Introduction to Sky Quality Measurement

The science behind quantifying darkness - from professional instruments to smartphone analysis

For most of human history, we described the night sky qualitatively: dark, bright, clear, hazy. Astronomers might note how many stars they could see in a constellation, but there was no universal standard for "how dark is dark?"

That changed in the late 20th century with the development of scientific instruments to measure sky brightness. Today, we have multiple methods - from expensive professional equipment to citizen science approaches like SkyQI.

This article explores the science of sky quality measurement.

Why Measure Sky Quality?

For Astronomy

Astronomers need to know sky brightness because it determines:

  • What objects they can observe
  • What exposure times they need
  • Whether observations will be scientifically useful
  • Whether an observing site is suitable

For Environmental Monitoring

Light pollution is an environmental issue. To track it, we need:

  • Baseline measurements
  • Change detection over time
  • Spatial mapping of pollution sources
  • Data to inform policy

For Research

Scientists studying light pollution's effects need quantifiable data:

  • Correlation with wildlife impacts
  • Health outcome studies
  • Urban planning research
  • Climate/energy studies

Units of Measurement

Magnitudes Per Square Arcsecond (mag/arcsec²)

The standard unit for sky brightness is magnitudes per square arcsecond.

Breaking it down:

  • Magnitude - The astronomical brightness scale. Each magnitude step is ~2.512x brightness difference. Lower numbers = brighter.

  • Arcsecond - A tiny unit of angular measurement. One degree = 3600 arcseconds.

  • Square arcsecond - A tiny patch of sky, 1 arcsecond × 1 arcsecond.

So "20.0 mag/arcsec²" means: "Each tiny square-arcsecond patch of sky has a surface brightness equivalent to a magnitude 20 star."

The scale:

  • 22+ mag/arcsec² = pristine darkness
  • 16 mag/arcsec² = heavily polluted urban sky

Higher numbers = darker sky (because magnitude scale is inverted).

Other Units

Candelas per square meter (cd/m²)

  • SI unit for luminance
  • Used in lighting industry
  • Conversion: mag/arcsec² ≈ 12.58 − 2.5 × log₁₀(cd/m²)

Lux

  • Measures illumination (light falling on surface)
  • Related but different from sky brightness
  • Natural starlight: 0.0001-0.001 lux
  • Full moon: 0.1-0.3 lux
  • Street lighting: 10-50 lux

Traditional Measurement Methods

Naked Eye Limiting Magnitude (NELM)

How it works: Observers count how many stars they can see in specific constellations (often the Little Dipper) and determine the faintest visible star.

Pros:

  • No equipment needed
  • Historical method with long baseline data
  • Intuitive (directly measures what you can see)

Cons:

  • Highly subjective (depends on observer's vision)
  • Limited precision (±0.5 mag at best)
  • Requires trained observers
  • Can't measure very bright skies accurately

NELM to SQM conversion: Roughly, SQM ≈ 2.1 × NELM + 8.5 (varies with conditions)

Star Counts

How it works: Count stars visible in a defined area of sky.

Globe at Night uses this method - observers compare their sky to reference charts showing different levels of star visibility.

Pros:

  • Simple for citizen science
  • No equipment needed
  • Good for education

Cons:

  • Coarse resolution (typically 1 Bortle class steps)
  • Subjective
  • Depends on observer knowledge

Electronic Measurement Devices

Sky Quality Meter (SQM)

The Unihedron SQM is the de facto standard instrument for amateur and semi-professional sky quality measurement.

How it works:

  • Photodiode sensor measures light
  • Internal processing converts to mag/arcsec²
  • Averages over ~80° field of view (SQM-L: ~20°)
  • Battery-powered handheld device

Specifications:

  • Range: ~16-22+ mag/arcsec²
  • Precision: ±0.1 mag/arcsec²
  • Cost: roughly ₹12,000–35,000, depending on the model

Pros:

  • Standardized, widely used
  • Quick measurements
  • Decent accuracy
  • Portable

Cons:

  • Expensive for citizen science
  • Single-point measurements
  • Limited spectral information
  • Requires purchase and maintenance

CCD/CMOS Imaging

Professional observatories use calibrated CCD cameras for sky brightness measurements.

How it works:

  • Calibrated camera images the sky
  • Software analyzes pixel values
  • Converts to absolute brightness units
  • Can create sky brightness maps

Pros:

  • Very accurate when calibrated
  • Spatial resolution (can map variations)
  • Spectral information available
  • Research-grade data

Cons:

  • Expensive equipment
  • Complex calibration required
  • Requires expertise
  • Not practical for citizen science

Satellite Observations

Satellites like VIIRS (Visible Infrared Imaging Radiometer Suite) on the Suomi NPP satellite measure light emissions from space.

Pros:

  • Global coverage
  • Regular temporal sampling
  • No ground equipment needed
  • Excellent for large-scale mapping

Cons:

  • Measures light going up, not sky brightness
  • Coarse resolution (roughly 500–750 m pixels)
  • Doesn't directly measure what observers see
  • Cloud interference

Smartphone-Based Measurement

The Challenge

Smartphones aren't designed for scientific measurement. Their cameras:

  • Have automatic processing that changes with each shot
  • Use different sensors across manufacturers
  • Apply noise reduction that can remove stars
  • Have limited dynamic range
  • Lack calibration data

So how can SkyQI produce useful measurements?

The SkyQI Approach

Rather than trusting any single number from a phone, SkyQI combines several pieces of evidence:

1. Camera Exposure Settings

  • Most phones save the shutter time, ISO and aperture with every photo
  • SkyQI uses them to convert the sky's pixel brightness into mag/arcsec², the same unit an SQM reports
  • This is what makes photos from different phones comparable
  • If a photo has lost these settings (forwarded or edited copies often have), the reading starts from a published world atlas of light pollution for that location, is adjusted using what the photo shows, and is labelled Estimated

2. Star Detection

  • Custom flood-fill algorithm identifies point sources
  • Distinguishes stars from noise/artifacts
  • Star count correlates with sky darkness

3. Horizon Glow Detection

  • Light pollution creates characteristic gradients
  • Much brighter near the horizon than overhead
  • A naturally dark sky is far more even

4. Color Temperature Analysis

  • Light-polluted skies often look orange or yellow
  • Natural dark skies look blue-black
  • Colour helps flag artificial light

5. Cross-Validation

  • Sky brightness, star count and horizon glow are checked against each other
  • When they agree, confidence is high; when they disagree, it drops
  • The reading is still reported, together with its confidence level

Calibration

For now, all phone models share one calibration. As photos with full camera settings accumulate, SkyQI is built to learn a correction for each phone model. SkyQI has not yet been compared side by side with a professional SQM; that is the next test we want to run.

Accuracy

On a set of 104 reference night-sky images with known Bortle classes, an earlier version of SkyQI was within one Bortle class about 65% of the time. That is far less precise than an SQM (±0.1 mag/arcsec²), but:

  • Good enough for a rough Bortle classification
  • Useful for spotting change when many readings are taken over time
  • Useful for mapping when many people contribute
  • Available to anyone with a smartphone

The democratization of measurement is worth the precision trade-off.

The Bortle Scale: Bridging Quantitative and Qualitative

John Bortle created his scale in 2001 to bridge objective measurements and subjective experience.

How It Works

Each Bortle class corresponds to:

  • A range of SQM/NELM values
  • Specific observable phenomena
  • Practical observing conditions
Class SQM Range Key Indicators
1 22.0+ Zodiacal band, gegenschein visible
2 21.89-21.99 Zodiacal light obvious, M33 easy naked eye
3 21.69-21.89 Milky Way very impressive, M33 visible
4 20.49-21.69 Milky Way obvious, zodiacal light visible
5 19.50-20.49 Milky Way washed out but visible
6 18.94-19.50 Milky Way only visible overhead
7 18.38-18.94 Milky Way invisible, sky grey
8-9 <18.38 Only brightest stars visible

Why This Matters

The Bortle scale translates numbers into meaning:

  • "SQM 19.2" is abstract
  • "Bortle 6 - you can barely see the Milky Way overhead" is concrete

For public communication and citizen science, this translation is essential.

Data Quality Considerations

Sources of Error

Instrumental:

  • Sensor calibration
  • Temperature effects
  • Aging/drift

Environmental:

  • Airglow (natural sky brightness variation)
  • Moonlight and twilight
  • Clouds and haze
  • Galactic position (Milky Way overhead vs. not)

Methodological:

  • Observer technique
  • Measurement position/direction
  • Sampling time

Making Measurements Comparable

For scientific comparison, record:

  • Date and time (UTC preferred)
  • Location (GPS coordinates)
  • Direction measured (zenith vs. horizon)
  • Moon phase and position
  • Weather conditions
  • Equipment used
  • Any anomalies

SkyQI captures most of this automatically through metadata.

The Future of Sky Quality Measurement

Trends

More sensors:

  • IoT networks of permanent SQM-like sensors
  • Real-time sky brightness maps
  • Integration with weather/air quality networks

Better satellites:

  • Higher resolution
  • Multi-spectral analysis
  • More frequent coverage

Improved citizen science:

  • Better smartphone algorithms
  • More participants = more data points
  • Machine learning improving accuracy

Standardization:

  • International measurement protocols
  • Data sharing frameworks
  • Policy-relevant metrics

The Citizen Science Revolution

Historically, light pollution data came from:

  • Scattered professional measurements
  • Expensive equipment deployments
  • Limited geographic coverage

Today, citizen science platforms can generate:

  • Thousands of measurements daily
  • Global geographic coverage
  • Temporal density impossible with traditional methods

This data complements (not replaces) professional measurements, creating a more complete picture.

How You Can Contribute

Quality citizen science data requires:

  1. Consistent methodology - Follow photo guidelines
  2. Good documentation - Include metadata
  3. Regular measurements - Track your location over time
  4. Data sharing - Upload to platforms like SkyQI

Your photos become part of a global dataset that:

  • Maps light pollution extent
  • Tracks changes over time
  • Identifies dark sky refuges
  • Informs policy decisions

Every measurement matters. Contribute yours at skyqi.in.

Further Reading

For those wanting to dive deeper:

  • Light Pollution Handbook - Kohei Narisada & Duco Schreuder
  • The End of Night - Paul Bogard
  • Globe at Night - globeatnight.org
  • International Dark-Sky Association - darksky.org
  • Journal of Quantitative Spectroscopy & Radiative Transfer - academic research

This is part of SkyQI's educational series. Visit www.skyqi.in to measure your sky.

Featured Image: images/featured_7_measurement_science.jpg

Tags: #Science #Measurement #SQM #Astronomy #Research #CitizenScience #SkyQI

Category: Science

Reading Time: 9 minutes

Slug: sky-quality-measurement-guide