Aboveground Biomass Maps for Howland Forest from LiDAR and Modeling

Howland Research Forest in central Maine has aboveground biomass estimates at 10-meter spatial resolution for 2012, 2015, 2017, 2021, and 2023. The dataset was created by ORNL_CLOUD using a random forest model calibrated with in-situ plot measurements and simulated biomass from the LANDIS-II model, applied to airborne LiDAR data from USGS 3DEP and NASA G-LiHT projects. Data are provided as cloud-optimized GeoTIFF files.

Use Cases

• Calibrating and validating regional or global biomass models using high-resolution (10-m) reference maps.
• Analyzing temporal changes in forest carbon stocks across five distinct years (2012-2023).
• Studying the relationship between LiDAR-derived metrics and ground-measured biomass from 50 fixed-area plots.
• Assessing forest structure in a transitional ecosystem dominated by spruce, hemlock, and northern hardwoods.
• Supporting the LANDIS-II forest landscape model with empirical data for simulation of ecosystem dynamics.

Strengths

• Model calibration uses 50 fixed-area ground plots with a 7.98m radius, collected during 2021, 2023, and 2024.
• Provides wall-to-wall coverage at a high 10-meter spatial resolution for the entire research forest.
• Integrates multiple LiDAR sources, including leaf-off (USGS 3DEP) and leaf-on (NASA G-LiHT) data, across a 12-year time range.

Limitations

• Exact file size, row count, and specific column names are not provided by any source.
• Conflicting 'last updated' dates exist between platforms (2026-03 13 vs. 2023-12-31).
• License is specified as 'other-license-specified' but the exact terms are not detailed.

Provenance

Source

ORNL_CLOUD

Collection Method

Random forest model applied to airborne LiDAR data, calibrated with in-situ plot measurements and LANDIS-II simulated biomass.

Time Range

2012 to 2023

Freshness

2026-03-13 20:11:19.033411

Geography

Howland Research Forest, central Maine, USA