Time SeriesStatistics
Atmospheric CO2 Forecasting
Time series analysis of Keeling Curve data with ARIMA/SARIMA modeling and 26-year forecast validation.
The Problem
Context & Challenge
Climate policy depends on accurate CO2 projections. The challenge was building forecasting models from historical Keeling Curve data and validating their long-term accuracy against decades of observed measurements.
The Approach
Architecture & Implementation
Analyzed 40 years of Mauna Loa CO2 data (1958-1997) using ARIMA/SARIMA models with seasonal decomposition and ADF stationarity testing. Validated forecasts against 26 years of actual observations (1998-2024) to quantify model drift.
The Results
Impact & Metrics
Models predicted 420 ppm threshold by 2030, but actual levels crossed 420 ppm in 2022—8 years early. Demonstrated importance of periodic model retraining as acceleration patterns change.
Key Result
Validated model against 26 years of observed data; actual 420 ppm crossed 8 years early
Technologies & Methods
RARIMASARIMATime Series AnalysisADF TestingForecast Validationggplot2