Antoine Collin

What optech’s bathymetric LiDAR sees underwater

This article presents early results of the FUDOTERAM project using bathymetric LiDAR data acquired with the SHOALS-3000, the latest bathymetric LiDAR system from Optech. The survey area is in the coastal zone along the northern shore of Chaleurs Bay, in the western Gulf of St. Lawrence, Canada. The project aimed to apply the SHOALS- 3000 to geological mapping, sedimentary process monitoring and marine habitat mapping. This paper focuses on the sedimentological part of the study and presents the early raw data obtained to produce a bottom type classification based on some simple parameters, roughness, slope angle and direction. Two methods are evaluated for analysis of the SHOALS-3000 waveforms, the Moment Method and the Gaussian Mixture Model, and the latter is used as an approach to model the bottom type signal.

What is the value added of 4 bands within the submetric remote sensing of tropical coastscape? QuickBird-2 vs WorldView-2

The seamless tropical coastscape was investigated in comparing remotely-sensed by-products derived from the spaceborne 8 band WorldView-2 (WV-2) and the 4 band QuickBird-2-like (QB-2-like) sensors. The value added of the 4 bands allowed both land and water surveys to be significantly improved. The WV-2-derived NDVI means were more correlated (R2=0.96) with in situ “greenness” than those of QB-2-like (R2=0.89), and the WV-2-derived NDVI standard deviations were systematically inferior to those of QB-2-like. The cross-validation of the bathymetric products indicated a better performance of WV-2 (R2=0.76; maximum depth, 9.4 m) than QB-2-like (R2=0.74; maximum depth, 8.9 m). Results of classification showed that land cover was best mapped by WV-2 with the Neural Network algorithm (8 classes, kappa=0.92), and seafloor cover (resulting from an underwater radiative transfer model) was best classified by WV-2 trained with Support Vector Machine (6 classes, kappa=0.76). These improvements were then discussed based upon their spectral properties.

Coral reef structural complexity provides important coastal protection from waves under rising sea levels

If coral reefs continue to degrade, waves on coastlines may substantially increase, leading to greater coastal erosion. Coral reefs are diverse ecosystems that support millions of people worldwide by providing coastal protection from waves. Climate change and human impacts are leading to degraded coral reefs and to rising sea levels, posing concerns for the protection of tropical coastal regions in the near future. We use a wave dissipation model calibrated with empirical wave data to calculate the future increase of back-reef wave height. We show that, in the near future, the structural complexity of coral reefs is more important than sea-level rise in determining the coastal protection provided by coral reefs from average waves. We also show that a significant increase in average wave heights could occur at present sea level if there is sustained degradation of benthic structural complexity. Our results highlight that maintaining the structural complexity of coral reefs is key to ensure coastal protection on tropical coastlines in the future.

Benthic Classifications Using Bathymetric LIDAR Waveforms and Integration of Local Spatial Statistics and Textural Features

Abstract The scope of this research is to assess benthoscape discrimination by airborne light detection and ranging (LIDAR) bathymetry (ALB) on the basis of statistical parameters derived from the LIDAR waveforms, textural information, and local spatial statistics. Analysis of the underwater camera stations allowed clustering of the stations into groups on the basis of their habitat composition (β-diversity). Twelve descriptive statistics describing the shape of the bottom part of the waveform, also called 12 benthic parameters, were used for discriminating four benthic habitats. A K-means classification and a supervised method based on the Support Vector Machine (SVM) were applied to this dataset, and overall accuracies of 67.7% and 89.9% were obtained, respectively. Geostatistical analyses, using 11 textural measures, defined by the gray-level occurrence matrix (GLOM) and the gray-level co-occurrence matrix (GLCM), and three local spatial statistics were then applied to the 12 benthic parameters to enhance the SVM classification performance. The assessment of the contribution of geostatistics into benthic class segmentation was achieved by computation of separability distance. Mean (from the GLOM), mean (from the GLCM) and the local Getis-Ord statistic yielded the best rates of discrimination. These added metrics, integrated with bands related to the 12 benthic parameters, showed that the rate of correct (supervised) classification was thereby improved and increased by 5.3%. Finally, the first four principal components (PCs) (i.e., 90.41% of the 12 parameter variances, boosted by the three best geostatistics) brought out an overall accuracy of 93.3%, showing evidence for optimizing the classification processing.

Mapping social-ecological vulnerability to inform local decision making

An overarching challenge of natural resource management and biodiversity conservation is that relationships between people and nature are difficult to integrate into tools that can effectively guide decision making. Social-ecological vulnerability offers a valuable framework for identifying and understanding important social-ecological linkages, and the implications of dependencies and other feedback loops in the system. Unfortunately, its implementation at local scales has hitherto been limited due at least in part to the lack of operational tools for spatial representation of social-ecological vulnerability. We developed a method to map social-ecological vulnerability based on information on human-nature dependencies and ecosystem services at local scales. We applied our method to the small-scale fishery of Moorea, French Polynesia, by combining spatially explicit indicators of exposure, sensitivity, and adaptive capacity of both the resource (i.e., vulnerability of reef fish assemblages to fishing) and resource users (i.e., vulnerability of fishing households to the loss of fishing opportunity). Our results revealed that both social and ecological vulnerabilities varied considerably through space and highlighted areas where sources of vulnerability were high for both social and ecological subsystems (i.e., social-ecological vulnerability hotspots) and thus of high priority for management intervention. Our approach can be used to inform decisions about where biodiversity conservation strategies are likely to be more effective and how social impacts from policy decisions can be minimized. It provides a new perspective on human-nature linkages that can help guide sustainability management at local scales; delivers insights distinct from those provided by emphasis on a single vulnerability component (e.g., exposure); and demonstrates the feasibility and value of operationalizing the social-ecological vulnerability framework for policy, planning, and participatory management decisions.

VHR coastal bathymetry using WorldView-3: colour versus learner

ABSTRACT Satellite imagery constitutes an affordable solution to map coastal bathymetry at very high resolution (VHR). Bathymetry retrieval is commonly based on regression analysis linking depth response with remotely-sensed spectral predictors. Most studies have focused on a single regressor using conventional three visible bands over clear waters. However the interdependence of added visible bands in respect to linear and non-linear regressors is poorly studied and strongly lacks for turbid waters. Here we investigate the single and joint contribution of both spectral bands (visible 1.24 m WorldView-3, WV-3) and regressor types on the performance of retrieval. A case study over Saint-Malo (Brittany, France) megatidal turbid waters enables to quantify the accuracy gain related to Coastal and yellow bands as well as ordinary least squares (OLS), generalized linear model (GLM) and artificial neural network (ANN). Correlation analysis reveals that both Coastal and yellow bands do not significantly increase conventional performance for the three regressors. ANN surpassed GLM and OLS for both conventional and boosted WV-3 visible spectral datasets, reaching 0.94 correlation coefficient and 0.52 m accuracy. Comparisons’ significance indicates that selecting a robust regression method (including parameterization) is more efficient than adding spectral bands for mapping VHR bathymetry of coastal turbid waters.

Very high resolution mapping of coral reef state using airborne bathymetric LiDAR surface-intensity and drone imagery

ABSTRACT Very high resolution (VHR) airborne data enable detection and physical measurements of individual coral reef colonies. The bathymetric LiDAR system, as an active remote sensing technique, accurately computes the coral reef ecosystem’s surface and reflectance using a single green wavelength at the decimetre scale over 1-to-100 km2 areas. A passive multispectral camera mounted on an airborne drone can build a blue-green-red (BGR) orthorectified mosaic at the centimetre scale over 0.01-to-0.1 km2 areas. A combination of these technologies is used for the first time here to map coral reef ecological state at the submeter scale. Airborne drone BGR values (0.03 m pixel size) serve to calibrate airborne bathymetric LiDAR surface and intensity data (0.5 m pixel size). A classification of five ecological states is then mapped through an artificial neural network (ANN). The classification was developed over a small area (0.01 km2) in the lagoon of Moorea Island (French Polynesia) at VHR (0.5 m pixel size) and then extended to the whole lagoon (46.83 km2). The ANN was first calibrated with 275 samples to determine the class of coral state through LiDAR-based predictors; then, the classification was validated through 135 samples, reaching a satisfactory performance (overall accuracy = 0.75).

High-energy events, boulder deposits and the use of very high resolution remote sensing in coral reef environments

ABSTRACT Collin, A., Etienne, S. and Planes, S., 2013. High-energy events, coastal boulder deposits and the use of very high resolution remote sensing. In recent years, coastal boulders have become a trendy proxy in studying high-energy marine inundation events. From their morphology and spatial distribution, authors are able to characterize a singular event in terms of intensity (wave height, flow velocity). Recent post-catastroph studies (e.g. Indian Ocean 2004 and Japan 2011 tsunamis) have demonstrated also the interest of boulder deposits in reconstructing the event kinematics through a multi-proxy approach. But as boulder studies require a statistically robust dataset they are field-time consuming and sometimes fieldwork takes place in remote areas with low facilities. The use of very high resolution remote sensing could overcome some of these limits. In this paper, we evaluate the possibility to identify meter-size coral boulder thrown on a reef flat during the hit of a tropical cyclone. Image analysis allows for the discrimination of major geographical object encountered on coralline islands: submerged coral boulder, emerged coral boulder, perched reef, sand beach. Within emerged boulder population, specific bands available with WorldView-2 images (i.e. red and NIR2 band) allow the spectral discrimination and mapping of fresh and weathered elements.

Very high-resolution mapping of emerging biogenic reefs using airborne optical imagery and neural network: the honeycomb worm ( Sabellaria alveolata ) case study

ABSTRACT Biogenic reefs provide a wide spectrum of ecosystem functions and services, such as biodiversity hotspot, coastal protection, and fishing practices. Honeycomb worm (Sabellaria alveolata) reefs, in the Bay of Mont-Saint-Michel (France), constitute the largest intertidal bioconstruction in Europe but undergo anthropogenic pressures (aquaculture-stemmed food/space competition and siltation, fishing-driven trampling). Very high-resolution (VHR) airborne optical data enable cost-efficient biophysical measurements of reef colonies, strongly expected for conservation approaches. A synergy of remotely sensed airborne optical imagery, calibration/validation photoquadrat ground-truth (202/101, respectively), and artificial neural network (ANN) modelling is first used to map S. alveolata relative abundance, over the largest bioconstruction in Europe. The best prediction of S. alveolata abundance was reached with the infrared–red–green (IRRG) spectral combination and ANN model structured with six neurons (R2 = 0.72, RMSE = 0.08, and r = 0.85). The six hyperbolic tangent formulas were applied to the three input spectral bands (IRRG) in order to build six hidden neuronal images, resulting in VHR digital S. alveolata abundance model (6547 × 6566 pixels with 0.5 m pixel size). The innovative model revealed undescribed spatial patterns, namely a reef polarization (perpendicular to the shoreline) of S. alveolata abundance: high abundance on forereef and low abundance on backreef.

Fine resolution remote sensing of species in terrestrial and coastal ecosystems

Detailed and accurate information on the spatial distribution of individual species over large spatial extents and over multiple time periods is critical for rapid response and effective management of environmental change. Although remote sensing has been used to map species for decades, the long-standing challenge is that the accuracy of species maps is often too low to meet the requirements of management, or the methods are too complex or location-specific to be used in routine mapping. On the other hand, in the 21st century, we have witnessed a rapid development in both fine resolution remote sensors and statistical theories and techniques, which hold great potential for improved accuracy of species mapping. This special issue is a collection of eight studies that present cutting-edge research on using fine resolution remotely sensed data for mapping species or communities in terrestrial and coastal ecosystems. Several important findings emerge from these studies: