Neural network-based methods for ocean surface wave measurement using submarine distributed acoustic sensing (DAS)
收藏DataONE2026-02-10 更新2026-02-14 收录
下载链接：
https://search.dataone.org/view/sha256:b1cadeac220126a31af8a66564e4bf6b2a6b6e85cb9e160642d9eb8602cd1cdc
下载链接
链接失效反馈官方服务：
资源简介：
The data accompanying the article present two new neural network-based methods for estimating ocean surface waves from distributed acoustic sensing (DAS) submarine cable strain rate. Models were trained using supervised machine learning on a 10-day dataset collected offshore of Oliktok Point, Alaska, in late summer. The new models were trained on target data from seafloor pressure moorings at three sites spaced evenly along 27.1 km of cable and were benchmarked against an empirical transfer function method previously used to estimate waves from DAS. This dataset contains both hourly and half-hourly DAS and mooring data used to train the neural networks described in the article. The data span 22 August 2023 to 22 September 2023, excluding the period in which DAS measurements were paused (30 August to 18 September) for repairs to the submarine fiber-optic cable. Trained model weights and normalizations are provided in PyTorch state dict format.
, , # Data from: Neural network-based methods for ocean surface wave measurement using submarine distributed acoustic sensing (DAS)
Dataset DOI: [10.5061/dryad.brv15dvnz](https://doi.org/10.5061/dryad.brv15dvnz)
## Description of the data and file structure
### Files and variables
DAS and mooring data are provided in netCDF (.nc) format and contain self-describing global attributes, variable descriptions, and units.
**oliktok_das_mooring_hourly_dataset.nc**
This dataset contains hourly DAS features (depth, cosine-square-wave direction, strain-rate spectra) and target mooring seafloor pressure spectra for training the spectral neural network model. Hourly strain rate and pressure spectral density estimates were computed from demeaned DAS and mooring records, respectively. Spectra were estimated using Welch's method with 512-point (256-s) Hann-windowed segments and 50% overlap. Frequencies above 0.5 Hz were trimmed to remove much of the noise floor, and every 4 adjacent frequencies were...,
创建时间：
2026-02-11
