Sky & Telescope - July 2019 - 37

The Queensland Astrofest presented a unique opportunity
to test experienced astronomers under exceptionally dark
conditions. Held every August, this 10-night festival is held
at Camp Duckadang, some 160 kilometers northwest of Brisbane, the capital of Queensland, Australia. Skies are exceptionally dark, with a limiting magnitude of 6 or better, and
the core of the Milky Way overhead often casts a shadow!
Given that the test protocol would take up to 90 minutes
per astronomer, the 10 nights gave me sufﬁcient time to test
15 (very patient) astronomers. Prior to testing I conﬁrmed
they didn't suffer from color blindness, poor visual acuity,
diabetes, or other exclusion criteria. I also determined their
limiting magnitudes by asking them to count how many stars
they could reliably see with dark-adapted, direct vision in a
speciﬁc area of the constellations of Centaurus and Crux, as
deﬁned by the International Meteor Organization.
During the actual test, once participants were fully darkadapted, they read an optometry visual acuity chart from a
laptop computer, comprising black lettering on variously colored backgrounds. They could adjust the brightness as much
as they needed to read ﬂuently out loud at their baseline
value. I determined the screen brightness using a lux meter.
After 1 minute of viewing the chart, I turned the screen off.
I then reassessed their limiting magnitudes 1 and 4 minutes later to see how reading the chart affected their dark
adaptation. I used two metrics: The change in star count with
direct vision served to test their cones, while the ability to see
both the Coalsack dark nebula, which lay within the test ﬁeld,
and the nearby Small Magellanic Cloud were a proxy for rods.
We did the test in turn using red, orange, green, white,
and ﬁnally bright red (4 lux) backgrounds, with time to
fully readapt between each color test. (For comparison, a
white sheet of paper illuminated by the full Moon is 0.1 lux.)
Viewed in the daytime, all these screens are virtually black!
I also conducted each test at the same time each night, with
essentially identical sky conditions.

ber of stars visible at different magnitude limits. The decrease
in the number of stars visible after exposure to the various
colored screens shows orange lighting to have a clear advantage under the dark skies we enjoy at Queensland Astrofest.
Exposure to green, white, or bright red light decreases the
total visible star count by around 50%.
Thirteen of the 15 participants could detect the Coalsack
and Small Magellanic Cloud 1 minute after exposure to
orange light. This fell to 11 participants after red light exposure and only six after exposure to green light.
Dark re-adaptation was essentially complete by 4 minutes
after exposure to red and orange light, and within 10 minutes
after green, white, or bright red light.

Decrease in Limiting
Magnitude at 1 Minute

Decrease in magnitude limit

The Trial

0.40
0.60
0.80
1.00
Red

Orange

Green

White

Bright red

Color

Cumulative Number of Stars
vs. Magnitude Limit
5,000
4,500

Full dark adaptation

4,000
3,500

Number

THE AU THOR (2), CU M UL ATIV E STA RS G R A PH A DA PTED FRO M NICK LO MB,
DATA FRO M N ATION A L SOL A R OBSERVATORY SACR A MEN TO PE A K

0.20

1.20

Outcome
Red screens needed to be three times as bright as green
screens for participants to read at their baseline visual acuity
level (facing page). The mean brightness of the orange screens
was brighter than the green screens but dimmer than the red
screens. Orange light caused the least loss of dark adaptation
for both cone and rod vision, at both 1 minute and 4 minutes
after exposure. As expected, green and white caused the most
impairment, but a bright red screen was as damaging to night
vision for both rod and cone vision as a green or white screen,
conﬁrming night vision is not immune to red light. Green
light, even though it was signiﬁcantly dimmer than other
colors, was the most damaging to rod vision.
How this translates to practical viewing can be seen in the
diagram comparing the number of stars visible versus limiting magnitude at right. The curved line, developed by Nick
Lomb of Sydney Observatory, represents the theoretical num-

0.00

3,000
2,500

Orange
Red
Green, white, bright red

2,000
1,500
1,000
500
0
-1

3
4
Magnitude

p RESULTS: STAR COUNTS Top: The faintest magnitude observers could see after light exposure changed depending on the
color (and brightness) of the test screen. Bottom: Translated to
number of stars visible, even a small decrease in limiting magnitude has a significant effect.

s k y a n d t e l e s c o p e . c o m * J U LY 2 0 1 9

http://www.skyandtelescope.com

Sky & Telescope - July 2019

Table of Contents for the Digital Edition of Sky & Telescope - July 2019