Ar − 11.8 ± 0.7, 5.5 ± 0.6, and 0.25 ± 0.1, respectively (von Zahn et al. 1983; Donahue & Pollack 1983). Conversely, ‘unmatched cases’ do not satisfy all three isotopic ratios simultaneously. The X-axis represents the initial nebular hydrogen content in the Venusian atmosphere and the Y -axis represents the initial H2O content in the atmosphere. The horizontal dashed line represents an initial H2O inventory of 1.05 × 104 moles/cm2 and CO2 inventory of 2.3 × 103 moles/cm2 (about 90 bar of CO2 on present Venus). Cases above the line assume higher initial CO2 , while those below assume lower initial CO2 . The vertical dashed line represents an initial H2 content of 210 moles cm−2 (equal to 0.00004 wt%). For cases with initial nebular-source H2 below 210 moles cm−2, we coupled the radiactive-convective atmosphere model to the escape model and updated the escape height (where escape flux works) at each time step, assuming an initial 20Ne/36Ar ratio of 0.33 (i.e. a strongly fractionated value compared to the solar value). For cases with initial nebular-source H2 above 210 moles cm−2, our calculation applied only the escape model, keeping r0 fixed over time and assuming an initial 20Ne/36Ar ratio of 44.7 (i.e. the solar value). |
3 Results
Based on previous studies, we innovated in three main aspects: firstly, we introduced a radiative-convective atmosphere model to approximately update the height of the base of flow r0 during the hydrodynamic escape process; secondly, we considered the EUV-driven escape under three different types of EUV evolution patterns; thirdly, we tested a series of initial Venusian atmospheres composed of different combinations of H2 and H2O-CO2 . The combination of these three aspects has allowed us to discover new possibilities for the composition of the initial Venusian atmosphere.
3.1 Evidence for the low initial nebular-source hydrogen
Lichtenegger et al. (2016) and Odert et al. (2018) both assumed that the primitive Venus had a 2000 km-depth magma ocean, releasing a steam atmosphere composed of 458 bar H2O and 101 bar CO2 (i.e. 2.27×104 moles cm−2 H2O and 5.01×103 moles cm−2 CO2). We followed this assumption and used it as the reference for the steam atmosphere composition. Taking twice (4.55×104 moles cm−2 H2O, 1.00×104 moles cm−2 CO2) as the upper limit and 0.1 times (2.27×103 moles cm−2 H2O and 5.01×102 moles cm−2 CO2) as the lower limit.
As is shown in Fig. 1, we calculated a series of escape histories for initial Venusian atmospheres composed of nebular- source H2 below 5.29×107 moles cm−2 (equal to 10 wt%). We applied the atmospheric structure-escape coupling model as the initial nebula-source H2 ≤ 210 moles cm−2. For the initial nebula-source H2 > 210 moles cm−2, since hydrogen is not included in the radiative-convective atmosphere model (i.e. hydrogen cannot be the main component of our simulated atmosphere), we only calculated the escape process applying the energy-limited hydrodynamic escape method, which means that r0 does not change as the escape progresses. In this way, we took 1.05 RVenus as the lower bound, r0,min, and 2.55 RVenus as the upper bound, r0,max, of r0 values, and took r0,min, 0.55 r0,max, 0.7 r0,max, 0.85 r0,max, and r0,max as the base of escape flow r0.
From the Fig. 1, we can see that all the matched cases occur when the initial H2 is below 210 moles cm−2. This is because the constraint of highly fractionated 20Ne/36Ar (0.33), compared to the solar value (44.7), only becomes reasonable when the initial H2 is no longer the dominant atmospheric component. This can be explained by the loss of an initially H2 -rich atmosphere, which results in significant elemental fractionation and minor isotopic fractionation, as greater mass differences generally lead to more pronounced fractionation effects. It is only if the 20Ne/36Ar was already highly fractionated, compared to the solar value at the onset of escape (i.e., from 44.7 to 0.33) that the secondary atmosphere could, through hydrodynamic escape, achieve the present-day observed range of 20Ne/36Ar fractionation. More details of these matched cases results are shown in Table 2. We also applied the escape model only to calculate cases with H2 ≤ 210 moles cm−2 but we did not succeed in finding any matched cases.
The matched cases only appeared in H2 O-dominated secondary atmosphere scenarios, specifically under the EUV law assumption for slow rotators. Therefore, we did not present cases for other assumptions in the main text. To provide a more comprehensive view of atmospheric escape under different assumptions, we present the results in Figs. B.1, B.2, and B.3. These figures show the escape processes of secondary atmospheres dominated by H2O and H2 under different EUV law assumptions. Additionally, they illustrate the isotopic fractionation processes of Ne and Ar during this period.
For matched cases with initial H2 ≤ 210 moles cm−2, we found that if the young Sun was a slow rotator (Fig. 1), when the initial nebula-source H2 was below 2.6 moles cm−2 and the degassing-source initial H2O ranged from 2.27×103 to 6.82×103 moles cm−2, corresponding to degassing-source CO2 ranging from 500 to 1500 moles cm−2. We can always obtain20 Ne/22 Ne, 36Ar/38Ar, and 20Ne/36Ar values that fall within the range of the observational values (Table 2). However, for the moderate and fast rotator, no matched cases were identified that could suggest the initial nebula-source H2 level approaching zero.
From the distribution of matched cases, we can see that in the regime where the initial nebular-source H2 content is below a certain number, the degassing-source H2O (rather than nebular-source H2) begins to dominate the hydrodynamic escape process. Taking an example of initial 6.82×103 moles cm−2 H2O, if there were initial nebula-source H2 lower than 11 moles cm−2, escape always ends after 8 Myr, left 356 moles cm−2 O and 1.5×103 moles cm−2 CO2 accumulated. 20Ne/22Ne and36 Ar/20 Ne will always be fractionated to 11.44 and 3.591, respectively. Ar has not experienced escape, thus the 36Ar/38Ar ratio remains at the solar value of 5.50. This is why the distribution interval where the initial nebular hydrogen content can approach zero in an infinite way comes up as a matched result.
Results of matched cases of initial atmosphere with H2 less than 210 moles cm−2.
3.2 Evidence for no initial nebular-source hydrogen
The results presented above assume a nebula lifetime of approximately 10 Myr, which marks the onset of the EUV-driven hydrodynamic escape process. However, the nebula can dissipate in as short a time as 4 Myr (Mamajek 2009; Wang et al. 2017; Lammer et al. 2020; Weiss et al. 2021). Under the assumption of a 10 Myr escape-starting time, we identified matched cases with nearly depleted initial nebula-source H2. To further explore this, we tested nebula lifetimes ranging from 4 to 13 Myr, with 0.2 Myr intervals. As shown in Fig. 2, more matched cases were found, suggesting that the initial H2 sourced from the nebula could be nearly exhausted. No matched cases were identified under the moderate and fast rotator assumptions. As t0 shifts to earlier times, the scenarios generally exhibit an increasing trend in initial H2O, reaching up to 1.33 oceans around 4 Myr and decreasing to 0.03 oceans by times after 12 Myr. At times after 12 Myr, no matched cases were identified, and by 13 Myr, Ar remained entirely retained. This trend suggests that the timing of hydrodynamic escape onset significantly impacts the retention of H2O in the atmosphere, with later t0 values generally leading to lower initial H2O content and reduced volatile escape.
4 Discussion
4.1 Comparing the results of traditional and improved model
As can be seen from Fig. 3, the escape process coupled with atmospheric structure, accompanied by a decrease in r0 height, significantly alters the Ar isotope fractionation calculated by the energy-limited hydrodynamic escape model. The 20Ne/22Ne ratio decreases from above to within the observed value range. The degree of 36Ar/38Ar and 20Ne/36Ar fractionation remains almost unchanged. In the scenario where r0 remains constant (left panel), at 1.2 Myr after the onset of escape, the crossover mass for 20Ne falls below its own mass, signifying the termination of20 Ne/22 Ne fractionation. In the case where r0 decreases with ongoing escape (right panel), the reduction in r0 accelerates the decrease in the crossover mass for 20Ne. As a result, at 1.9 Myr after the onset of escape, 20Ne exhibits a crossover mass below its own mass, concluding the 20Ne/22Ne more fractionated compared to the left panel.
By comparing the atmospheric escape processes for an equal amount of initial hydrogen, but with different sources in Fig. 4, it can be seen an equal amount of H2 source hydrogen is replaced by H2O source hydrogen; namely, this is the case where there is less initial H2 and more initial H2O, while the total amount of initial hydrogen remains the same and the amount of initial O increases. As a result, the EUV energy that originally drove hydrogen escape is more evenly distributed among O, leading to a slower rate of hydrogen escape. At the same time, the degree of fractionation of Ne increases, with Ne fractionation shifting from being too high to fall within the observed range. Therefore, the relative abundance of nebular source H2 and degassing source H2O in Venusian atmosphere at the beginning of the hydrodynamic escape process (i.e. the main sources of hydrogen) are important factors affecting the early evolution of Venusian atmosphere.
![]() |
Fig. 2 Results illustrated under varying assumptions of the initial hydrodynamic escape onset time, t0, within the context of the slow rotator scenario. Each circle represents a ‘matched case’; namely, a scenario that satisfies the constraints from three isotopic ratios of Ne and Ar in Venus’ atmosphere. The colour bar indicates the timescale required for the complete loss of hydrogen. All cases shown correspond to scenarios with no residual H2 in the secondary atmosphere. The variable t0 represents the time, in million years (Myr), at which the EUV-driven hydrodynamic escape begins. The dashed line marks an initial H2O inventory of 1.05×104 moles/cm2 and an initial CO2 inventory of 2.3×103 moles/cm2 (approximately equivalent to 90 bars of CO2 on present Venus). Matched cases above the dashed line assume an initial CO2 inventory higher than the present value, while those below assume a lower initial CO2 inventory. |
4.2 Evidence for the low initial nebular hydrogen regime
The matched cases show that hydrodynamic escape of a H2O-rich atmosphere can leave up to 30 bar of CO2 (≈1500 moles cm−2) in the Venusian atmosphere. In this study, we assumed the initial CO2 content to be approximately 20% of the initial H2O. However, the processes of magma ocean outgassing and the extent to which impact degassing contributed to the H2O and CO2 content during the formation of the secondary atmosphere remain uncertain (Bower et al. 2022). Better constraints on the H2O and CO2 content could further refine our understanding of the hydrodynamic escape scenario for Venus’ secondary atmosphere.
Our matched cases with the initial H2 lower than 50 moles cm−2 (i.e. ≈1 bar) for a young Sun with a slow rotator correspond to initial atmospheres with H2 similar to scenario of 10−4 < PH < 1 bar in Charnoz et al. (2023). They used a magma ocean-atmosphere coupling model that includes physical–chemical equilibrium models to calculate the surface of a terrestrial planet covered by a magma ocean and found that when the surface temperature is 1800 K: if 10−4 < PH < 1 bar, then the atmosphere is dominated by H2 and H2O. Early Venus could have not been surrounded by a dense nebula-source hydrogen atmosphere (Lammer et al. 2014) or lost massive nebula-source hydrogen in a few Myrs (Donahue 1986; Valencia et al. 2010; Lopez et al. 2012). Atmospheric erosion caused by giant impacts, sometimes taken as the reason for Venus’ slow rotation, could also be the reason for low initial nebula-source gas. Only a small amount of nebular source H2 was present in the initial Venusian atmosphere, also suggesting that the proto- planetary disk may have undergone a rapid dissipation process. Venus would have lost most of its nebular hydrogen envelope during an atmospheric loss process (i.e., ‘boil-off’) at a much higher rate than in hydrodynamic escape (Owen & Wu 2016; Owen et al. 2020). In addition, it is also possible that a significant amount of H2 dissolved into the magma ocean and possibly even reached deep into Venus (Hier-Majumder & Hirschmann 2017; Bower et al. 2022; Young et al. 2023). This scenario also supports the idea that the early Venusian atmosphere may have been a H2 -poor atmosphere, so that during the magma ocean stage of Venus, due to the low-pressure environment on the surface, hundreds of bar of dense H2O-dominated steam atmosphere could be released.
4.3 Limitation of the model
There is still room for further improvement in the selfconsistency of our energy-limited method. Our calculations assume a non-viscous system (Watson et al. 1981), which tends to overestimate the efficiency of hydrodynamic escape. Modirrousta-Galian & Korenaga (2023) demonstrated that the mass-loss rate estimated by the diffusion-limited method can be significantly lower, by an order of magnitude, compared to the energy-limited method.
Zahnle & Kasting (2023) considered the escape of hydrogen and oxygen from a steam atmosphere. For oxygen, they included surface sinks as well as escape. They found that O would build up such that the H:O ratio above the homopause was roughly 1:2. In our study, we assumed an H:O ratio of 2:1 based on the composition of H2O molecules. If the actual H:O ratio is lower than our assumption, meaning more O is involved in the escape process, the additional energy required for O to escape might lead us to overestimate the escape flux of H.
We set the heating efficiency η as 15%, a typical value for hydrogen-dominated atmosphere. Additionally, the accumulation of CO2 during the late stage of hydrodynamic escape can lead to an overestimation of the total mass-loss rate. Overall, CO2 is effective at cooling the thermosphere, thus reducing mass-loss efficiency (Yoshida & Kuramoto 2020). However, since only a little fraction of CO2 escaped, we assume it settled down to the lower atmosphere. Ignoring CO2’s cooling effect might not significantly impact the EUV heating efficiency of the hydrogen-dominated thermosphere.
There are several massive overall atmospheric escape mechanisms during the early stages, which can also lead to an underestimation of our total mass-loss rate. We simulated a very early evolution of Venusian atmosphere without considering the impact of frequent collisions in the young solar system. Impacts can induce significant atmospheric loss through shock waves, potentially removing up to 10% of an atmosphere on terrestrial planets (Genda & Abe 2005; Schlichting et al. 2015). Moreover, impact may cause substantial thermal heating, leading to additional atmospheric escape driven by interior energy (Biersteker & Schlichting 2019, 2021; Modirrousta-Galian & Korenaga 2023). In extreme cases, if giant impacts increase the planetary surface temperature to around 10 000 K, the resulting mass-loss rate could be orders of magnitude higher than that the mass loss driven by X-ray and EUV (Modirrousta- Galian & Korenaga 2023). However, these overall atmospheric escape mechanisms may not cause isotopic fractionation of such elements as Ne and Ar.
![]() |
Fig. 3 Variations in column abundances of H, O, and CO2 at the surface, the evolution of isotope ratios of Ne and Ar, and the evolution of escape altitude, shown from top to bottom. The left panel shows the results of the escape scenarios calculated using the radiative-convective atmosphere model only at the initial time step for the escape base height, r0 , followed by calculations based on the energy-limited escape model. The right panel shows the results of our atmospheric structure-escape coupling model, with r0 updated at each time step. The purple and light red shaded regions represent the current Venusian atmosphere ranges of 20Ne/22Ne, 36Ar/38Ar, and 20Ne/36Ar – 11.8 ± 0.7, 5.5 ± 0.6, and 0.25 ± 0.1, respectively (von Zahn et al. 1983; Donahue & Pollack 1983). Note: the 20Ne/36Ar ratio is assumed to evolve from 0.33, a value that is strongly fractionated compared to the solar value 44.7. Also, z0 = r0 –RVenus represents the height from the base of the escape flow to the surface of Venus. |
5 Conclusion
We conducted extensive simulations of the hydrodynamic escape process of Venus’ primordial atmosphere, exploring over 500 000 scenarios with varying initial H2-H2O atmospheric compositions, types of young Sun rotators, and the evolution of the base of the escape flow (r0). By coupling an energy-limited atmospheric escape model with a 1D radiative-convective atmospheric model (Marcq et al. 2017), we simultaneously calculated the evolution of both the lower and upper atmospheres to constrain r0 during hydrodynamic escape.
Our results indicate that Venus’ secondary atmosphere likely contained only a small fraction of nebula-derived H2 at its formation, with the H2O content not exceeding 1.4 ocean equivalents. Specifically, when the hydrodynamic escape onset time (t0) is set to 10 Myr, Ne and Ar isotope-matched scenarios are hydrogen-poor, involving initial H2 levels ranging from 50 moles cm−2 down to zero and initial H2O levels between 9.10×102 and 6.82×103 moles cm−2 (approximately 0.06–0.42 ocean equivalents). If the EUV-driven hydrodynamic escape began earlier, at around 4 Myr, the maximum initial H2O content could increase to 1.33 ocean equivalents.
These findings suggest that Venus likely lost its primary hydrogen-rich atmosphere before forming its secondary, H2O- rich atmosphere. An H2O-dominated steam atmosphere, formed through degassing processes from impactors, a magma ocean, or both, was sufficient to be photolyzed, producing enough hydrogen atoms to drive the hydrodynamic escape process. Such an H2O-dominated steam atmosphere is typically considered an effective trigger for Venusian runaway greenhouse climate, leading to extreme surface temperatures and inhibiting the condensation of water (Kasting 1988; Salvador et al. 2023). This process resulted in the current 20Ne/22Ne, 36Ar/38Ar, and 20Ne/36Ar ratios observed in Venus’ atmosphere. The low surface pressure on early Venus facilitated the formation of this dense steam atmosphere during magma ocean outgassing, which played a key role in the subsequent hydrodynamic escape processes.
![]() |
Fig. 4 Escape process for an atmosphere with an initial number of hydrogen atoms of 1.36×104 moles cm−2, but with different relative compositions of initial hydrogen. Compared to the left panel, the right panel replaces 200 moles cm−2 of the initial H2 source hydrogen with H2O source hydrogen. |
Data availability
The code of energy-limit hydrodynamic escape model used in the calculations is available at https://osf.io/k3mu2/.
Acknowledgements
We are deeply grateful to the reviewer Kevin J. Zahnle for refining our research methodology, which has significantly enhanced this manuscript. The project is supported by the National Key Research and Development Program of China No. 2024YFF0807500. P.O. acknowledges the Austrian Science Fund (FWF): 10.55776/I5711. Y.Z. is supported by the Sichuan Provincial Natural Science Foundation (grant number 2023NSFSC0278) and the National Science Foundation of China (NSFC grant No. 4224114). We thank Dr. Darius Modirrousta-Galian for constructive discussion that helped to improve the manuscript.
Appendix A Numerical method
We assumed a series of primordial atmosphere composed of different initial H2 and H2O, corresponding to different initial H/O ratios. Different initial compositions of primordial atmospheres lead to different isotope fractionation processes. Therefore, we can utilize the isotope fractionation results to constrain the initial compositions. As a minor species, H2 always runs out in a very short period, resulting in a H2O-dominated atmosphere. We take the 1 µbar level as the base of the flow of escaping particles. As the escape progresses, the particle numbers of atmospheric species become significantly reduced, which results in lower 1 µbar level. Thus, before the atmospheric escape calculation at each time step, we refresh the 1 µbar level by atmospheric structure model. Considering EUV-photodissociation, we simply calculate the escape of H and O to represent for the mass-loss rate of H2 and H2O. Thus, the atmospheric escape model output H and O particle numbers, we take the particle number of less one of O and 1/2 H to present for the number of H2O particles in lower atmosphere. The loop process of the model is illustrated in Fig.A.1.
![]() |
Fig. A.1 Flow chart of the iterative logic between the atmospheric structure model and the escape model. |
Appendix B Comparison between H2-poor cases and H2-rich cases
Since all the matched cases in the main text are concentrated under the assumption of the slow rotator for the EUV law in H2 -poor conditions, we present here the H2-poor and H2- dominated cases under different EUV law assumptions. For the H2-dominated cases under the slow rotator assumption (right panel in Fig.B.1), the crossover masses of Ne and Ar during the late stage of escape fall below their own masses, resulting in their isotopic compositions no longer evolving with the escape of other components. However, the fractionation of 38Ar/20Ne is insufficient to match the observed data. Similarly, the H2- dominated cases shown in Figs. B.2 and B.3 also fail to achieve sufficient fractionation of 36Ar/20Ne.
It should be noted that the constant r0 value is derived from the characteristics of an H2O-dominated atmosphere. For H2- dominated cases, the mass-loss rates shown in Figs. B.1, B.2, and B.3 are closer to the minimum, as an H2-dominated atmosphere undergoes significant expansion when heated by EUV.
![]() |
Fig. B.1 H2-poor cases and H2-rich cases under the EUV law following slow rotator assumption. |
![]() |
Fig. B.2 H2-poor cases and H2-rich cases under the EUV law following moderate rotator assumption. |
![]() |
Fig. B.3 H2-poor cases and H2-rich cases under the EUV law following fast rotator assumption. |
References
- Abe, Y., & Matsui, T. 1986, J. Geophys. Res. Solid Earth, 91, E291 [Google Scholar]
- Benedikt, M. R., Scherf, M., Lammer, H., et al. 2020, Icarus, 347, 113772 [NASA ADS] [CrossRef] [Google Scholar]
- Biersteker, J. B., & Schlichting, H. E. 2019, MNRAS, 485, 4454 [NASA ADS] [CrossRef] [Google Scholar]
- Biersteker, J. B., & Schlichting, H. E. 2021, MNRAS, 501, 587 [Google Scholar]
- Bower, D. J., Hakim, K., Sossi, P. A., & Sanan, P. 2022, Planet. Sci. J., 3, 93 [NASA ADS] [CrossRef] [Google Scholar]
- Charnoz, S., Falco, A., Tremblin, P., et al. 2023, A&A, 674, A224 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
- Claire, M. W., Sheets, J., Cohen, M., et al. 2012, ApJ, 757, 95 [Google Scholar]
- Donahue, T. M. 1986, Icarus, 66, 195 [NASA ADS] [CrossRef] [Google Scholar]
- Donahue, T. M., & Pollack, J. B. 1983, in Venus, eds. D. M. Hunten, L. Colin, T. M. Donahue, & V. I. Moroz (Tucson: University of Arizona Press), 1003 [Google Scholar]
- Elkins-Tanton, L. T. 2008, Earth Planet. Sci. Lett., 271, 181 [Google Scholar]
- Erkaev, N. V., Kulikov, Y. N., Lammer, H., et al. 2007, A&A, 472, 329 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
- Erkaev, N. V., Lammer, H., Odert, P., Kulikov, Y. N., & Kislyakova, K. G. 2015, MNRAS, 448, 1916 [NASA ADS] [CrossRef] [Google Scholar]
- Genda, H., & Abe, Y. 2005, Nature, 433, 842 [NASA ADS] [CrossRef] [Google Scholar]
- Gillmann, C., Chassefière, E., & Lognonné, P. 2009, Earth Planet. Sci. Lett., 286, 503 [CrossRef] [Google Scholar]
- Grimberg, A., Baur, H., Bochsler, P., et al. 2006, Science, 314, 1133 [NASA ADS] [CrossRef] [Google Scholar]
- Guo, J. H. 2019, ApJ, 872, 99 [NASA ADS] [CrossRef] [Google Scholar]
- Haar, L. 1984, NBS/NRC Steam Tables (CRC Press) [Google Scholar]
- Hier-Majumder, S., & Hirschmann, M. M. 2017, Geochem. Geophys. Geosyst., 18, 3078 [Google Scholar]
- Hunten, D. M., Pepin, R. O., & Walker, J. C. G. 1987, Icarus, 69, 532 [NASA ADS] [CrossRef] [Google Scholar]
- Kasting, J. F. 1988, Icarus, 74, 472 [NASA ADS] [CrossRef] [Google Scholar]
- Kasting, J. F., & Pollack, J. B. 1983, Icarus, 53, 479 [NASA ADS] [CrossRef] [Google Scholar]
- Lammer, H., Stökl, A., Erkaev, N. V., et al. 2014, MNRAS, 439, 3225 [NASA ADS] [CrossRef] [Google Scholar]
- Lammer, H., Leitzinger, M., Scherf, M., et al. 2020, Icarus, 339, 113551 [Google Scholar]
- Lange, M. A., & Ahrens, T. J. 1982, J. Geophys. Res. Solid Earth, 87, A451 [Google Scholar]
- Lange, M. A., Lambert, P., & Ahrens, T. J. 1985, Geochim. Cosmochim. Acta, 49, 1715 [NASA ADS] [CrossRef] [Google Scholar]
- Lebrun, T., Massol, H., Chassefière, E., et al. 2013, J. Geophys. Res. Planets, 118, 1155 [NASA ADS] [CrossRef] [Google Scholar]
- Lichtenegger, H. I. M., Kislyakova, K. G., Odert, P., et al. 2016, J. Geophys. Res. Space Phys., 121, 4718 [Google Scholar]
- Lodders, K. 2020, Solar Elemental Abundances, in The Oxford Research Encyclopedia of Planetary Science (Oxford University Press) [Google Scholar]
- Lopez, E. D., Fortney, J. J., & Miller, N. 2012, ApJ, 761, 59 [Google Scholar]
- Mamajek, E. E. 2009, AIP Conf. Proc., 1158, 3 [NASA ADS] [CrossRef] [Google Scholar]
- Marcq, E., Salvador, A., Massol, H., & Davaille, A. 2017, J. Geophys. Res. Planets, 122, 1539 [Google Scholar]
- Matsui, T., & Abe, Y. 1986a, Nature, 319, 303 [NASA ADS] [CrossRef] [Google Scholar]
- Matsui, T., & Abe, Y. 1986b, Nature, 322, 526 [CrossRef] [Google Scholar]
- Meshik, A., Hohenberg, C., Pravdivtseva, O., & Burnett, D. 2014, Geochim. Cosmochim. Acta, 127, 326 [NASA ADS] [CrossRef] [Google Scholar]
- Modirrousta-Galian, D., & Korenaga, J. 2023, ApJ, 943, 11 [NASA ADS] [CrossRef] [Google Scholar]
- Odert, P., Lammer, H., Erkaev, N. V., et al. 2018, Icarus, 307, 327 [NASA ADS] [CrossRef] [Google Scholar]
- Oksana, T., & Dmytro, G. 2021, in Clean Water and Sanitation, eds. W. Leal Filho, A. M. Azul, L. Brandli, A. Lange Salvia, & T. Wall (Springer Cham), 1 [Google Scholar]
- Owen, J. E., & Wu, Y. 2016, ApJ, 817, 107 [NASA ADS] [CrossRef] [Google Scholar]
- Owen, J. E., Shaikhislamov, I. F., Lammer, H., Fossati, L., & Khodachenko, M. L. 2020, Space Sci. Rev., 216, 1 [CrossRef] [Google Scholar]
- Pepin, R. O. 1991, Icarus, 92, 2 [NASA ADS] [CrossRef] [Google Scholar]
- Ribas, I., Guinan, E. F., Güdel, M., & Audard, M. 2005, ApJ, 622, 680 [Google Scholar]
- Salvador, A., Avice, G., Breuer, D., et al. 2023, Space Sci. Rev., 219, 51 [NASA ADS] [CrossRef] [Google Scholar]
- Sasaki, S., & Nakazawa, K. 1988, Earth Planet. Sci. Lett., 89, 323 [Google Scholar]
- Schlichting, H. E., Sari, R., & Yalinewich, A. 2015, Icarus, 247, 81 [NASA ADS] [CrossRef] [Google Scholar]
- Sekiya, M., Nakazawa, K., & Hayashi, C. 1980, Earth Planet. Sci. Lett., 50, 197 [Google Scholar]
- Sekiya, M., Hayashi, C., & Nakazawa, K. 1981, Prog. Theor. Phys., 66, 1301 [CrossRef] [Google Scholar]
- Tian, F. 2009, ApJ, 703, 905 [NASA ADS] [CrossRef] [Google Scholar]
- Tian, M., & Heng, K. 2024, ApJ, 963, 157 [NASA ADS] [CrossRef] [Google Scholar]
- Tu, L., Johnstone, C. P., Güdel, M., & Lammer, H. 2015, A&A, 577, L3 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
- Turbet, M., Bolmont, E., Chaverot, G., et al. 2021, Nature, 598, 276 [CrossRef] [Google Scholar]
- Valencia, D., Ikoma, M., Guillot, T., & Nettelmann, N. 2010, A&A, 516, A20 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
- von Zahn, U., Komer, S., Wiemann, H., & Prinn, R. 1983, in Venus, eds. D. M. Hunten, L. Colin, T. M. Donahue, & V. I. Moroz (Tucson: University of Arizona Press), 297 [Google Scholar]
- Wang, H., Weiss, B. P., Bai, X.-N., et al. 2017, Science, 355, 623 [NASA ADS] [CrossRef] [Google Scholar]
- Watson, A. J., Donahue, T. M., & Walker, J. C. G. 1981, Icarus, 48, 150 [NASA ADS] [CrossRef] [Google Scholar]
- Weiss, B. P., Bai, X.-N., & Fu, R. R. 2021, Sci. Adv., 7, eaba5967 [NASA ADS] [CrossRef] [Google Scholar]
- Yoshida, T., & Kuramoto, K. 2020, Icarus, 345, 113740 [NASA ADS] [CrossRef] [Google Scholar]
- Young, E. D., Shahar, A., & Schlichting, H. E. 2023, Nature, 616, 306 [NASA ADS] [CrossRef] [Google Scholar]
- Zahnle, K. J., & Kasting, J. F. 1986, Icarus, 68, 462 [NASA ADS] [CrossRef] [Google Scholar]
- Zahnle, K. J., & Kasting, J. F. 2023, Geochim. Cosmochim. Acta, 361, 228 [CrossRef] [Google Scholar]
- Zahnle, K. J., Kasting, J. F., & Pollack, J. B. 1990, Icarus, 84, 502 [NASA ADS] [CrossRef] [Google Scholar]
All Tables
All Figures
![]() |
Fig. 1 Distribution of matched cases in the entire case range is shown. Matched cases (i.e. red circles) represent the escape scenarios that can produce present Venusian atmospheric 20Ne/22Ne, 36Ar/38Ar and 20Ne/36Ar − 11.8 ± 0.7, 5.5 ± 0.6, and 0.25 ± 0.1, respectively (von Zahn et al. 1983; Donahue & Pollack 1983). Conversely, ‘unmatched cases’ do not satisfy all three isotopic ratios simultaneously. The X-axis represents the initial nebular hydrogen content in the Venusian atmosphere and the Y -axis represents the initial H2O content in the atmosphere. The horizontal dashed line represents an initial H2O inventory of 1.05 × 104 moles/cm2 and CO2 inventory of 2.3 × 103 moles/cm2 (about 90 bar of CO2 on present Venus). Cases above the line assume higher initial CO2 , while those below assume lower initial CO2 . The vertical dashed line represents an initial H2 content of 210 moles cm−2 (equal to 0.00004 wt%). For cases with initial nebular-source H2 below 210 moles cm−2, we coupled the radiactive-convective atmosphere model to the escape model and updated the escape height (where escape flux works) at each time step, assuming an initial 20Ne/36Ar ratio of 0.33 (i.e. a strongly fractionated value compared to the solar value). For cases with initial nebular-source H2 above 210 moles cm−2, our calculation applied only the escape model, keeping r0 fixed over time and assuming an initial 20Ne/36Ar ratio of 44.7 (i.e. the solar value). |
In the text |
![]() |
Fig. 2 Results illustrated under varying assumptions of the initial hydrodynamic escape onset time, t0, within the context of the slow rotator scenario. Each circle represents a ‘matched case’; namely, a scenario that satisfies the constraints from three isotopic ratios of Ne and Ar in Venus’ atmosphere. The colour bar indicates the timescale required for the complete loss of hydrogen. All cases shown correspond to scenarios with no residual H2 in the secondary atmosphere. The variable t0 represents the time, in million years (Myr), at which the EUV-driven hydrodynamic escape begins. The dashed line marks an initial H2O inventory of 1.05×104 moles/cm2 and an initial CO2 inventory of 2.3×103 moles/cm2 (approximately equivalent to 90 bars of CO2 on present Venus). Matched cases above the dashed line assume an initial CO2 inventory higher than the present value, while those below assume a lower initial CO2 inventory. |
In the text |
![]() |
Fig. 3 Variations in column abundances of H, O, and CO2 at the surface, the evolution of isotope ratios of Ne and Ar, and the evolution of escape altitude, shown from top to bottom. The left panel shows the results of the escape scenarios calculated using the radiative-convective atmosphere model only at the initial time step for the escape base height, r0 , followed by calculations based on the energy-limited escape model. The right panel shows the results of our atmospheric structure-escape coupling model, with r0 updated at each time step. The purple and light red shaded regions represent the current Venusian atmosphere ranges of 20Ne/22Ne, 36Ar/38Ar, and 20Ne/36Ar – 11.8 ± 0.7, 5.5 ± 0.6, and 0.25 ± 0.1, respectively (von Zahn et al. 1983; Donahue & Pollack 1983). Note: the 20Ne/36Ar ratio is assumed to evolve from 0.33, a value that is strongly fractionated compared to the solar value 44.7. Also, z0 = r0 –RVenus represents the height from the base of the escape flow to the surface of Venus. |
In the text |
![]() |
Fig. 4 Escape process for an atmosphere with an initial number of hydrogen atoms of 1.36×104 moles cm−2, but with different relative compositions of initial hydrogen. Compared to the left panel, the right panel replaces 200 moles cm−2 of the initial H2 source hydrogen with H2O source hydrogen. |
In the text |
![]() |
Fig. A.1 Flow chart of the iterative logic between the atmospheric structure model and the escape model. |
In the text |
![]() |
Fig. B.1 H2-poor cases and H2-rich cases under the EUV law following slow rotator assumption. |
In the text |
![]() |
Fig. B.2 H2-poor cases and H2-rich cases under the EUV law following moderate rotator assumption. |
In the text |
![]() |
Fig. B.3 H2-poor cases and H2-rich cases under the EUV law following fast rotator assumption. |
In the text |
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.