Using a UAV-Mounted Multispectral Camera for the Monitoring of Marine Macrophytes

Alejandro Román,Antonio Tovar-Sánchez,Gabriel Navarro,Irene Olivé

doi:10.3389/fmars.2021.722698

Abstract

Marine macrophytes constitute one of the most productive ecosystems on the planet, as well as one of the most threatened by anthropogenic activities and climate change. Their monitoring is therefore essential, which has experienced a fast methodological evolution in recent years, from traditional in situ sampling to the use of satellite remote sensing, and subsequently by sensors mounted on unmanned aerial vehicles (UAV). This study aims to advance the monitoring of these ecosystems through the use of a UAV equipped with a 10-band multispectral camera, using different algorithms [i.e., maximum likelihood classifier (MLC), minimum distance classifier (MDC), and spectral angle classifier (SAC)], and using the Bay of Cádiz Natural Park (southern Spain) as a case of study. The results obtained with MLC confirm the suitability of this technique for detecting and differentiating seagrass meadows in a range of 0–2 m depth and the efficiency of this tool for studying and monitoring marine macrophytes in coastal areas. We inferred the existence of a cover of 25452 m2 of Cymodocea nodosa, and macroalgae species such as Caulerpa prolifera, covering 22172 m2 of Santibañez (inner Bay of Cádiz).

Highlights

We examine the benefits of improved spectral detection over red green blue (RGB) sensors from the use of a 10-band multispectral camera, which allows us (i) to detect the potential of these sensors for taxonomic separation of the different species of macroalgae and seagrasses in the intertidal and subtidal zones; and (ii) to analyze the effectiveness of geoprocessing tools for the generation of supervised classification, allowing us to monitor the evolution of these coastal ecosystems in a faster, more precise, and economically viable way
The detection of vegetation is significantly improved if reflection in the near infrared is measured, since vegetation spectrum typically absorbs in the red and blue wavelengths, reflects in the green wavelength, and strongly reflects in the near infrared (NIR) wavelength, so that it is marked in a clear red color if a false color composite is used
AR, AT-S, IO, and GN participated in the field trips in which unmanned aerial vehicles (UAV) flights and in situ samplings were carried out

Summary

Introduction

Marine macrophytes, including seagrass meadows and macroalgae, constitute a coastal habitat of great value due to the ecosystem services they provide, which makes them one of the most productive ecosystems on the planet, comparable to coral reefs or tropical forests (Beca-Carretero et al, 2020; Veetil et al, 2020; Zoffoli et al, 2020) They act as large sinks of atmospheric CO2 (De los Santos et al, 2020; Egea et al, 2020; Zoffoli et al, 2020), but they are considered biodiversity hotspots providing refuge and food for numerous animal species, some of which are protected, such as turtles or other marine mammals (Brodie et al, 2018; Egea et al, 2020; Veetil et al, 2020). They are poorly protected due to the scarce media attention they receive compared to other ecosystems, which makes them all the more vulnerable (Brun et al, 2015)

Objectives

Methods

Results

Conclusion