# The Well: Helmholtz Staircase, Acoustic Scattering by a Periodic Surface

## Facts

- Provider: Polymathic AI (The Well); Fruzsina Julia Agocs, Flatiron Institute CCM and University of Colorado Boulder
- License: CC-BY-4.0 (https://arxiv.org/html/2412.00568)
- Size: 56 files
- Format: HDF5 (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/polymathic-ai/helmholtz_staircase
- Access: public, readable without an account
- Category: Physics
- Source: https://arxiv.org/abs/2310.12486
- Page: https://scigantic.com/archive/687e4b61-9ee2-5d27-a7e8-accadf0384b7
- Keywords: Physics, HDF5, Hugging Face

## About

High-order accurate solutions of the Helmholtz equation for acoustic scattering of a non-periodic point source by an infinite periodic staircase surface, relevant to waveguides, diffraction gratings, antennae and photonic or phononic crystals. Computed with a Floquet-Bloch transform (array scanning) combined with boundary integral equation methods, with Neumann conditions at the surface. Each trajectory holds 50 time steps of 1024 by 256 images over one period 2 pi / omega. Fields are the real and imaginary parts of acoustic pressure and a stationary staircase mask. The datasheet lists 512 trajectories: 16 frequencies omega times 32 source positions. Listing data/ (measured 2026-11-13) shows train/ (16 files, 44 GB, one per frequency), valid/ and test/ (16 files, 5.4 GB each); about 55 GB in the repository. Files are HDF5 in the common The Well format; open them with the_well package (WellDataModule with base path hf://datasets/polymathic-ai/) or directly with h5py. Part of The Well (Ohana et al., NeurIPS 2024 Datasets and Benchmarks, arXiv:2412.00568); released under CC-BY-4.0, cite The Well and the simulation's source papers.

## Load the data

```python
from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="polymathic-ai/helmholtz_staircase", filename="data/test/helmholtz_staircase_omega_006.hdf5", repo_type="dataset")
print(path)
```

## Data structure

File types in sampled files: HDF5

Example file: `data/test/helmholtz_staircase_omega_006.hdf5`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/687e4b61-9ee2-5d27-a7e8-accadf0384b7
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/687e4b61-9ee2-5d27-a7e8-accadf0384b7
- Over MCP: `get_schema_card` with id `687e4b61-9ee2-5d27-a7e8-accadf0384b7` at https://api.scigantic.com/mcp

## Related topics

- [HDF5 datasets](https://scigantic.com/topics/hdf5)
- [Physics datasets](https://scigantic.com/topics/physics)
