The Jovian Vortex Hunter (JVH), which was launched on June 20, 2022 on the Zooniverse platform, and completed on Dec 23, 2023! We had more than 6000 registered volunteers performing more than a million classifications on 68000 images.
I would like to thank every single volunteer who participated in this project! It has been a wonderful experience chatting with you on the Talk threads, and going through your classifications. This project would not exist without you, and I would like to express my gratitude for your efforts!
So, what did we learn from this project?
Overview of the project
The goal of the project was to look through images from the Juno spacecraft and identify those that contained vortices, turbulent filaments and cloud bands. The images with vortices were moved to a second workflow where the goal was to annotate the vortices by color.
From the volunteer annotations, we cluster the individual classifications to create an “aggregated” vortex, as you see below. You can find more details in our separate blog post here!
Using these aggregated vortices, we can generate a catalog which has the following properties:
- Location (Latitude and Longitude)
- Perijove (or time that the vortex was observed by the Juno spacecraft)
- Size
- Shape (i.e., the aspect ratio — or ratio of height to width) of the vortex
- Color (which is given by the consensus from all the volunteers’ color label)
By the numbers
The power of this project is in the amount of data that it has generated. Here are some statistics on the project:
Classification
There are over 6000 registered volunteers who participated. We had over a million classifications across the two workflows! 141 people contributed more than 1000 classifications to the project!
In the chart below, each colored line is the contribution from a single volunteer, and the width represents what fraction of the total classifications they contributed. There were a few volunteers who contributed a significant fraction of the total classification! There were also many volunteers who contributed only a few classifications, but their classifications made up a significant component.
Talk:
We have over 400 participants on Talk with more than 12000 unique posts and over 18000 comments.
There are over 400 unique tags on subjects, and atleast 70 of them have been applied to more than 10 subjects, and 22 have been applied to more than 100 subjects! You can see some of our most popular ones in the figure below
Vortices
From the classification, we derived over 83000 unique ellipses from our subjects, which corresponded to roughly 7000 unique vortices on the planet. Each vortex had at least 8 classifications and several of them (especially the large ovals, like the Great Red Spot) had over a 100, across many subjects!
Let’s see what we can do with this catalog
Distribution of features
Firstly, we can simply plot the distribution of the vortices across the planet. On Jupiter, most of the dynamics changes in the meridional (i.e., north-south) direction. This is because there are different jet streams in the meridional direction, and as the wind speed changes, so too does the dynamics. These jet streams are actually what lead to the colorful bands! The change from a belt (which is brown in the image above) to a zone (bright white band) happens at the peak of a jet stream.
From our data, we see that vortices and folded filamentary regions (FFRs, which are turbulent regions with multiple swirls) are not found throughout the atmosphere! They are mostly concentrated in the 20° and 80° north and south! This is actually expected due to the ways that the atmosphere responds to instabilities. Vortices are formed when the atmosphere becomes unstable (like how hurricanes form on Earth!), and closer to the Equator, the Coriolis force is smaller and the vortices get stretched out easily. Closer to the poles, they are able to retain their structure more effectively! Our data only goes as high as 75°, and so you see a drop-off at the very high latitudes, but other observations have shown that there are a lot of vortices right next to the poles (including some that are permanently circling it!)
Structure of Jupiter’s atmosphere
Distribution of vortices
The distribution of vortices has been previously observed by other studies, but what is unique is the pure volume of data, which means that our statistics is much more accurate. Specifically, we are able to clearly see the transition between the zonal (i.e., banded east-west flow) to the atmosphere becoming more turbulent (filled with vortices and FFRs) as we go away from the equator. We can also break this up by color:
As you can see, each color is unique. The white and dark ovals are more prominent in the higher latitudes, while the brown vortices are in the mid-latitudes
Atmospheric stability
Vortices are formed when the atmosphere becomes unstable. They are essentially a turbulent way of the atmosphere readjusting itself to become stable. As such, to understand the structure of Jupiter’s atmosphere, we can use the properties of the vortices to determine when/how the atmosphere becomes unstable!
The fluid dynamics at the core of this study is very long and convoluted, but if you are interested, I recommend these papers (Jupiter-style Jet Stability and The turbulent dynamics of Jupiter’s and Saturn’s weather layers: order out of chaos?). Both go deep into the fluid dynamics and mathematics of the fluid dynamical instabilities.
Briefly, the current theory is that Jupiter’s atmosphere is neutrally stable. What this means is that the structure of the jet streams on Jupiter ensure that the winds will not get disrupted easily and devolve into a turbulent mess, but it is easy for small perturbations (maybe from heat released from the deep, hot interior) to cause brief periods of instability.
These instabilities can be of many forms (as seen in the GIF above): two typical ones are convection (these are mainly in the vertical direction, like thunderstorms on Earth), or driven by wind shear (these are mostly horizontal, like the small swirls that form when you mix cream into your coffee or tea). Vortices on Jupiter are formed by several types of instabilities, and so we can only derive basic conditions for instability, which give us only minimum or maximum bounds for vortex properties. Knowing the specific nature of the instability will allow us to get better estimates, but finding a vortex in the forming process is very rare!
Therefore, we can use a criterion known as Arnol’d’s second stability (Arnol’d-II) which relates the wind shear in the horizontal direction (how wind speed is changing with latitude) with the depth and stability of the atmosphere. Essentially what this tells us is how deep do we need the atmosphere to be in order for it to maintain a steep change in wind speed. If this depth or the wind speed changes in a way that breaks Arnol’d’-II, then an instability will develop, and it will most likely be vortical in nature. What we get is interesting:
There are two takeaways from this: First, the depth of the fluid is between 50 to 100km below the cloud tops! Note: that this is not actually the depth of Jupiter’s atmosphere — it is a measure of how deep the atmosphere affecting the clouds is! The approximations we have made in this calculation means that we are essentially treating the atmosphere as two separate fluids: one that forms the clouds and is turbulent, and the second deep layer which is fixed and completely stable!
The second is that the atmosphere needs to be deeper near the poles for the atmosphere to be stable, especially for the white and dark ovals to form. This means that vortices closer to the poles are likely deep rooted, which would explain why they are so long-lived!
Where next?
Our results from this study are being finalized for publication, and we plan to send this out soon. There are still several perijoves worth of data that we can process, but for now, our goal is to finish the publication before we start another round of data. This project has created a wonderful dataset and we hope to increase the volume of this data through another round of the JVH project, soon! This would also likely enable the first-of-its-kind machine learning analysis on a large planetary atmosphere dataset, which we hope will result in several unique insights into this amazing planet!
Jovian Vortex Hunter 