If stars were uniformly bright across their surfaces, the analysis of transit observations would require only the subtraction of this bright background from the observed data, resulting in the transit light curve and the absorption spectrum produced by atoms and molecules in the exoplanet’s atmosphere. Even this unrealistically simple observing scenario is a data analysis challenge, because the area of a transiting planet may be only 1% or smaller than that of the host star, and the area of a terrestrial exoplanet’s atmosphere seen in absorption against the stellar disk may be only 0.01% that of the stellar disk area. Thus high signal/noise data are essential to minimize random noise.