Skip to content

Global Offshore Wind Infrastructure: Deployment and Operational Dynamics from Dense Sentinel-1 Time Series

Home / Papers / Global Offshore Wind Infrastructure: Deployment and Operational Dynamics from Dense Sentinel-1 Time Series

Global Offshore Wind Infrastructure: Deployment and Operational Dynamics from Dense Sentinel-1 Time Series

Authors: Thorsten Hoeser, Felix Bachofer, Claudia Kuenzer Date: 2026-04-22 Paper ID: openalex:2604.20822

Summary

This paper addresses the gap in temporally dense, semantically annotated data for monitoring offshore wind infrastructure by providing a global, decade-long Sentinel-1 SAR time series corpus. The authors compile over 14 million analysis-ready 1D SAR backscatter profiles representing construction and operational phases. By releasing both the large-scale corpus and a curated expert-annotated benchmark, this work provides a standardized foundation for evaluating time series classification and change detection methods in the renewable energy sector.

Key Contributions

  • Introduces a global Sentinel-1 SAR time series corpus of 15,606 offshore wind infrastructure sites containing 14.8 million event-labeled backscatter profiles.
  • Provides an expert-annotated benchmark dataset of 553 time series with 328,657 event labels for training and validating infrastructure monitoring models.
  • Establishes a baseline rule-based classifier for semantic event labeling of SAR profiles, achieving a macro F1 score of 0.84.

Open Questions & Future Work

Archivist Review

The paper introduces a significant domain-specific dataset for monitoring offshore infrastructure using SAR time series. I have approved the primary corpus as a dataset note and an open question regarding early-onset detection, as these provide lasting value beyond the specific benchmarking results. Other candidates were rejected as sub-components or for being too closely tied to the specific paper’s internal evaluation tasks.

Approved Open Questions

  • Early-Onset Offshore Infrastructure Detection: Robust early-onset detection is essential for proactive monitoring of renewable energy deployment and requires moving beyond static site-based time series analysis.

Rejected Candidates

  • [dataset] Sentinel-1 expert-annotated benchmark dataset (sentinel-1-benchmark-dataset) - subcomponent_of_broader_mechanism: This is a sub-component of the broader Sentinel-1 offshore wind corpus, and I am prioritizing the latter as the primary dataset note.

Datasets

Metadata & Links

url
https://arxiv.org/abs/2604.20822
paper_id
2604.20822
paper_source
openalex
domain
time-series
tags
architectures
datasets
sentinel-1-offshore-wind-corpus
concept_slugs
dataset_slugs
sentinel-1-offshore-wind-corpus
skill
TimeSeriesSkill
created_at
2026-04-25T05:37:28Z