[3] The NTBs Jet in 2007 and 2008: Evidence on the structure of the jet and the nature of global upheavels.

John Rogers,

British Astronomical Association

Last year’s dramatic outbreak of the NTBs jetstream, displaying the super-fast speed that we call North Temperate Current D (NTC-D), was extremely similar to the last such outbreak, in 1990, but was even better observed by ground-based amateur imagers and by the Hubble Space Telescope (HST).  Here I report a new observation which seems to extend the parallels between them and to provide much better temporal and spatial resolution of the return of the jetstream to its more normal state.  The observation is the presence of faint, narrow ovals (‘lozenges’) on the NTBs.  These may be incipient anticyclonic ovals, i.e. jetstream vortices, such as appeared 21 months after the 1990 NTC-D outbreak with the standard speed that we call North Temerate Current C (NTC-C) [footnote 1].  This time they have been detected only 11 months after the 2007 NTC-D outbreak, with a speed intermediate between NTC-D and NTC-C. The rapidity of this change supports a model in which the jetstream has a permanent NTC-D speed below the cloud-tops, but the cloud-top wind speed alternates between the NTC-D speed (when outbreaks driven from below are frequent) and NTC-C speed at other times (sometimes, or perhaps always, when jetstream vortices are present).

Between 1991 and 2002 the NTBs carried a set of 7 stable jetstream vortices with a virtually constant speed of DL1 ~ -60 deg/mth (u3 = 125 m/s), and the peak jet speed alongside them was 135 m/s; this state comprises NTC-C. (See Rogers et al., 2006, & refs. therein.) The parameters of the 1990 and 2007 NTC-D outbreaks, and of the NTBs jet as observed before and after the outbreaks, are listed in Table 1.  These data are from the amateur observations (my book, & JUPOS charts for 2007, and the present observations).  Sanchez-Lavega et al.(2008) report much the same results from the amateur images, and also analyse HST images, showing that the latest outbreak actually began on 2007 March 25, and measuring the latitudinal profile of the jet on that date and again in June.


New observations:

The 2007 outbreak led to the revival of a prominent orange NTB(S), which is still very strong.  In 2008, the NTBs edge carries at least 6 pale orange lozenges, each about 8 deg. long (Figure 1).  Measurements on 3 of them give speeds as follows (DL1, deg/month):

-80 (+/-4; May 2-15); -78 (+/-6; May 3-13); -60 (+/-15; May 7-13).

Thus we can adopt a consensus speed DL1 = -78 (+/-6) deg/mth (u = 133 m/s). 

They developed during solar conjunction: at least some were present, with similar speed, in the earliest adequate images of the year (by A. Wesley on 2008 Feb.25-26).

This observation is relevant to two questions that arose after these outbreaks.

  Implications (1) How did the jetstream vortices arise? 

In both 1990 and 2007, dark patches with DL1 ~ -25 deg/mth appeared during the NTC-D outbreak with speeds of ~-25 deg/mth, close to NTC-C, so it was conjectured that these might have developed into the jetstream vortices.  However this seems unlikely.  In each outbreak, the dark patches ceased to be visible as the NTB revived as a smooth featureless belt, before the NTC-C vortices appeared.  Thus in the 1990/91 apparition, no such spots were visible.  In 2007, we were able to study the evolution of the dark patches with much better images: they never developed anticyclonic forms, and all but one of them disappeared before the end of the apparition, as the smooth reddish NTB stabilised.  The lozenges in 2008 seem to be new objects, with a different shape and a faster speed. 

It is likely that they have been seen sooner after this outbreak only because of the much better quality of the images.  They still have very low contrast.  I suggest that they are still developing, and within a year they may show a more compact form, and greater visibility (depending on the evolving cloud cover of the NTB) and slower speed (typical NTC-C).


Implications (2) How deep does the NTBs jet extend?

Historically, the NTBs has alternated between ‘NTC-C’ and ‘NTC-D’ states (book pp.106-110).  Only NTC-C spots were seen from 1891-1943, and from 1991-2002. Conversely NTC-D outbreaks were seen in 1880 and every 5 years from 1970 to 1990 [footnote 2], distinguished by superfast speeds, brilliant white spots, and subsequent orange colour.  These states were confirmed by spacecraft measurements of the peak wind speed of the jet between outbreaks: NTC-D in 1979 (Voyager) and NTC-C in 1995-2000 (HST and Cassini).  The question thus arose:  Did the whole NTBs jet change its speed between these states?  Or were the changes confined to the cloud-top level?

There is already evidence that the change is confined to cloud-top level, with the jet extending much deeper with permanent NTC-D speed (at least 170 m/s).  The main evidence is adduced by Sanchez-Lavega et al.(2008):

1) Modelling the white plumes that move with NTC-D suggests they arise by convection from well below the visible clouds;

2) Modelling the whole outbreak indicates that NTC-D white plumes and NTC-C vortices can arise only if the jet extends deep with a permanent speed at least as fast as NTC-D;

3) It is implausible that a deep jet could change its speed as rapidly as observed.

Our new observations add further weight to this last point, since the jet appears to be returning to a NTC-C state only 11 months after the latest NTC-D outbreak.

Indeed it is possible to compile a detailed record of the change in the peak cloud-top jet speed before and after the 2007 outbreak (Table 2 and Figure 2).  This is the speed of small cloud tracers, not plumes nor vortices.  It starts with the stable NTC-C state that prevailed up to 2002 (135 m/s: HST, Cassini). Then from 2002 to 2005, as the vortices disappeared, JUPOS analysis of amateur images was able to track gradual acceleration of the jet itself (Rogers et al., 2006).  In 2007, the jet was fortuitously tracked by New Horizons in Jan. (159.5 m/s: Cheng et al.,2008), and by HST on the day the outbreak appeared (160 m/s), and 3 months later (145 m/s) (Sanchez-Lavega et al., 2008).  The present speed of the jet can be estimated as 143 m/s if we assume that the orange lozenges are indeed jetstream vortices and that the peak jet speed is 10 m/s faster, as it was before the outbreak.

The present model of the NTBs jet can therefore be outlined as follows. There is a permanent deep jet with a speed of at least 170 m/s, but its surface expression alternates between two stable states.  In the NTC-C state, jetstream vortices develop (as on other jets) and limit the speed of the jet; I conjecture that they must be shallow structures, and when fully developed they impose a strict limit of ~135 m/s on the speed (presumably determined by the maximum stable wind shear and the width of the jet and the vortices).  When the vortices broke up after 2002, the jet at cloud-top level was free to accelerate towards the faster speed maintained at deeper level, reducing the vertical shear (Rogers et al., 2006).  Only when this acceleration passed a critical threshold, was it was possible for a deep convective storm to erupt all the way to the cloud-tops in the form of the brilliant plumes moving with NTC-D (Sanchez-Lavega et al., 2008).  From 1970 to 1990, this NTC-D state was maintained continuously; but after 1990 and 2007, the cloud-top speed decelerated again within months, allowing NTC-C vortices to re-establish themselves.


  Implications (3)  What are global upheavals?

The trigger for the deep convective storm may have been a global one, which is not understood but manifests in the global upheaval (book p.249-252), in which the NTBs jet outbreak and the SEB Revival outbreak are both typical components.  Now we can see surprising similarity between these two outbreaks.  Both start with a convective upsurge which can rise up from a deep level (probably the hypothetical water cloud layer) only when special conditions prevail in the cloud-top layer.  But there is still a remarkable difference betwwen them, that the NTC-D outbreak is in the peak of the fastest jet, whereas the SEB outbreak is in a slow-moving belt.

The now-well-established pattern of NTC-D outbreaks also sharpens the definition of the global upheaval (book p.250).  A NTBs spot outbreak is one of the most important elements in a global upheaval, but was thought to begin within a wide range of time, more than a year after the SEB Revival outbreak.  But as we update our review of global upheavals to 2007, we now see that most of them involved a NTC-D outbreak, which often occurred within just months before or after the SEB Revival outbreak (1970, 1975, 1990, 2007). In contrast, two ‘classic’ global upheavals (1928-29, 1938-39) now look unusual in that the NTBs outbreaks displayed NTC-C only and started more than a year after the SEB Revival outbreak.  I wonder if these, like the recent NTC-C outbreaks, were actually triggered about a year earlier, by an event unobserved during solar conjunction [Footnote 3].  So we can now see that the SEB Revival outbreak and the NTBs jet outbreak typically occur much closer in time than was previously thought. 



Observers can conclude that hi-res images over the next year or two will be of great value, to test the predictions made above: namely that the orange lozenges will develop into distinct vortices and decelerate to the standard NTC-C speed (DL1 ~ -60 deg/mth). If this does not occur, the present observations have nevertheless provided a valuable measurement of the NTBs jetstream speed at a unique time-point in the aftermath of a NTC-D outbreak.



[See these papers for earlier references. ]

Rogers JH, Mettig H-J & Peach D (2006)  Icarus 184, 452-459.  Renewed acceleration of the 24 deg.N jet on Jupiter.

Sanchez-Lavega A et 24 al. (2008 Jan.24)  Nature 451, 437-440.  Depth of a strong jovian jet from a planetary-scale disturbance driven by storms.

Cheng AF et al (2008 June) Astronomical Journal 135 2446-2452. Changing characteristics of Jupiter's Little Red Spot.



[1]  North Temperate Currents A,B,C,D refer to different speed ranges.  For completeness: NTC-B (intermediate between Systems 1 and 2)  refers to ‘rifts’ in the NTB, often seen 1-2 years after a NTC-C or –D outbreak.  NTC-A (DL2 ~ +17 deg/mth) is the permanent slow current of the NTB and NTZ.

[2]  This 5-year interval assumes that an outbreak occurred during solar conjunction in early 1985; no spots were seen but the region showed a vivid orange colour during the 1985 apparition.

[3]  However any such trigger for the 1929 and 1939 NTC-C outbreaks was probably not a NTC-D outbreak in the manner of 1990-91 (and possibly 2007-08). Even though a NTC-D outbreak could have been missed during solar conjunction, it would have left reddish colour in the NTB, which was not present. Actually, strong reddish colour was present in 1927, following a NTBs outbreak in 1926, so that might have involved an unobserved NTC-D outbreak.  There was no such activity preceding 1939.


Table 1: NTBs outbreaks: comparison of 1990 and 2007








Increased jet speed before outbreak

Voyager, 1979:

JUPOS, 2005:



Speed (u3) (m/s)





Bright plumes (NTC-D outbreak):





Start date

1990 Feb.10

2007 Mar.27

[Mar.25 (HST)]

Speed (DL1) (deg/mth)

-169, -160

-162, -154



Speed (u3) (m/s)

173, 169

170, 167



Dark patches following: (DL1)


-21 to -51



Jetstream spots (NTC-C outbreak):





First sighting after conj'n

1991 Nov.

2008 Feb.




(21 months)

(11 months)



Speed (DL1)





Speed (u3)















Table 2: Measurements of peak speed of NTBs jet (u3, m/s):









HST; Cassini; JUPOS



2003 Jan-Apr.


Rogers et al.2006



2004 Feb-May


Rogers et al.2006



2005 Mar-Jun


Rogers et al.2006



2007 Jan


Cheng et al.,2008



2007 March 25


S-L. et al., 2008



2007 June 5


S-L. et al., 2008



2008 May


this paper




Figure 2 and this Table are a continuation of the chart in Rogers et al.(1996).

*I thank A. Cheng and A. Simon for the exact figure from New Horizons (they note there is variation).

**The jet speed in 2008 is assumed to be 10 m/s faster than the drift rate of the putative vortices.