From the NVIDIA edition of October 26, 2026
StormScope paper by NVIDIA authors: generative nowcasting beats HRRR up to 4 hours but authors do not claim state of the art for 0 to 2 hour radar-only nowcasting
NVIDIA (paper authors) · Weather and climate AI research · Continental United States (evaluation domain)
- Publisher
- arXiv
- Original publication
- 2026-01-24 (arXiv v1)
- Source retrieved
- 2026-10-27
- Event date
- 2026-01-24
What happened
The StormScope preprint, by NVIDIA authors, reports that the generative nowcasting model outperforms the HRRR operational model for forecasts up to about 4 hours and is comparable around 6 hours, at 6 km resolution. The authors state they do not claim state-of-the-art performance on the 0 to 2 hour timescale for radar-only nowcasting. The paper does not evaluate hail. It is included now because CSU's Oct 26 hail work builds on this model.
Why it matters
Gives agencies and universities the base model's documented limits when assessing CSU's hail adaptation.
Evidence and measured results
Preprint with quantitative comparison against HRRR; not yet peer reviewed as accessed; authored by the vendor's researchers.
Limitations and uncertainty
Vendor-authored preprint; no hail evaluation; resolution differs from CSU target.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-10-27; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Agencies need to know what the base model behind new hail work can and cannot do.
- Stakeholders
- State weather and emergency agencies
- University researchers
- Discovery questions
- Which forecast horizon matters most for your hail decisions?
- Value hypothesis
- Interpretation: the model may add value at 2 to 6 hours; short-horizon value is not claimed.
- Potential engagement
- Briefing on documented limits during weather AI discussions.
- Unsupported claims
- Hail skill
- Skill at 1.3 km
Pre-sales engineering
Role takeaway
- Fit
- Research use for multi-hour convective nowcasting.
- Architecture and integration
- Generative nowcasting model compared with HRRR at 6 km.
- Prerequisites
- GPU compute
- Radar and satellite inputs
- Constraints
- No state-of-the-art claim for 0 to 2 hours radar-only
- No hail evaluation
- 6 km resolution
- Security
- None specific.
- Proof of value
- Verify at target resolution
- Evaluate hail-specific output
Delivery
Role takeaway
- Work
- Use as background for evaluating derived products.
- Dependencies
- Derived product verification
- Ownership
- Agency forecasting lead.
- Skills and adoption
- Forecasters need to understand horizon-dependent skill.
- Governance checkpoints
- Require verification at used resolution and horizon
- Acceptance criteria
- Proposed: 0 to 2 hour skill compared with current radar nowcasting before operational use
- Risks
- Transferring base-model skill claims to adapted products
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Generative model at 6 km; CSU's reported 1.3 km models differ markedly from the evaluated configuration.
Governance
Who approves, reviews and stays accountable for outcomes?
Agencies should distinguish base-model skill from adapted-model skill when reviewing claims.
Security and privacy
What data, permissions and controls need testing?
None specific.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
None specific.
Procurement
What should contracts, pricing and exit terms secure?
No procurement implication on its own.
Operating model
Which teams own the service once it runs?
Supports using AI nowcasts as one input alongside operational models, not a replacement for 0 to 2 hour radar nowcasting.
Publication history
- 2026-10-26NVIDIA · Issue 518 resources
Stable resource ID: nvidia-stormscope-nowcasting-paper-limits-20260124