Model vs. Observation Matchups

Overview

This pillar analyzes the divergence between leading climate model projections (like CMIP6) and actual observed temperature data. It helps determine if models are running hot or cold, providing an edge for markets on climate milestones.

What It Does

The analysis systematically tracks the historical performance of the CMIP6 climate model ensemble against the ERA5 reanalysis dataset, a high-quality record of observed temperatures. It calculates the difference between projected and actual warming over various time scales. This identifies persistent biases that can inform predictions about future performance.

Why It Matters

Climate models are the basis for long-term climate policy and market expectations. By quantifying how well these models match reality, this pillar provides a data-driven edge to position against consensus forecasts if models show a consistent positive or negative bias.

How It Works

First, we aggregate the multi-model mean temperature projections from the CMIP6 archive. Second, we gather the corresponding observed global mean temperatures from the Copernicus ERA5 dataset. Finally, we calculate the rolling divergence between the two datasets to identify and track trends in model accuracy over time.

Methodology

The core calculation is a rolling 10-year mean difference between the CMIP6 multi-model ensemble mean for global surface temperature anomalies and the ERA5 reanalysis dataset. The anomaly baseline is standardized to 1981-2010 for accurate comparison. The pillar also tracks shifts in the consensus range for Equilibrium Climate Sensitivity (ECS) from major scientific reports.

Edge & Advantage

This pillar provides a quantitative edge by identifying systematic model drift before it becomes common knowledge, allowing you to position against market prices based on potentially biased model outputs.

Key Indicators

• Model vs. Observation Divergence

  high

  The rolling difference between the climate model ensemble average and observed temperature data from ERA5.

• ECS Estimate Trend

  medium

  Tracks the consensus range for Equilibrium Climate Sensitivity, a key variable influencing long-term model projections.

• Ensemble Spread

  low

  The range of outcomes predicted by different models within the CMIP6 ensemble, which indicates forecast uncertainty.

Data Sources

• Copernicus Climate Change Service (C3S)

  Provides the ERA5 global reanalysis dataset for observed atmospheric and surface conditions.

• Coupled Model Intercomparison Project (CMIP)

  Provides the multi-model ensemble projections from global climate modeling centers, specifically the CMIP6 dataset.

• IPCC Assessment Reports

  Publishes reports summarizing the state of climate science, including consensus ECS estimates.

Example Questions This Pillar Answers

• → Will the 5-year global mean temperature anomaly first exceed 1.5°C before 2030?
• → Will the 2030-2040 average global temperature be above the CMIP6 SSP2-4.5 median projection?
• → Which year will be the first to have a global mean temperature anomaly of +1.7°C or higher?

Tags