# TIDE Strongly Rotating Turbulence (Coriolis Force): 3D Incompressible Navier-Stokes DNS at 256^3 (Dai et al.)

## Facts

- Provider: Yilong Dai, Yiming Sun, Yiheng Chen, Shengyu Chen, Peyman Givi
- License: CC-BY-4.0 (https://huggingface.co/datasets/ydai17/TIDE-rotating_ro0p2_256_fp64)
- Size: 2,415 files
- Stored at: Hugging Face, https://huggingface.co/datasets/ydai17/TIDE-rotating_ro0p2_256_fp64
- Access: public, readable without an account
- Category: Physics
- Source: https://arxiv.org/abs/2608.04222
- Page: https://scigantic.com/archive/cf7a4d75-3d42-536d-8515-1bb6d9ec122a
- Keywords: Physics, benchmark, Hugging Face

## About

Direct numerical simulation (DNS) data for strongly rotating turbulence: the forced system with one added ingredient, a Coriolis (rotation) term in the momentum balance, in the strong-rotation case of the two rotating configurations, about 222 GB on disk. Part of TIDE (Turbulent Incompressible DNS Ensembles), a 3D turbulence corpus of 15 configurations of one incompressible Navier-Stokes solver on a periodic 256-cubed grid, each with 8 to 16 independent realizations that passed statistical acceptance gates and equation-level residual checks (Dai et al., arXiv:2608.04222). The store is a chunked zarr (zstd, one frame per chunk) holding velocity [N,3,256,256,256], pressure [N,256,256,256], per-frame time, k_max*eta resolution and seed; fields are computed in double precision and stored as float32, with frames exported every 0.05 large-eddy turnover times (about one Kolmogorov time) and every frame resolved to k_max*eta >= 1.5. Open with zarr or xarray; the index repository ydai17/TIDE holds the datasheet and the train/validation/test split manifest, and the GitHub repository Dyloong1/TIDE-dataset-benchmark holds the solver and benchmark. Use it to train and test neural operators and surrogate models for 3D turbulence. Released under CC BY 4.0; cite Dai et al. (2026), arXiv:2608.04222, and the datasheet record DOI 10.5281/zenodo.21589489.

## Load the data

```python
from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="ydai17/TIDE-rotating_ro0p2_256_fp64", filename="k_max_eta/.zarray", repo_type="dataset")
print(path)
```

## Data structure

Example file: `k_max_eta/.zarray`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/cf7a4d75-3d42-536d-8515-1bb6d9ec122a
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/cf7a4d75-3d42-536d-8515-1bb6d9ec122a
- Over MCP: `get_schema_card` with id `cf7a4d75-3d42-536d-8515-1bb6d9ec122a` at https://api.scigantic.com/mcp

## Related topics

- [Benchmark datasets](https://scigantic.com/topics/benchmark)
- [Physics datasets](https://scigantic.com/topics/physics)
