1. Sustaining eruptions on Enceladus Edwin Kite (U.Chicago), Allan Rubin (Princeton) Geysers on Enceladus draw water from a subsurface ocean, but the sustainability of conduits linking ocean and surface is not understood Today, use a model of tidally flexed slots that puncture the ice shell to interpret the tidal phase curve of the eruptions, the surprising persistence of fissure eruptions, and the total power output of the tiger stripe terrain.

2. ocean surface habitability astrobiology Today: How are curtain eruptions sustained on 10 1 -10 6 yr timescales? ice shell Parkinson et al. 2008 Tsou et al. 2012 McKay et al. 2014 Not addressed today: How is the ocean sustained on 10 7 – 10 10 yr timescales? Meyer & Wisdom 2007, O’Neill & Nimmo 2008, Behounkova et al. 2012 Kite & Rubin, in prep. conduit = crack? Prevailing view. e.g. Hurford et al. 2007, Nimmo et al. 2007, Olgin et al. 2011, Smith-Konter & Pappalardo 2008 water source = sub-ice ocean e.g. Postberg et al. 2009, 2011; Hu et al. 2015; Iess et al. 2014; Porco et al. 2014; Waite et al. 2009; Nimmo & Spencer 2013; Zolotov, 2007

3. ? phase shift 55° (Q ~ 1) Challenge for the prevailing view: Understanding tidal modulation of eruptions (all four tiger stripes erupt as “curtains” throughout orbit; Spitale et al. accepted) smaller plume grains larger plume grains (period: 1.3 days) Crack models are hard to reconcile with curtain eruptions maintained throughout orbit NASA/JPL/Hedmanet al. 2013 Hurford LPSC 2015

4. Challenge for the prevailing view: energy balance at water table x z water table (35±5) km Iess et al. 2014 Kite & Rubin, in prep. (4.4±0.2) GW Howett et al. 2014

5. Alternative: Melted-back slot Compression Tension ocean supersonic plume water level falls water level rises x z Attractive properties: Slot width lags tidal cycle Slot does not close Turbulent dissipation heats slot Pumping disrupts ice formation Kite & Rubin, in prep.

6. Daily tidal cycle of water in tiger stripes Change in slot width Width change due to water flow from ocean into slot Width change due to tides from Saturn Width change due to back-force from ice shell elasticity Slots interact elastically: use boundary- element method Kite & Rubin, in prep. Tiger stripes can be approximated as straight, parallel and in-phase

7. Daily cycle of tidal flow of water in slots Wide slots track tidal forcingNarrow slots lag tidal forcing by 8 hours width change (m) volume change (fractional) volume change (fractional) width change (m) 1.3 days forcing 5m wide at zero stress ½ m wide at zero stress Kite & Rubin, in prep.

8. T urbulent dissipation of tidally-pumped vertical flow inside tiger stripes explains power output, phase lag and sustainability of the eruptions Colored disks = diurnal width ratio Increasing slot width Observed power Wide slots Narrow slots Observed phase lag Changing “active slot” length Kite & Rubin, in prep.

9. Long-lived water-filled slots have tectonic consequences COLD, STRONG ICE WARM, WEAK ICE x z Kite & Rubin, in prep.

10. Long-lived water-filled slots have tectonic consequences COLD, STRONG ICE WARM, WEAK ICE x z Kite & Rubin, in prep. Tectonic feedback between subsidence and melt-back buffers Enceladus power to 5 GW (no tuning, no free parameters)

11. Summary: Slot model explains and links sustainability of volcanism on 10 yr - 10 6 yr timescales Gravity (Iess et al. 2014) Observed power Predicted Myr-average power output matches observed phase lag slot aperture varies by >1.5 No tuning No free parameters Kite & Rubin, in prep.

12. Testable predictions Endogenic thermal emission will be absent between tiger stripes. There should be no correlation between the magnitude of emission and local tiger-stripe orientation, a prediction that distinguishes the slot model from all crack models. The slot model predicts a smooth distribution of thermal emission, in contrast to the spotty emission near jets that would be expected if flow is concentrated in pipes. The pattern of spatial variability should be steady, in contrast to bursty hypotheses (e.g. Matson et al. 2012).Vapor flux should covary with ice-grain flux. For late 2015 / early 2016 Cassini flybys: Kite & Rubin, in prep. For numerical experiments: How does changing water level and conduit width affect gas-dynamic flow in vent? (Ingersoll & Pankine, 2010) What testable consequences would long-lived tiger stripes have for “ropy terrain” in between tiger stripes?(Barr & Pruess, 2010)

13. Conclusions Paul Schenk / LPI Turbulent dissipation may explain the phase curve of Enceladus and the maintenance of fissure eruptions over geological timescales. With long-lived slots, Myr-average power is buffered by tectonic feedback to ~5 GW, equal to observed power (suggesting long-term stability). Testing habitability on Enceladus (or Europa) ultimately requires access to ocean materials - easier if turbulent dissipation props active fissures open for ≫ Kyr. We thank Terry Hurford, Karl Mitchell, Alyssa Rhoden, Joseph Spitale, Robert Tyler, and Steve Vance. http://geosci.uchicago.edu/~kite

14. Supplementary slides

15. Tectonic feedback between subsidence and meltback buffers the South Polar terrain to 5 GW Equilibrium power output 5 km 100 km Ice shell thickness: Gravity constraint of shell thickness:

16. Highly simplified ice flow: assuming cools the ice shell Initial temperature profile (conductive) Steady-state temperature profile (Subsidence-perturbed) Initial ice inflow rate Steady-state ice inflow rate Tectonic feedback between subsidence and meltback buffers South Polar terrain power to 5 GW Kite & Rubin, in prep.

17. Open questions Source: What is the water source for Enceladus’ eruptions? Ocean water is exposed to space, raising energy balance problems Plumbing system: How can conduits between ocean and surface avoid freezing shut? (w/ Allan Rubin) Tidal heating is the only plausible candidate, but location of heating is poorly constrained. Engine: What powers Enceladus volcanism? Sodium and nano-silica tell us that the source is a subsurface ocean, not clathrates or sublimation Tidal heating is the only plausible candidate Turbulent dissipation within tiger stripes can sustain the phase curve of Enceladus’ eruptions.

19. Meyer & Wisdom, Icarus, 2008 History of the Enceladus:Dione resonance

20. Clathrate-source and solid-sublimation explanations face insuperable challenges Salt Nano-silica I:G ratio Thermal

21. Pipe model and slot model Option: Array of unresolved pipes Area changes by 10^-4 Option: Each tiger stripe is one slot Area changes by order unity under 1-bar pressure cycle Apertures < 10 m (not to scale) 130 km 35 km

22. 1 tiger stripe Extensional normal stress (Pa)

23. Rudolph & Manga, Icarus 2009 inferred from gravity

24. Models of tidal modulation Nimmo et al., Astron. J. 2014 supersonic ascent time

25. Crack models are falsified by eruptions at periapse

26. All existing models fail! Model A Model B Model C Porco et al. Astron. J. 2014 Should erupt Should not erupt Should erupt Should not erupt Should erupt Should not erupt

27. ancient, cratered South polar terrain: no craters 4 continuously- active “tiger stripes” Cryo-volcanism on Enceladus has deep tectonic roots Density = 1.6 g/cc water source = subsurface ocean Postberg et al. 2009, 2011; Hu et al. 2015 Iess et al. 2014; Porco et al. 2014; Waite et al. 2009; Nimmo & Spencer 2013

28. Closest distance to SaturnFurthest distance from Saturn Enceladus period = 1.3 days Enceladus orbital eccentricity = 0.0047 Tidal stress amplitude ~ 1 bar tiger stripes = time-averaged shape eccentricity tide only thin-shell approximation k 2 appropriate for global ocean Looking down on South Pole Crack models are hard to reconcile with curtain eruptions at periapse

29. Paul Schenk / LPI / USRA 130 km 35 km Slots interact elastically: used displacement-discontinuity method Kite & Rubin, in prep. Crouch & Starfield, 1983 Rubin & Pollard, 1988

30. (4.4±0.2) GW excess thermal emission from surface fractures Spencer & Nimmo AREPS 2013 Spitale et al., accepted Porco et al. Astron. J. 2014 Abramov & Spencer 2009 South polar projection Hotspots up to 200K No liquid water at surface Latent heat represented by plumes < 1 GW to Saturn 130 km Howett et al. 2014

31. Zolotov, GRL 2007 Salt composition matches expectations for hydrothermal leaching Age of interaction unconstrained