Skip to main content
CTrees Logo
  • Tree Level

US urban tree canopy height data

Estimate tree canopy height for urban areas in the U.S.

Lead scientist

This dataset is available by request. See data for select U.S. cities in a Google Earth Engine app.

Tree canopy in Austin, TX

Canopy Data for US Cities

The Urban Tree Canopy Height dataset from CTrees provides tree canopy height in meters for urban areas across the conterminous United States (CONUS). The dataset is based on National Agriculture Imagery Program (NAIP) images acquired during the agricultural growing season (April–August), using the most recent years available. The dataset is currently available for 100 major urban areas located across CONUS. For other areas, it can be processed on demand.

Derived datasets include the Urban Tree Canopy Cover image, generated by applying a minimum height threshold of 3.2 meters and sharing the same specifications as the tree canopy height dataset, as well as the Urban Tree Canopy Polygons, which contain attributes such as a unique identifier for each polygon, canopy height statistics (median, 5th, and 95th percentiles; meters), planimetric area of the tree canopy polygon (square meters), and a binary flag indicating whether the tree canopy is spatially isolated from neighboring canopies.

CTrees’ approach to producing this dataset represents the frontier of remote sensing and deep learning modelling. Our high-resolution, tree-level datasets have applications in use cases such as utility asset management, urban resilience, and insurance. Based on the availability and quality of additional training data, we offer bespoke modeling training and validation tailored to specific user requirements for geographies, temporal frequency, and accuracy.

Urban tree canopy in the U.S.

Data at a Glance

Units

Meters(m)

Spatial resolution

60 centimeters (cm)

Data type, scale, offset
  • Urban Tree Canopy Height: Float32; Scale: 1; Offset: none
  • Urban Tree Canopy Cover: Binary; Scale: 1; Offset: none
File format
  • Urban Tree Canopy Height: Geotiff
  • Urban Tree Canopy Cover: Geotiff
  • Urban Tree Canopy Polygons: Geopackage

A U-Net model for urban canopy height, validated with LiDAR

input data

Remote sensing inputs (NAIP imagery)

  • Four-band aerial imagery (Red, Green, Blue, Near-Infrared) at 0.6-meter resolution
  • Source: USDA, USGS
  • Orthorectified, radiometrically corrected
  • Spatial reference: UTM (per tile)


Calibration and validation data

  • Airborne LiDAR-derived Canopy Height Models

MODELING APPROACH

The dataset is derived from CTree’s Canopy Height Model (Wagner et al., 2024), a U-Net–based deep learning model trained on nearly one million image patches using aerial LiDAR data from NEON campaigns in California (2018–2023) and USGS LiDAR datasets across 16 additional states, combined with NAIP optical imagery (2020–2023).

The model was trained using a Mean Squared Error (MSE) loss function, with data augmentation applied through random flips, rotations, and brightness variations. Post-processing using a vegetation index was performed to reduce commission errors associated with heterogeneous urban features.

VALIDATION APPROACH AND RESULTS

Validation across 28,337,598 points showed strong agreement (R² = 0.81), a mean absolute error (MAE) of 1.73 m, and a small negative bias (-0.78 m). Canopy height was evaluated by comparing predicted values with randomly selected USGS LiDAR LAS tiles from 2019 and 2024, at a common 1 meter resolution, across 35 urban areas chosen to represent diverse ecoregions and population sizes.

Use cases and applications

Applications of this tree-level dataset include but are not limited to:

  • Aboveground biomass estimation
  • Urban forestry management
  • Urban climate modeling
  • Detection and identification of trees in proximity to utilities or infrastructure
  • Habitat modeling
  • Defensible space analysis

References

Wagner, F.; Roberts, S.; Ritz, A. L.; Carter, G.; Dalagnol, R.; Favrichon, S.; Hirye, M.C.M.; Brandt, M.; Ciais, P.; Saatchi, S. (2024). Sub‐meter tree height mapping of California using aerial images and LiDAR‐informed U‐Net model. Remote Sensing of Environment, v. 305. DOI 10.1016/j.rse.2024.114099 (opens in a new tab).

Team

Data Access

Page last updated Nov 16, 2026

Related Datasets

Stay Connected

Subscribe to our monthly newsletter and be the first to know about new data releases, research, and events.

By submitting your email address, you agree to receive communications from CTrees. You can opt-out at any time.