API src

Annotated boulder occurrence dataset using quartzite block detection derived from multibeam bathymetry processed data of SV DVOCEAN, Rhine River, Germany

Description: ANNOTATED BOULDER ABSTRACT This dataset comprises high-resolution multibeam echosounder (MBES) bathymetry and backscatter data collected in a 7.5 km segment of the River Rhine (km 747.0–754.5, Düsseldorf, Germany) between July 24th and 31st, 2024. The survey, conducted with a Kongsberg EM2040P MKII multibeam echosounder operated at a frequency of 300 kHz, achieved a point density of ≥ 160 points/m² and met the IHO Special Order standards (horizontal uncertainty ≤ 15 cm, vertical uncertainty ≤ 3 cm). Furthermore, the dataset includes 8,654 manually annotated quartzite blocks (mean area: 1.90 m², mean height above riverbed: 0.28 m), which were used to train a U-Net convolutional neural network (CNN) for an automated detection. The dataset comprises the raw (.db) bathymetric data, the applied sound velocity profiles (.csv), the pre-processed and filtered bathymetric data (0.25 m grid as .xyz and .tif) with their uncertainty metrics (.xyz, .tif), the computed and cleaned backscatter mosaics (available as 0.20 m and 0.25 m pixel size mosaics in .xyz and .tif format) and the manually digitized polygon annotations (.shp). The data are suited for hazard mapping, geomorphological studies, AI/ML model training (e.g., semantic segmentation) as well as the validation of automated boulder detection methods.

Global identifier:

Doi(
    "10.1594/PANGAEA.997228",
)

Types:
DataMeasurements(
    DataMeasurements {
        domain: Unspecified,
        station: None,
        measured_variables: [
            "Binary Object",
            "Binary Object (File Size)",
            "Binary Object (Media Type)",
            "Event label",
            "File content",
        ],
        methods: [],
    },
)

Tags: Fluss ? Deutschland ? Gewässervermessung ? Studie ? Daten ? Multibeam Echosounder ? River Rhine ?

Bounding boxes: 6.74983° .. 6.75219° x 51.2504° .. 51.2512°

License: Creative Commons Namensnennung 4.0 (Data access requires registration)

Language: Englisch/English

Organisations

Persons

Modified: 2026-09-07

Last harvest: 23.09.2026 02:03

Time ranges: 2024-07-20 - 2024-07-30

Resources

Status

Quality score

Accessed 1 times.