“…On the other hand, for velocity and pressure approximation, the discontinuous Galerkin formulation with P 1 − P 0 space is stable [85]. Because of its flexibility for mesh/polynomial refinement, localizability, and stability, the discontinuous Galerkin method has been widely extended and applied to solve many partial differential equations, such as the local discontinuous Galerkin (LDG) method [17,22,51,97], the interior penalty discontinuous Galerkin (IPDG) method [1,24,25,78,86,96,101], the hybridizable DG method [10,20,43,48,76], reduced order DG method [58][59][60], and many others [15, 49, 61, 68, 72-74, 81, 92, 95, 108]. One important feature of the IPDG method is its capability to easily incorporate hp local refinement.…”