Lesson 5 of 6 · atmospheric measurements
Separate atmospheric evidence from competing signals
10 minute lesson · 3 checks along the way
Your goal for this lessonInterpret transmission spectra, quantify a limited contrast and select observations that can distinguish planetary from stellar explanations.
Follow a linked name to explore its archive card. “Back to your lesson” brings you to the same part of the reading.
Words with a dotted underline open a short definition right here in the lesson.
Different experiments sample different light
During a transit, a thin part of the star’s light passes through the planet’s atmospheric edge. Molecules absorb some wavelengths more strongly, changing the apparent across the spectrum. Emission observations instead examine light from the planet, sometimes by measuring what disappears when it passes behind the star. Reflected light probes another contribution. These experiments sample different regions and conditions rather than one complete atmospheric cross-section.
The James Webb Space Telescope measures infrared spectra with several instruments. A spectrum is a set of calibrated wavelength-dependent measurements, not a photograph showing individual molecules. Models connect absorption patterns to gases, temperatures, clouds, and geometry. Broad wavelength coverage can test whether several features support the same interpretation. An inferred gas abundance remains conditional on the assumed atmosphere and on how the stellar signal was treated.
Check 1 of 3
10 pointsUse the reading above. You can try again without losing points.
Clouds and stars can complicate the explanation
Clouds and hazes can mute features or obscure deeper layers. A small feature can reflect a small abundance, a compact atmospheric scale height, obscuration, or another combination of properties. also have spectra different from the rest of the stellar disk. If the transited region differs from the average star, the apparent planetary spectrum can acquire features that need not originate in the planet’s atmosphere.
A May 2023 release of GJ 486 b observations illustrated this problem: a water-rich planetary model and a stellar-spot model both fit the measured infrared region. Their predictions separated at shorter wavelengths, suggesting a discriminating follow-up. This is a dated example of what those data could establish, not a claim about the latest complete evidence. The illustrated planet behind the published graph was an artist’s concept, not a resolved view.
The May 1, 2023 GJ 486 b release compares Webb transmission measurements with a water-rich atmosphere model and a stellar-spot model. White points are data; colored curves and shaded bands represent models. The faint planet behind the graph is an artist’s concept, not a resolved observation.
Look closely
Compare the part of the graph containing white data points with the shorter-wavelength predictions to the left. Where would an additional measurement most directly distinguish the two illustrated explanations?
See what to notice
The models overlap through much of the measured infrared region but separate strongly at shorter wavelengths, particularly left of about one micron. Measurements where their predictions diverge, together with monitoring of the host star, test the unresolved origin of the signal. This dated graph does not by itself establish an ocean or life.
Work through a decision
Worked contrast: a difference needs its uncertainty
In an invented experiment, one wavelength channel gives a transit depth of 2,500 and a comparison channel gives 2,400. Assume their random errors are independent, each with a standard deviation of 30 parts per million. The difference is 100, with combined equal to the square root of 30 squared plus 30 squared, approximately 42.4 parts per million.
The contrast is therefore about 2.36 combined standard uncertainties. It is not a secure molecular identification by itself: one pair of channels does not establish a full absorption pattern, and the calculation has not included instrumental or stellar systematic errors. would also change the uncertainty of the difference. Repeated observations and additional predicted features test whether the signal persists in ways the proposed molecule requires.
Check 2 of 3
10 pointsUse the reading above. You can try again without losing points.
Choose the next observation for a reason
When two models overlap in one band and differ in another, extending wavelength coverage can be more useful than repeatedly measuring only their shared prediction. Monitoring the host star tests activity; observations at several transit epochs test repeatability; independent reductions test sensitivity to data handling. A planet can also look different at different orbital phases, so the geometry and region sampled should be stated before comparing spectra.
A water-vapor feature is evidence about a chemical species in an observed light path, not proof of a liquid ocean or biology. Turning it into a global atmospheric inventory requires additional assumptions about mixing, temperature, pressure, and clouds. Good reporting identifies the actual measurement, the models compared, the uncertainty included, and the unresolved alternatives. This approach makes a partial result scientifically useful without filling the unseen world with invented detail.
Check 3 of 3
10 pointsUse the reading above. You can try again without losing points.
Put it together
Your lesson progress
Each check connects to a part of the story. Complete all 3 to earn your 25-point lesson bonus.
Go to the source
Explore the evidence behind this lesson. These optional readings and videos open in a new tab.