Due to global changes, UAV flights are of great interest for the surveillance of coastal environments. Some UAVs are fitted with multispectral sensors to better discriminate coastal eco-geo-systems such as saltmarsh meadows and sandy dunes. This research evaluates photogrammetry-based horizontal (XY) and vertical (Z) accuracies derived from four separate channels (Green, G, Red, R, Red-Edge, RE, and Near-Infrared, NIR). The best X and Y accuracy is achieved by the NIR channel (0.07 m and 0.09 m, respectively). The best Z vertical accuracy (0.13 m) is reached with the full set of channels (R-G-RE-NIR). The individualized investigations of the saltmarsh vegetation and the sand dune show that the best vertical accuracies are attained with the NIR and RE with 0.10 m and 0.11 m, respectively. Our findings witness that the visible spectrum, represented by the G and R bands, provide lower performance than the infrared gamut.