We present the fourth Fermi Large Area Telescope catalog (4FGL) of γ-ray sources. Based on the first eight years of science data from the Fermi Gamma-ray Space Telescope mission in the energy range from 50 MeV to 1 TeV, it is the deepest yet in this energy range. Relative to the 3FGL catalog, the 4FGL catalog has twice as much exposure as well as a number of analysis improvements, including an updated model for the Galactic diffuse γ-ray emission, and two sets of light curves (one-year and two-month intervals). The 4FGL catalog includes 5064 sources above 4σ significance, for which we provide localization and spectral properties. Seventy-five sources are modeled explicitly as spatially extended, and overall, 358 sources are considered as identified based on angular extent, periodicity, or correlated variability observed at other wavelengths. For 1336 sources, we have not found plausible counterparts at other wavelengths. More than 3130 of the identified or associated sources are active galaxies of the blazar class, and 239 are pulsars.

The latest Fermi-LAT gamma-ray catalog, 4FGL–DR3, presents a large fraction of sources without clear association to known counterparts, i.e. unidentified sources (unIDs). In this paper, we aim to classify them using machine learning algorithms, which are trained with the spectral characteristics of associated sources to predict the class of the unID population. With the state-of-the-art CatBoost algorithm, based on gradient boosting decision trees, we are able to reach a 67 per cent accuracy on a 23–class dataset. Removing a single of these classes –blazars of uncertain type– increases the accuracy to 81 per cent. If interested only in a binary AGN/pulsar distinction, the model accuracy is boosted up to 99 per cent. Additionally, we perform an unsupervised search among both known and unID population, and try to predict the number of clusters of similar sources, without prior knowledge of their classes. The full code used to perform all calculations is provided as an interactive Python notebook.

We analyze 11 years of Fermi-Large Area Telescope (LAT) data corresponding to the sky regions of seven dwarf irregular (dIrr) galaxies. DIrrs are dark matter (DM)-dominated systems, proposed as interesting targets for the indirect search of DM with gamma rays. The galaxies represent interesting cases with a strong disagreement between the density profiles (core versus cusp) inferred from observations and numerical simulations. In this work, we addressed the problem by considering two different DM profiles, based on both the fit to the rotation curve (in this case, a Burkert cored profile) and results from $N$-body cosmological simulations (i.e., Navarro-Frenk-White cuspy profile). We also include halo substructure in our analysis, which is expected to boost the DM signal by a factor of 10 in halos such as those of dIrrs. For each DM model and dIrr, we create a spatial template of the expected DM-induced gamma-ray signal to be used in the analysis of Fermi-LAT data. No significant emission is detected from any of the targets in our sample. Thus, we compute upper limits on the DM annihilation cross section versus mass parameter space. Among the seven dIrrs, we find IC10 and NGC6822 to yield the most stringent individual constraints, independently of the adopted DM profile. We also produce combined DM limits for all objects in the sample, which turn out to be dominated by IC10 for all DM models and annihilation channels, i.e., $b\overline{b}$, ${\tau }^{+}{\tau }^{-}$, and $W^{+}{W}^{-}$. The strongest constraints are obtained for $b\overline{b}$ and are at the level of $⟨\sigma v⟩\sim 7\times {10}^{-26}{\mathrm{cm}}^3{\mathrm{s}}^{-1}$ at $m_{\chi }\sim 6\mathrm{GeV}$. Though these limits are a factor of $\sim 3$ higher than the thermal relic cross section at low weakly interacting massive particles masses, they are independent from and complementary to those obtained by means of other targets.