---
title: Magnitudes, Fluxes, and the Distance Modulus
draft: false
module: Observational Foundations
moduleNumber: 2
lessonNumber: 1
order: 201
summary: >
  The brightness of a star reaches us as a radiant flux that falls off as the
  inverse square of distance. The magnitude scale encodes flux logarithmically
  through the Pogson ratio; the apparent and absolute magnitudes differ by the
  distance modulus, which converts a measured brightness into a distance. The
  bolometric correction folds a filtered magnitude into a total luminosity, and
  the difference of two magnitudes in different bands, the color index, measures
  surface temperature.
topics: [Observational Foundations]
sources:
  - book: Carroll & Ostlie
    ref: "Ch. 3 — The Continuous Spectrum of Light; §3.1 Stellar Parallax, §3.2 The Magnitude Scale, §3.6 The Color Index"
  - book: Maoz
    ref: "Ch. 2 — Basic Physics of Stars"
---

Every quantitative statement about a star begins with a measurement of how much
energy from it crosses a detector. That measured quantity is the **radiant flux**,
and the historical unit for reporting it, the magnitude, is logarithmic, inverted,
and older than the physics that explains it. This lesson fixes the definitions
that the rest of the course uses without comment: flux and its inverse-square
dilution, the magnitude scale and the Pogson ratio, the split between apparent and
absolute magnitude through the distance modulus, the bolometric correction to a
total luminosity, and the color index as a temperature proxy.

## Radiant flux and luminosity

The **luminosity** $L$ of a star is the total power it radiates, integrated over
all wavelengths and emitted isotropically into $4\pi$ steradians. Energy is
conserved as the light propagates through empty space, so the power crossing every
sphere centered on the star is the same $L$. At distance $r$ that power is spread
over an area $4\pi r^2$, and the **radiant flux** — power per unit area reaching
the detector — is

$$
F = \frac{L}{4\pi r^2}.
$$

Flux carries SI units of $\text{W}\,\text{m}^{-2}$. The inverse-square dependence
is the single most important relation in observational astronomy: it is what makes
a measured flux, combined with an independent distance, yield a luminosity, and
what makes a known luminosity (a standard candle) plus a measured flux yield a
distance.[^co-flux]

$$
% caption: The same luminosity L crosses every sphere; because the area grows as
% r squared, the flux through a fixed patch falls as one over r squared.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
\fill (0,0) circle (3pt);
\node[anchor=north] at (0,-0.15) {source};
% three arcs at increasing radius
\draw[thick] (1.6,-1.6) arc (-45:45:2.26);
\draw[thick] (3.2,-2.3) arc (-45:45:3.2);
\draw[thick] (4.8,-2.9) arc (-45:45:4.07);
% radial rays
\foreach \a in {-38,-19,0,19,38}
  \draw[black] (0,0) -- (\a:5.2);
% patch markers
\node[black!70, anchor=west] at (1.7,1.9) {area grows as r squared};
\node[black!70, anchor=south] at (1.65,1.6) {r};
\node[black!70, anchor=south] at (3.25,2.35) {2r};
\node[black!70, anchor=south] at (4.85,3.0) {3r};
\draw[<->, black] (0.1,-1.75) -- (1.5,-1.75);
\draw[<->, black] (0.1,-2.5) -- (3.1,-2.5);
\node[black, anchor=north] at (0.8,-1.8) {r};
\node[black, anchor=north] at (1.6,-2.55) {2r};
\node[black, anchor=west] at (4.9,0.2) {brightness: full, quarter, ninth};
\end{tikzpicture}
$$

The Sun sets the scale for both quantities. Its luminosity is
$L_\odot = 3.828 \times 10^{26}\ \text{W}$, and the flux it delivers at the top of
Earth's atmosphere, the **solar constant**, is
$F_\odot = 1361\ \text{W}\,\text{m}^{-2}$. These two numbers and the astronomical
unit are consistent through the inverse-square law, and stellar luminosities are
routinely quoted in units of $L_\odot$.

## The magnitude scale

Hipparchus ranked naked-eye stars into six classes, first magnitude for the
brightest and sixth for the faintest visible. The eye responds roughly
logarithmically to flux, so equal steps in this ranking correspond to equal
_ratios_ of flux, not equal differences. In 1856 Pogson fixed the scale
quantitatively by defining a difference of five magnitudes to be exactly a factor
of 100 in flux. One magnitude is therefore the ratio

$$
100^{1/5} = 10^{2/5} = 2.512,
$$

the **Pogson ratio**. Two stars with fluxes $F_1$ and $F_2$ have apparent
magnitudes related by

$$
\frac{F_1}{F_2} = 100^{(m_2 - m_1)/5},
\qquad
m_1 - m_2 = -2.5 \log_{10}\!\left( \frac{F_1}{F_2} \right).
$$

The scale runs backwards — a smaller magnitude is a brighter star — and it is
open-ended, extending to negative values for the brightest objects and past $m=+30$
for the faintest detections. The Sun sits at $m_\odot = -26.83$, the full Moon near
$-12.7$, Sirius at $-1.46$, and the naked-eye limit at about $+6$. Each five-step
descent multiplies the flux by 100; the range from the Sun to the faintest galaxies
imaged spans more than 57 magnitudes, a flux ratio exceeding $10^{22}$.

$$
% caption: Every five magnitudes is a factor of 100 in flux, so one magnitude is
% the fifth root of 100, about 2.512; brighter objects have smaller magnitudes.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
% horizontal magnitude axis
\draw[->, black] (-0.4,0) -- (10.2,0) node[right, black!70] {magnitude};
\foreach \x/\m in {0/0,2/1,4/2,6/3,8/4,10/5}
  \draw[black] (\x,0.08) -- (\x,-0.08) node[below, black] {\m};
% flux ratio at the endpoints (integers only; intermediate values are decimals)
\node[anchor=south] at (0,0.35) {ratio 1};
\node[anchor=south] at (10,0.35) {ratio 100};
\foreach \x in {2,4,6,8} \node[anchor=south] at (\x,0.35) {x};
\node[anchor=south, font=\scriptsize] at (5,1.15) {one step is a constant brightness ratio};
% brace connecting the 5-step span
\draw[black] (0,-0.7) -- (0,-0.95) -- (10,-0.95) -- (10,-0.7);
\node[black, anchor=north] at (5,-1.0) {5 magnitudes give a factor of 100};
\end{tikzpicture}
$$

Because the scale is a ratio, it needs a zero point. Historically the star Vega
defined $m=0$ in every band; the modern Vega system keeps this convention for
broadband photometry, while the AB system instead ties the zero point to an
absolute spectral flux density, so that a magnitude maps directly onto
$\text{W}\,\text{m}^{-2}\,\text{Hz}^{-1}$. The choice of zero point is a bookkeeping
convention; the Pogson ratio between two magnitudes is universal.

## Apparent versus absolute magnitude and the distance modulus

The apparent magnitude $m$ mixes two effects: how luminous the star is and how far
away it is. To separate them, define the **absolute magnitude** $M$ as the apparent
magnitude the star would have if placed at the reference distance of
$10\ \text{pc}$. Because both magnitudes describe the same star, they encode the
same luminosity; they differ only by the ratio of fluxes at the true distance $d$
and at $10\ \text{pc}$. Applying the Pogson relation to those two fluxes,

$$
m - M = -2.5 \log_{10}\!\left( \frac{F_d}{F_{10}} \right)
      = -2.5 \log_{10}\!\left( \frac{10\ \text{pc}}{d} \right)^{2}
      = 5 \log_{10}\!\left( \frac{d}{10\ \text{pc}} \right),
$$

where the inverse-square law supplies $F_d/F_{10} = (10\ \text{pc}/d)^2$. The
combination $m - M$ is the **distance modulus**, and it depends only on distance:

$$
m - M = 5 \log_{10}(d) - 5 \qquad (d \text{ in parsecs}).
$$

Inverting gives the distance directly from the two magnitudes,
$d = 10^{(m-M+5)/5}\ \text{pc}$. A distance modulus of $0$ means $10\ \text{pc}$;
$+5$ means $100\ \text{pc}$; $+10$ means $1\ \text{kpc}$; the modulus to a nearby
galaxy such as the Large Magellanic Cloud is about $+18.5$, or roughly
$50\ \text{kpc}$.

> **Worked example.** A Cepheid variable in a distant galaxy has apparent
> magnitude $m = 22.0$. The period-luminosity relation (a later rung of the
> distance ladder) fixes its absolute magnitude at $M = -5.0$. The distance
> modulus is $m - M = 27.0$, so
> $$
> d = 10^{(27.0 + 5)/5}\ \text{pc} = 10^{6.4}\ \text{pc} \approx 2.5\ \text{Mpc}.
> $$
> The star sits about 2.5 megaparsecs away, comparable to the distance of the
> Andromeda galaxy.

Real measurements must also account for **interstellar extinction**: dust between
the star and the detector absorbs and scatters light, dimming the star by $A$
magnitudes in the observing band. The relation generalizes to

$$
m - M = 5 \log_{10}(d) - 5 + A,
$$

so neglecting extinction overestimates the distance. Extinction is wavelength
dependent, which is what makes the color index below both a temperature diagnostic
and a way to measure the reddening.

## Filters, apparent magnitudes, and photometric systems

A detector never records the total flux. It records the flux transmitted through a
**filter**, a passband that admits light over a limited wavelength range. A
magnitude is therefore always a magnitude _in a band_. The standard Johnson-Cousins
**UBVRI** system defines broad passbands in the ultraviolet ($U$, centered near
$365\ \text{nm}$), blue ($B$, $445\ \text{nm}$), visual ($V$, $551\ \text{nm}$),
red ($R$, $658\ \text{nm}$), and near-infrared ($I$, $806\ \text{nm}$). The flux in
a band is the stellar spectral flux $F_\lambda$ weighted by the filter transmission
$S(\lambda)$,

$$
F_{\text{band}} = \int_0^\infty F_\lambda\, S(\lambda)\, \d\lambda,
$$

and the corresponding magnitude follows the Pogson rule against a band-specific
zero point. Photometry in several bands samples the spectral energy distribution at
a few points and is far cheaper than full spectroscopy, which is why colors carry
so much of the observational load.

$$
% caption: The Johnson-Cousins passbands sample a stellar spectrum at successive
% wavelengths; a magnitude in each band integrates the spectrum against its filter.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
\draw[->, black] (0,0) -- (10.4,0) node[right, black!70] {wavelength};
\draw[->, black] (0,0) -- (0,4.2) node[above, black!70] {brightness};
% stellar spectrum (blackbody-like peak in the blue-green)
\draw[acc, very thick] (0.3,0.4) .. controls (1.6,0.9) and (2.0,3.6) .. (2.9,3.7)
  .. controls (4.6,3.9) and (7.5,1.2) .. (10.0,0.5);
\node[acc, anchor=south west] at (3.2,3.3) {stellar spectrum};
% filter passbands as trapezoids along the axis
\foreach \x/\lab in {1.1/U, 2.6/B, 4.1/V, 5.9/R, 7.9/I}{
  \draw[black] (\x-0.55,0) -- (\x-0.35,1.0) -- (\x+0.35,1.0) -- (\x+0.55,0) -- cycle;
  \node[black!70, anchor=north] at (\x,-0.05) {\lab};
}
\end{tikzpicture}
$$

## Bolometric magnitude and the bolometric correction

The luminosity in the physics of stars is the total power, but a magnitude in a
single band captures only a slice of the spectrum. The **bolometric magnitude**
$m_{\text{bol}}$ (or $M_{\text{bol}}$ for the absolute version) is the magnitude
that would be measured by an ideal detector sensitive to all wavelengths; it maps
directly onto the total flux and hence the luminosity. The link between an absolute
bolometric magnitude and luminosity uses the same Pogson relation against a defined
zero point,

$$
M_{\text{bol}} = M_{\text{bol},\odot} - 2.5 \log_{10}\!\left( \frac{L}{L_\odot} \right),
\qquad M_{\text{bol},\odot} = 4.74.
$$

The IAU fixes the bolometric zero point at a luminosity of
$3.0128 \times 10^{28}\ \text{W}$, which reproduces $M_{\text{bol},\odot} = 4.74$
for the adopted solar luminosity.

The **bolometric correction** $BC$ converts an observed band magnitude — almost
always the $V$ band — into the bolometric magnitude,

$$
BC = m_{\text{bol}} - V = M_{\text{bol}} - M_V.
$$

By convention $BC$ is negative for most stars, because $V$ captures only part of
the emitted power and the missing light always makes the total brighter (a smaller
bolometric magnitude). The size of the correction depends on where the spectrum
peaks relative to the $V$ band. A star whose spectrum peaks in the visible, near
the Sun's temperature, has a small correction, $BC \approx -0.08$. A hot O star
radiates most of its power in the ultraviolet, far from $V$, so $|BC|$ is large,
several tenths to over three magnitudes. A cool M star radiates mostly in the
infrared and also has a large correction. The bolometric correction is minimized
near spectral type F, where the peak sits inside the $V$ band.

$$
% caption: The bolometric correction is smallest for stars whose spectra peak in
% the V band and grows for hot UV-bright and cool IR-bright stars.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
\draw[->, black] (0,0) -- (9.4,0) node[right, black!70] {hotter to cooler};
\draw[->, black] (0,0) -- (0,3.8) node[above, black!70] {size of correction};
% U-shaped BC magnitude curve (large at both ends, min in the middle)
\draw[acc, very thick] (0.5,3.4) .. controls (2.2,0.9) and (3.6,0.35) .. (4.6,0.35)
  .. controls (5.8,0.35) and (7.2,1.3) .. (8.8,3.3);
% spectral type ticks
\foreach \x/\lab in {0.8/O, 2.2/B, 3.4/A, 4.6/F, 5.8/G, 7.0/K, 8.4/M}
  \node[black, anchor=north] at (\x,-0.05) {\lab};
\node[black!70, anchor=south] at (4.6,0.4) {smallest near F};
\node[acc, anchor=east] at (1.5,3.0) {UV-bright};
\node[acc, anchor=west] at (7.6,3.0) {IR-bright};
\end{tikzpicture}
$$

## The color index as a temperature proxy

The difference between a star's magnitudes in two bands is its **color index**. The
two most-used indices are

$$
U - B = m_U - m_B, \qquad B - V = m_B - m_V.
$$

Because a magnitude difference is a flux ratio, a color index measures the _shape_
of the spectrum between two wavelengths, independent of the star's distance (the
inverse-square dimming cancels in the difference). For an approximately blackbody
spectrum, that shape is set by temperature. A hot star peaks in the blue, so it is
bright in $B$ relative to $V$: $m_B$ is small, $m_V$ is larger, and $B - V$ is
small or negative. A cool star peaks in the red, is faint in $B$ relative to $V$,
and has a large positive $B - V$. The color index therefore runs monotonically with
effective temperature, decreasing as the star gets hotter.[^co-color]

| Spectral type | $T_e$ (K) | $B - V$ | $U - B$ |
| --- | --- | --- | --- |
| O5 | 42,000 | $-0.33$ | $-1.19$ |
| B0 | 30,000 | $-0.30$ | $-1.08$ |
| A0 | 9,500 | $0.00$ | $0.00$ |
| F0 | 7,200 | $+0.30$ | $+0.03$ |
| G2 (Sun) | 5,800 | $+0.65$ | $+0.13$ |
| K0 | 5,200 | $+0.81$ | $+0.45$ |
| M0 | 3,900 | $+1.40$ | $+1.22$ |

The zero points of $UBV$ are fixed so that an A0 main-sequence star such as Vega
has $U - B = B - V = 0$; this is why the table crosses zero at A0. An approximate
inversion for main-sequence stars, valid across the middle of the range, relates
color to temperature through

$$
T_e \approx \frac{7090\ \text{K}}{(B - V) + 0.62},
$$

which recovers about $5800\ \text{K}$ at the solar color $B - V = 0.65$. Real
calibrations use tabulated relations rather than a single formula, and they must
correct for reddening, since interstellar dust makes a star appear redder (larger
$B - V$) than it is. The reddening-corrected color, the **intrinsic color**, is the
true temperature diagnostic; the difference between observed and intrinsic color is
the **color excess** $E(B-V)$, which measures the dust column.

## The color-magnitude diagram

Plotting a magnitude (usually $V$ or an absolute magnitude $M_V$) against a color
index (usually $B - V$) produces a **color-magnitude diagram**, the observational
form of the Hertzsprung-Russell diagram. Color replaces temperature on the
horizontal axis and magnitude replaces luminosity on the vertical axis, with both
axes inverted relative to the physical quantities: bluer (hotter) to the left,
brighter (smaller magnitude) at the top. For a single star cluster every member
lies at essentially the same distance, so the distance modulus is a common additive
constant; the apparent-magnitude diagram then has the same shape as the
absolute-magnitude one, shifted vertically. This is what makes clusters the natural
laboratories for stellar evolution: the main sequence, the turnoff, and the giant
branch all appear at once for a coeval population.

$$
% caption: A cluster color-magnitude diagram: bluer stars to the left, brighter
% stars at the top; the main sequence runs diagonally with giants to the upper right.
\begin{tikzpicture}[scale=1.0, font=\footnotesize]
\definecolor{acc}{HTML}{4A6FA5}
\draw[->, black] (0,5.0) -- (0,0) node[below, black!70] {} ;
\node[black!70, anchor=south, rotate=90] at (-0.5,2.5) {brighter};
\draw[->, black] (0,5.0) -- (8.6,5.0) node[right, black!70] {redder, cooler};
\node[black, anchor=south east] at (0.1,5.05) {bluer};
% color index axis ticks (B minus V); integer marks only
\node[black, anchor=south, font=\scriptsize] at (0.9,5.05) {0};
\node[black, anchor=south, font=\scriptsize] at (5.1,5.05) {1};
\node[black, anchor=south, font=\scriptsize] at (4,5.55) {color index B minus V};
% main sequence band running upper-left to lower-right
\draw[acc, very thick] (0.8,0.6) -- (6.4,4.3);
\node[acc, rotate=33, anchor=south] at (3.3,2.2) {main sequence};
% turnoff and giant branch
\draw[thick] (3.2,1.9) .. controls (4.0,1.4) and (5.2,1.1) .. (6.8,1.1);
\node[anchor=west] at (6.0,1.0) {giant branch};
\fill[black] (3.2,1.9) circle (2pt);
\node[black!70, anchor=south east] at (3.2,1.95) {turning point};
% cluster scatter dots along the main sequence
\foreach \x/\y in {1.2/0.9,1.9/1.4,2.5/1.75,3.0/2.0,3.6/2.4,4.2/2.9,5.0/3.5,5.8/4.0}
  \fill (\x,\y) circle (1.3pt);
\end{tikzpicture}
$$

The vertical shift equals the distance modulus. Fitting a cluster's observed
main sequence onto a calibrated absolute-magnitude template — **main-sequence
fitting** — reads off $m - M$ and therefore the cluster's distance, one of the
rungs assembled in [the cosmic distance
ladder](/astrophysics-cosmology/observational-foundations/the-cosmic-distance-ladder).
The color axis simultaneously encodes the temperature, so the same diagram fixes
each star's place on the [Hertzsprung-Russell
diagram](/astrophysics-cosmology/orientation/the-sun-and-stars) and, through the
turnoff point, the cluster's age.

The three quantities of this lesson — flux through the inverse-square law,
magnitude through the Pogson ratio, and color through the ratio of two band fluxes
— convert raw detector counts into the distance, luminosity, and temperature that
every later analysis assumes. The color index anticipates the physics of the next
lesson, where the [stellar
spectrum](/astrophysics-cosmology/observational-foundations/stellar-spectra-and-spectral-classification)
and its lines, not just its broadband shape, classify a star and pin its
temperature far more precisely.

[^co-flux]: Carroll & Ostlie, §3.2 — The Magnitude Scale: radiant flux, the inverse-square law, luminosity, and the definition of apparent magnitude.
[^co-color]: Carroll & Ostlie, §3.6 — The Color Index: UBV magnitudes, the color index as a temperature indicator, bolometric magnitude, the bolometric correction, and interstellar reddening.