1. Scale, Light, and Evidence

Space question

How can a small amount of light support a claim about an object that is trillions of kilometres away?

Astronomy is unusually honest about a central scientific problem: almost everything is remote. We cannot place a star on a laboratory bench. We collect messengers—light, particles, gravitational waves—and reason backward.

1.1 Start with scale

Useful reference scales:

• Earth radius: $6.37\times10^6$ m;
• astronomical unit (AU), roughly Earth–Sun distance: $1.496\times10^{11}$ m;
• light-year: $9.461\times10^{15}$ m;
• parsec: $3.086\times10^{16}$ m.

Scientific notation is not cosmetic. It makes multiplicative comparisons visible. Light takes about 8.3 minutes to travel 1 AU but about 4.24 years to arrive from the nearest stellar system.

Estimation as a scientific tool

Before using code, estimate. If an answer is ten orders of magnitude away from your estimate, either you found new physics or, much more often, mixed units.

1.2 Flux is not luminosity

Luminosity $L$ is power emitted by the source. Flux $F$ is power received per unit collecting area. For isotropic emission:

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

The $d^{-2}$ dependence is geometry, not a special property of stars. A sphere has area $4\pi d^2$, so the same power is diluted over more area.

Worked example: move the Sun

If a Sun-like star is placed at twice its original distance, its observed flux becomes

$$\frac{F_2}{F_1}=\left(\frac{d_1}{d_2}\right)^2=\frac14.$$

Without distance, a dim object could be intrinsically faint or simply far away. This degeneracy is why the astronomical “distance ladder” matters so much.

1.3 Angular size and resolution

For small angles in radians,

$$\theta \approx \frac{D}{d},$$

where $D$ is physical size and $d$ is distance. A telescope records angular structure. Converting it into physical size requires distance.

For a circular aperture, diffraction gives an approximate angular resolution:

$$\theta_{\rm min}\approx 1.22\frac{\lambda}{D_{\rm aperture}}.$$

Larger apertures and shorter wavelengths reveal smaller angular detail. Atmosphere, optics, sampling, and noise add further limits.

1.4 The evidence chain

Suppose a paper states: “This star contains sodium.”

1. A detector recorded counts at different pixel positions.
2. Calibration mapped pixels to wavelength and counts to relative flux.
3. A dip appeared near wavelengths associated with sodium transitions.
4. A forward model included temperature, pressure, motion, instrumental broadening, and candidate atomic lines.
5. The sodium model explained the pattern better than alternatives.

The detector did not directly display the word “sodium.” The conclusion is an inference whose strength depends on calibration, model adequacy, and competing explanations.

1.5 Three levels of knowledge

Use more precise language than “known/unknown”:

• Observed robustly: repeated measurements agree on a phenomenon.
• Inferred within a model: the interpretation follows if the model and assumptions are adequate.
• Open: multiple explanations remain viable or the required observation has not yet been made.

Example: the universe’s expansion is observed through several independent distance and redshift probes. Accelerated expansion is robustly supported. Calling the cause “dark energy” names the explanatory problem; it does not tell us the underlying physics.

Used in space science

• spacecraft navigation uses units, reference frames, and light-time;
• stellar luminosities require flux plus distance;
• telescope design trades wavelength, aperture, and resolution;
• galaxy sizes and exoplanet radii require geometry plus a model;
• cosmological surveys are fundamentally exercises in calibration and controlled inference.

Journey to the frontier

Modern frontier questions often arise because observations are strong while the interpretation is incomplete. Galaxy motion and gravitational lensing support extra gravitating mass, yet the nature of dark matter remains unknown. Cosmic acceleration is measured, yet dark energy could reflect a cosmological constant, a changing component, modified gravity, or something not yet formulated.

The foundation beneath these mysteries is not exotic: scale, flux, angles, calibration, uncertainty, and models.

Connections

• Next, vectors and differential equations turn gravity into an orbit.
• Chapter 3 treats a calibrated spectrum as data that linear algebra can decompose.
• Chapter 4 asks whether a faint repeating signal is distinguishable from noise.

Check yourself

1. A source moves three times farther away. By what factor does flux change?
2. Why can two stars with equal observed flux have different luminosities?
3. Label “we observe a flat galaxy rotation curve, therefore a new particle exists” as observation or inference. What is missing?

Answers: $1/9$; different distances (and attenuation); inference—the observation supports missing gravity/mass under a model, but does not uniquely identify its physical cause.