transported by processes other than wind ( 28 ),
but the effect of these processes would fall
within the uncertainties of our wind speed
measurement for 2MASS J1047+21.
We detect a positive (eastward) wind speed
at >98% confidence (2.1s). As with Jupiter, the
IR period of 2MASS J1047+21 is shorter than
its radio period, indicating an atmosphere
that is rotating faster than the interior. The
wind speed on 2MASS J1047+21 is higher than
on the gas giant planets in the Solar System
( 16 , 29 ). Analytic theory predicts that larger
atmospheric heat fluxes lead to higher wind
speeds ( 7 ). Three-dimensional numerical sim-
ulations show that zonal winds of hundreds
of meters per second can occur when strong
convective forcing and/or weak damping (either
radiative or frictional) promote the formation
of atmospheric jet streams ( 9 ).
Our method for determining the wind speed
can in principle also be applied to exoplanets,
which have rotation rates and periodic varia-
bility similar to those of brown dwarfs ( 1 , 4 ).
Exoplanets with masses similar to Jupiter’s
have magnetic field strengths of around 100
G( 30 ), which is weaker than the kilogauss
fields of brown dwarfs ( 26 ). Because the fre-
quency at which ECMI emission is detected
is proportional to the magnetic field strength,
we expect exoplanets to emit at lower radio
frequencies.
REFERENCES AND NOTES
- I. A. G. Snellenet al.,Nature 509 , 63–65 (2014).
- S. A. Metchevet al.,Astrophys. J. 799 , 154 (2015).
- B. A. Billeret al.,Astron. J. 155 , 95 (2018).
- Y. Zhouet al.,Astron. J. 157 , 128 (2019).
5. J. Radigan, D. Lafrenie’re, R. Jayawardhana, E. Artigau,
Astrophys. J. 793 , 75 (2014).
6. Y. Kaspiet al.,Nature 555 , 223–226 (2018).
7. A. P. Showman, Y. Kaspi,Astrophys. J. 776 , 85 (2013).
8. G. Lee, C. Helling, I. Dobbs-Dixon, D. Juncher,Astron.
Astrophys. 580 , A12 (2015).
9. A. P. Showman, X. Tan, X. Zhang,Astrophys. J. 883 ,4
(2019).
10. T. Louden, P. J. Wheatley,Astrophys. J. 814 , L24 (2015).
11. J. Radiganet al.,Astrophys. J. 750 , 105 (2012).
12. D. Apaiet al.,Science 357 , 683–687 (2017).
13. P. K. Seidelmannet al.,Celestial Mech. Dyn. Astron. 98 ,
155 – 180 (2007).
14. C. A. Jones,Icarus 241 , 148–159 (2014).
15. T. Guillotet al.,Nature 555 , 227–230 (2018).
16. J. Tollefsonet al.,Icarus 296 , 163–178 (2017).
17. E. Bergeret al.,Astrophys. J. 695 , 310–316 (2009).
18. G. Hallinanet al.,Astrophys. J. 684 , 644–653 (2008).
19. J. D. Nicholset al.,Astrophys. J. 760 , 59 (2012).
20. M. M. Kaoet al.,Astrophys. J. 818 , 24 (2016).
21. H. Geet al.,Astron. J. 157 , 89 (2019).
22. M. Gillonet al.,Astron. Astrophys. 555 , L5 (2013).
23. Materials and methods are available as supplementary
materials.
24. J. C. Filippazzoet al.,Astrophys. J. 810 , 158 (2015).
25. P. K. G. Williams, E. Berger,Astrophys. J. 808 , 189
(2015).
26. M. M. Kao, G. Hallinan, J. S. Pineda, D. Stevenson, A. Burgasser,
Astrophys. J. 237 (suppl.), 25 (2018).
27. J. M. Vos, K. N. Allers, B. A. Biller,Astrophys. J. 842 , 78
(2017).
28. D. S. Choi, A. P. Showman, A. R. Vasavada, A. A. Simon-Miller,
Icarus 223 , 832–843 (2013).
29. E. García-Melendo, S. Pérez-Hoyos, A. Sánchez-Lavega,
R. Hueso,Icarus 215 , 62–74 (2011).
30. P. W. Cauley, E. L. Shkolnik, J. Llama, A. F. Lanza,Nature
Astronomy 3 , 1128–1134 (2019).
31. P. K. G. Williams, K. N. Allers, J. M. Vos, B. A. Biller, Dynamic
spectrum of 2MASSW J1047539+212423 observed by the
Karl G. Jansky Very Large Array, as a Numpy save file, Zenodo
(2020).
ACKNOWLEDGMENTS
This study was based on observations made with the Spitzer
Space Telescope, which is operated by the Jet Propulsion
Laboratory, California Institute of Technology, under a contract
with NASA. The National Radio Astronomy Observatory is a
facility of the National Science Foundation operated under
cooperative agreement by Associated Universities, Inc. This
work benefited from the Exoplanet Summer Program in the Other
Worlds Laboratory (OWL) at the University of California,
Santa Cruz, funded by the Heising-Simons Foundation. We
acknowledge S. Beiler, E. Berger, M. Kao, L. Lanwermeyer,
M. Marley, S. Metchev, B. Pantoja, D. Powell, E. Shkolnik, A. Showman,
X. Tan, J. Tolman, and X. Zhang for useful conversations. We also
thank the anonymous reviewers, whose thoughtful comments
improved this manuscript.Funding:Support for Program numbers
14188 and 14686 was provided by NASA through a grant from
the Space Telescope Science Institute, which is operated
by the Association of Universities for Research in Astronomy,
Incorporated, under NASA contract NAS 5-26555. J.M.V.
acknowledges funding support from the National Science
Foundation under award AST-1614527 and Spitzer Cycle 14
Caltech/JPL Research Support Agreement 1627378.Author
contributions:K.N.A. planned and proposed the Spitzer Space
Telescope observations, conducted an independent check of the
data reduction, calculated the wind speeds, and wrote the
manuscript. J.M.V. led the reduction and analysis of the Spitzer
data, including the MCMC modeling, production of Figs. 1
and figs. S1 to S5, and associated text. B.A.B. performed the
Lomb-Scargle analysis, produced Fig. 2 and fig. S6, and wrote part
of the manuscript. P.K.G.W. planned, proposed, and reduced the
VLA data, conducted the radio pulse time-of-arrival (TOA) analysis,
produced Figs. 3 and 4, and wrote the corresponding text.
Competing interests:We declare no competing interests.Data
and materials availability:The VLA data are available via the
NRAO Science Data Archive http://archive-new.nrao.edu/ under
Project Code 18A-427. The Spitzer data are available via the
Spitzer Heritage Archive http://sha.ipac.caltech.edu using
Program IDs 13031 and 13231. The light curve data and code
used for analysis and figure production are available at
https://github.com/johannavos/BDwindspeeds/ and archived
at https://doi.org/10.5281/zenodo.3700897. The VLA
dynamic spectrum (an intermediate data product) is archived
at Zenodo ( 31 ).SUPPLEMENTARY MATERIALS
science.sciencemag.org/content/368/6487/169/suppl/DC1
Materials and Methods
Supplementary Text
Tables S1 to S3
Figs. S1 to S6
References ( 32 – 58 )
26 September 2019; accepted 12 March 2020
10.1126/science.aaz2856172 10 APRIL 2020•VOL 368 ISSUE 6487 sciencemag.org SCIENCE
Fig. 4. Alignment of pulses for possible periods of the radio data.
(AtoE) The data (points) from Fig. 3 are folded by periods that differ by
0.007 hours for comparison of the phase (fraction of a rotation period) of radio
pulses. Each color (arbitrarily) indicates the data from a separate rotation.
1 suncertainties are plotted as vertical lines for each data point. (C) The radio
pulses folded at the period preferred by our analysis ( 23 ). At this period,
the pulses from all of our observed rotations are aligned at a phase of
around 0.5. The pulses are aligned at a phase of around 0.5 for periods of
1.751 to 1.765 hours (B to D), whereas the phases of the pulses are not aligned
for periods outside of this range (A and E).RESEARCH | REPORTS