Context. Detailed characterizations of exoplanets are moving to the forefront of planetary science. Temperature is a key marker for understanding atmospheric physics and chemistry. Aims. We aim to retrieve temperatures of N2-O2 dominated atmospheres from secondary eclipse spectroscopic observations of the thermal emission of Earth-like exoplanets orbiting G-, K-, and M-stars using large future space telescopes. Methods. Line-by-line radiative transfer was used to generate synthetic thermal infrared (TIR) observations. Atmospheric temperature is approximated by an expansion with base vectors defined by a singular value decomposition of a matrix comprising representative profiles. Nonlinear least squares fitting is used to estimate the expansion coefficients. Results. Analysis of the $4.3 \rm\,\mu m$ and $15 \rm\,\mu m$ CO2 bands in the TIR permits inference of temperatures even for low signal-to-noise (S/N) ~5 at medium resolution. Deviations from the true temperature in the upper troposphere and stratosphere are usually a few Kelvin, with larger deviations in the upper atmosphere and, less often, in the lower troposphere. Although the performance of the two bands is equivalent in most cases, the longwave TIR is more favorable than the shortwave due to increased star-planet contrast. A high spectral resolution, as provided by JWST instruments, is important for retaining sensitivity to the upper atmosphere. Furthermore, the selection of an appropriate set of base functions is also key. Conclusions. Temperature in the mid-atmosphere can be suitably characterized by IR spectroscopy with a resolution of at least 1000 (ideally ~2500). Obtaining the necessary S/N could be feasible with future missions, such as the Origins Space Telescope or the Large Interferometer for Exoplanets. Meanwhile a least squares fitting with appropriate base functions is also applicable for other classes of planets.

We investigate the detectability of atmospheric spectral features of Earth-like planets in the habitable zone (HZ) around M dwarfs with the future James Webb Space Telescope (JWST). We use a coupled 1D climate-chemistry-model to simulate the influence of a range of observed and modelled M-dwarf spectra on Earth-like planets. The simulated atmospheres served as input for the calculation of the transmission spectra of the hypothetical planets, using a line-by-line spectral radiative transfer model. To investigate the spectroscopic detectability of absorption bands with JWST we further developed a signal-to-noise ratio (S/N) model and applied it to our transmission spectra. High abundances of CH$_4$ and H$_2$O in the atmosphere of Earth-like planets around mid to late M dwarfs increase the detectability of the corresponding spectral features compared to early M-dwarf planets. Increased temperatures in the middle atmosphere of mid- to late-type M-dwarf planets expand the atmosphere and further increase the detectability of absorption bands. To detect CH$_4$, H$_2$O, and CO$_2$ in the atmosphere of an Earth-like planet around a mid to late M dwarf observing only one transit with JWST could be enough up to a distance of 4 pc and less than ten transits up to a distance of 10 pc. As a consequence of saturation limits of JWST and less pronounced absorption bands, the detection of spectral features of hypothetical Earth-like planets around most early M dwarfs would require more than ten transits. We identify 276 existing M dwarfs (including GJ 1132, TRAPPIST-1, GJ 1214, and LHS 1140) around which atmospheric absorption features of hypothetical Earth-like planets could be detected by co-adding just a few transits. We show that using transmission spectroscopy, JWST could provide enough precision to be able to partly characterise the atmosphere of Earth-like TESS planets around mid to late M dwarfs.

Infrared solar occultation measurements are well established for remote sensing of Earth's atmosphere, and the corresponding primary transit spectroscopy has turned out to be valuable for characterization of extrasolar planets. Our objective is an assessment of the detectability of molecular signatures in Earth's transit spectra. To this end, we take a limb sequence of representative cloud-free transmission spectra recorded by the space-borne ACE-FTS Earth observation mission (Hughes et al., ACE infrared spectral atlases of the Earth's atmosphere, JQSRT 2014) and combine these spectra to the effective height of the atmosphere. These data are compared to spectra modeled with an atmospheric radiative transfer line-by-line infrared code to study the impact of individual molecules, spectral resolution, the choice of auxiliary data, and numerical approximations. Moreover, the study serves as a validation of our infrared radiative transfer code. The largest impact is due to water, carbon dioxide, ozone, methane, nitrous oxide, nitrogen, nitric acid, oxygen, and some chlorofluorocarbons (CFC11 and CFC12). The effect of further molecules considered in the modeling is either marginal or absent. The best matching model has a mean residuum of 0.4 km and a maximum difference of 2 km to the measured effective height. For a quantitative estimate of visibility and detectability we consider the maximum change of the residual spectrum, the relative change of the residual norm, the additional transit depth, and signal-to-noise ratios for a JWST setup. In conclusion, our study provides a list of molecules that are relevant for modeling transmission spectra of Earth-like exoplanets and discusses the feasibility of retrieval.