Reconfigurable intelligent surface (RIS) is an emerging technology for enabling the manipulation of wireless environments, which shows great potential for the coverage and capacity enhancements of wireless networks. In this letter, a hybrid deployment scheme of terrestrial and aerial RISs (TRIS, ARIS) is proposed to avoid the obstruction of light-of-sight (LoS) communication in dense urban. The objective is to maximize the minimum average rate of users located in different blind zones of dense urban by jointly optimizing transmit beamforming at BS, the reflecting coefficients of TRISs and ARIS, and the height deployment of ARIS. Due to the non-convexity of this problem, an effective iterative method including fractional programming, projected subgradient method, and local region optimization is proposed to divide and conquer each subproblem. Numerical simulations indicate the effectiveness of the joint optimization on the passive beamformings of multi-hierarchical RISs and the height deployment of the ARIS for enhancing system performance, and demonstrate a clear gain of the proposed scheme of hybrid deployment compared with the benchmark schemes.