RT journal article
T1 Evaluating the Performance of High Spatial Resolution UAV-Photogrammetry and UAV-LiDAR for Salt Marshes: The Cadiz Bay Study Case
A1 Curcio, Andrea Celeste
A1 Peralta González, Gloria
A1 Aranda García, María
A1 Barbero González, Luis Carlos
A2 Biología
A2 Ciencias de la Tierra
K1 salt marshes
K1 light detection and ranging (LiDAR)
K1 photogrammetry
K1 multispectral
K1 high resolution
K1 unmanned aerial vehicle (UAV)
K1 digital models
AB Salt marshes are very valuable and threatened ecosystems, and are challenging to studydue to their difficulty of access and the alterable nature of their soft soil. Remote sensing methodsin unmanned aerial vehicles (UAVs) offer a great opportunity to improve our knowledge in thistype of complex habitat. However, further analysis of UAV technology performance is still requiredto standardize the application of these methods in salt marshes. This work evaluates and tunesUAV-photogrammetry and UAV-LiDAR techniques for high-resolution applications in salt marshhabitats, and also analyzes the best sensor configuration to collect reliable data and generate thebest results. The performance is evaluated through the accuracy assessment of the correspondinggenerated products. UAV-photogrammetry yields the highest spatial resolution (1.25 cm/pixel)orthomosaics and digital models, but at the cost of large files that require long processing times,making it applicable only for small areas. On the other hand, UAV-LiDAR has proven to be apromising tool for coastal research, providing high-resolution orthomosaics (2.7 cm/pixel) andhigh-accuracy digital elevation models from lighter datasets, with less time required to processthem. One issue with UAV-LiDAR application in salt marshes is the limited effectiveness of theautoclassification of bare ground and vegetated surfaces, since the scattering of the LiDAR pointclouds for both salt marsh surfaces is similar. Fortunately, when LiDAR and multispectral dataare combined, the efficiency of this step improves significantly. The correlation between LiDARmeasurements and field values improves from R2 values of 0.79 to 0.94 when stable reference points(i.e., a few additional GCPs in rigid infrastructures) are also included as control points. Accordingto our results, the most reliable LiDAR sensor configuration for salt marsh applications is the nadirnon-repetitive combination. This configuration has the best balance between dataset size, spatialresolution, and processing time. Nevertheless, further research is still needed to develop accuratecanopy height models. The present work demonstrates that UAV-LiDAR technology offers a suitablesolution for coastal research applications where high spatial and temporal resolutions are required.
PB MDPI
SN 2072-4292
YR 2022
FD 2022-08
LK http://hdl.handle.net/10498/27954
UL http://hdl.handle.net/10498/27954
LA eng
DS Repositorio Institucional de la Universidad de Cádiz
RD 08-nov-2026