With the development and maturity of the photon-counting ldiar technology, photon-counting ldiars are increasingly used in marine mapping. Based on the JONSWAP wave spectrum theory and micro-face model theory, this paper first establishes a photon-counting lidar simulation model via the Monte Carlo method. By substituting the system parameters of a spaceborne lidar into the proposed model, the photon distributions reflected by sea surfaces with different wind speed conditions are simulated and verified by comparing the averaged numbers of photons measured by the ICESat-2 spaceborne lidar. Then, the ranging errors under different wind speed conditions are discussed. The results indicate that the elevation value measured by photon counting lidars will be smaller than the actual elevation of sea surfaces, and the bias and standard deviation of photon counting lidars will increase when the wind speed is rising. When the wind speed is 10 m/s, the cumulative pulse number is 100, the elevation bias is approximately -2.5 cm and the standard deviation is 3.6 cm. The proposed simulation model and analysis results have important reference for optimizing the system parameter design of a photon-counting lidar for sea surface observations and correcting the bias on sea surface observations.