climate-analytics-lab / climate-analytics-lab/ClimateBench2

Tier III: Perfect-model experiments (CESM2, MPI-ESM, GISS)

Open
#95 0 comments 0 reactions 0 assignees View on GitHub
metric tier-iii
Dominant language
Shell
Stars
1
Forks
1
PR merge metrics
No merged PRs in 30d

Description

## Description
Test whether data-driven models can predict the future climate of existing ESMs when trained on their present-day output.

## Specification
- Train on ESM historical output (pre-2015), predict SSP2-4.5 future
- **CESM2** (ECS ≈ 5.2 K): required for ML models — hardest test
- **MPI-ESM** (ECS ≈ 3.63 K): extended
- **GISS ModelE2** (ECS ≈ 2.31 K): extended
- Evaluate using same Tier II metrics
- Models used for training must have been initialized before the test period (no data leakage)

## Required simulations
Perfect model × 3 ESMs (~85 years each)

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.