Home / Learn / Science / Astronomy / Lesson 5.1
All lessons

Module 5: Doing real research · Lesson 5.1 · about 25 min

Aperture photometry: measuring a star's brightness

Add up the light inside a circle around a star, subtract the sky measured in a ring around it, and compare with stars of known brightness. That simple recipe gives magnitudes good to a few thousandths with care.

Key points, meanings and formulas on one page: choose "Save as PDF" in the print box.

Module 5 · about 25 minutes · read first: 1.5 Flux, luminosity and the inverse square law, Sky watching 6.2 Exposure, noise and signal-to-noise, Sky watching 6.3 Calibration frames: darks, flats and bias · maths: sums and logarithms. You’ll be able to measure a star’s brightness on an image, choose the aperture and sky ring sizes, and turn counts into a calibrated magnitude using comparison stars.

To measure a star's brightness, add up the light inside a circle around it, subtract the sky measured in a ring, and compare with stars of known magnitude in the same image.

1. Why this matters

Photometry, measuring brightness, is the backbone of a huge amount of research: variable stars, exoplanet transits, supernova light curves, asteroid rotation. It is also where amateurs with modest cameras make real contributions, because careful photometry needs patience more than a big telescope.

2. The recipe

  1. Find the star’s centre on a calibrated image (Sky watching 6.3).
  2. Aperture: add up every pixel’s counts inside a circle of radius r.
  3. Sky ring (annulus): take the median of pixels in a ring well outside the star.
  4. Subtract: star counts = aperture sum − (number of aperture pixels × sky per pixel).
  5. Instrumental magnitude: minst=−2.5log10⁡(counts).
  6. Compare with stars of known magnitude in the same image.
Try it: measure a star
star counts (sky removed)
signal-to-noise
instrumental magnitude
share of the star's light caught
A made-up 41 × 41 pixel image: one star (true total 20,000 counts, star width FWHM 4 px) on a sky of 100 counts per pixel, with random photon noise. The sky ring is 5 px wide. Too small an aperture misses starlight; too big adds sky noise. The best signal-to-noise usually comes at a radius near the FWHM.

Things to notice:

  • A tiny aperture misses part of the star; a big one adds sky noise. The best signal-to-noise comes near the star’s width (FWHM).
  • Moving the sky ring onto the star makes the sky too high and the star too faint.

Measure one star in a calibrated image:

Find the centre. Locate the star's centre to a fraction of a pixel.
The aperture. Add every pixel inside a circle of radius about 1.5 times the star's blur width. Say 120 pixels total 85,000 electrons.
The sky ring. In a ring further out, the median sky is 150 electrons per pixel. The aperture holds 120 × 150 = 18,000 of sky.
Subtract. Star alone: 85,000 − 18,000 = 67,000 electrons.
Compare. A comparison star of magnitude 11.20 gives 100,000 electrons. m = 11.20 − 2.5 log₁₀(67,000/100,000) = 11.63.
Maths and principle

Sky-subtracted counts, with aperture sum Σ, n pixels and sky b per pixel:

S=Σ−nb

Signal-to-noise (Sky watching 6.2), ignoring read noise and with nsky pixels in the ring:

SNR=SS+nb(1+nnsky)

Differential magnitude against a comparison star c of known magnitude mc:

m=mc−2.5log10⁡SSc

Uncertainty in magnitudes: σm≈1.086/SNR.

3. What it means physically

A star’s light is spread over several pixels by the air and optics (Sky watching 3.3), so you must gather it all back up. The sky isn’t black: air glow, light pollution and moonlight add light to every pixel, including those under the star. Comparing with stars in the same frame is powerful because thin cloud or changing altitude dims all of them by the same fraction, and a ratio cancels it out.

S = Σ − n b
Star signal: aperture sum minus sky per pixel times pixels.
SNR
Signal-to-noise: how much bigger the signal is than its random scatter.
m = mc − 2.5 log₁₀(S/Sc)
Differential magnitude against a comparison star in the same image.
Physical meaning
It is like weighing a parcel on a scale while the box is still on it: weigh the box alone (the sky) and subtract. Comparing with stars in the same picture cancels clouds, airmass and the camera's sensitivity, because they affect every star in the frame alike.

“A bigger aperture always gives a better measurement.” A bigger circle catches a little more starlight but much more sky noise. For faint stars, an aperture about 1 to 1.5 times the blur width gives the best signal-to-noise.

4. Worked example

The target gives 15,400 counts in its aperture of 50 pixels; the sky ring median is 120 counts per pixel. A comparison star of magnitude 11.20 gives 25,000 sky-subtracted counts. What is the target’s magnitude?

S=15,400−50×120=9,400
m=11.20−2.5log10⁡9,40025,000=11.20−2.5×(−0.425)=11.20+1.06=12.26

SNR ≈9400/9400+6000=76, so σm≈1.086/76=0.014 mag.

5. Practice

More practice (answers below):

  1. An aperture of 30 pixels sums 8,000 counts; sky is 150 per pixel. What are the star counts and the instrumental magnitude?
  2. Why should comparison stars have a similar colour to the target?
  3. A measurement has SNR 200. What is its uncertainty in magnitudes?
Answers
  1. S=8000−4500=3500; minst=−2.5log10⁡3500=−8.86.
  2. The air dims blue light more than red (Sky watching 2.6), so stars of different colours would change by different amounts as altitude changes.
  3. 1.086/200=0.005 mag.
  • Aperture photometry: sum inside a circle, subtract the sky from a ring.
  • S = Σ − n b; noise grows with both star and sky light.
  • Differential photometry against comparison stars cancels clouds and airmass.
  • Magnitude error ≈ 1.086 / SNR.

Try it with real numbers: open your own images in the FITS viewer, check focus with the star FWHM checker, and convert results with the magnitude converter.

New words: photometry, aperture, annulus, instrumental magnitude, differential photometry, comparison star.

Check yourself

1Why measure the sky in a ring around the star?
2Why use the median of the sky ring rather than the mean?
3Differential photometry compares the target with:

Sources

  • Howell, S. B. (2006), Handbook of CCD Astronomy, Cambridge University Press (method described, not copied)

Your ticks stay in this browser. Save them as a file on the course home.

Updated 2026-11-22. Numbers in the examples are made up to show the idea unless a source is given. Simulations run in your browser on made-up data; nothing you type is sent anywhere. Astronomy only: nothing here is astrology.