Glad Juno has made it and appears to be working OK. Not the most amazing image, better ones are routinely taken by amateurs from earth, but the crescent can only be seen by going there as Jupiter lies outside our orbit. We'll see much closer views from Juno when it's moved to a closer orbit in October.
This isn't a photo. It a photoshop fiction of 400 different photos blended together. There's a big difference between that and a single frame from a spacecraft. Comparing the two seems silly.
Hardly fictional, it's a pretty accurate representation of what Jupiter actually looks like. The main reason for using this technique is because our turbulent atmosphere blurs any image. Taking a video allows only a small percentage of the clearest frames to be combined. A spacecraft doesn't have this problem. The technique is called lucky imaging: https://en.wikipedia.org/wiki/Lucky_imaging
Stacking improves signal-to-noise ratio but does not introduce details that aren't present. A stacked image can be more accurate than a single frame as camera artifacts are reduced.
Really well said! Also, before reading your explanation, I was a little hazy on how the image stacking thing worked. Now I understand a lot better. It's like averaging to get all of the details to the forefront without the oddities!
Glad you found it useful. If you haven't come across him before, this guy takes some of the best planetary images: http://www.damianpeach.com/index.htm
Of course it's a photo. All photography involves varying degrees of processing and subjectivity with respect to the information that is presented or discarded. Stacking may be somewhat less intuitive than a single frame, but conceptually it's no different from a long exposure - it's just selecting the moments of best seeing (minimal distortion from the atmosphere) to include in that long exposure. 400 frames of video at 60 fps is roughly analogous to a 6.7 second exposure in terms of signal vs noise.
Your typical single-frame exposure involves a great number of processing operations and filtering that may be unknown to most people, but still define the "look" of the end product (which is nowhere close to a 1:1 representation of the light entering the lens).
To complement what others have said, stacking photos is mathematically the same thing as using a long exposure time and blocking the iris whenever the image would have been fuzzy, so it doesn't seem any less of a "photo" than anything else. You are still integrating across the stream of photons you want while discarding ones you don't.
It's not a fiction, nor is it "photoshop", it's actual data.
The basic problem of all astronomy from Earth's surfaces is that the atmosphere impedes seeing. So much so that even small backyard telescopes rapidly reach the limit of atmospheric seeing before hitting the limits of diffraction. There are several ways around this. One is to leave the atmosphere entirely, resulting in the stunning capabilities of space based observatories like Hubble. Another is adaptive optics, which relies on various techniques to read the atmospheric disturbance causing degraded seeing and precisely counteracting it by interposing a reactive optical element.
The technique in question is basically a sort of poor man's adaptive optics related to speckle imaging (another technique employed for the same purpose). For bright objects (like planets) an image exposure using modern CCDs need only be a fraction of a second. Which provides the opportunity to collect many of them over a period of time, which is something that many cameras already do quite well by recording video. The alternative of counteracting poor atmospheric seeing is to hope you get lucky and capture a moment where atmospheric distortion is at a minimum. Such moments are rare and fleeting, but over the course of hundreds or thousands of images from a video stream, there will be a few. The trick is to find the best moments of seeing for a given portion of the image from within the frames and combine all of them together into a single image. You then have essentially a "dream team" of atmospheric seeing conditions for every part of the image. Every part is real, and the overall image is a true representation, not a fantasy.
However, your point about the futility of comparing the two is accurate. Comparing a single image from Juno to the best thing produced from a backyard astronomer ever is not a good comparison. Especially since Juno will collect a great many more pictures and is not at all optimized for taking pictures at its current distance from Jupiter. The best pictures of Jupiter from Juno will outclass anything we've taken with any instrument from any observatory or spacecraft so far, but it'll be a while yet before we have those.
If I'm not mistaken, they use a similar technique to makes satellite images of earth that don't have cloud cover, or maps without cars and stuff, yet we know it is highly likely these were present when pictures were taken. So it's not fiction.
Stacking allows errors caused by our atmosphere to be cancelled out. It's an attempt to get it closer to what a spacecraft with no atmosphere to contend with would see.
Are we going to get amazing Pluto-like images, like that chilling closeup? They really impressed me A LOT. Or is Jupiter close enough for us to already know enough about its surface?
Jupiter either doesn't have a solid surface, or it lies very deep inside (like 4/5ths of the way to the center). So there isn't any surface to see, just clouds.
Juno will be passing within 3,000 miles of the tops of Jupiter's clouds. Since Jupiter's radius is 43,000 miles, it will fill the entire frame. The only question is how high-quality the camera is.
The camera is for public relations only (no science) and I don't know the resolution. But this photo was taken from 2.7M miles, so you can expect the resolution to get roughly 9 thousand times better.
EDIT:
> Jupiter itself will only appear to be 75 pixels across from JunoCam when Juno reaches the furthest point of its orbit around the planet. At its closest approaches JunoCam could achieve 15 km/pixel resolution from 4300 km, while Hubble has taken images of up to 119 km/pixel from 600 million km....The camera uses a Kodak image sensor, the KODAK KAI-2020, capable of color imaging at 1600 x 1200 pixels. It has a field of view of 18 x 3.4 degrees with three filters to provide color imaging.
> At its closest approaches JunoCam could achieve 15 km/pixel resolution from 4300 km
In comparison, the highest resolution photos of Pluto from New Horizons achieved 80 meters per pixel from a closest approach of 12500 km. Its amazing how good the camera was on New Horizons.
No doubt New Horizon's camera was phenomenal for the mission, but I'm actually not sure how good it was compared to something you could do with a terrestrial lens. (There are lots of compromises when you go to space.) Juno's camera is very modest by design: It's a wide-field of view and only a few megapixels, so basically a cell-phone camera.
> It's a wide-field of view and only a few megapixels, so basically a cell-phone camera.
It's not similar to a cell phone camera at all. Cell phone cameras are usually much higher resolution (current iPhone is 8 megapixels vs the 2 for Juno), but with a much smaller sensor (iPhone cam has a 1.5µm pixel size vs 7.4µm for Juno).
Also light thresholds need to be taken into consideration. Jupiter is 5x[1] further from the Sun then Earth, so there is much less light to capture in an image.
It's the opposite of cellphone camera. It has low megapixels and a large sensor size. This will allow a lot of light onto the sensor and it won't suffer from the noisiness that high megapixels on a smaller sensor suffer from.
The point is that, on the spectrum of all imaging devices you might put on a satellite, both Juno and a cell phone camera are in the same rough neighborhood. They are dwarfed by telephoto lens, or 200 megapixel sensors. Remember especially that the topic we are discussing is the resolution (km/pixel) at a particular distance (see above), and for this the lens are much more important than the particular CCD chosen.
So now it's the lens that you don't like, not the sensor? The imaging camera has a wide angle (11.7mm) because it's going to be so close to the surface that a wide angle is optimal. That's actually a lot wider than cell phones (iPhone 6s is 31mm equivalent). There are also other instruments on board that look through the gas with telescopic reach.
Resolution has very little to do with it. Many images from NASA are stitched, even the ones from New Horizons. The important part is light sensitivity (in the correct wavelengths!) and making sure that it can work in the radiation of Jupiter.
You can go buy a camera with the same sensor if you'd like, but they're expensive:
I believe it's the same sensor that flew on the Curiosity mission and we all know those photos are fantastic. They are also flight tested, invaluable for missions like this where you only get one shot at it.
Yes, I think the casual "basically a cell-phone camera" comment you made, way above, has been taken to mean "it's a POS".
But, as you have made clear, you were just making a statement about the combination of wide field-of-view (i.e., the optics) and the number of pixels, which both figure in to resolution. (By contrast with the New Horizons LORRI instrument, which had a very narrow FOV, because they did not approach Pluto closely.)
If one takes a breath and re-reads what you wrote, this meaning is evident. On the other hand, it's very easy to mis-interpret what you wrote.
> As you'll see if you look above, it was you who brought about the size of the sensor when critiquing my comment, which was addressing resolution.
Because sensor size is what's important in digital photography, you can always stitch together multiple images to get more resolution. A large sensor like what's flying on Juno is orders of magnitude better than a camera phone sensor regardless of the resolution. In general cell phone cameras (which you decided to compare Juno's to) are high resolution, but with small a sensor (everything in a phone is small!). Juno is the opposite, low resolution and a large sensor. If you want to take pictures of stuff in space you most certainly want a large sensor.
> You also seem to feel the need to defend Juno, as if I was attacking it somehow, so I don't think this conversation can be very productive.
Well you did attack it by saying its camera was the same quality as a camera phone (aka cheap and shitty). It's not similar to a camera phone whatsoever, you were just plain wrong.
That's actually way cheaper than I expected it to be. Though I supposed an actual spaceflight-worthy camera with that sensor would cost far more than the one you linked.
In fact I believe Juno has ice-penetrating radar, intended to probe for the mythical surface. We may be seeing images of its real surface for the first time soon
I do not believe radar is part of the instrument set.
There is a "gravity sensor" which is really, I think, an extremely careful Doppler measurement which will allow us to infer the mass distribution within Jupiter.
>Are we going to get amazing Pluto-like images, like that chilling closeup?
No. The camera on Juno was literally an after thought. Originally it wasn't even going to have one. It's main job is studying magnetic fields and radiation belts.
Yes, the optical-wavelength camera was not a science driver for Juno. It's just making images of cloud tops. The cloud tops aren't that interesting, it's what's going on underneath or beyond.
In addition to fields and radiation instruments you mention, there are two other imaging spectrographs, one an IR instrument and one in UV. The IR camera can look ~70 km down below the cloud tops.
Importantly, there is a microwave radiometer, which can peer ~500km beneath the clouds to determine atmospheric composition as a function of time and space.
They have put Juno in a polar orbit, so it will scan the atmosphere through dozens of orbits before the radiation kills the instruments.
We should get some extraordinary views of Jupiter's polar aurora activity. Also much higher resolution of gas dynamics phenomena because of Juno's perijove altitude of ~3,000 miles.
Doubt we'll see depth, unless it's through fortuitous breaks in high clouds. Other (non-imaging) instruments (microwave sounders, etc.) will tell us what's going on in Jupiter's interior. Metallic hydrogen? Rocky core? Juno may tell us.
I am astonished that humans can do this technology. Here I am looking at these pics a flying robot snapped from 4M+ km away from Jupiter?!
I wish they mark these pics with some pictographs regarding the scale of things. Like something comparing the orbital distances between the moons and Jupiter itself to that of Earth's orbital distance to the Sun or something. I cannot comprehend 4 million kilometers (which I know is Juno's distance to Jupiter, not the moons' orbital distance..). I have trouble wrapping my mind around just 4 km.
For scale, Io is almost the same distance from Jupiter as the Moon is from Earth - about 400,000 kilometers.
That's about 0.3% the distance of Earth to the Sun.
For a good sense of how enormous these differences are, try the video "Riding Light" - https://vimeo.com/117815404 - Jupiter shows up at the 43:10 mark (Earth's at 8:17, for size comparison).
This is cool. Jupiter is easy to spot with the naked eye, as it's the second brightest object in the night sky after Venus (obviously, excluding the moon when lit). You can even see some of its moons with some basic optics.
To think there is a probe there now is a pretty powerful image to inspire children. Something we couldn't do with the recent Pluto visit, as awesome as that was.
On a visit with the nieces, I took the eldest outside and using a night sky tracker app, pointed it out to her. "You know what that is?" {teenager sigh}"no. A star?" "No, that's a planet. It's Jupiter." She got a little more interested after that, but a career in astro-anything wasn't to be, looks like.
My daughter's massively into space, and having a mid-size telescope (Celestron NexStar 6SE) has been truly wonderful. There's something really different about seeing it for yourself even if the picture isn't as good as one from a probe.
I was camping under a clear sky with friends and asked, "has anyone seen another planet with their own eyes?" and explained I meant seen any detail through a telescope. Nobody had. I resolved start some amature astronomy then.
The sidebar of https://www.reddit.com/r/astronomy has a bunch of useful articles on getting started. Just be warned that it's one of those hobbies that can get really expensive if you lack impulse control - "I need a new eyepiece! And a few more inches of aperture! And a better mount!" starts to pile up if you really enjoy it. :-D
It's also not something you can really save money on, as I have discovered, alas --- the bottom end telescopes are terrible: wobbly mounts and feeble optics. Frequently you get shoddy magnification eyepieces that are far too powerful for the light collection ability of the telescope and as a result you get this tiny, smeary image that won't stay in focus and every time you try to adjust anything you lose your target.
However, you can do great things with a good pair of binoculars (and a steady hand). You can resolve planetary discs, pick out the moons of Jupiter, spot Neptune and Uranus if you're good (although I never have) and exploring the surface of the moon is endlessly fascinating. And if you're really desperate, you can look at birds through them.
See whether there's an observatory near you that has public viewing nights. I've seen Jupiter and a couple of its moons at the Hampstead Observatory. It's a much less frustrating experience than a backyard telescope when you're just getting started, IME, and you can meet some cool people too.
I understand the importance of this mission, but after the Curiosity skycrane, the Pluto flyby and the Rosetta orbiter, I'm really jonesing for something cool.
Solar Probe Plus will fly within 8.5 solar radii of the Sun into the hottest region of the sun, the corona, approaching speeds of 200 km/s and withstanding temperatures of 1400C.
We don't know why the corona of the sun is hotter than the surface of the sun. SPP will provide data to help us to better understand and create better models of coronal heating and the solar wind.
This brings to mind a moment in the movie Apollo 13, where one of the supporting characters says something to the effect of: "Leave it to NASA to make a moon landing boring".
I think it says a lot of great things about NASA that people consider putting a satellite in orbit around Jupiter to be routine. Further great things from the fact that they really want NASA to stop being "boring" and invent some crazy new thing and push the boundaries of what humans think is realistically possible again - after all its been like a whole year since they blew our minds last time.
All for the low price of ~.03 militaries. (Not including the fact that NASA does some stuff for the military...)
A late 2020s Europa mission has been given development money. It will take about five years to launch and another five to arrive. Europa has a large interior ocean under the ice and maybhave life. It has enough energy to keep water liquid at certain depths and drive life.
A Europa mission might resemble Cassini and Juno: orbiting the main planet with frequent flybys of the moon. This requires less fuel and keeps the probe in the dangerous radio belts less of the time.
SpaceX might launch a Mars lander during the 2018 low energy window. Though I am guessing more likely the 2020 window following that.
NASA is fully cooperating with SpaceX by doing things like providing detailed data of previous Mars landings. NASA is SpaceX largest customer and financier. Many ex-NASA people workmat SpaceX. These data are public. But it helps to have good communication channels.
>SpaceX might launch a Mars lander during the 2018 low energy window.
Is this budgeted? Is it on a launch docket? Is there a rocket ready for it? Is the lander completed? 2018 is right around the corner and frankly I'm getting sick of "Spacex might" comments. There's a big difference between "might" and "do." When you see a mission on the NASA roster, its ready to go and unless there's some serious technical issue, it'll happen.
I'd like to see more "do" from SpaceX that isn't LEO flights and landing on barges.
It's being built right now, the test is in December.
> Is the lander completed?
It's being built right now, the test is soon.
> 2018 is right around the corner and frankly I'm getting sick of "Spacex might" comments.
SpaceX has completed every single thing they have ever said they would do, except for things that remain in the future. I don't know how you could be sick of "SpaceX might" because it generally means "SpaceX will".
> I'd like to see more "do" from SpaceX that isn't LEO flights and landing on barges.
They're planning on colonizing Mars by sending 80,000 people per year until there are millions living on Mars. They have been working steadily towards that goal for 15 years. They are extremely busy and making the most rapid progress in spaceflight since the beginning of spaceflight itself. Take a step back and realize how much SpaceX is actually doing. If this isn't enough "do" for you, then I think nothing in the world or anywhere in the solar system would satisfy whatever you're looking for.
So in other words no rocket and no lander, but will be in the air in 18 months? Pardon my skepticism, but I think you're buying into "Elon Musk Time" which promises x but comes out y or never.
SpaceX is impressive, but I think its fandom is out of control. Repeating marketing and PR pieces to me isn't convincing. I wish them the best of luck but a 2018 launch isn't happening. I'd be surprised if any of this happens before 2022-2025.
I think you're mixing up Falcon Heavy / Crew Dragon with the proposed 2018 Mars mission, "Red Dragon" (which will probably launched by a Falcon Heavy, but will involve a heavily modified Dragon v2). My understanding is we won't have any details on Red Dragon until September.
The "Red Dragon" mission will indeed involve a Falcon Heavy and a modified Crew Dragon, craft which are mostly complete and will be tested in the near future.
The September announcement at the International Astronautical Congress will be their full proposal for sending humans to Mars, including BFR, MCT, ISRU, the whole works.
For more info, check out the SpaceX forum on Reddit. I know Reddit has a reputation, but the SpaceX fans run a top-notch community there.
Elon Musk time means "2018" is almost certainly going to be 2020 or 2022 (even with NASA's promise of technical support). That said, "landing on barges" is potentially a massive step towards reusable, cheap launch vehicles, and it's one their competitors mocked them for even attempting.
> On later orbits, Juno will continue photographing the poles, but where else the camera will focus will be open to a popular vote by the public.
NASA has been pretty active in public outreach. The article indicates that Juno takes something like 4 photos an hour. I wonder whether it will be 1 photo for the public to 3 photos for science, or totally different?
I understand JunoCam is a public outreach tool. But if it provides much worse resolution than Hubble from inside Jupiter's orbit, it's basically useless even for that.
I don't know if the cam is the problem or the transmission bandwidth, one thing is clear, when serios science starts there will be much less bandwidth available to send sharp and cool photos.
That's what you get from a solar powered probe. We waste billions of dollars and unique opportunities because some people have gotten in their minds that RTG power is evil and dangerous. NASA even canceled the ASRG, based on the Stirling engine, which was 5 times more efficient in terms of radioactive material and would have allowed powering probes for the whole solar system. Imagine a Pluto rover !
>gotten in their minds that RTG power is evil and dangerous
RTG usage is curbed due to NASA's low supply of plutonium-238. Until more is made it makes sense to design for solar powered systems. Also these are risky missions with high pricetags and erring on the side of caution using well known technologies isn't some big conspiracy. Its to avoid high-profile fails. We can try new and risky stuff in lower profile missions.
Heck, Curiosity was launched just a few years ago and has 11lbs of plutonium powering its RTG. I hardly see a conspiracy here. Also RTGs add weight and cost to projects. If you don't need one, you probably shouldn't be using one - even ignoring their rarity. Missions like Juno get significant weight savings using solar.
>I don't know if the cam is the problem or the transmission bandwidth,
The quality of the photo has to do with the distance from Jupiter. It has nothing to do with "power." This is a spacecraft, not a gaming PC. Yes more power would mean a higher bitrate, but the bitrate it uses is good enough for the mission.
Juno is the first use of solar panels for a probe going that far out. You're right that they want to err on the side of caution using well-known technology to minimize risk. But you're missing the fact that RTGs are that well-known risk-minimized technology.
If solar panels saved weight, they would have been used on past missions like Galileo. Better solar panel technology will eventually shift the balance, but we're not there yet. Juno's panels weigh 340kg. Galileo's two RTGs, which produced about the same amount of power at Jupiter, weighed 57kg each.
I was referring to the new stirling engine RTG. According to wikipedia, NASA doesn't see it being used in a mission for another decade or two due to further testing and certification.
I wasn't aware solar was so heavy. Thanks for the info.
The fact is, it could be better, and it really should. It's just silly to use Solar panels in the outer Solar System. This mission was dominated by politics and not science. I think that is a very real issue. Look at Curiosity, I think it's very clear that Rover is a massive upgrade over the preceding solar powered missions.
Well, Curiosity is huge and at that point needed a power source that wasn't solar. I really am not buying that these are political decisions. I think there's more emphasis on solar and less RTG in some scenarios because everyone is worried about the plutonium situation. No use wasting plutonium on missions that frankly just don't need it.
The probe is 2.7 million miles from Jupiter, headed further out (its orbit is highly eccentric). That's nearly 11 times the distance from the Earth to the Moon.
At closest approach, Juno will be 2,200 miles from Jupiter, less than 1% of the distance between the Earth and Moon.
Space exploration is expensive, and a very large part of that expense is mass. Putting a 2.7 million mile scope on a probe that's going to be spending much of its time far closer in to interesting targets costs mass, and fuel, and other sensing apparatus you can't add. So there's that.
It's possible that there are higher-resolution capabilities for the probe as well, or that post-processing or higher data rates will improve current images.
It's also possible that visual reconnaissance wasn't a high priority for this mission, though I'm not familiar with the sensing platforms included.
Given orbital eccentricity, Juno can acquire and store high-quality data, and transmit that at leisure during the outbound parts of its orbit.
Strictly speaking, the probe will be spending most of its time at distances like these. That's the nature of elliptical orbits. But it'll spend enough time up close to get some great pictures.
And yes, photos weren't a priority at all. If it were just a matter of returning scientifically useful data, they wouldn't have included this camera at all. But NASA likes pretty pictures too, so they included one.
Juno is in a highly elliptical orbit, which will take it within 5000km of Jupiter's cloud tops. This photo was taken from much further away than that. In short, check back in about a month. :)
Agreed, and to add more detail for comparison - the current distance is around 2 million kilometers, so resolution should jump by a factor of about 500 (per dimension, not number of pixels) at closest range.
EDIT: For a comparison to Hubble - at closest approach JunoCam will get 5 kilometers per pixel [1], while Hubble gets in the region of 120 kilometers per pixel [2]. So, giant jump in quality, though the field of view from up close will be limited.
Juno's solar panels produce as much power at Jupiter as Galileo's RTGs did. The quality of this picture has nothing to do with solar power. It's a bad picture because Juno is on a highly elliptical orbit for reasons of efficiency and longevity (there's tons of radiation closer in) and this picture was taken when the probe was very far away.
There is some basic info on the vehicle and its payload at [1]. "The spacecraft will also carry a color camera, called JunoCam, to provide the public with the first detailed glimpse of Jupiter's poles." - so like you say, not a priority. More on JunoCam at [2].
This photo was taken from 2.7M miles, and Juno will approach closer than 3,000 miles of the cloud tops, so you can expect the resolution to get roughly 9 thousand times better.
I came across this on Reddit the other day, taken with a smartphone and 8" telescope: https://www.reddit.com/r/astrophotography/comments/4kegyv/sm...