Juno Perijove 56, November 22, 2023 Options

[mcaplinger]

Some of the images have been posted on missionjuno.

You will immediately notice that there is something weird about them. Yes, we know about this, and I'll say more about it when I can. In the meantime, anything people can do to process them would be appreciated. I've had some success using https://github.com/chunglabmit/pystripe

[Kevin Gill]

Some of the images have been posted on missionjuno.

You will immediately notice that there is something weird about them. Yes, we know about this, and I'll say more about it when I can. In the meantime, anything people can do to process them would be appreciated. I've had some success using https://github.com/chunglabmit/pystripe

Oh, yeah, this is pretty bad. And subtracting the noise does not leave a lot of room for good color fidelity in the remaining values.

[Antdoghalo]

The camera color really took a beating with this encounter. Really hope we get more info on this soon.

[mcaplinger]

From https://www.missionjuno.swri.edu/junocam/processing

One of the biggest challenges for Juno is Jupiter's intense radiation belts, which are expected to limit the lifetime of both Juno’s engineering and science subsystems. JunoCam is now showing the effects of that radiation on some of its parts. PJ56 images show a reduction in our dynamic range and an increase in background and noise. We invite citizen scientists to explore new ways to process these images to continue to bring out the beauty and mysteries of Jupiter and its moons.

Maybe more detail will be released at some point, and maybe not, I can't tell. We are exploring some possible mitigations, but if anyone has any ideas we'd love to hear them.

[Explorer1]

Interesting that it's such a major change in one encounter after relative stability for many years with little evidence of cumulative damage until now, all of a sudden.
Some external factor could be the case, such as going through a particularly hot region or at a bad time (we are close to solar maximum, after all)?

[stevesliva]

We are exploring some possible mitigations, but if anyone has any ideas we'd love to hear them.

Are you talking about recovering this data, or operating a solid-state camera in a radiation environment?

[mcaplinger]

Are you talking about recovering this data, or operating a solid-state camera in a radiation environment?

Both, really, although I don't expect an idea about how to operate going forward that is something we haven't thought of.

[volcanopele]

Interesting that it's such a major change in one encounter after relative stability for many years with little evidence of cumulative damage until now, all of a sudden.
Some external factor could be the case, such as going through a particularly hot region or at a bad time (we are close to solar maximum, after all)?

This orbit had a pretty deep dive into the Io Plasma Torus.

[Antdoghalo]

As much as I'd like to see perhaps a social media post asking for suggestions, this may just lead to a needle in a haystack of noise.

[mcaplinger]

The approach images have been posted so one can watch the problem develop on the inbound leg. The partial frames are, we believe, unrelated to this issue; those look more like the data loss problems we've encountered before.

[mcaplinger]

As much as I'd like to see perhaps a social media post asking for suggestions...

We did post an explicit invitation for new ways to process the data on missionjuno, see above.

I've said elsewhere that there are very few knobs we have to turn on the camera. We can't change, or even measure, any of the internal voltages. We could adjust the companding parameters and that's something we are considering. We can change the temperature of the camera, but it's much easier to make it hotter than colder, and in general hotter is worse.

[Kevin Gill]

Single GRS frame from the approach movie.

[Kevin Gill]

And all the RGB frames from the approach movie. The background noise really picks up on the last two or three frames:

[Gerald]

Some of my early thoughts about the striping artefacts below. Presumably not too much new insight thus far.
The superposed signal appears to be mostly low-frequency along x.
There should be ways to filter out most of it. Here a very ad-hoc example of a cropped draft of JNCE_2023326_56C00140_V01 without and with some filtering along x:

It essentially removes wavelengths of more than ~100 pixels along x, kind of high-passing along x.

I'm trying to learn more about the spectrum of the addiional signal. That may help to find a suitable filter.
Some spectral features seem to be fairly constant after a first informal glance. A more formal analysis is work in progress. We'll see whether the properties of the stripes can be nailed down a little better.

A high dark level may truncate the signal numerically during companding. That may limit the recovery of the original data.
Combining/stacking multiple images over all three color channels may improve S/N.

That's roughly the ideas I'm going to persue.

[Gerald]

...We can change the temperature of the camera, but it's much easier to make it hotter than colder, and in general hotter is worse.

There is plenty of literature that says that radiation damage of semiconductors can be annealed by heating in at least some cases.

Two of several examples:
G. Kim et al. "A study of radiation damage and heat annealing effect on the irradiated 3T active pixel sensor"

The most well-known effect of the x-ray CIS due to the radiation damage are increments in the reset voltage and dark current. These effects cause the quality of image to degrade. To overcome these problems, many sensor recovery methods are studied. Annealing is the best method among many other methods. For the assembled sensor, the heat annealing is most suitable.

C. J. Marshall et al. "Hot Pixel Annealing Behavior in CCDs Irradiated at -83°C"

Hot pixel annealing began below -40 °C and the anneal process was largely completed by the time the detector reached +20°C.

How well this applies to the JunoCam electronics is mostly outside of my access.
All I can say is that the presumably radiation-induced reddening effect seems to have temporarily reversed a little bit after the chip was heated after, I think, PJ 49 or PJ50.
For some reason I thought that this heating option for annealing was also considered for JunoCam by design. But I don't find the according reference any more.
I'm aware of the experience that annealing will never be complete.

I think that in doubt it's easier to cope with the thermal dark current than with severe radiation damage in case the heating is maintained during the flyby. If it's properly calibrated and reliable, the companding function could be offset by almost the level of the dark current. I know that we would see a lot more hot pixels. But that's something more or less systematic and reprodicible.