Benthic Habitat Classification Research Articles

The Camera-Based Assessment Survey System (C-BASS): A towed camera platform for reef fish abundance surveys and benthic habitat characterization in the Gulf of Mexico

An ongoing challenge for fisheries management is to provide cost-effective and timely estimates of habitat stratified fish densities. Traditional approaches use modified commercial fishing gear (such as trawls and baited hooks) that have biases in species selectivity and may also be inappropriate for deployment in some habitat types. Underwater visual and optical approaches offer the promise of more precise and less biased assessments of relative fish abundance, as well as direct estimates of absolute fish abundance. A number of video-based approaches have been developed and the technology for data acquisition, calibration, and synthesis has been developing rapidly.Beginning in 2012, our group of engineers and researchers at the University of South Florida has been working towards the goal of completing large scale, video-based surveys in the eastern Gulf of Mexico. This paper discusses design considerations and development of a towed camera system for collection of video-based data on commercially and recreationally important reef fishes and benthic habitat on the West Florida Shelf. Factors considered during development included potential habitat types to be assessed, sea-floor bathymetry, vessel support requirements, personnel requirements, and cost-effectiveness of system components. This regional-specific effort has resulted in a towed platform called the Camera-Based Assessment Survey System, or C-BASS, which has proven capable of surveying tens of kilometers of video transects per day and has the ability to cost-effective population estimates of reef fishes and coincident benthic habitat classification.

Integrating dual frequency side-scan sonar data and multibeam backscatter, angular response and bathymetry, for benthic habitat mapping in the Laganas Gulf MPA, Zakinthos Isl., Greece

The preferred procedure nowadays for benthic habitat mapping is combining marine acoustic and ground truthing methods, with the former ones tending to be the swath sonars, such as the Multi Beam Echo Sounders (MBES) and the Side-Scan Sonars (SSS). Both above acoustic systems, in conjunction with an extensive underwater video footage, were employed to map in detail benthic habitats in the marine area of Laganas Gulf in the National Marine Park of Zakynthos Isl., Greece, including key protected habitats such as P.oceanica beds and coralligenous formations. Object oriented seafloor classification was achieved taking advantage of the multi-layer information available, including the two individual frequencies of SSS and the MBES backscatter, angular response, and bathymetry data. The extracted statistical derivatives regarded: (1) textural analysis of the multi-frequency backscatter imagery, (2) angular range analysis of the MBES backscatter, as well as (3) application of various bathymetric indices. Those derivatives were classified individually or fuzzed together, to explore the rate of improvement when choosing one system or another and to investigate their best combinations and practices towards valid seafloor classification. Benthic habitat classification has been comprehended using the NATURA and EUNIS classification schemes to be compatible with the EU regulations.

Impact of Microphytobenthos Photosynthesis on the Characteristics of the Echo Signal from Baltic Sandy Sediments

Abstract The understanding the influence of biological processes on the characteristics of the signals backscattered by the sea floor is crucial in the development of the hydroacoustical benthic habitat classification techniques. The impact of the microphytobenthos photosynthesis on the acoustical backscattering properties of the Atlantic sandy sediments was previously demonstrated by Holliday et al. (2004) and Wildman and Huettel (2012). To account for the sensitivity of the hydroacoustical classification techniques to the backscattering properties of local marine sediments, it is important to understand the microphytobenthos photosynthesis impact for the Baltic Sea where the techniques are being actively developed now. This is the main motivation of the paper. In the paper the influence of the microphytobenthos photosynthesis on the characteristics of the echo signals reflected by sandy sediments in the typical Baltic temperature and the salinity conditions is discussed. The interdisciplinary multiday laboratory experiment was conducted to study the impact of benthic microalgal photosynthesis on the characteristics of the echo signal reflected by sandy sediments. Hydroacoustical data were collected under controlled constant light, temperature and salinity conditions. The oxygen content at different levels of the water column was simultaneously monitored.

IDENTIFICATION AND CLASSIFICATION OF LAKE BOTTOM SURFACE AND AQUATIC VEGETATION IN LAUT TAWAR LAKE, ACEH

Identification and classification of benthic habitats in Lake of Laut Tawar, Aceh by using hydro acoustic method can provide data and information on types of substrate and aquatic vegetation in a short time and wide spatial coverage, as done in the present work. Data acoustic collection was performed in 2013 using quantitative echosounder with split beam frequency of 120 kHz, and through a visual observation. The later is destined to look at the bottom types and macrophytes that lie on the line transect of acoustic survey. Analysis of data is to extract the value of bottom volume backscattering for each transect of 0.5-1 km. Classification of the bottom type was done based on the value of Sv using geospatial models. Results show the interval value of Sv for soft bottom ranged between -24.00 dB and -32.00 dB, the type of hard bottom (e.g. rocks, rocky sand substrate) ranged between -14.00 dB and -22.00 dB, whereas the Sv value of macrophyte ranged between- 45.00 dB and -54.00 dB. The percent covers were about 42.90%, 44.71% and 12.93% for hard bottom type, soft bottom and macrophytes, respectively. The types of aquatic vegetation commonly found in the lake were two genera belonging Hydrocharitaceaea and Gramineae. The current work is still lack of information on the classification of organisms into genera scales. Therefore, more signal verification and algorithms verification would be needed in order to estimate macrophytes biomass by comparing with other visual observation.

A comparison of top-down and bottom-up approaches to benthic habitat mapping to inform offshore wind energy development

Recent interest in offshore renewable energy within the United States has amplified the need for marine spatial planning to direct management strategies and address competing user demands. To assist this effort in Rhode Island, benthic habitat classification maps were developed for two sites in offshore waters being considered for wind turbine installation. Maps characterizing and representing the distribution and extent of benthic habitats are valuable tools for improving understanding of ecosystem patterns and processes, and promoting scientifically-sound management decisions. This project presented the opportunity to conduct a comparison of the methodologies and resulting map outputs of two classification approaches, “top-down” and “bottom-up” in the two study areas. This comparison was undertaken to improve understanding of mapping methodologies and their applicability, including the bottom-up approach in offshore environments where data density tends to be lower, as well as to provide case studies for scientists and managers to consider for their own areas of interest. Such case studies can offer guidance for future work for assessing methodologies and translating them to other areas.The traditional top-down mapping approach identifies biological community patterns based on communities occurring within geologically defined habitat map units, under the concept that geologic environments contain distinct biological assemblages. Alternatively, the bottom-up approach aims to establish habitat map units centered on biological similarity and then uses statistics to identify relationships with associated environmental parameters and determine habitat boundaries. When applied to the two study areas, both mapping approaches produced habitat classes with distinct macrofaunal assemblages and each established statistically strong and significant biotic–abiotic relationships with geologic features, sediment characteristics, water depth, and/or habitat heterogeneity over various spatial scales. The approaches were also able to integrate various data at differing spatial resolutions. The classification outputs exhibited similar results, including the number of habitat classes generated, the number of species defining the classes, the level of distinction of the biological communities, and dominance by tube-building amphipods. These results indicate that both approaches are able to discern a comparable degree of habitat variability and produce cohesive macrofaunal assemblages. The mapping approaches identify broadly similar benthic habitats at the two study sites and methods were able to distinguish the differing levels of heterogeneity between them.The top-down approach to habitat classification was faster and simpler to accomplish with the data available in this study when compared to the bottom-up approach. Additionally, the top-down approach generated full-coverage habitat classes that are clearly delineated and can easily be interpreted by the map user, which is desirable from a management perspective for providing a more complete assessment of the areas of interest. However, a higher level of biological variability was noted in some of the habitat classes created, indicating that the biological communities present in this area are influenced by factors not captured in the broad-scale geological habitat units used in this approach.The bottom-up approach was valuable in its ability to more clearly define macrofaunal assemblages among habitats, discern finer-scale habitat characteristics, and directly assess the degree of macrofaunal assemblage variability captured by the environmental parameters. From a user perspective, the map is more complex, which may be perceived as a limitation, though likely reflects natural gradations in habitat structure and likely presents a more ecologically realistic portrayal of the study areas. Though more comprehensive, the bottom-up approach in this study was limited by the reliance on full-coverage data to create full-coverage habitat classes. Such classes could only be developed when sediment data was excluded, since this point-sample dataset could not be interpolated due to high spatial heterogeneity of the study areas. Given a higher density of bottom samples, this issue could be rectified.While the top-down approach was more appropriate for this study, both approaches were found to be suitable for mapping and classifying benthic habitats. In the United States, objectives for mapping and classification for renewable energy development have not been well established. Therefore, at this time, the best-suited approach primarily depends on mapping objectives, resource availability, data quality and coverage, and geographical location, as these factors impact the types of data included, the analyses and modeling that can be performed, and the biotic–abiotic relationships identified.

Classifying the Baltic Sea Shallow Water Habitats Using Image-Based and Spectral Library Methods

The structure of benthic macrophyte habitats is known to indicate the quality of coastal water. Thus, a large-scale analysis of the spatial patterns of coastal marine habitats enables us to adequately estimate the status of valuable coastal marine habitats, provide better evidence for environmental changes and describe processes that are behind the changes. Knowing the spatial distribution of benthic habitats is also important from the coastal management point of view. A big challenge in remote sensing mapping of benthic habitats is to define appropriate mapping classes that are also meaningful from the ecological point of view. In this study, the benthic habitat classification scheme was defined for the study areas in the relatively turbid north-eastern Baltic Sea coastal environment. Two different classification methods—image-based and the spectral library—method were used for image classification. The image-based classification method can provide benthic habitat maps from coastal areas, but requires extensive field studies. An alternative approach in image classification is to use measured and/or modelled spectral libraries. This method does not require fieldwork at the time of image collection if preliminary information about the potential benthic habitats and their spectral properties, as well as variability in optical water properties exists from earlier studies. A spectral library was generated through radiative transfer model HydroLight computations using measured reflectance spectra from representative benthic substrates and water quality measurements. Our previous results have shown that benthic habitat mapping should be done at high spatial resolution, owing to the small-scale heterogeneity of such habitats in the Estonian coastal waters. In this study, the capability of high spatial resolution hyperspectral airborne a Compact Airborne Spectrographic Imager (CASI) sensor and a high spatial resolution multispectral WorldView-2 satellite sensor were tested for mapping benthic habitats. Initial evaluations of habitat maps indicate that image-based classification provides higher quality benthic maps compared to the spectral library method.

Spatial distribution of marine benthic habitats in the Estonian coastal sea, northeastern Baltic Sea

INTRODUCTION Annex I of the Habitat Directive (Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora) defines the habitat types of higher importance and in need of conservation within Europe. Together with the Birds Directive this directive forms a solid base for the European nature conservation policy and the Natura 2000 network. In order to implement the requirements of the directive and to resolve the existing needs for local conservation purposes, a benthic habitat classification system EBHAB (Eastern Baltic marine benthic HABitats) was developed in the frame of the multinational EU LIFE project 'Marine Protected Areas in the Eastern Baltic Sea (Baltic MPAs)'. In the current study a detailed inventory of the EBHAB habitat classes was performed within six study areas covering the most valuable nearshore marine areas and including several protected areas, i.e. the Natura 2000 areas in the Estonian coastal range. In general, prior to this inventory the knowledge on the distribution of habitats and associated biota in the coastal areas was very poor including only a few publications each containing a handful sampling sites only (e.g. Martin, 2000; Kotta & Orav, 2001; Kotta et al., 2008, Kovtun et al., 2009). Moreover, the published background information was very scattered and owing to differences in methodologies there was no way to systematize such knowledge. Thus, there was a need to examine the species composition and identify typical communities within their environment. The current inventory allowed us for the first time (1) to summarize the range of large-scale variability of the studied habitats in terms of benthic biodiversity, species composition, and dominance structure; (2) to provide a knowledge base for follow-on biological and ecological studies; and (3) thus, to provide important information for managing our marine resources. MATERIAL AND METHODS Study area Altogether six areas representing various environmental and climatic conditions in the northeastern Baltic Sea were studied in detail with the aim of classifying the observed benthic habitats according to the recently established EBHAB classes (Fig. 1). An overview of the study areas is given below. [FIGURE 1 OMITTED] Eastern and western Gulf of Finland areas The average depth of the Gulf of Finland is 37 m and the maximum depth 123 m. The coastal slope is quite steep and depths reach over 30 m in all bays of the study area. Sand, silt, or sandy clay bottoms dominate in deeper areas, boulders and stones prevail near the coast. The coastline is diverse and disjuncted by peninsulas, many small islands occur in the area. The eastern gulf receives runoff from a huge drainage area and the western gulf is a direct continuation of the Baltic Proper. Therefore the gulf has a permanent east-west gradient of salinity. The salinity range of the study area is 4.5-6.2. The area is strongly influenced by diffuse and point-source nutrient loads. West-Estonian Archipelago Sea The area comprises shallow bays with many small islets and mudflats. The average depth of the area is less than 4 m with the maximum close to 20 m. The bottom morphology of the area is flat, with gentle slopes towards deeps. The whole water basin is semi-exposed. Sand and sandy clay sediments prevail in the entire area of the archipelago. Due to the shallowness and clayey sediments already moderate winds result in strong resuspension of bottom sediments and poor underwater light conditions. Salinity varies from 0 to 7, being the lowest in the estuary of the Kasari River in the eastern part of the area and around 6-7 in the western, open-sea area. Haapsalu and Matsalu bays are the most eutrophied bays within the West-Estonian Archipelago Sea. Western Saaremaa Island Similarly to the previous study area this area consists of small shallow bays with many scattered islets. …

SUPPORT VECTOR CLASSIFICATION OF LAND COVER AND BENTHIC HABITAT FROM HYPERSPECTRAL IMAGES

This paper presents a novel wavelet and support vector machine (SVM) based method for hyperspectral image classification. A 1-D wavelet transform is applied to the pixel spectra, followed by feature extraction and SVM classification. Contrary to the traditional method of using pixel spectra with SVM classifier, our approach not only reduces the dimension of the input pixel feature vector but also improves the classification accuracy. Texture energy features computed in the spectral dimension are mapped using polynomial kernels and used for training the SVM classifier. Results with AVIRIS and other hyperspectral images for land cover and benthic habitat classification are presented. The accuracy of the method with limited training sets and computational burden is assessed.

Land cover and benthic habitat classification using texture features from hyperspectral and multispectral images

A texture-based method for classification of land cover and benthic habitats from hyperspectral and multispectral images is presented. The features considered in this work are a set of statistical and multiresolution texture features, including a 2-level wavelet transform that uses an orthonormal Daubechies filter. These features are computed over spatial extents from each band of the image. A stepwise sequential feature selection process is applied that results in the selection of optimal features from the original feature set. A supervised classification is performed with a distance metric. Results with AVIRIS hyperspectral and IKONOS multispectral images show that texture features perform well under different land cover scenarios and are effective in characterizing the texture information at different wavelengths. Results over coastal regions show that wavelet texture features computed over the reflectance spectrum can accurately detect the benthic classes.

A review of approaches for classifying benthic habitats and evaluating habitat quality

We have assessed the current state of knowledge relative to methods used in assessing sub-tidal benthic habitat quality and the classification of benthic habitats. While our main focus is on marine habitat, we extensively draw on knowledge gained in freshwater systems where benthic assessment procedures are at an advanced stage of maturity. We found a broad range of sophistication/complication in terms of the methods applied in assessing and mapping benthic habitats. The simplest index or metric involved some assessment of species richness, while the most complicated required utilizing multi-variate analysis. The simplest mapping attempts equated physical substrate with benthic habitat while the most sophisticated relied on extensive environmental preference and groundtruth data for species of concern. The leading edge of methods for benthic habitat mapping involves combining the advances in optical and acoustic methods that allow for routine classifying and mapping of the seafloor with biological and habitat data for species of concern. The objective of this melding of dispirit methods is to produce benthic habitat maps with broad system wide coverage and sound biological underpinning. It is clear that the disparity in information density between the physical and biological sides of the equation currently hinder applicability and acceptability of benthic habitat mapping efforts. In addition to the lack of basic information on the biological and environmental tolerances of targeted species, the proliferation of metrics for characterizing and assessing biological conditions further clouds the usefulness of any broad scale mapping attempt. The problem of data density mismatch between physical and biological methods will likely not be solved until acoustic methods can routinely resolve the elusive biological components that make a physical substrate a habitat.

Benthic habitats in the northwest Mediterranean characterised by sedimentary organics, benthic macrofauna and sediment profile images

A total of 36 benthic stations were studied in four areas in the oligotrophic Gulf of Lions and in one area of the severely disturbed Berre Lagoon in the northwest Mediterranean in October 2001. The study included analysis of sediment characteristics, particularly organic matter, and analyses of macrofauna and sediment profile images (SPIs). In the Gulf of Lions, values of superficial sediment biochemical characteristics were similar to available literature data, with the important exception of enzymatically hydrolysable amino acids (EHAA) and enzymatically to total amino acid ratios (EHAA/THAA) off the Rhône River and the city of Saint-Cyprien. All biochemical concentrations, but not EHAA/THAA ratios, were higher in the Berre Lagoon. A total number of 190 taxa were recorded, generally occurring in low density and biomass. In most SPIs from the Gulf, a deep infaunal bioturbation was observed, i.e. burrows and feeding voids were recorded in 85 and 62% of the images, respectively. This infaunal activity was associated with a mean depth of the apparent redox potential discontinuity (aRPD) in the sediment at 2.1–3.5 cm in the gulf and 0.2 cm in the lagoon. The benthic habitat quality (BHQ) index, summarising surface structures, subsurface structures and the depth of the aRPD, assigned most benthic stations in the Gulf to the benthic successional stages II or III, i.e. fairly undisturbed habitats. Demersal trawling, however, probably was a disturbance to most muddy superficial sediments in the Gulf. Multivariate analysis (MDS) of the macrofauna and the SPIs both separated three groups of stations: muddy, sandy and those that were disturbed from the lagoon, but a few stations showed dissimilarities to this general pattern. Principal component analysis (PCA) resulted in the best correlation between faunal abundance and biomass and enzymatically available amino acids in the sediment. PCA grouped almost all stations into three separate groups: muddy, sandy and severely disturbed habitats, and the muddy habitat group was associated with high aRPDs and BHQ indices. It is concluded that SPI is useful for in situ analysis of infaunal bioturbation activity and redox conditions, and for benthic habitat classification in the oligotrophic northwestern Mediterranean and the disturbed Berre Lagoon.