# swmitgcm

**Repository Path**: arcfrontier/swmitgcm

## Basic Information

- **Project Name**: swmitgcm
- **Description**: Arctic high resolution simulations
- **Primary Language**: Unknown
- **License**: MIT
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-03-11
- **Last Updated**: 2026-04-30

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

## 🚀 High-Resolution pan-Arctic Ocean Modeling
This open-source code enables ultra-high-resolution (~500m) simulations of a sea ice-ocean coupled model across the pan-Arctic region, where the model solves the problem with more than 15 Billion grids.
## 🚀 Code Refactoring for Domestic Supercomputers
We have comprehensively refactored the original MITgcm code, including dynamic and I/O routines, to optimize performance on a heterogeneous many-core supercomputer built with domestic technology.
## 🚀 Example Output
The figure below illustrates simulated ice-ocean interactions in the Arctic, highlighting small-scale processes resolved by the model. The tearing of sea ice resembles the process of a black hole devouring a planet in the universe.(海冰撕裂类似宇宙中黑洞吞噬行星)

![The tearing of sea ice resembles the process of a black hole devouring a planet in the universe. (海冰撕裂类似宇宙中黑洞吞噬行星)](doc/sic.2018.8.15.JPG)

## 🚀 MOSAiC Expedition Alignment
Simulations are aligned with the MOSAiC expedition timeframe to evaluate and validate physical processes against observational data, bridging gaps between model physics and real-world measurements.
## 🚀 Visualization of Results
A video showcasing the extraordinary Arctic simulation results is provided here: [Video of Arctic ultra-high-res (500m) simulation](https://gitee.com/arcfrontier/swmitgcm/blob/master/doc/simulation.500m.mp4)
![Snapshot](doc/vlcsnap.png)

## 🚀 Data Accessibility
As the simulation data exceeds several petabytes (PB), we are actively exploring solutions to ensure broad public access, in a way that could be fetched using tools like jupyter-notebook. It however demands large supports from Cloud infrastructure and technical work. But it will be coming soon, we have tested it already in our local network.

## 💡 Code Status
Note that this repository does not reflect the latest ongoing developments.