The article outlines the development of a global map detailing carbon removal (CR) rates in naturally regenerating forests. This involved combining field measurements of aboveground carbon (AGC) stocks with various environmental factors that influence carbon density over time.
Key Steps:
-
Data Assembly: A comprehensive dataset was compiled from national and local forest inventory data across multiple countries, including Australia, Canada, the Netherlands, Sweden, and the U.S. Additionally, the research incorporated literature data from systematic reviews to enhance geographic coverage.
-
Modeling AGC Density: The AGC densities for up to 100-year-old naturally regenerating forests were estimated using a spatial prediction model that integrated a global dataset with 66 environmental variables. A random forest model was employed, and the data were validated to mitigate spatial autocorrelation.
-
CR Curve Calculation: The model generated a global carbon density map for different age classes. An S-shaped growth function was fitted to these data points, refining parameters related to carbon accumulation based on detailed local historical conditions.
- Verification and Comparison: The growth curve estimates were validated against reserved datasets and compared with existing remote sensing-derived curves and IPCC default rates, emphasizing a conservative approach to estimating carbon removal.
Findings:
- The highest carbon densities were observed in tropical forests, while boreal forests exhibited lower values.
- The methodology improved geographic coverage significantly, particularly in underrepresented tropical regions, and ensured consistency across diverse data sources.
Overall, the study provides a sophisticated model for estimating forest carbon dynamics, significantly enhancing the understanding of carbon storage potential across the globe.
