Voyager mosaics and images of Jupiter, A fresh look at some ancient stuff Options

[jovianjohn]

Another mosaic of Jupiter's southern hemisphere, capturing some white ovals and some turbulence they're creating in their wake.


South Temperate Zone Ovals and Turbulence by Justin Cowart, on Flickr

Congratulations and thanks for this superb mosaic. It really shows how good the Voyager imagery is, when optimally processed. --John R.

[jccwrt]

I was putting together a high-resolution Voyager 1 mosaic taken late on March 4, when I noticed a little spot. At first I thought it was just a bad reseau removal on the OPUS calibrated images, but turns out it was a real object! My best guess is that this is the silhouette of Adrastea sailing over Jupiter's cloud tops, but I haven't found a way to rule out its shadow as the culprit. Io's shadow is following it at a slightly more leisurely pace.



Image times at 20:23:48z, 20:25:24z, 20:27:00z, and 20:28:36z.

I made this using the narrow-angle mosaic as the base image. All four positions of Adrastea's shadow are taken directly from narrow-angle images taken as part of the mosaic. Io's shadow only appeared in two narrow-angle images, so I resized a crop of the shadow in the wide-angle shot to match the narrow-angle frame. Then I figured out how much Io's shadow moved between the first two frames, and extrapolated forward in time. Aside from some issues with geometry (the imaged area was rotating towards the center of the disk, so it would have been distorting in shape a little) and movement of the wave clouds (these might have actually been propagating at speeds that would have shown up in a true video), this should provide a realistic view of the scene.

Here's the full mosaic:



(This is actually a pair of separate 2x2 mosaics. They would have overlapped a lot better had an image in the second mosaic not been completely blown out.)

[Bjorn Jonsson]

Here is an unusual high resolution Voyager 1 view of Jupiter:



This is a mosaic of two color frames. The color and contrast should be fairly accurate even though the source data consists of green and violet filtered images which is not optimal for producing highly accurate true color images. The images were obtained when Voyager 1 was ~840,000 km from Jupiter's center and the resolution of the original images is ~8 km/pixel (slightly oversampled here).

What's most interesting in this image is that just outside the limb, Jupiter's blue sky is clearly visible. It was somewhat difficult to align the original images (green and violet) accurately at the limb but this should be rather accurate. I got nearly identical results when I first aligned the images visually in Photoshop and when I later made another version where I aligned the images at the limb using color images of the Earth's limb as a guide (I compared the appearance of the Earth's limb in the red, green and blue channels and then compared this to the appearance of Jupiter's limb in green and violet light).

In this image, the phase angle is relatively low (~19°). This means that sunlight illuminates Jupiter's cloudtops at a fairly low angle near the limb. This is one factor that contributes to the limb darkening and makes Jupiter's sky brighter relative to the limb than it would be at higher phase angles (i.e. less pronounced limb darkening).

[Bjorn Jonsson]

With the first Juno high resolution imaging soon coming up I decided several weeks ago to take a look at Voyager color images at various resolutions - mainly high resolution though (I ended up using 7-24 km/pixel). The idea was to get at least some idea of what might be expected. This resulted in several mosaics and what follows are several posts, one post per mosaic. The posts are ordered by resolution, starting with the lowest resolution mosaic. In all cases the distance from Jupiter's center and resolution is indicated. The approximate distance for JunoCam to achieve comparable resolution is also shown. This applies to JunoCam's resolution at the nadir - due to the very wide angle field of view the resolution will not necessarily be uniform across the hi-res images (this is different from Voyager which has a narrow angle camera).

With the exception of one mosaic (a white oval mosaic), the source images are in all cases green and violet filtered images. Most of the hi-res Voyager color observations were performed using this filter combination. This is not optimal for constructing true color images but I think I managed to get resonably realistic color. Here is a quick and dirty global image I used as a test - the two source images are orange and violet filtered.



Not bad but I suspect the color could be improved a bit. The color processing I used for the hi-res mosaics is comparable to what I did in this global view. The features shown in the mosaics are semi-randomly selected. In particular I'm not attempting to show all possible types of cloud features - I simply selected images where mosaics could be constructed, color was available and the images looked interesting in some way.

In all cases two versions of a mosaic are shown, one approximately true color/contrast version and a version where the color and contrast has been exaggerated and the image sharpened to better show various details. The contrast stretch varies for the different mosaics and depends on things like scene brightness and contrast. North is up in all cases. The mosaics are 10-15% bigger than the original data, i.e. they are slightly oversampled to avoid losing details. In all cases the indicated resolution applies to the original data. As usual I reprojected the images to simple cylindrical projection, did all of the mosaicking and color processing there and then rendered an image using the viewing geometry close to the mid-time of the mosaic's imaging sequence.

The image in the preceding post (posted two days ago) can be considered an 'appetizer'.

[Bjorn Jonsson]

And here is the first mosaic:



This is a nine frame Voyager 1 mosaic showing mainly the North Equatorial Belt (NEB), the North Tropical Zone (NTrZ) and their turbulent boundary. This mosaic shows lots of various features. A jet stream is visible in central NTrZ and examples of the NEB plumes are also visible. Gravity waves can be seen near (1900,1980) in the mosaic. Notice that boundaries where there is a major change in the color and/or brightness of the clouds usually appear sharp at this resolution.

Range: ~2.4 million km from Jupiter's center. Resolution of original data: ~24 km/pixel.
JunoCam, corresponding height above Jupiter's cloudtops: ~42,000 km.

[Bjorn Jonsson]

Mosaic 2:



This Voyager 1 mosaic shows the southern edge of the North Tropical Zone at the top and then the NTrZ/NEB boundary.
And this wide angle context image shows the location of the mosaic:



Range: ~1.4 million km from Jupiter's center. Resolution of original data: ~14 km/pixel.
JunoCam, corresponding height above Jupiter's cloudtops: ~25,000 km.

[Bjorn Jonsson]

Mosaic 3:



A high resolution Voyager 2 mosaic showing the area near a brown barge in Jupiter's northern hemisphere. The barge is partially visible at upper right. These barges usually appear at the northern edge of the North Equatorial Belt (or sometimes within the North Equatorial Belt ) near 16°N.

Range: ~1.3 million km from Jupiter's center. Resolution of original data: ~13 km/pixel.
JunoCam, corresponding height above Jupiter's cloudtops: ~24,000 km.

[Bjorn Jonsson]

Mosaic 4:

Click for enhanced version (too big to post here)
Click for approximately true color/contrast (too big to post here)

About 20 hours before closest approach Voyager 1 took a 'cross sequence' of images across the Great Red Spot. This was among other things a supporting observation for the IRIS instrument which was observing the GRS at the same time. The IRIS observations can be seen in an image from a paper in the JGR special Voyager issue (the first page of the paper indicates that it is not subject to US copyright so I can safely post this here):



The 'cross' appears somewhat different from my mosaic above because it consists of both green and violet images but in the mosaic only areas covered by both green and violet are included.

Before and after the cross sequence the white oval southeast of the GRS was imaged together with a part of the GRS; this was used to determine wind vectors within the white oval. The oval is one of three similar ovals that were visible during the Voyager encounters - this one was named oval BC. At this high resolution the cloud features change noticeably in 30 minutes or less. This made it necessary to warp the oval images slightly to make it possible to mosaic them seamlessly into the GRS cross sequence. The result is a 28 frame mosaic showing most of the GRS, all of the white oval and some of the areas near the GRS.

Range: ~1.25 million km from Jupiter's center. Resolution of original data: ~12.5 km/pixel.
JunoCam, corresponding height above Jupiter's cloudtops: ~23,000 km.

EDIT: And here is a schematic view showing the size and location of the above mosaic on Jupiter's disc:

[Bjorn Jonsson]

Mosaic 5:



Range: ~930,000 km from Jupiter's center. Resolution of original narrow angle camera data: ~9 km/pixel.
JunoCam, corresponding height above Jupiter's cloudtops: ~16,000 km.

And here is a wide angle context image:



Voyager 1 obtained this mosaic of a southern oval about 12 hours before closest appoach. In addition, two color frames obtained almost one hour earlier were used for the rightmost part of the image to get complete coverage of the easternmost part of the reddish oval that is conspicuous in this mosaic. The oval is located near latitude 60°S and is about 5000 km across. Like the Great Red Spot it is anticyclonic.

Farther south the south polar hood/haze is obvious. Its northern edge is usually near 66°S. It is bluish in the enhanced image here, both due to the stronger blue color (relative to red/green) in the polar hood and also because of Rayleigh scattering near the limb. Observations have shown that the top of the main cloud layer is lower in the polar regions than elsewhere.

When Voyager 1 obtained these images it was less than 1 degree south of Jupiter's equatorial plane while the oval is located near latitude 60°S. This makes the oval highly foreshortened; in reality it is much closer to a circular shape than the images above suggest. Here is an image showing the oval from directly above:

[Bjorn Jonsson]

Mosaic 6. Here several high resolution (~8 km pixel) green filtered images have been merged into a lower resolution color image that Voyager 1 obtained with its wide angle camera:



Range: ~800,000 km from Jupiter's center. Resolution of original narrow angle camera data: ~8 km/pixel.
JunoCam, corresponding height above Jupiter's cloudtops: ~14,000 km.

About 10 hours before closest approach Voyager 1 acquired three 1x3 narrow angle green filtered mosaics of one of the three big, white ovals that were present in the South Temperate Zone at latitude 33°S during the Voyager flybys. These ovals formed in 1939-1941 and had been shrinking since then. They were named oval BC, oval DE and oval FA. In 1998, ovals DE and BC merged into a single oval that was named oval BE. In 2000, oval BE absorbed oval FA to form what was named oval BA. In 2006, the color of oval BA changed from white to red, similar to the Great Red Spot. It still has a strong, orange color. I'm hoping to see hi-res Juno images of this oval; its 'ancestors' were well imaged by Voyager.

This mosaic shows white oval DE. This oval is visible at lower right in this global mosaic of Jupiter.

Here is an approximately true color/contrast wide angle context view:



And this is an enhanced version of the high resolution mosaic of narrow angle green filtered images that was merged into the wide angle data to produce the color image at the top of this post.

　

A single narrow angle violet image was also obtained showing the oval's southern boundary. This makes it possible to construct a higher quality color view of this area.



Notice that at this high resolution (~8 km/pixel) some of the boundaries between cloud features are getting more fuzzy compared to the lower resolution (~24 km/pixel) in mosaic 1.

[Bjorn Jonsson]

Mosaic 7. This is the last mosaic, at least for the time being.



A Voyager 2 mosaic showing Jupiter's limb. Here the version with approximately true color and contrast actually looks more interesting/beautiful because hints of Jupiter's blue sky are visible above Jupiter's limb. The blue sky is less conspicuous relative to the cloudtops at the limb than in the Voyager 1 mosaic I posted two days ago. This is probably mainly because of the higher phase angle which in this case results in a higher solar elevation angle at the limb and more brightly lit cloudtops relative to the blue sky than in the more limb-darkened Voyager 1 view.

Range: ~745,000 km from Jupiter's center. Resolution of original data: ~7 km/pixel.
JunoCam, corresponding height above Jupiter's cloudtops: ~13,000 km.

EDIT: Later I added one final mosaic (mosaic 8) to this series of posts.

[Gerald]

Outstanding work!

Just another 2 weeks, and we might see this in stereo.
Another push of motivation to squeeze out the full quality of the JunoCam images. This is so exciting! Thanks!

[Brian Burns]

These are great! I really like the white oval image - I've been looking through the Voyager 1 images also and have been amazed at how many beautiful cloud pictures there were. I don't have the ability to make the colors accurately so they've just been mixes of reds, purples, blues, and greens. Fitting them together must be quite a puzzle.

Incidentally, I recently came across some images of the original mosaic process here - i.e. tape pictures to a board - maybe they've been posted before...

http://imgur.com/a/SttsW

https://www.ucl.ac.uk/mathematical-physical...history/jupiter

Things have come a ways since then...

[Ian R]

Bjorn,

I spent a whole half-hour sitting in a darkened room last night, exploring your Jupiter imagery on my Kindle. It was a spine-tingling experience, and I have never seen the King of the Planets look so good!!

I was inspired to work on geometrically correcting the PDS calibrated imagery—utilising your custom-made flatfields—to produce the best Jupiter composites I've ever rendered. Thanks to the tiepoint info, an Excel macro, and the Unwarp plugin for ImageJ, I finally succeeded in warping the cleaned CALIB frames to match the GEOMED products.

This is the final result: a color view made from orange and violet frames shuttered by Voyager 1 on March 2nd, 1979:

[JRehling]

My best guess is that this is the silhouette of Adrastea sailing over Jupiter's cloud tops, but I haven't found a way to rule out its shadow as the culprit.

Perhaps the angular velocity between frames could answer this. Adrastea's shadow would move faster near Jupiter's terminator than Adrastea itself. Near the sub-Voyager point on Jupiter, the moon and its shadow would move at about the same angular velocity. Left as an exercise for the reader!

Amazing work! Beautiful images!