# Why is it hard to take a high quality image of far away objects (like Pluto) using a high-tech space telescope?

Why did NASA send a probe so close to Pluto instead of using a high-tech telescope to capture images of its surface?

The Icy Mountains of Pluto. Image credit: New Horizons / Johns Hopkins Applied Physics Laboratory, July 15, 2015

I always imagined its possible to take at least low-quality pictures of tiny little things from far far away using a space telescope. But to be honest, I got confused seeing latest pictures of Pluto taken by New Horizons, and I actually don't understand why we couldn't take such low-quality images without sending an entire probe close to Pluto?

I understand that it's impossible to take such picture from this angle from Earth or telescopes close to Earth, but the image quality is poor comparing to Hubble images (the ones taken from deep space) or other similar images of other planets (like Mars, Saturn or Moon or even planets from other solar systems).

• The image quality is excellent. – gerrit Jul 16 '15 at 13:35
• This image has a resolution of something like 0.4km per pixel, if I recall correctly. That is pretty good resolution compared to what Hubble can accomplish. – TylerH Jul 16 '15 at 14:14
• "Space is big. Really big. You just won't believe how vastly, hugely, mind-bogglingly big it is. I mean, you may think it's a long way down the road to the chemist, but that's just peanuts to space." – T.J. Crowder Jul 17 '15 at 9:40
• @MichelKogan Yep - the overlay one is earlier. However, it shows the best that earth bound space telescopes can do and the same features with New Horizons (at a distance). – user5892 Jul 17 '15 at 15:15
• You think that photo of Pluto is bad? I couldn't believe this is the best photo they took of Hydra: nasa.gov/sites/default/files/thumbnails/image/nh-hydra_1_0.jpg – pacoverflow Jul 18 '15 at 18:28

I think a lot of folks see these gorgeous photos of distant galaxies, with fine detail on dust lanes and spiral arms and assume that since they’re so far away, seeing Pluto would be easy. But while these galaxies are far away, they’re also huge.

The (relative) detail that can be seen in any given telescope will be found by the object’s size divided by the distance from the telescope. In other words, the angular size in the sky.

$$\theta = \frac{\mathrm{size}}{\mathrm{distance}}$$

Let’s compare the angular size of a distant galaxy with Pluto. A nice spiral galaxy (like ours) is about 100,000 light years in extent. We’ll make it old and put it at 10 billion light years away. How big is it in a telescope?

$$\theta = \frac{10^5\thinspace\mathrm{ly}}{10^9\thinspace\mathrm{ly}}$$ $$\theta = 10^{-5}\thinspace\mathrm{radians}$$

Now, how much can we see Pluto? It has a diameter of ~2400km, and when farthest from Earth is around 48AU in distance. So it has an angular size of $$\theta = \frac{2.4\times 10^6\thinspace\mathrm{m}}{7.2\times 10^{12}\thinspace\mathrm{m}}$$ $$\theta = 3.3\times 10^{-7}\thinspace\mathrm{radians}$$ $$\text{linear ratio} = 30:1$$ $$\text{area (square) ratio} = 900:1$$

This means that compared to a fuzzy distant galaxy, Pluto covers about 3 orders of magnitude less angular area in the sky. No wonder that it’s more difficult to image.

Moreover, at these angular scales you run into the diffraction limit. For visible light and for an aperture the size of Hubble’s, you can’t expect to resolve features that are significantly smaller than about $2.5 \times 10^{-7}$ radians across, which is comparable to the angular size of Pluto itself. To do better from Earth orbit, you'd need to get a much larger mirror.

In fact, if you look at this image from NASA showing the Hubble extreme deep field:

at full size (976 pixels wide), then by my calculations the most zoomed in portion in the upper right is about $2.5 \times 10^{-7}\thinspace\mathrm{radians/pixel}$. In such a picture, Pluto would be just a bit bigger than 1 pixel.

• You hit the point @BowlOfRed ... Thanks for the great answer and nice formulas – Michel Gokan Khan Jul 16 '15 at 19:55
• When you put it that way it's almost surprising we discovered Pluto in 1930. – Random832 Jul 17 '15 at 4:08
• Our ability to detect something depends on the intensity of light from it, not on the angular size. Even this small, Pluto has a disk far larger than most stars, but that doesn't mean the stars are hard to see, just hard to see differences across the surface. – BowlOfRed Jul 17 '15 at 4:13
• Re "...need to get a much larger mirror": Or multiple mirrors separated by some distance. – jamesqf Jul 17 '15 at 6:01
• @njzk2 Conveniently, theta for astronomical objects -- even the sun or the moon -- is always quite small. – Russell Borogove Jan 9 '17 at 5:11

Distance. Size of the target. Its poor albedo at such distance to its only source of illumination, the Sun, compared to closer celestial bodies. And movement of the target and the vantage point in their orbits preventing advanced image interpolation techniques combining multiple observations of same side of Pluto at same lighting conditions.

Pluto is currently 31.9 AU (astronomical units, 1 AU is average distance between the Earth and the Sun) away from Earth, where currently highest spatial resolution orbital optical telescope, the Hubble Space Telescope (HST) orbits around it. In other words, Pluto is nearly 32 times farther away from us, and the Hubble, than the Sun is:

Distance to Pluto to scale. Image source: Solar System Scope, using realistic (non-orrery) model and sizes to scale

HST's highest resolution imager, ESA's Faint Object Camera has spatial resolution of 0.0072 arcsecond pixels and can produce roughly 16x16 pixel images of Pluto from such a distance:

ESA Faint Object Camera's views of Pluto and computed interpolations (Credit: ESA)

Of course, even those 16x16 pixels are badly stricken by Rayleigh limit of HST's Cassegrain type optics, but I'll spare you with technical details. Point is, that for anything better than what HST can produce, we'd either require an even larger telescope than the Hubble, or what we've done with New Horizons, move smaller ones much, much closer to its target.

Also, is the image you attach to your question really of poor quality? It resolves features that are smaller than a mile (1.6 km) across. And it also took other images resolving detail up to 70 meters across. We'll just have to wait for them. With its transmission rate of ~ 2 kb/s, it will take nearly 1 and a half years for it to send all the data it took during the flyby back to Earth.

• You explained the technical part pretty well ... actually I imagined for much better quality, like the ones we have from Mars and Moon surface (the ones we can access via Google Earth for Mars/Moon) ... However as you said better quality ones are on their way. Thanks for the great answer. – Michel Gokan Khan Jul 16 '15 at 14:19
• The mars and moon google maps were created by satellites in orbit around the objects they were mapping and at much closer distances than those of the pluto flyby < 500 miles as opposed to 7000+ miles – Mauro Jul 16 '15 at 14:54
• @MichelKogan Kepler doesn't take pictures of exoplanets. It measures brightness of the parent star over time and then that data is analyzed for possible transiting planets by looking for periodic dips in brightness as that planet completes orbits around its star and blocks some of its light when between the star and Kepler. Direct imaging of exoplanets is problematic for a plethora of reasons, e.g. exoplanet's dim albedo compared to its nearby star and the need for distant occulting objects in between the star and observer. But that's a completely different questions, so please ask it as such. – TildalWave Jul 16 '15 at 20:51
• @DrZ214 You're reading things that aren't there. We would need an even larger telescope. That can then be anything. Including JWST that will have much, much higher resolving power (diameter of 6.5 m), albeit not in the visible light spectrum. But currently, HST is the largest diameter (optical or otherwise) space telescope we have. E.g. Kepler is only 0.95 m, Spitzer 0.85 m, WISE is 0.4 m,... And frankly, I personally wouldn't launch on a Proton anything that's prohibitively costly to replace. It has a rather poor reliability record with some spectacular failures. Some recent. – TildalWave Jul 16 '15 at 20:59
• To get anywhere near the resolution of the New Horizons image, you would need a telescope with a mirror diameter of several kilometers wide. I doubt any rocket will be big enough to put that into outer space. – Rob Jul 17 '15 at 8:37

For getting really high resolution stuff, you need to get outside of the atmosphere. The best instrument to do this is the Hubble Space Telescope. The resolution of Hubble is about 0.05 arc seconds. Pluto is currently around 3.5 billion miles. That gives a resolution of 850 miles or so. This is limited by the size of the mirror that Hubble has, which is about 2m wide. This is all very rough, mind you. Pluto is a rather small object, of 1470 miles, give or take. So Pluto is only a few pixels from Hubble. This was further explained by the Planetary Society.

The absolute best image that came before New Horizons was done by watching Pluto for a long time, and doing some very complex stuff, to come up with the following:

All in all, I heard it cited that until 9 days from the closest approach, Hubble could do a better job of imaging Pluto than New Horizons. (Although that is with the super resolution image, as Emily Lakdawalla points out, it actually started near the beginning of this year. The super resolution with Hubble is easier because we have more data from it, and more accurate data.) And most of the images we've seen so far have been from before that period of time. Even then, while some of the features from the Hubble composite are visible (In particular, the Heart), the detail has been better for a number of days for New Horizons.

The images that were captured, and will be sent to Earth soon, are of much higher resolution than what has been seen so far. It will take as long as 18 months to get all of the data from the flyby to the ground. Some of the high resolution images are already showing up, more of them will come.

The summary is that, yes, we have amazing cameras near Earth, but the distance is astronomical to Pluto, requiring sending a spacecraft close to get a good image.

• Feel free to accept it then with the checkbox;-) – PearsonArtPhoto Jul 16 '15 at 14:57
• "the distance is astronomical" <-- literally – loneboat Jul 16 '15 at 15:23
• Is the bright feature in the 180 view the "heart"? – Random832 Jul 17 '15 at 4:07
• I believe so. It's fun to compare for sure. – PearsonArtPhoto Jul 17 '15 at 9:43

To add to some already great answers, I'd like to toss in the basic physics of photography that you can experience here on Earth.

When you take a picture of something, you're collecting information in the form of light that has been reflected off of it. In a bright, well-lit environment where you're taking pictures up-close or have a huge lens, you can collect enough light to make the detail of the image apparent in a fraction of a second. When you shoot in low-light, the aperture of the camera has to stay open longer to collect enough light to make the detail discernible (for an easy example, think of how it takes time for your eyes to adjust to low light enough for you to see well in it - imagine a camera shutter having to stay open that long to collect that much light to see clearly as well), so the object you're shooting has to stay perfectly still during that time or the picture will end up blurred. The same effect applies when you're shooting something very far away, even if it's well-lit, because a smaller amount of reflected light is making its way to your lens (the light isn't all reflecting directly to you, it's going in many directions and spreading out, weakening).

With Pluto we're combining both of those problems, extremely low light and extremely long distance. We receive such a tiny amount of reflected light from Pluto here on Earth that it's not even visible without excellent optics, and then virtually only as a speck of light. In order to get a "higher-resolution" image from our vantage point, we'd have to make a single exposure of Pluto last for weeks or longer (compared to the fraction of a second required to take pictures here), but over the course of those weeks, the Earth is orbiting the sun, Pluto is orbiting the sun, Pluto is rotating, etc. so that basically nothing is holding still for your shot. We can do some amount of correction for this but it's still limited in its effectiveness.

There are only two solutions. Build a bigger telescope to collect more reflected light in less exposure time, or get closer. The size of telescope required to take higher-resolution images would be enormous and vastly less cost-effective than simply bolting a camera on a metal box and slinging it at Pluto as fast as humanly possible.

Apologies to the New Horizons team for the astonishing over-simplification of their mission... I'm very much in awe of their work.

• "[Pluto is] not even visible without excellent optics in orbit" Pluto was discovered in 1930. I'm not aware of any space-based telescopes in that era. – user Jul 17 '15 at 12:17
• I stand corrected! – thanby Jul 17 '15 at 14:38

For over forty years, there have been man made objects sitting on the moon that are not exactly tiny and it is only quarter of a million miles away, but even our best telescopes have been unable to get a decent photo of them. Pluto is between two and a half, and four and a half billion miles away.