Inference of morphology and dynamical state of nearby Planck-SZ galaxy clusters with Zernike polynomials

We analyse the maps of the Sunyaev-Zel'dovich (SZ) signal of local galaxy clusters ($z<0.1$) observed by the $Planck$ satellite in order to classify their dynamical state through morphological features. To study the morphology of the cluster maps, we apply a method recently employed on mock SZ images generated from hydrodynamical simulated galaxy clusters in THE THREE HUNDRED (THE300) project. Here, we report the first application on real data. The method consists in modelling the images with a set of orthogonal functions defined on circular apertures, the Zernike polynomials. From the fit we compute a single parameter, $\mathcal{C}$, that quantifies the morphological features present in each image. The link between the morphology of 2D images and the dynamical state of the galaxy clusters is well known, even if not obvious. We use mock $Planck$-like Compton parameter maps generated for THE300 clusters to validate our morphological analysis. These clusters, in fact, are properly classified for their dynamical state with the relaxation parameter, $\chi$, by exploiting 3D information from simulations. We find a mild linear correlation of $\sim 38\%$ between $\mathcal{C}$ and $\chi$ for THE300 clusters, mainly affected by the noise present in the maps. In order to obtain a proper dynamical-state classification for the $Planck$ clusters, we exploit the conversion from the $\mathcal{C}$ parameter derived in each $Planck$ map in $\chi$. A fraction of the order of $63\%$ of relaxed clusters is estimated in the selected $Planck$ sample. Our classification is then compared with those of previous works that have attempted to evaluate, with different indicators and/or other wavelengths, the dynamical state of the same $Planck$ objects. The agreement with the other works is larger than $58\%$.