Astronomy · Our view of the Milky Way

Mapping stars and gas through dust

Use complementary tracers and a known-distance two-band example to distinguish cumulative attenuation from the location and density of intervening material.

Continue in the approved chapter order. Necessary vocabulary, quantities and assumptions are explained in the lesson.

Back to the chapter

After reading, you can

Explain what infrared starlight and radio gas measurements add to an internal Galactic map.
Bracket a synthetic dust screen with known-distance sources and declared band calibrations.
Distinguish integrated extinction, local dust density and assumptions in a three-dimensional reconstruction.

Seeing farther still requires interpretation

Dust absorbs and scatters radiation out of a direct beam. Its effect depends on wavelength and material properties, so an optical image can hide stars that contribute to an infrared view. Reduced extinction is useful, but infrared does not mean dust-free or equally transparent in every direction.

Infrared starlight can therefore add distance and population information where visible measurements are difficult. The source spectrum, band response, crowding and selection still influence what is measured. Infrared emission from dust is also a different signal from background starlight transmitted through it.

Radio spectral lines add gas tracers. The hydrogen line commonly called H I 21 centimetres traces neutral atomic hydrogen; carbon monoxide lines can trace molecular material under an appropriate emission and chemical model. These signals do not count all interstellar phases in one interchangeable way.

A measured line intensity is not automatically total gas mass. Excitation, opacity, chemical abundance and unresolved structure can affect the conversion. A line’s Doppler shift adds a line-of-sight motion component, but converting that motion into a position requires the further kinematic assumptions examined in the next lesson.

A background source samples the path in front of it

For direct-beam attenuation in a fixed band b, define transmission T_b as received flux divided by the flux that the same source would have at the same distance without attenuation. With positive transmission, extinction is A_b = −2.5 log₁₀(T_b) magnitudes. Additional emitted or scattered light entering the aperture must be treated separately.

A more distant source can sample additional intervening material, but an observed colour also depends on its intrinsic spectrum. Comparing two differently coloured stars without a source model does not isolate reddening. Reliable distance and intrinsic-colour information help separate these contributions.

The extinction toward a star is cumulative along its path. A change in that cumulative quantity between different distance probes can indicate where extra attenuation enters. One integrated value is not a local density at the star’s position.

Green and collaborators’ three-dimensional dust study combines stellar photometry with distance, source and dust assumptions. Its treatment of angular variation and prior-dominated regions illustrates why a reconstructed dust profile must carry resolution and uncertainty, rather than become an exact layer-by-layer photograph.

Locate an added attenuation without inventing its exact distance

Consider three separate synthetic standard candles on one ideal pencil-beam direction at distances L0, 2L0 and 4L0. L0 is an arbitrary positive length. They have identical intrinsic spectra and known distances. In fictional fixed bands X and Y, let F_X0 and F_Y0 be the respective unattenuated received fluxes at L0, and choose the band zero points so that each candle’s unattenuated X−Y colour is zero.

The two band normalizations have their own physical flux units; they are not assumed equal. In either band the dimensionless unattenuated flux is (L0/d)². Stipulate one zero-thickness uniform screen strictly between 2L0 and 4L0, with transmissions T_X = 1/4 and T_Y = 1/2 and no other attenuation or light entering the aperture.

The first two sources retain zero colour excess. The third has received fluxes F_X/F_X0 = 1/64 and F_Y/F_Y0 = 1/32. Its screen extinctions are A_X ≈ 1.505149978 magnitudes and A_Y ≈ 0.752574989 magnitudes; the colour excess E(X−Y) = A_X − A_Y is approximately 0.752574989 magnitudes.

Under the stipulated one-screen model, the extra attenuation lies beyond 2L0 and before 4L0. Nothing in these three probes places it exactly at 3L0 or measures a finite thickness. The screen is an original teaching device, with no claim about a real cloud or a universal extinction law.

Original known-distance screen experiment. Flux columns are separately normalized in their own bands; X and Y are fictional. Extinction and colour excess concern the supplied direct-beam transmissions, not distance dimming.
Distance d/L0	F_X/F_X0	F_Y/F_Y0	A_X (mag)	A_Y (mag)	E(X−Y) (mag)
1	1	1	0	0	0
2	1/4	1/4	0	0	0
4	1/64	1/32	1.505149978	0.752574989	0.752574989

Sparse or biased sampling limits a dust map

Real sources do not provide perfectly known distances and identical intrinsic spectra. Distance and reddening can be correlated in the inference. A broad or shifted stellar distance estimate can broaden or move an apparent increase in cumulative extinction.

Stars within a finite angular pixel also look through slightly different columns. A narrow dust structure can be hidden by an angular average, or increase apparent scatter among sources assigned to the same direction. Angular resolution and the number of useful depth probes therefore matter together.

Strong attenuation can remove the very background sources that would constrain material farther away. A smooth inferred continuation behind the last useful source may be supplied mainly by a prior. A blank detected region establishes neither an empty stellar volume nor a precisely known dust distribution.

Cross-checking independent photometry, spectra, distance information and dust emission can test parts of an inference. Each comparison has its own calibration and forward model. Agreement between maps that share the same source assumptions is less independent evidence than agreement between genuinely different measurement channels.

A tracer map is more than a coloured image

A gas survey can record two sky coordinates and a spectral line profile for each angular element. Features at different line-of-sight velocities may overlap in the same sky direction. That data cube is valuable, but its third measurement coordinate is velocity, not automatically distance.

A circular Galactic rotation model can associate some directions and relative velocities with possible radii and distances. It can also produce multiple distances for one measurement. Noncircular motion, finite line widths and missing tracers further limit this reconstruction.

The strongest interpretation states which component was measured, what is integrated along the ray, and what assumptions add the inferred depth or mass. A visible image, a cumulative extinction map and a radio velocity map can describe the same region while displaying different physical quantities.

The next lesson works through a near–far distance ambiguity explicitly. The detailed phases, emission physics and evolution of interstellar gas and dust are developed later; here the central question is what our internal observations can place in space.

Worked examples

Keep distance dimming separate from extinction

Why does the far source’s X−Y colour change when the nearer sources’ colours do not?

The inverse-square factor 1/16 applies to both separately normalized bands at 4L0, leaving intrinsic normalized colour zero without the screen.
The screen multiplies that factor by 1/4 in X and 1/2 in Y, giving received fractions 1/64 and 1/32.
Subtract its band extinctions: E(X−Y) = −2.5 log₁₀(1/4) + 2.5 log₁₀(1/2) ≈ 0.752574989 mag.

The supplied wavelength-dependent transmission changes the colour; distance alone would dim both normalized bands equally in this example.

Report a bracket rather than an invented midpoint

What screen-distance claim follows from these three exact probes under the supplied one-screen model?

The source at 2L0 has no added attenuation, so the screen is beyond it.
The source at 4L0 has the stipulated additional attenuation, so the screen is in front of it.
No further probe fixes a position within that interval or supplies a thickness.

The screen is strictly between 2L0 and 4L0. The midpoint 3L0 is not a measured or derived location.

Try the reasoning

Does infrared observation eliminate extinction and selection effects?
Show the answer and reasoning
No.
It often reduces obscuration relative to visible light, while wavelength dependence, crowding and source selection remain.
Is extinction toward one star a direct measurement of local dust density at that star?
Show the answer and reasoning
No.
It concerns cumulative attenuation along the foreground path.
Is the spectral coordinate of a radio line data cube automatically a geometrical distance?
Show the answer and reasoning
No.
It measures a line profile and calibrated line-of-sight motion information; a distance interpretation needs additional modelling.
Does a smooth inferred profile beyond the last useful background probe guarantee a well-measured dust distribution there?
Show the answer and reasoning
No.
That continuation may be dominated by prior assumptions rather than constraining data.

What to carry forward

Different wavelengths and tracers add complementary information while retaining their measurement models.
Known-distance sources can bracket added attenuation along a ray.
Cumulative extinction, local density and a velocity coordinate are distinct quantities.

References

Teaching resources and research supporting this lesson’s explanations and examples.