Dynamics of Saturn s South Polar Vortex

Transcription

1 Dynamics of Saturn s South Polar Vortex Ulyana A. Dyudina 1, Andrew P. Ingersoll 1,Shawn P. Ewald 1, Ashwin R. Vasavada 2, Robert A. West 2, Anthony D. Del Genio 3, John M. Barbara 3, Carolyn C. Porco 4, Richard K. Achterberg 5, F. Michael Flasar 5, Amy A. Simon-Miller 5, Leigh N. Fletcher 2,6 1 Division of Geological and Planetary Sciences, California Institute of Technology, , Pasadena, CA 91125, USA. 2 Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109, USA. 3 Goddard Institute for Space Studies, NASA, 2880 Broadway, New York, NY 10025, USA. 4 Space Science Institute, 4750 Walnut Street, Suite 205, Boulder, CO 80301, USA. 5 NASA Goddard Space Flight Center, Code 693, Greenbelt, MD 20771, USA. 6 Department of Physics, University of Oxford, Clarendon Laboratory, Parks Road, Oxford OX1 3PU, UK. To whom correspondence should be addressed; ulyana@gps.caltech.edu. 1

2 We present observations of Saturn s south polar vortex (SPV) showing that it shares some properties with terrestrial hurricanes - cyclonic circulation, warm central region (the eye) surrounded by a ring of high clouds (the eyewall), and convective clouds outside the eye. The polar location and absence of an ocean are major differences. It also shares properties with the polar vortices on Venus - polar location, cyclonic circulation, warm center, and long lifetime, but the Venus vortices have cold collars and are not associated with convective clouds. The SPV s combination of properties is unique among vortices in the solar system. 2

3 Most planets with an atmosphere have large vortices. Here we present observations of Saturn s south polar vortex (SPV) showing that it has a unique combination of properties, resembling some vortices in some respects but not any other vortex in all respects. Our data are from observations over three hours by Cassini on 11 October A falsecolor image of cloud heights (Fig. 1A) shows a dark, red central eye indicating a nearly cloud-free upper atmosphere above lower, tropospheric clouds. The blue-green ring outside the eye indicates high clouds and haze, which is consistent with uplifted air. The eye has two concentric boundaries. The eyewall clouds cast shadows that followed the sun in a counterclockwise direction as the planet turned. (See Fig. 1(B-D)). From the shadow lengths, we estimate that the outer wall is 40±20 km high and that the inner wall is 70±30 km high, about twice the pressure scale height of Saturn s atmosphere. The eyewall clouds seem to extend up to the tropopause, which is at the 100 mbar level (1). We tracked the motion of individual cloud features (2). The peak zonal velocity ū was 150±20 m s 1 near the outer eyewall. Absolute vorticity consists of two parts - a part ζ due to motion relative to the planet and a part f due to the planet s rotation (3). Up to latitude -85 the measured ū increased slightly faster than a constant absolute vorticity profile. Poleward of -85 ū increased more slowly. Constant absolute vorticity is consistent with horizontal stirring by eddies. The angular momentum in Saturn s south polar vortex decreased toward its center. We observed no poleward or equatorward mean motion. The relative vorticity ζ estimated from the measured ū was close to zero up to the edge of the eyewall. The puffy red clouds in Fig. 1A are anticyclones (2), with a vorticity of 1 ± s 1, which is 1/3 the magnitude of the planetary vorticity f but of opposite sign. This is consistent with a convective origin, since parcels rising from the convective interior should have ζ + f =0when they spread out in the upper troposphere (3). 3

4 Cassini CIRS data show that the vortex is anomalously warm, particularly just beneath the tropopause (by 5K) but also in the stratosphere (by 3-4K) (1). The warm central core means that the central low pressure, and with it the cyclonic circulation, should weaken with altitude if the flow is balanced. We searched for this weakening, but did not find it (2). The failure of the wind to weaken means that the centrifugal force at high altitudes is not completely balanced by the inward pressure force. This unbalanced force could drive an outward flow. The SPV is a warm-core feature with cyclonic relative vorticity. Like a terrestrial hurricane, it has an eye, eyewall clouds, and multiple convective clouds outside the eye. However hurricanes exist in the tropics, are not stationary, and derive their energy from interaction with the underlying ocean. The SPV is different from Jupiter s Great Red Spot and white ovals, which are anticyclones with uniformly high clouds at their centers (4). Observations do not cover the poles of Jupiter well enough to detect a possible vortex there. In some respects, the SPV resembles the polar vortices on Venus, which are cyclonic and have warm features at the poles, although the features are dipole-shaped, have cold collars, and are not surrounded by convective clouds (5). Neptune s atmosphere is warm poleward of 70 at altitudes near 100 mbar (6). The SPV is different from Earth s Arctic and Antarctic polar vortices, which are cold-core features that are not associated with clouds and/or convection. 4

5 References and Notes 1. L. Fletcher, et al., Science 319, 79 (2008). 2. The details of the cloud elevations, the eyewall height measurement, cloud tracking, vorticity measurements, and the cloud movie are available as supporting online material (SOM) at Science Online. 3. J. R. Holton, An Introduction to Dynamic Meteorology (Elsevier Academic Press, Amsterdam, ed. 4, 2004). 4. F. Bagenal, ed., Jupiter - The Planet, Satellites and Magnetosphere (Oxford University Press, 2001). 5. G. Piccioni, et al., Nature 450, 637 (2007). 6. G. S. Orton, T. Encrenaz, C. Leyrat, R. Puetter, A. J. Friedson, A & A 473, L5 (2007). 7. This research was supported by the NASA Cassini Project. Supporting Online Material SOM text Fig. S1 Movie S2 References 5

6 Figure 1: Images of Saturn s south polar clouds taken by the Cassini imaging science subsystem (ISS). The images have been map projected using polar stereographic projection. Latitudes are planetocentric. (A) False-color image from light at 889 nm, 727 nm, and 750 nm projected as blue, green, and red, respectively. In the original images sunlight was attenuated by a factor of e ( ) at the 80 mbar and 300 mbar levels for light at 889 and 727 nm, respectively. Sunlight passes through to deeper levels for light at 750 nm. Thus clouds below 300 mbar appear red, and high thin clouds appear blue or green (2). The eyewalls can be seen in all three color planes, and thus extend above the 80 mbar surface. (B-D) Time sequence showing shadows (the dark crescent-shaped areas inside the walls). The first map is taken on 11 October 2006 at 19 hr 42 min. The maps are labeled by the time lapsed since the first map. The white arrow shows the direction of propagation of the incident sunlight. 6

7

8 Supporting online material Earth-based telescopic observations (S1) revealed a hot spot at Saturn s south pole in Cassini imaging observations (S2, S3) revealed cyclonic rotation around the spot in Our observations from 11 October 2006 have spatial resolution of 20 km/pixel, about ten times the resolution of the previous ISS observations (S2, S3). The SPV eye in Fig. 1 has two boundaries. The inner boundary is oblong (major axis = 2400 km); the outer one is circular (diameter = 4200 km). The 889 nm, 727 nm, and 750 nm ISS filters cover different methane gas absorption bands (S4, S5). In the original images the sun was 15 above the horizon at the pole. We accounted for this slant illumination in calculating the attenuation due to methane absorption, which is used to estimate cloud heights (see also modeling results in (S3)). To reduce the effect of varying solar illumination across the image, each color plane in Fig. 1A is high-pass filtered at the spatial scale of 150 pixels (which is 300 km, or 0.3 of latitude). Figure 1(B-D) shows a sample of three images out of nine that we used to measure the wall heights. To obtain the eyewall height we multiplied each shadow length by the tangent of the solar elevation angle above the horizon, assuming the clouds inside the eye are flat and horizontal. In Fig. S1A, the points represent the zonal velocity (positive eastward) of individual cloud features. We tracked the clouds on 14 images in the continuum band filter at 750 nm taken within a 3-hour period. The average of the points is the mean zonal velocity ū. Without frictional losses by eddies, rings of air moving poleward would produce a profile with constant angular momentum, which is a much steeper curve than the curve in Fig. S1A and did not fit the data. Instead angular momentum decreased toward the pole. In Fig. S1B, the smooth curve is the mean relative vorticity ζ computed from the measured ū in Fig. S1A. Each point is the 1

9 relative vorticity of a puffy red cloud in Fig. 1A. We searched for the weakening of the cyclonic circulation with altitude using a 4-frame color movie S2 of images like the one in Fig. 1A, and found no difference in the wind with altitude, at least at -84 where there were features in the blue-green haze suitable for tracking. 2

10 Figure S1: Profiles of zonal velocity (eastward) and cyclonic vorticity (clockwise) around Saturn s south pole. The dashed vertical lines indicate the inner and outer eyewalls. (A) Zonal velocity measured by tracking clouds in a sequence of images over a 3-hour period. The solid curves are for constant absolute vorticity ζ + f starting at latitude φ 0 (values labeled on the curves) with ū =0and ζ =0at that point. (B) Relative vorticity ζ. The solid curve is a spline fit to the velocity data of Fig. S1A. The points are the puffy red clouds of Fig. 1A. To determine the relative vorticity of a puffy red cloud, we measured its angular velocity of rotation relative to the rotating planet. Twice this angular velocity is the vorticity of the cloud. We repeated the procedure three to four times for each cloud and assigned error bars from the residuals. 3

11 Movie caption Movie S2: A four-step movie combined from the color images similar to Fig. 1A taken within approximately 2 hr 20 min during the 11 October 2006 ISS observation. References S1. G. S. Orton, P. A. Yanamandra-Fisher, Science 307, 696 (2005). S2. A. R. Vasavada, et al., J. Geophys. Res. (Planets) 111, 5004 (2006). S3. A. Sánchez-Lavega, R. Hueso, S. Pérez-Hoyos, J. F. Rojas, Icarus 184, 524 (2006). S4. C. C. Porco, et al., Space Sci. Rev. 115, 363 (2004). S5. E. Karkoschka, Icarus 133, 134 (1998). 4