Assessing aboveground tropical forest biomass using Google Earth canopy images

Area of Science:

• Forestry
• Remote Sensing
• Climate Change Science

Background:

• National forest resource assessment is vital for climate change mitigation strategies like Reducing Emissions from Deforestation and Forest Degradation (REDD).
• Tropical regions face challenges in biomass estimation due to complex forest structures, limited access to high-resolution imagery, and insufficient inventory data.

Purpose of the Study:

• To evaluate the Fourier-based FOTO method for predicting tropical forest aboveground biomass using Google Earth imagery.
• To compare the effectiveness of free Google Earth images against commercial IKONOS data for biomass assessment.
• To identify research needs for cost-effective, accurate regional biomass estimation.

Main Methods:

• Applied the FOTO method to analyze canopy texture from 1436 Google Earth images (125x125m) in India's Western Ghats.
• Correlated texture gradients with forest structure parameters using 15 1-ha ground plots.
• Benchmarked Fourier spectra against simulated canopy scenes and a 3-D light interception model.

Main Results:

• Both Google Earth and IKONOS imagery, analyzed with the FOTO method, showed strong correlations with field-observed stand structure (tree density, basal area, biomass).
• The texture-biomass relationship using Google Earth data exhibited a low average relative error (15%) and no saturation at high biomass values.
• Generated the first reliable map of tropical forest aboveground biomass derived from free Google Earth imagery.

Conclusions:

• The FOTO method, when applied to readily available Google Earth imagery, presents a powerful and cost-effective tool for tropical forest biomass estimation.
• This approach offers significant potential for supporting REDD initiatives by enabling accurate, large-scale forest carbon stock assessments.
• Further research should focus on refining affordable methods for regional biomass mapping.