TY - JOUR

T1 - Assessment of sand nourishment dynamics under repeated storm impact supported by machine learning-based analysis of UAV data

AU - Tiede, Jan

AU - Lovell, Joshua Leon

AU - Jordan, Christian

AU - Moghimi, Armin

AU - Schlurmann, Torsten

N1 - Publisher Copyright: Copyright © 2025 Tiede, Lovell, Jordan, Moghimi and Schlurmann.

PY - 2025/4/28

Y1 - 2025/4/28

N2 - Understanding beach dynamics and the long-term evolution of beach nourishment projects is critical for sustainable coastal management, particularly in the face of rising sea levels and increasingly variable storm climates. This study examines the development of a large-scale sand nourishment (600,000 m³) in the southwestern Baltic Sea over 25 months (October 2021–November 2023) using UAV-derived digital surface models (DSMs) and machine learning (ML). High-frequency, multi-temporal UAV surveys enabled detailed analyses of the development of the nourished beach and dune. Results revealed that the volumetric impact of the 100-year flood in October 2023 was comparable to the cumulative effects of the October 2022–January 2023 storm season. This demonstrates that both episodic extreme events and the cumulative impacts shape the morphological evolution of the nourishment. The study also highlights sediment transport reversals under easterly winds, promoting longer-term stability by retaining sediment within the system. By standardizing volumetric analyses using tools equipped with ML, this research provides actionable insights for adaptive management and establishes a framework for comparable, accurate assessments of nourishment lifetime. In particular, these methods efficiently capture subtle variations in coastline orientation, wave incidence angles, and resulting alongshore beach dynamics, offering valuable insights for optimizing nourishment strategies. These findings underscore the importance of continuous, high-resolution monitoring in developing sustainable strategies for storm-driven erosion and sea level rise.

AB - Understanding beach dynamics and the long-term evolution of beach nourishment projects is critical for sustainable coastal management, particularly in the face of rising sea levels and increasingly variable storm climates. This study examines the development of a large-scale sand nourishment (600,000 m³) in the southwestern Baltic Sea over 25 months (October 2021–November 2023) using UAV-derived digital surface models (DSMs) and machine learning (ML). High-frequency, multi-temporal UAV surveys enabled detailed analyses of the development of the nourished beach and dune. Results revealed that the volumetric impact of the 100-year flood in October 2023 was comparable to the cumulative effects of the October 2022–January 2023 storm season. This demonstrates that both episodic extreme events and the cumulative impacts shape the morphological evolution of the nourishment. The study also highlights sediment transport reversals under easterly winds, promoting longer-term stability by retaining sediment within the system. By standardizing volumetric analyses using tools equipped with ML, this research provides actionable insights for adaptive management and establishes a framework for comparable, accurate assessments of nourishment lifetime. In particular, these methods efficiently capture subtle variations in coastline orientation, wave incidence angles, and resulting alongshore beach dynamics, offering valuable insights for optimizing nourishment strategies. These findings underscore the importance of continuous, high-resolution monitoring in developing sustainable strategies for storm-driven erosion and sea level rise.

KW - 100-year flood

KW - co-alignment

KW - machine learning

KW - RTK-UAV

KW - sand nourishments

UR - http://www.scopus.com/inward/record.url?scp=105004764878&partnerID=8YFLogxK

U2 - 10.3389/fmars.2025.1537066

DO - 10.3389/fmars.2025.1537066

M3 - Article

AN - SCOPUS:105004764878

VL - 12

JO - Frontiers in Marine Science

JF - Frontiers in Marine Science

SN - 2296-7745

M1 - 1537066

ER -