UAV-Derived Multispectral Bathymetry

Abstract

Bathymetry is considered an important component in marine applications as several coastal erosion monitoring and engineering projects are carried out in this field. It is traditionally acquired via shipboard echo sounding, but nowadays, multispectral satellite imagery is also commonly applied using different remote sensing-based algorithms. Satellite-Derived Bathymetry (SDB) relates the surface reflectance of shallow coastal waters to the depth of the water column. The present study shows the results of the application of Stumpf and Lyzenga algorithms to derive the bathymetry for a small area using an Unmanned Aerial Vehicle (UAV), also known as a drone, equipped with a multispectral camera acquiring images in the same WorldView-2 satellite sensor spectral bands. A hydrographic Multibeam Echosounder survey was performed in the same period in order to validate the method’s results and accuracy. The study area was approximately 0.5 km2 and located in Tuscany (Italy). Because of the high percentage of water in the images, a new methodology was also implemented for producing a georeferenced orthophoto mosaic. UAV multispectral images were processed to retrieve bathymetric data for testing different band combinations and evaluating the accuracy as a function of the density and quantity of sea bottom control points. Our results indicate that UAV-Derived Bathymetry (UDB) permits an accuracy of about 20 cm to be obtained in bathymetric mapping in shallow waters, minimizing operative expenses and giving the possibility to program a coastal monitoring surveying activity. The full sea bottom coverage obtained using this methodology permits detailed Digital Elevation Models (DEMs) comparable to a Multibeam Echosounder survey, and can also be applied in very shallow waters, where the traditional hydrographic approach requires hard fieldwork and presents operational limits.

BathyNet: A Deep Neural Network for Water Depth Mapping from Multispectral Aerial Images

“…While it is possible to reduce the depth estimation errors to a certain extent by means of multispectral algorithms (Lyzenga et al 2006;Legleiter and Fosness 2019), reflectance variations still pose a fundamental problem, especially in the very shallow littoral zone where submerged vegetation is the main source of spectral differences rather than the homogeneity of the water bottom substrate (Heblinski et al 2011). In addition, the existence of underwater vegetation is also problematic for the generation of reference data, for which echo sounding or laser bathymetry are the most prominent capturing techniques (Lyzenga et al 2006;Song et al 2015;Kasvi et al 2019;Rossi et al 2020;Brown et al 2011;Kogut and Bakuła 2019). Both methods are generally capable of penetrating (loose) vegetation and therefore often provide reference depths related to the bottom whereas image based techniques tend to deliver the topmost surface of the vegetation canopy (Ressl et al 2016).…”

“…Muzirafuti et al (2020) use log-band ratio and OBRA methods for deriving shallow water bathymetry from multispectral QuickBird satellite images. Rossi et al (2020) apply the methods of Lyzenga et al (2006) and Stumpf et al (2003) to multispectral Unmanned Aerial Vehicle (UAV) images featuring the same spectral bands as the WorldView-2 satellite sensor. For mapping bathymetry of inland running waters, Gentile et al (2016) use UAV hyperspectral images and apply empirical models for depth retrieval, which are applicable under a range of specific field conditions.…”

Going Batty: The Challenges and Opportunities of Using Drones to Monitor the Behaviour and Habitat Use of Rays

“…Determining the location of areas that are only inundated at high tide is particularly of interest, as these areas usually represent important feeding grounds for the rays [5]. Mapping bathymetry using multispectral drone imagery may prove fruitful for this purpose [79], however may not be necessary as the structure of the intertidal zone could be mapped to higher detail using an RGB sensor at times of the tidal cycle when it is exposed.…”

Application of Drone Technologies in Surface Water Resources Monitoring and Assessment: A Systematic Review of Progress, Challenges, and Opportunities in the Global South

“…Meanwhile, results showed that more efforts from the community of practice were widely exerted towards mapping water quality in relation to water quantity. Specifically, only fourteen studies assessed the level of water, whereas thirty-seven studies assessed water quality parameters based on drone remotely sensed data [44,[47][48][49][61][62][63][64][65][66][67][68][69][70][71]. A few examples of studies that mapped water levels included Ridolfi and Manciola [62] who used a method that was based on the Ground Control Points (GCPs) to detect water levels, where water level values were measured using drone-derived data.…”