4.1.2. Ground Control Point Effect

At a more local scale, three scenarios can explain the results that were obtained. The first is the number of GCP that were extracted from the 2018–LiDAR as calibration points for our topographic models. The tests that were performed without GCP, with one, or with three GCP, showed that when we added GCP points, the model is more accurate [27]. However, The RPC files that were delivered with the Pleiades–1 images could be more powerful. The RPC files integrate all of the parameters that are related to the photographs in order to correct for the satellite images. This approach could be investigated in a future study.

#### 4.1.3. Information Reflected by the LiDAR Wavelengths

Two laser sensors were used for the whole LiDAR dataset: a mixed topo–bathymetric laser in the NIR and green spectra, and a stronger bathymetric laser in the green spectrum, which was specifically used for deeper areas. Depending on the nature (albedo) of the coastal habitat, the wavelength that is used by the LiDAR does not return the same information, being more or less reflected. Thus, the reflectance in the NIR is strong for eco–geo–systems with high chlorophyll dominance, and conversely, it is fully absorbed by habitats with a high water content, such as seawater, for example [28].

The density of the LiDAR points may also play a crucial role. The intertidal and coastal land area is denser in a number of points per square meter than marine area (deep water area) is. The detection of features such as trees is easy when the density of measured points is high [29].
