A sample design for globally consistent biomass estimation using lidar data from the Geoscience Laser Altimeter...

Area of Science:

• Forestry
• Remote Sensing
• Geospatial Analysis

Background:

• Geosciences Laser Altimeter System (GLAS) lidar data (2002-2008) offers potential for global forest biomass inventory.
• GLAS lidar shots correlate strongly with aboveground forest biomass.
• GLAS data's non-random spatial pattern hinders its use in global forest sampling.

Purpose of the Study:

• To propose a method for selecting a subset of GLAS data that can be treated as a simple random sample.
• To enable model-based biomass estimation using GLAS data for global forest inventory.

Main Methods:

• Developed a method to identify a subset of GLAS data with uniform spatial distribution and locally arbitrary positioning.
• Applied model-based estimation using GLAS data and compared it with the US national forest inventory (NFI).

Main Results:

• Demonstrated model-based estimation using GLAS data in California, yielding a biomass estimate of 211 Mg/hectare.
• The GLAS-based estimate had a 1.4% standard error, comparable to the NFI estimate.
• The cost of the GLAS-based estimate was negligible compared to the NFI estimate.

Conclusions:

• The proposed method provides a globally extensible approach for generating a simple random sample from GLAS data.
• This enables consistent and comparable international biomass estimates through standardized methods and a coherent sample frame.
• Facilitates global forest inventory activities using GLAS data.