Stellar Properties and the H–R Diagram
A star is a glowing ball of gas so far away that, even through the finest telescope, it is
never more than a point of light. We can't land on one, weigh it, or push a thermometer into
its surface. And yet astronomers will tell you, with confidence, that Betelgeuse is a cool
red super-giant some 700 times the radius of the Sun, while Sirius B is a searingly hot white
dwarf no bigger than the Earth. How can anyone possibly know such things about an
object they can only ever see as a speck?
The astonishing answer is that a star's light is a message, and it carries almost everything
we need. Two numbers unlock the rest: the star's colour (which fixes its
surface temperature) and its total power output (its
luminosity). This page shows how the two physical laws behind those
numbers — Wien's displacement law and Stefan's law — let us
read a star like a book, and how plotting temperature against luminosity for thousands of
stars reveals one of the most important pictures in all of astronomy: the
Hertzsprung–Russell diagram.
A star glows like a black body
Heat something up and it glows. A poker in a fire runs from dull red, to bright orange, to
yellow-white as it gets hotter; the glowing element of a toaster is a deep cherry red. This is
thermal radiation, and any hot object emits it across a whole spread of
wavelengths at once. A
star is a very good approximation to an idealised black body — a perfect
absorber and emitter — so its radiation follows a smooth, characteristic curve whose
shape depends on only one thing: the temperature of its surface.
The key feature of that curve is where it peaks — the wavelength
\lambda_{\max} at which the star pours out the most energy. A cool
star peaks far into the red; a hot star peaks in the blue or even the ultraviolet. So the
colour your eye (or a spectrometer) sees is a direct read-out of the star's
temperature — hotter means bluer, cooler means redder. That relationship has
an exact form.
Wien's law: colour tells you temperature
The peak wavelength and the surface temperature are locked together by Wien's
displacement law. As the temperature T rises, the peak
wavelength \lambda_{\max} shifts (is "displaced") to
shorter values — the two are inversely proportional:
\lambda_{\max}\,T = 2.9\times10^{-3}\ \text{m K}.
Rearranged for whichever quantity you want:
T = \frac{2.9\times10^{-3}}{\lambda_{\max}} \qquad\text{or}\qquad \lambda_{\max} = \frac{2.9\times10^{-3}}{T}.
Worked example — the Sun's temperature
The Sun's spectrum peaks at about \lambda_{\max} = 500\ \text{nm}
(in the green-yellow). What is its surface temperature?
T = \frac{2.9\times10^{-3}}{500\times10^{-9}} = \frac{2.9\times10^{-3}}{5\times10^{-7}} = 5800\ \text{K}.
Just under 6000 K — which is indeed the temperature of the Sun's visible surface. Notice the
power of this: we measured a temperature 150 million kilometres away purely from the
colour of the light, and nothing else.
Worked example — a cool red giant
Betelgeuse peaks at roughly \lambda_{\max} = 830\ \text{nm}, out in
the near-infrared. Its temperature is
T = \frac{2.9\times10^{-3}}{830\times10^{-9}} \approx 3500\ \text{K}.
Far cooler than the Sun — which is exactly why it glows a deep red. A longer peak wavelength
always means a lower temperature.
Luminosity: how much power a star pours out
Temperature is only half the story. A star's luminosity
L is the total power it radiates — the energy
leaving its whole surface every second, measured in watts. (Don't confuse it with how
bright a star merely appears from Earth: a very luminous star can look faint
simply because it is far away. Luminosity is the true, intrinsic output.)
Because a star radiates like a black body, its luminosity is fixed by just two properties: its
radius r (a bigger surface radiates from more
area) and its surface temperature T (a hotter
surface radiates far more fiercely per square metre). The exact relationship is
Stefan's law:
L = 4\pi r^2 \sigma T^4,
where 4\pi r^2 is the surface area of the (spherical) star and
\sigma = 5.67\times10^{-8}\ \text{W}\,\text{m}^{-2}\,\text{K}^{-4}
is the Stefan constant. The single most important thing to notice is the
T^4: luminosity depends only on the square of the radius, but on the
fourth power of the temperature. Temperature is the dominant lever.
Worked example — the luminosity of the Sun
The Sun has radius r = 7.0\times10^{8}\ \text{m} and surface
temperature T = 5800\ \text{K}. Then
L = 4\pi (7.0\times10^{8})^2 \,(5.67\times10^{-8})\,(5800)^4.
Take it in pieces: 4\pi r^2 = 6.16\times10^{18}\ \text{m}^2, and
(5800)^4 = 1.13\times10^{15}\ \text{K}^4. Multiplying,
L \approx 6.16\times10^{18} \times 5.67\times10^{-8} \times 1.13\times10^{15} \approx 3.9\times10^{26}\ \text{W}.
Nearly 4\times10^{26} watts — the accepted value for the Sun. We
call this one solar luminosity, L_\odot, and use it as a yardstick
for other stars.
Turning it around: finding a star's size
Stefan's law is at its most powerful run backwards. If we already know a star's
luminosity (from its brightness and distance) and its temperature (from Wien's law), we can
rearrange for the one thing we can never measure directly — its radius:
r = \sqrt{\dfrac{L}{4\pi\sigma T^4}}.
It is often cleanest to compare with the Sun. Dividing a star's Stefan equation by the Sun's,
the constants cancel and we get a tidy ratio:
\frac{r}{r_\odot} = \sqrt{\frac{L/L_\odot}{(T/T_\odot)^4}} = \sqrt{\frac{L}{L_\odot}}\left(\frac{T_\odot}{T}\right)^2.
Worked example — how big is a red giant?
A red giant has luminosity L = 10\,000\,L_\odot but a cool surface
of T = 3500\ \text{K} (against the Sun's 5800 K). How large is it?
\frac{r}{r_\odot} = \sqrt{10\,000}\times\left(\frac{5800}{3500}\right)^2 = 100 \times (1.66)^2 \approx 100 \times 2.75 \approx 275.
The giant is about 275 times the radius of the Sun. Here is the crucial
lesson: this star is cooler than the Sun, so each square metre of it radiates
less than the Sun's surface — yet it is dazzlingly luminous. The only way to reconcile
those facts is that it must be enormous. A cool star that is still very
luminous is telling you, through Stefan's law, that it is gigantic.
The Hertzsprung–Russell diagram
Around 1910, Ejnar Hertzsprung and Henry Norris Russell independently had the same brilliant
idea: take thousands of stars, and for each one plot its luminosity up the
vertical axis against its surface temperature along the horizontal axis. If
stars were scattered randomly, the plot would be a shapeless smear. They are not. Instead the
stars fall into a few sharply defined regions — a pattern so revealing that the
Hertzsprung–Russell (H–R) diagram became the single most useful chart in
stellar astronomy.
Two quirks of the axes trip up every newcomer, so fix them now:
-
Temperature runs backwards. For historical reasons the horizontal axis has
temperature increasing to the left — hot blue stars sit on the left, cool red stars
on the right. It is the one axis in physics that is habitually drawn in reverse.
-
Luminosity is on a logarithmic scale. Stars span an incredible range —
from about 10^{-4} to 10^{6} times the
Sun's output — so each step up the axis is a factor of ten, not a fixed amount.
Explore the diagram below. Use the selector to spotlight each family in turn, and find where
our Sun sits.
Reading the regions
Three great groupings dominate the diagram, and — armed with Stefan's law — you can work out
the story behind each just from where it lies:
-
The main sequence. The broad diagonal band, sweeping from hot-and-luminous
at the top-left down to cool-and-dim at the bottom-right, holds about
90% of all stars — including the Sun. Every star here is doing the same
thing: steadily fusing
hydrogen into helium in its core. A star's position along the band is set by
its mass: the heaviest stars are the hot, brilliant O and B stars at the
top-left, the lightest are the cool, dim red dwarfs at the bottom-right.
-
Red giants and super-giants. Up in the top-right sit
stars that are cool (so, to the right) yet hugely luminous (so, high up).
As we saw, cool-but-luminous can only mean vast — these are stars that have
swollen enormously late in life. Super-giants like Betelgeuse occupy the very top.
-
White dwarfs. Down in the bottom-left lie stars that are
hot (to the left) yet faint (low down). Hot-but-dim can only mean
tiny — these are the exposed, Earth-sized dead cores of former Sun-like
stars, radiating fiercely per square metre but with almost no surface area to do it from.
The diagram is not just a snapshot — it is a map a star moves across as it
ages. A star spends most of its life sitting almost still on the main sequence. When its core
hydrogen runs out it swells and cools, so it migrates up and to the right
into the red-giant region. Finally, a Sun-like star sheds its outer layers and its hot little
core drops down to the bottom-left to become a white dwarf, slowly cooling
and fading along the way. The whole
life cycle of a star
is a journey traced out on the H–R diagram.
For a star treated as a black body of surface temperature T and
radius r:
-
Wien's displacement law. The wavelength of peak emission is inversely
proportional to temperature:
\lambda_{\max}\,T = 2.9\times10^{-3}\ \text{m K}. Hotter stars
peak at shorter (bluer) wavelengths; cooler stars peak at longer (redder) ones.
-
Stefan's law. The total power radiated is
L = 4\pi r^2 \sigma T^4, with the Stefan constant
\sigma = 5.67\times10^{-8}\ \text{W}\,\text{m}^{-2}\,\text{K}^{-4}.
Luminosity grows with the square of the radius and the fourth power of the
temperature.
Four traps catch almost everyone on the H–R diagram:
-
Temperature increases to the LEFT. The horizontal axis is reversed. The
hottest stars are on the far left, the coolest on the far right — the opposite of every
normal graph. Always check which way you are reading it.
-
A blue star is HOTTER than a red one. It feels backwards — we think of red
as "hot" and blue as "cold" from taps and flames the wrong way round. But blue light is
short-wavelength light, and by Wien's law short \lambda_{\max}
means high temperature. Blue = hot, red = cool.
-
A red giant is luminous because it is HUGE, not because it is hot. It is
actually cooler than the Sun. Its enormous output comes entirely from its vast
surface area (Stefan's law, big r), overwhelming the low
temperature.
-
Luminosity is not apparent brightness. How bright a star looks
depends on its distance too. The H–R diagram uses the star's true power output,
not how faint or dazzling it happens to appear from Earth.
It seems like magic, but it is just these two laws working together. First, split the star's
light into a spectrum and find the wavelength where it is brightest — Wien's law instantly
converts that peak into a surface temperature. Second, measure how bright
the star appears and combine it with its distance (found by other means, such as parallax) to
get its true luminosity. Now feed both numbers into Stefan's law rearranged
for radius, r = \sqrt{L / (4\pi\sigma T^4)}, and out drops the
star's size.
That is genuinely how we know Betelgeuse is a bloated super-giant hundreds of times wider
than the Sun. And it is why its mysterious Great Dimming in 2019–2020 —
when the star suddenly faded by a third — caused such excitement: astronomers wondered
whether they were watching its temperature or size change in the run-up to a supernova. (It
turned out to be a giant belch of dust partly hiding the star.) Every clue came from
decoding its light with the very laws on this page — no spaceship required.