In this paper, a comparison between two array geometries, including URA arrays and L-shaped arrays, for two-dimensional DOA estimation is presented. For each geometry, algorithm which achieved the more accurate 2D-DOA estimates is considered. Indeed, conjugate symmetry based subspace and 2D-ESPRIT estimators are, respectively, chosen for the estimation of the L-Shaped and URA structures. Various criteria are taken into consideration such us the angular spread. A comparative study of the performance of geometries is carried out in terms of RMSE over a wide range of scenarios. The advantages and drawbacks of each geometry and the respective DOA estimation approaches are illustrated through a numerical comparison.