This article proposes a new bathymetry in coastal areas based on multi-spectral satellite images (Sentinel-2 satellite images) and GEBCO (General Bathymetric Chart of the Oceans) data. To measure the water depth of the coastal area, an empirical model has selected, and a new model has obtained by interpreting the relationship between each element of the model and the spectral band of the Sentinel-2 satellite image. Also, a sunlight reflection correction model has established to overcome the effect of sunlight reflection, one of the main factors that limit the accuracy of remotely sensed data. Linear regression analysis is performed between the results derived from the new model and the GEBCO data to determine the coefficients and then used the raster calculator to get the desired results. Field bathymetry data is expensive and difficult to obtain, but Sentinel-2 satellite images and GEBCO data can bedownloaded for free from the website at any time. Therefore, it is easy to measure a wide range of coastal water depths in a short time using this method. The proposed bathymetry can quickly realize using QGIS spatial analysis tools.