Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 9 min ago
PreprintCC BYEarth System Science Data Discussions

BenthiCat: An opti-acoustic dataset for advancing benthic classification and habitat mapping

  • Hayat Rajani1
  • Valerio Franchi2
  • Borja Martinez-Clavel Valles
  • Raimon Ramos3
  • Rafael Garcia2
  • Nuno Gracias2
  1. 1Universitat de Girona (Girona, Spain)
  2. 2Computer Vision and Robotics Research Institute, University of Girona, Spain
  3. 3Tecnoambiente SLU, Spain
Open at sourceDOICite

Abstract

Benthic habitat mapping is fundamental for understanding marine ecosystems, guiding conservation efforts, and supporting sustainable resource management. Yet, the scarcity of large, annotated datasets limits the development and benchmarking of machine learning models in this domain. This paper introduces a thorough multi-modal dataset, comprising about a million side-scan sonar (SSS) tiles collected along the coast of Catalonia (Spain), complemented by bathymetric maps and a set of co-registered optical images from targeted surveys using an autonomous underwater vehicle (AUV). Approximately 36000 of the SSS tiles have been manually annotated with segmentation masks to enable supervised fine-tuning of classification models. All the raw sensor data, together with mosaics, are also released to support further exploration and algorithm development. To address challenges in multi-sensor data fusion for AUVs, we spatially associate optical images with corresponding SSS tiles, facilitating self-supervised, cross-modal representation learning. Accompanying open-source preprocessing and annotation tools are provided to enhance accessibility and encourage research. This resource aims to establish a standardized benchmark for underwater habitat mapping, promoting advancements in autonomous seafloor classification and multi-sensor integration.

Related

  1. Anthropogenically induced shifts in sediment vegetation impact coastal benthic microbiomes
    Borja Aldeguer-Riquelme, Meagan A Beatty, Luis H Orellana +5
    HALPreprintPublisher termsFull textEcologyMicrobiologyHAL
  2. Persistent seasonal deoxygenation alters benthic metabolic processes and accelerates hypoxia feedbacks in the Baltic Sea
    Shuba Varshini Alampalli, Stefanie Böhnke-Brandt, Nicole Adam-Beyer +11
    Communications Earth & EnvironmentPeer-reviewedCC BY 4.0Full textOceanography & cryosphereCrossrefDOAJ
  3. Spatio-Temporal Patterns and Ecological Implications of Microplastic Contamination in Benthic Fish from the Tongkou River, China.
    Jiayun Wu, Qingqing Liao, Yingying Ren +7
    Animals : an open access journal from MDPIPeer-reviewedCC BYFull textEnvironmental science & pollutionEurope PMC
  4. The effects of estuary salinization on benthic assemblages: a richer and treacherous future
    Francisco Barros, Yuri Costa, Marcos Krull +5
    Royal Society Open SciencePeer-reviewedCC BY 4.0Full textEcologyOceanography & cryosphereCrossrefEurope PMCPMC
  5. Intra- and inter-specific variability in benthic foraminifera modulates surface sediment reworking
    Manon Doutrelant, Vincent M. P. Bouchet, Emmanuelle Geslin +4
    Royal Society Open SciencePeer-reviewedCC BY 4.0Full textGeneral biologyEnvironmental science & pollutionCrossrefHAL

Cite

BibTeXRIS
BibTeX
@misc{rajani2026benthicat,
  title = {{BenthiCat: An opti-acoustic dataset for advancing benthic classification and habitat mapping}},
  author = {Hayat Rajani and Valerio Franchi and Borja Martinez-Clavel Valles and Raimon Ramos and Rafael Garcia and Nuno Gracias},
  howpublished = {Earth System Science Data Discussions},
  year = {2026},
  month = aug,
  doi = {10.5194/essd-2026-565},
  url = {https://doi.org/10.5194/essd-2026-565},
  note = {CC BY},
}
RIS
TY  - UNPB
TI  - BenthiCat: An opti-acoustic dataset for advancing benthic classification and habitat mapping
AU  - Hayat Rajani
AU  - Valerio Franchi
AU  - Borja Martinez-Clavel Valles
AU  - Raimon Ramos
AU  - Rafael Garcia
AU  - Nuno Gracias
PY  - 2026
DA  - 2026/08/10
PB  - Earth System Science Data Discussions
DO  - 10.5194/essd-2026-565
UR  - https://doi.org/10.5194/essd-2026-565
AB  - Benthic habitat mapping is fundamental for understanding marine ecosystems, guiding conservation efforts, and supporting sustainable resource management. Yet, the scarcity of large, annotated datasets limits the development and benchmarking of machine learning models in this domain. This paper introduces a thorough multi-modal dataset, comprising about a million side-scan sonar (SSS) tiles collected along the coast of Catalonia (Spain), complemented by bathymetric maps and a set of co-registered optical images from targeted surveys using an autonomous underwater vehicle (AUV). Approximately 36000 of the SSS tiles have been manually annotated with segmentation masks to enable supervised fine-tuning of classification models. All the raw sensor data, together with mosaics, are also released to support further exploration and algorithm development. To address challenges in multi-sensor data fusion for AUVs, we spatially associate optical images with corresponding SSS tiles, facilitating self-supervised, cross-modal representation learning. Accompanying open-source preprocessing and annotation tools are provided to enhance accessibility and encourage research. This resource aims to establish a standardized benchmark for underwater habitat mapping, promoting advancements in autonomous seafloor classification and multi-sensor integration.
SN  - 1866-3591
ER  -