# Hybrid physics-ML weather climate simulation on ERA5 reanalysis 1979–2019

Source: [Hybrid physics-ML weather climate simulation on ERA5 reanalysis 1979–2019](<https://www.sota2.com/research/sota/hybrid-physics-ml-weather-climate-simulation-on-era5-reanalysis-1979-2019>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-23T10:14:03.390Z

Evaluation of hybrid physics-machine learning models designed for comprehensive weather and climate simulation using the full ERA5 reanalysis dataset spanning 1979 to 2019. Metrics focus on model infrastructure and training efficiency.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Hybrid physics-ML weather climate simulation |
| Dataset (including any stated split) | ERA5 reanalysis 1979–2019 |
| Metric | Model Size (Parameters) |
| Unit | SIZE |
| Direction | Lower is better |
| Results loaded | 1 of 1 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Model Size (Parameters)](<https://www.sota2.com/research/sota/hybrid-physics-ml-weather-climate-simulation-on-era5-reanalysis-1979-2019.md?metric=Model%20Size%20(Parameters)>): Lower is better; unit: SIZE.

- [Training Duration](<https://www.sota2.com/research/sota/hybrid-physics-ml-weather-climate-simulation-on-era5-reanalysis-1979-2019.md?metric=Training%20Duration>): Lower is better; unit: TIME.

## Recorded results

| Method | Model Size (Parameters) | Result date | Report paper |
| --- | --- | --- | --- |
| NeuralGCM | 11.5 | Not recorded | 68fa7c501e209131cea8d7fa |

### NeuralGCM

Model Size (Parameters): 11.5. Result ID: 69c1123147f34ce924a90f2e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7c501e209131cea8d7fa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa47921e209131ce92913f.md>)

Recorded evaluation configuration: {"Grid Resolution":"0.7◦, 1.4◦, 2.8◦ dense grids.","Forecasting Horizon":"≤15 days (360 steps at 1-hour increments)","Input Variables":"9 atmospheric variables + extra features","Output Variables":"9 atmospheric states."}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.