{
  "abstract": "Accurate multi-horizon wind direction forecasting is critical for turbine yaw control and grid security. Rapid directional shear (turning $\\ge 90^\\circ$) challenges models via non-Euclidean geometry on $S^1$, multiscale dynamics, and regime-dependent uncertainty. Conventional discrete models and foundation models suffer from mid-frequency phase lag and turning misalignments. We show that directional predictability decays at disparate rates across frequency subbands, rendering monolithic mechanisms suboptimal. We propose a predictability-guided paradigm: slow synoptic drift $\\to$ deterministic regression; intermediate turning $\\to$ continuous latent differential flows; unresolved turbulence $\\to$ conditional residual diffusion; followed by causal recalibration. On a 10,000-sequence multi-year benchmark, our framework maintains calm-weather accuracy (Test MCE $38.48^\\circ$) while reducing extreme-turning error (Case 1 MCE $60.69^\\circ$ vs $70.42^\\circ$ for zero-shot foundation models). The circular CRPS reaches $22.36^\\circ$, with 93.88\\% coverage at nominal 95\\% (91.01\\% out-of-distribution). Density estimation further reveals near-antipodal bimodal structure under severe shear (13.39\\%--15.43\\% tail mass $\\ge 135^\\circ$), exposing a geometric bound where single-center calibration under-covers (81.56\\%), motivating multimodal circular manifold learning.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.16707",
  "author_total": 1,
  "authors": [
    {
      "name": "Hailong Shu"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/6r8bw7zjwh7fg.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.16707",
  "fulltext_url": null,
  "id": "6r8bw7zjwh7fg",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.16707",
  "language": null,
  "licence": "CC-BY-NC-ND-4.0",
  "licence_label": "CC BY-NC-ND 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by-nc-nd/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "physics.ao-ph"
    ],
    [
      "openalex",
      "2208|Energy Load and Power Forecasting"
    ]
  ],
  "openalex": "W7213392533",
  "pdf_url": "https://arxiv.org/pdf/2609.16707",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-15",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/6r8bw7zjwh7fg.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:21:41Z",
      "native_id": "2609.16707",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.16707"
    }
  ],
  "subjects": [
    {
      "field": "earth",
      "origin": "arxiv:physics.ao-ph",
      "subfield": "atmospheric"
    }
  ],
  "title": "Predictability-Guided Multiscale Probabilistic Forecasting of Wind Direction under Extreme Shear",
  "title_text": "Predictability-Guided Multiscale Probabilistic Forecasting of Wind Direction under Extreme Shear",
  "type": "preprint",
  "updated_at": "2026-10-30T12:21:41Z",
  "url": "https://research.draconitic.com/paper/6r8bw7zjwh7fg",
  "venue": "arXiv",
  "versions": []
}