Astrophotography

Astrophotography is photography where the subject is faint, the exposure is long, and the target is moving because the Earth is turning. Everything difficult about it follows from those three facts.

It splits into distinct disciplines that share almost no equipment. Wide-field work — the Milky Way over a landscape, star trails, constellations — needs a camera, a fast lens and a tripod, and is the most rewarding place to start by a wide margin. Deep-sky imaging of galaxies and nebulae needs accurate tracking and hours of accumulated exposure. Planetary and lunar imaging is the opposite problem: bright targets, thousands of very short frames, and software that picks the sharpest moments out of atmospheric turbulence.

The counterintuitive part, and the thing that saves beginners the most money: a mediocre telescope on an excellent mount produces better images than an excellent telescope on a mediocre mount. Optics gather light; the mount determines whether that light lands on the same pixels for the whole exposure. Almost every disappointed beginner bought the wrong one first.

Buy in this order

Mount, then camera, then opticsThe reverse of what feels natural, and the reason most beginner rigs disappoint.

Best possible start

A camera you own, a fast wide lens, a tripod

What limits you most

Cloud, then light pollution, then equipment

Where the image is actually made

In processingThe raw frames look like almost nothing. This is not cheating; it is the craft.

1. 1

   Step 1: Start with the camera and tripod you already have

   A wide lens at its widest aperture, focused manually on a bright star using live view at maximum zoom, ISO around 1600–3200, and an exposure short enough that stars stay points rather than streaks. That is a real astrophotograph and it costs nothing. Autofocus does not work on stars; expect to focus manually every session.

   You will know it worked when: A Milky Way or constellation image, and a sense of whether you enjoy standing outside in the cold.

2. 2

   Step 2: Learn how long you can expose before stars trail

   The sky moves. How long you can expose depends on your focal length and where you are pointing — wide lenses tolerate far longer than telephoto, and stars near the celestial pole move less than those on the equator. Find your limit empirically by taking a bracket and zooming in; the published rules of thumb are approximations that vary with sensor and with how fussy you are.

   You will know it worked when: You understand the constraint that everything else exists to remove.

3. 3

   Step 3: Learn to stack

   The central technique of the entire hobby. Take many identical frames and average them: the signal adds consistently and the random noise partially cancels, so twenty frames are dramatically cleaner than one. Free stacking software does this well. This is the step where beginners discover that faint detail was in their data all along.

   You will know it worked when: Images far better than any single exposure, from equipment you already own.

4. 4

   Step 4: Add a star tracker before you add a telescope

   A small motorised head that rotates the camera to follow the sky. It converts 20-second exposures into two-minute exposures, which is a larger improvement than any lens upgrade, for £250–£450. This is the highest-value purchase in the hobby.

   You will know it worked when: Long exposures with pinpoint stars, still using camera lenses.

5. 5

   Step 5: Take calibration frames

   Darks, flats and bias frames: extra images that let the software remove sensor noise, dust shadows and vignetting. They are tedious, they feel like bookkeeping, and they make more difference to a final image than most equipment upgrades.

   You will know it worked when: Clean backgrounds and even illumination.

6. 6

   Step 6: Only then consider a telescope and equatorial mount

   And when you do, spend the money on the mount. A short focal length refractor is far more forgiving than a long one, and forgiveness is what you want while learning polar alignment, guiding and framing all at once.

   You will know it worked when: A deep-sky rig you can actually get results from.

A single raw frame of a galaxy looks like a grey rectangle with a few stars. This alarms beginners, who assume something has gone wrong. Nothing has: the signal is genuinely there and it is buried under noise, and the entire discipline is about improving that ratio.

Two things improve it. Total integration time — the sum of all your exposures — increases the signal relative to random noise. Doubling your total exposure time does not double image quality, though; the improvement goes roughly as the square root, which is why serious deep-sky images represent many hours across multiple nights rather than one long session. And calibration removes the systematic errors that stacking cannot: sensor bias, thermal signal, dust shadows and vignetting.

Then the image is stretched. Astronomical data is recorded linearly and the interesting detail occupies a tiny fraction of the range, so it must be non-linearly stretched to be visible at all. This is the step that looks like it must be faking something and is not: the photons were recorded, and stretching is how you display data with an enormous dynamic range on a screen that has very little.

The techniques that make the largest difference:

• Polar alignment — Aligning the mount’s rotation axis with the Earth’s. Everything else depends on it, and poor alignment shows up as field rotation and drifting stars that no amount of processing fixes.
• Dithering — Shifting the pointing very slightly between frames so that sensor artefacts fall in different places and average away during stacking. Costs nothing and removes a whole class of pattern noise.
• Flat frames — Images of an evenly illuminated field, taken through the same optical train without changing focus, which let software divide out vignetting and dust. The calibration frame that most obviously improves an image.
• Narrowband imaging — Filtering to specific emission lines — hydrogen-alpha, oxygen III, sulphur II — so nebula signal passes and light pollution largely does not. This is what makes deep-sky imaging possible from suburbs.
• Lucky imaging — For planets and the Moon. Capture thousands of very short frames as video, then let software select and align the sharpest few per cent. It works because atmospheric turbulence briefly steadies, and it is why amateur planetary images now exceed what large observatories managed decades ago.
• Mosaicking — Shooting overlapping panels and stitching them, for targets larger than your field of view.

In the order that actually improves your images:

• A sturdy tripod — Every long exposure is a test of how still the camera is. A flimsy tripod is the cheapest thing to fix and the most commonly ignored.
• A fast wide lens — Something that opens to f/2.8 or wider. Aperture matters more than megapixels here, because you are collecting photons that are genuinely scarce.
• A star tracker — The transformative purchase. Needs rough polar alignment — pointing its rotation axis at the celestial pole — which takes a few minutes and becomes routine.
• An equatorial mount — The foundation of deep-sky work, and where the budget should concentrate. Mounts are rated by payload capacity and the standard advice is to load them to about half of it for imaging, because the published figure assumes visual use.
• A telescope — For imaging, a short focal length apochromatic refractor is the forgiving choice: less demanding on tracking, wider field, no collimation to maintain. Long focal lengths magnify both the target and every tracking error.
• A camera — A modern mirrorless or DSLR is entirely capable. Dedicated astronomy cameras add cooling — which reduces thermal noise substantially on long exposures — and mono sensors with filter wheels offer more control and considerably more work.
• An autoguider — A small secondary camera that watches a star and corrects the mount’s tracking in real time. The step that takes usable exposures from a couple of minutes to ten or more.
• Filters — Light pollution and narrowband filters isolate specific wavelengths and are what makes imaging from a town viable. Narrowband in particular can produce excellent nebula images under skies that are hopeless for broadband work.
• Dew control — A heated band around the optic. In Britain, dew ends more imaging sessions than cloud does at certain times of year, and it is entirely preventable for £25.

1. 1

   Step 1: Untracked wide-field

   Camera, fast lens, tripod. Constellations, the Milky Way, star trails.

   You will know it worked when: Real images for no outlay, and an honest sense of whether you like the cold.

2. 2

   Step 2: Stacking

   Many frames, free software, calibration frames. Nothing new to buy.

   You will know it worked when: A dramatic quality jump from equipment you already own.

3. 3

   Step 3: A star tracker

   Minutes-long exposures with camera lenses. Nebulae and galaxies become achievable.

   You will know it worked when: The single largest improvement per pound in the hobby.

4. 4

   Step 4: A small refractor on an equatorial mount

   Polar alignment, framing, and the first real deep-sky targets at proper scale.

   You will know it worked when: Images that look like the ones that made you want to try this.

5. 5

   Step 5: Guiding and narrowband

   Long sub-exposures and imaging through light pollution.

   You will know it worked when: Serious results from a suburban garden.

6. 6

   Step 6: Mono, filter wheels, multi-night projects

   Separate luminance and colour channels, tens of hours per target, remote operation.

   You will know it worked when: Work indistinguishable from professional images of a generation ago.

The hobby has an unusually strong data-sharing norm. Astrophotographers routinely publish their acquisition details — equipment, exposure lengths, total integration, filters, sky quality — alongside the image, and an image posted without them is regarded as slightly incomplete. That transparency is what makes the community useful to learn from, because you can see exactly what produced a result.

There is a live and good-natured argument about processing. Where the line sits between revealing recorded data and inventing it is genuinely debated, and it has sharpened with AI-based noise reduction and detail enhancement tools that can generate plausible structure that was never recorded. The prevailing norm is disclosure rather than prohibition: say what you used.

Amateur astrophotography also makes real contributions. Amateurs discover comets and supernovae, monitor variable stars, and image transient events that professional telescopes are not pointed at. The gap between a well-equipped amateur and a professional facility is smaller here than in almost any other science.

The rest is unglamorous and matters:

• Tell somebody where you are going — Dark sky sites are by definition remote and unlit, and you will be there alone at night for hours. A location, a return time and a charged phone.
• Cold is the real hazard — Standing still outdoors for four hours in winter is colder than any activity you have a reference point for. Layer far beyond what feels necessary, and account for the fact that you will not be moving.
• Use red light, and manage cables — Red preserves dark adaptation and is a courtesy at any shared site. Cables across dark ground are the commonest cause of both falls and pulled equipment.
• Check access and permissions — A field at 2am is somebody’s field. Established dark sky sites and Dark Sky Discovery locations exist partly to solve this, and turning up where you are entitled to be avoids an awkward conversation with a farmer.

• Widefield and nightscape — A camera on a tripod pointed at the Milky Way over a landscape, with no telescope anywhere. The cheapest branch, the one most people start in without realising it is a branch, and the one where the limiting factor is a dark site rather than equipment.
• Star-tracker widefield — The same shot with a small motorised mount cancelling the Earth’s rotation, which turns a 20-second exposure into a four-minute one. The single largest step in result for the money in the whole hobby, and it happens before a telescope is involved at all.
• Deep-sky imaging — Galaxies and nebulae through a telescope on an equatorial mount, usually guided, stacked from many sub-exposures. This is what most people mean by the term and it is the most demanding branch by a wide margin — the mount matters more than the optics, which is the opposite of what newcomers expect.
• Planetary and lunar — Short video runs of a bright object, with software keeping the sharpest frames and discarding the rest. Almost the inverse discipline: it rewards aperture and steady air rather than dark skies, works from a city centre, and a full Moon helps rather than ruins it.
• Solar — The Sun in white light or in a narrow hydrogen-alpha band, with dedicated filtration. A daytime hobby with its own equipment and its own absolute safety rule, covered in the safety section — the filter goes at the front of the telescope, never at the eyepiece.
• Smartphone and afocal — A phone held or clamped to an eyepiece. Genuinely capable on the Moon and the brighter planets, unhelpful on anything faint, and worth knowing about because it answers the question of whether the interest is real before any money is spent.
• Electronically assisted astronomy — Live stacking on a screen at the telescope, somewhere between observing and imaging. It has quietly become the way a great many people show faint objects to other people, and it sidesteps the processing that puts most beginners off.