Stochastic Snowball States in Planetary Climate

arXiv:2610.10922 · astro-ph.EP · Submitted 2026-10-07 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.

Vera: I'm Vera, and with me are Jocelyn and Subrahmanyan, guest researcher.

Jocelyn: Today's paper: "Stochastic Snowball States in Planetary Climate".

Vera: The gist The stochastic dynamics are quantitatively and qualitatively different than the deterministic dynamics apart from mean eccentricities less than 0.1,

Jocelyn: First, who's behind it and why it matters.

Title and authors: Vera: Now, let’s go back to the core summary of this paper, "Stochastic Snowball States in Planetary Climate". They are showing how orbital noise changes the fundamental physics of these climate states.

Jocelyn: Basically, they are taking the deterministic energy balance model and injecting stochasticity into the orbital eccentricity to see what that does.

Subrahmanyan: They found that apart from mean eccentricities less than zero point one, the resulting stochastic dynamics are quantitatively and qualitatively different from the deterministic ones >Ref:two thousand six hundred ten point one zero nine two two#pg6>. That’s the first major deviation they pinpoint.

Vera: It’s not just a tiny nudge; it’s a fundamental change in how the system behaves when those orbital fluctuations are active >Ref:two thousand six hundred ten point one zero nine two two#pg6>. It shows that if you have these orbital fluctuations, the physics of the climate response is different.

Jocelyn: And then they look at what happens when those stochasticity drives radiative flux forcing asymmetry and intermittency >Ref:two thousand six hundred ten point one zero nine two two#pg5>. That’s what ultimately increases the time it takes to reach a stable seasonal cycle.

Subrahmanyan: They quantified that going from zero to zero point two in annual eccentricity variance stretches that time from three years up to nine years >Ref:two thousand six hundred ten point one zero nine two two#pg5>. That’s a concrete number showing how much uncertainty stretches the system out.

Vera: So, for someone listening who doesn't deal with orbital mechanics every day, this paper is basically saying that if you add orbital noise to the energy balance equation, your climate system takes longer to settle into its long-term pattern >Ref:two thousand six hundred ten point one zero nine two two#pg3>.

Jocelyn: And they also explored what happens when the mean eccentricity is higher than zero point one >Ref:two thousand six hundred ten point one zero nine two two#pg6>. They found that the deterministic transition from Snowball to non-Snowball state occurs over a range of stellar flux ∆S˚ >Ref:two thousand six hundred ten point one zero nine two two#pg6>.

Subrahmanyan: This means the critical forcing threshold isn't a single number anymore, but a band of possibilities >Ref:two thousand six hundred ten point one zero nine two two#pg10>. It moves us away from that idea of a single deterministic threshold.

Vera: That’s really important for climate science—we might be looking for a fuzzy region for phase transitions rather than sharp lines >Ref:two thousand six hundred ten point one zero nine two two#pg6>.

Jocities: And they also showed that as the mean eccentricity increases and the variance of those annual eccentricity fluctuations grows, this range ∆S˚ gets bigger, which in turn decreases the probability of reaching a steady state and experiencing that Snowball to non-Snowball bifurcation >Ref:two thousand six hundred ten point one zero nine two two#pg6>.

Subrahmanyan: It means for highly eccentric orbits with high fluctuation noise, the system is less likely to find a stable equilibrium at all >Ref:two thousand six hundred ten point one zero nine two two#pg4>. That has some big implications for predicting long-term planetary behavior >Ref:two thousand six hundred ten point one zero nine two two#pg4>.

Vera: So, this paper summarizes the core finding is that orbital noise fundamentally alters the stability landscape of these climate states >Ref:two thousand six hundred ten point one zero nine two two#pg4>. It’s not just a small tweak to the math, it changes the physics of what's possible.

Jocelyn: And it shows that those transition points become smeared out rather than being sharp, which is a very important detail for anyone modeling planetary climate dynamics >Ref:two thousand six hundred ten point one zero nine two two#pg6>. That smearing means we can't just look for one perfect forcing number anymore.

Subrahmanyan: This is a lot of implications for how we model planetary evolution across different stability regimes >Ref:two thousand six hundred ten point one zero nine two two#pg4>.

The paper's summary: Vera: Now let’s look at the improvements they suggest—they aren't proposing a new way to model the planet itself, but rather how we should interpret and use this stochastic approach >Ref:two thousand six hundred ten point one zero nine two two#pg4>. They are suggesting that we need to adopt this view that orbital uncertainty is not just noise we filter out, but a driver of the system's transition behavior >Ref:two thousand six hundred ten point one zero nine two two#pg3>.

Jocelyn: Specifically, they suggest that for risk assessment, you should look at how the variance of annual fluctuations directly scales with the change in the critical forcing threshold ∆S˚ >Ref:two thousand six hundred ten point one zero nine two two#pg6>.

Subrahmanyan: That means instead of just finding one value for S˚, you calculate a probability distribution for S˚ based on the orbital noise parameters >Ref:two thousand six hundred ten point one zero nine two two#pg4>.

Vera: It moves us away from that idea that there’s a single deterministic threshold and toward understanding a probabilistic range of thresholds >Ref:two thousand six hundred ten point one zero nine two two#pg6>.

Jocitudin: And they also point out that the transition is smeared out over this range of ∆S˚ as the mean eccentricity increases, which suggests that abrupt deterministic transitions might actually be transitioning into a region in a noisy system >Ref:two thousand six hundred ten point one zero nine two two#pg10>.

Jocelyn: That's a significant shift in how we approach the stability of these kinds of planetary climates when orbital parameters are uncertain >Ref:two thousand six hundred ten point one zero nine two two#pg3>.

Subrahmanyan: This implies that we might have many high eccentricity planetary systems that are perpetually transient rather than strictly binary states >Ref:two thousand six hundred ten point one zero nine two two#pg4>. That's a new way to categorize these systems >Ref:two thousand six hundred ten point one zero nine two two#pg4>.

Vera: It really challenges the idea that we can neatly classify these systems into just snowball or non-snowball categories >Ref:two thousand six hundred ten point one zero nine two two#pg4>. That's a big shift in how we think about stability.

Jocelyn: This is a significant shift in how we approach the stability of these kinds of planetary climates when orbital parameters are uncertain >Ref:two thousand six hundred ten point one zero nine two two#pg3>.

Subrahmanyan: So, this implies that you might have many high eccentricity planetary systems in a perpetually transient state rather than strictly binary states >Ref:two thousand six hundred ten point one zero nine two two#pg4>. That's a new way to categorize these systems >Ref:two thousand six hundred ten point one zero nine two two#pg4>.

Subrahmanyan: The paper also points out that the effects of stochastic eccentricity on the bifurcation become negligible when the mean eccentricity is below zero point one >Ref:two thousand six hundred ten point one zero nine two two#pg6>. That’s a clear threshold for where we can trust those simpler deterministic models >Ref:two thousand six hundred ten point one zero nine two two#pg4>.

Vera: And they show that this effect of stochasticity on the bifurcation becomes negligible for mean eccentricities below zero point one >Ref:two thousand six hundred ten point one zero nine two two#pg6>. That’s a crucial boundary to remember >Ref:two thousand six hundred ten point one zero nine two two#pg6>.

Jocelyn: So, the paper is showing that the transition is "smeared out" over a range of ∆S˚ as the mean eccentricity increases, which is consistent with an abrupt deterministic transition transitioning into a region in a noisy system >Ref:two thousand six hundred ten point one zero nine two two#pg10>.

Subrahmanyan: This work implies that we might have many high eccentricity planetary systems in a perpetually transient state >Ref:two thousand six hundred ten point one zero nine two two#pg6>. It’s a significant finding for how we model planetary evolution across different stability regimes >Ref:two thousand six hundred ten point one zero nine two two#pg4>.

Vera: We've covered the main points of this paper, "Stochastic Snowball States in Planetary Climate", and it’s clear that orbital noise plays a distinct role depending on the mean eccentricity >Ref:two thousand six hundred ten point one zero nine two two#pg6>.

Jocelyn: It’s a reminder that when we think about climate stability, we need to incorporate those orbital fluctuations into our models for better forecasting >Ref:two thousand six hundred ten point one zero nine two two#pg4>.

Subrahmanyan: Thanks for joining us as we discussed "Stochastic Snowball States in Planetary Climate".

The paper's improvements: Vera: So, to recap this whole thing on "Stochastic Snowball States in Planetary Climate", the main conclusion is that adding orbital noise—that stochastic eccentricity—fundamentally changes how these planetary climate models behave >Ref:two thousand six hundred ten point one zero nine two two#pg4>.

Jocelyn: Exactly. It’s not just a small tweak; it means the dynamics are quantitatively and qualitatively different from what you get with just a fixed, deterministic orbit when the mean eccentricity is less than zero point one >Ref:two thousand six hundred ten point one zero nine two two#pg6>.

Subrahmanyan: And that noise amplifies the asymmetry and intermittency in radiative flux forcing, which means it takes longer for these systems to settle into a stable seasonal cycle >Ref:two thousand six hundred ten point one zero nine two two#pg5>.

Vera: Right, so if you have high variance in eccentricity, that time to reach a steady state jumps from three years up to nine years when the annual fluctuation variance is zero point two >Ref:two thousand six hundred ten point one zero nine two two#pg5>. That’s a pretty big stretch for climate response times >Ref:two thousand six hundred ten point one zero nine two two#pg3>.

Jocelyn: And they show that instead of a single stellar flux S˚ causing the transition from Snowball to non-Snowball state, it happens over a range of forcing values, ∆S˚ >Ref:two thousand six hundred ten point one zero nine two two#pg6>. That gives us a probabilistic window instead of just one sharp number >Ref:two thousand six hundred ten point one zero nine two two#pg10>.

Subrahmanyan: That's important because it means that as the mean eccentricity grows above zero point one, the probability of actually reaching that steady state and experiencing the Snowball to non-Snowball bifurcation actually decreases >Ref:two thousand six hundred ten point one zero nine two two#pg6>.

Vera: So they conclude that for any mean eccentricity as σyear increases, the probability of converging to a steady state within one hundred years in one thousand realizations drops and that bifurcation region ∆S˚ gets wider >Ref:two thousand six hundred ten point one zero nine two two#pg6>. That’s a pretty big stretch for climate response times >Ref:two thousand six hundred ten point one zero nine two two#pg3>.

Jocelyn: It’s suggesting that there might be many high eccentricity planetary systems that are perpetually transient, not just strictly binary states >Ref:two thousand six hundred ten point one zero nine two two#pg4>. That changes how we classify these worlds >Ref:two thousand six hundred ten point one zero nine two two#pg4>.

Vera: So, the main point of "Stochastic Snowball States in Planetary Climate" is that orbital noise fundamentally alters the stability landscape and smears out those sharp transition points >Ref:two thousand six hundred ten point one zero nine two two#pg6>.

Jocelyn: It’s a reminder that we need to incorporate those orbital fluctuations into our models for better forecasting when we're looking at planets with significant eccentricity variations >Ref:two thousand six hundred ten point one zero nine two two#pg4>.

Subrahmanyan: If you want to look at how orbital uncertainty impacts planetary climate stability, this paper provides the framework for that investigation >Ref:two thousand six hundred ten point one zero nine two two#pg4>.

Subrahmanyan: Thanks for joining us as we discussed "Stochastic Snowball States in Planetary Climate".

Conclusion: Vera: So we've seen how adding orbital noise fundamentally alters the stability landscape of these planetary climate models in "Stochastic Snowball States in Planetary Climate".

Jocelyn: Right, so it’s not just a small tweak; it means the dynamics are quantitatively and qualitatively different from what you get with just a fixed, deterministic orbit when the mean eccentricity is less than zero point one.

Subrahmanyan: And that noise amplifies the asymmetry and intermittency in radiative flux forcing, which means it takes longer for these systems to settle into a stable seasonal cycle.

Vera: If you have high variance in eccentricity, that time to reach a steady state jumps from three years up to nine years when the annual fluctuation variance is zero point two. That’s a pretty big stretch for climate response times.

Jocelyn: And they show that instead of a single stellar flux S˚ causing the transition from Snowball to non-Snowball state, it happens over a range of forcing values, ∆S˚. That gives us a probabilistic window instead of just one sharp number.

Subrahmanyan: That's important because as the mean eccentricity grows above zero point one, the probability of actually reaching that steady state and experiencing that Snowball to non-Snowball bifurcation actually decreases.

Jocelyn: It’s suggesting that there might be many high eccentricity planetary systems that are perpetually transient, not just strictly binary states. That changes how we classify these worlds.

Vera: So, the main point of "Stochastic Snowball States in Planetary Climate" is that orbital noise fundamentally alters the stability landscape and smears out those sharp transition points.

Jocelyn: It’s a reminder that when we think about climate stability, we need to incorporate those orbital fluctuations into our models for better forecasting when we're looking at planets with significant eccentricity variations.

Subrahmanyan: If you want to look at how orbital uncertainty impacts planetary climate stability, this paper provides the framework for that investigation.

Vera: We've covered the main points of "Stochastic Snowball States in Planetary Climate", and it’s clear that orbital noise plays a distinct role depending on the mean eccentricity.

Subrahmanyan: Thanks for joining us as we discussed "Stochastic Snowball States in Planetary Climate". Now, next time we’ll look at how collisionless plasma interacts with black holes.

Duc Thach Son Vu, John S. Wettlaufer

Department of Earth & Planetary Sciences, Yale University · Departments of Earth & Planetary Sciences, Mathematics and Physics, Yale University · Nordita, KTH Royal Institute of Technology and Stockholm University

astro-ph.EP

Submitted: 2026-10-07

Updated: 2026-10-07

Comments: 30 pages, 19 figures

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 77/100

The gist: The gist The stochastic dynamics are quantitatively and qualitatively different than the deterministic dynamics apart from mean eccentricities less than 0.1, and fluctuations in eccentricity amplify

Key concepts

Stochastic Dynamics: This refers to the random, unpredictable nature of orbital movements, specifically eccentricity fluctuations. Instead of smooth, predictable orbits, these fluctuations introduce noise into the climate model's energy balance equations. This noise causes the system's behavior to deviate from what a perfectly predictable (deterministic) model would show.
Snowball Bistability: This is a specific climate phenomenon where a planetary system can exist in two stable states: one where ice covers the planet (Snowball state) and one where it does not. The paper explores how random orbital changes affect this switch, showing that the conditions for this bistability change when eccentricity is not fixed.
Radiative Flux Forcing: This describes the imbalance between incoming solar energy (shortwave radiation) and outgoing planetary energy (longwave radiation). Fluctuations in eccentricity amplify how unevenly this forcing is distributed across the planet, making the climate response more erratic and intermittent.

Terminology

Outgoing Longwave Radiation — 2× in this paper
Stochastic Dynamics — 2× in this paper · explained in 2 episodes
Alternation — 1× in this paper · explained in 1 episode
Boltzmann Equation — 1× in this paper · explained in 2 episodes
CLM3 Global Climate Model — 1× in this paper
Core — 1× in this paper · explained in 2 episodes
Hysteresis — 1× in this paper · explained in 3 episodes

Summary

The gist The stochastic dynamics are quantitatively and qualitatively different than the deterministic dynamics apart from mean eccentricities less than 0.1, and fluctuations in eccentricity amplify the asymmetry and intermittency in radiative flux forcing, thereby increasing the time it takes to reach a stable seasonal cycle <ref:2610.10922#pg6>.

Motivation

The study is motivated by the recent finding that at moderate eccentricity (0.2–0.3) the planetary Snowball bistability vanishes <ref:2610.10922#pg5>. The authors introduced stochasticity of orbital eccentricity into a simplified version of the energy balance model of Eisenman and Wettlaufer (2009) to study a general Earth-like planet. This approach is motivated by the need to analytically demonstrate what cycles are important on which timescales and reveal the nature of the coupling between core phenomena operating on these time scales <ref:2610.10922#pg2>.

Energy Balance Model and Bifurcation

The model uses a first-order nonautonomous ordinary differential equation, dE/dt = f(t, E), where E represents energy stored in sea ice or ocean mixed layer as sensible heat <ref:2610.10922#pg3>. This equation involves the ice-albedo dependent difference between incoming shortwave radiation and outgoing longwave radiation, augmented by sensible and latent heat fluxes <ref:2610.10922#pg4>. A bifurcation occurs when the system converges to a stable solution that is a non-Snowball state, defined as the absolute difference in E between the beginning and end of a given orbital year not exceeding 0.1 Wm 2yr <ref:2610.10922#pg7>.

Analysis of Stochastic Eccentricity

The stochastic nature of orbits is introduced by using Kepler’s equation to approximate fluctuations in eccentricity, specifically a two-timescale model characterized by annual and daily variations with standard deviations σday and σyear <ref:2610.10922#pg4>. When variance σ squared increases, the stochastic and deterministic solutions deviate, amplifying the asymmetry and intermittency in radiative flux forcing <ref:2610.10922#pg5>. This amplification increases the time it takes to reach a stable seasonal cycle from 3 to 9 years as σyear increases from 0 to 0.2 <ref:2610.10922#pg5>.

Impact of Eccentricity and Flux

Key findings regarding the system's behavior include:

Apart from mean eccentricities less than 0.1, the stochastic dynamics are quantitatively and qualitatively different than the deterministic dynamics <ref:2610.10922#pg6>.
For a fixed mean eccentricity, fluctuations in eccentricity amplify radiative flux forcing asymmetry and intermittency, increasing the time to reach a stable seasonal cycle <ref:2610.10922#pg5>.
As the mean eccentricity increases above 0.1, the stochastic transition from Snowball to non-Snowball state occurs over a range of stellar flux ∆S˚ rather than at a particular stellar flux S˚ for the deterministic case <ref:2610.10922#pg6>.
As the mean eccentricity increases from 0.1 and variance of annual eccentricity fluctuations increases, ∆S˚ increases and the probability of reaching a steady state and experiencing a Snowball to non-Snowball bifurcation decreases <ref:2610.10922#pg6>.

Conclusion

The research concludes that for any mean eccentricity as σyear increases, the probability of convergence to a steady state within 100 years in 1000 realizations decreases and the bifurcation region ∆S˚ broadens, implying that there may be many high eccentricity planetary systems in a perpetually transient state <ref:2610.10922#pg6>. The stochasticity fundamentally influences the nature of deterministic steady-state solutions corresponding to perennial ice, seasonal ice and ice-free states <ref:2610.10922#pg4>. The results demonstrate that increasing the mean orbital eccentricity reduces the deterministic critical annual mean stellar flux S˚ for the bifurcation from Snowball to non-Snowball state <ref:2610.10922#pg10>. The effects of stochastic eccentricity on the bifurcation become negligible for mean eccentricities below 0.1 <ref:2610.10922#pg6>. The transition is smeared out over a range of ∆S˚ as the mean eccentricity increases, consistent with an abrupt deterministic transition transitioning to a region in a noisy system <ref:2610.10922#pg10>. The paper implies that there may be many high eccentricity planetary systems in a perpetually transient state <ref:2610.10922#pg6>. The authors chose the simplest treatment: a two time-scale eccentricity characterized by annual and daily variations with distinct standard deviations σday! σyear <ref:2610.10922#pg4>.

Acknowledgements

The authors gratefully acknowledge comments by Dorian Abbot and support from Yale University and the Swedish Research Council under Grant No.638-2013-9243 <ref:2610.10922#pg17>. The references listed include works on orbital dynamics, climate modeling, and exoplanet observations <ref:2610.10922#pg18>. The paper includes supplemental material detailing high-frequency fluctuations in eccentricity based on circumbinary planets like Kepler-16b, Kepler-34b and Kepler-35b <ref:2610.10922#pg26>. The NASA database provides histograms for exoplanet masses and eccentricities for reference. The paper also includes a detailed analysis of fluctuations in eccentricity using Bessel functions to approximate the orbital motion <ref:2610.10922#pg20>. The paper concludes with a discussion on the multiple timescale fluctuations in circumbinary systems, such as apsidal beats controlling the long term modulation <ref:2610.10922#pg27>. The paper provides critical values for S˚ for energy convergence and bifurcation thresholds in Table A.2 <ref:2610.10922#pg20>. The analysis of noise values shows that when σ is sufficiently small, most of the values of u are sharply centered about 1r squared and variance σU! 1 <ref:2610.10922#pg24>. The distribution of noise values exhibits symmetric normal distributions around the mean values obtained from Kepler’s equations and formulas for orbital distance <ref:2610.10922#pg24>. The paper illustrates the case with e = 0.3 and σ = 0.1 at M = 1.4π <ref:2610.10922#pg24>. The distribution of noise values shows slight skewness for r and r squared <ref:2610.10922#pg25>. The paper notes that when σ is larger, we do not obtain Gaussian distributions <ref:2610.10922#pg25>. The paper provides a quantitative argument for high-frequency fluctuations in eccentricity based on circumbinary planets Kepler-16b, Kepler-34b and Kepler-35b <ref:2610.10922#pg26>. The paper includes supplemental figures showing the percentage of 1,000 stochastic paths that converge when e = 0.1, 0.2, 0.3, 0.4, 0.5, and 0.6 shown as a color map <ref:2610.10922#pg28>. The paper includes supplemental figures showing the percentage of 1,000 stochastic paths that converge within 100 years when e = 0.1, 0.2, 0.3, 0.4, and 0.5 shown as a color map <ref:2610.10922#pg28>. The paper includes supplemental figures showing the percentage of bifurcations in 1,000 stochastic paths when e = 0.1 through e = 0.6 shown as a color map <ref:2610.10922#pg29>. The paper includes supplemental figures showing masses of exoplanets in the NASA (2025) database and eccentricities of exoplanets in the NASA (2025) database. The paper includes Figure C.19 showing eccentricity vs Earth Mass of exoplanets in the NASA (2025) database. The paper includes Figure C.18 showing eccentricities of exoplanets in the NASA (2025) database. The paper includes Figure C.17 showing masses of exoplanets in the NASA (2025) database. The paper includes Figure C.16 showing the percentage of bifurcations in 1,000 stochastic paths when e = 0.1 through e = 0.6 shown as a color map <ref:2610.10922#pg29>. The paper includes Figure C.15 showing the percentage of 1,000 stochastic paths that converge within 100 years when e = 0.1 through e = 0.6 shown as a color map <ref:2610.10922#pg28>. The paper includes Figure C.14 showing the percentage of 1,000 stochastic paths that converge when e = 0.1 through e = 0.6 shown as a color map <ref:2610.10922#pg28>. The paper includes Figure C.13 showing the distribution of noise values e, ψ, cos ψ,r,r squared and 1r squared with σ = 0.2 at M = 1.4π <ref:2610.10922#pg24>. The paper includes Figure B.13 showing the distribution of noise values of e, ψ, cos ψ,r,r squared and 1r squared with σ = 0.2 at M = 1.4π <ref:2610.10922#pg25>. The paper includes Figure B.12 showing the distribution of noise values e, ψ, cos ψ,r,r squared and 1r squared with σ = 0.1 at M = 1.

Improvements for AI systems

The improved system can predict that apart from mean eccentricities less than 0.1, the stochastic dynamics are quantitatively and qualitatively different than the deterministic dynamics, allowing for a more nuanced assessment of planetary stability beyond simple deterministic thresholds.
The system can quantify how fluctuations in eccentricity affect climate transitions by demonstrating that they amplify the asymmetry and intermittency in radiative flux forcing, thereby increasing the time it takes to reach a stable seasonal cycle. This enables better forecasting of climate response times under uncertain orbital conditions.
The improved model can predict that the stochastic transition from the Snowball to non-Snowball state occurs over a range of mean stellar flux ∆S˚ rather than at a particular stellar flux S˚ for the deterministic case, providing a probabilistic range for critical forcing thresholds instead of a single deterministic value.
The system can estimate that "as the mean eccentricity increases from 0.1 and the variance of the annual eccentricity fluctuations increases, ∆S˚ increases and the probability of the system reaching a steady state and experiencing a Snowball to non-Snowball bifurcation decreases." This allows for risk assessment regarding planetary stability when orbital parameters are uncertain or evolving.
The system can identify that there may be many high eccentricity planetary systems in a perpetually transient state, enabling the classification of certain exoplanetary systems as being in a perpetually transient state rather than strictly binary (Snowball/non-Snowball) states.

Abstract

Motivated by the recent finding of Ji and Abbot (2025) that at moderate eccentricity (0.2--0.3) the planetary Snowball bistability vanishes, we have introduced stochasticity of the orbital eccentricity into a simplified version of the energy balance model of Eisenman and Wettlaufer (2009) to study a general Earth-like planet. The stochastic eccentricity is approximated by two dominant time scales, annual and daily, and is introduced through an analysis of the Kepler equation. We find that, apart from mean eccentricities less than 0.1, the stochastic dynamics are quantitatively and qualitatively different than the deterministic dynamics. For a fixed mean eccentricity, fluctuations in eccentricity amplify the asymmetry and intermittency in radiative flux forcing, thereby increasing the time it takes to reach a stable seasonal cycle. As the mean eccentricity increases above 0.1, the stochastic transition from the Snowball to non-Snowball state occurs over a range of mean stellar flux ΔS* rather that at a particular stellar flux S* for the deterministic case. Moreover, as the mean eccentricity increases from 0.1 and the variance of the annual eccentricity fluctuations increases, ΔS* increases and the probability of the system reaching a steady state and experiencing a Snowball to non-Snowball bifurcation decreases. This implies that there may be many high eccentricity planetary systems in a perpetually transient state.