EarthDefine's Tree Map is the most detailed and up-to-date tree-cover dataset for the United States. It maps the presence or absence of trees at 60cm resolution, capturing individual trees and small gaps in the canopy, and is refreshed annually to reflect deforestation, urbanisation, land-use change, and natural disasters.
We have leveraged advances in computer vision and AI to solve geospatial classification challenges, deploying a proprietary tree classification model on hundreds of terabytes of summer leaf-on imagery every year to create and update the Tree Map.
The National Land Cover Database (NLCD) tree-canopy layer provides nationwide estimates, but at 30-metre resolution it is not suited to analyses that need fine spatial detail. Tree Map classifies canopy at 2,500 times the spatial resolution and updates annually, capturing even small individual trees and accurately mapping the edges of forest patches.
Tree Map has an overall accuracy of 96.6%, rising to 97.3% in census-defined urban areas. Accuracy was assessed using 48,000 random points (1,000 per state).
EarthDefine has partnered with PlanIT Geo to deliver pre-analysed tree-cover data so you can plan and manage your urban forest with ease. The TreePlotter Canopy solution lets users jump straight into analysis without specialised GIS staff or software, with tree-equity scores, ecosystem benefits, and interactive maps for community engagement.
The Tree Map AI model can map tree canopy from aerial imagery of previous years. For most of the US we have historical imagery back to 2005, so multiple snapshots of tree cover can be derived for change analysis.
High-resolution tree-canopy data supports a wide range of mapping and analytical applications, including:
Tree Map is produced on demand from the latest NAIP acquisition cycle, with seamless coverage across the contiguous United States and historical snapshots available from our imagery archive.
Download the datasheet, or grab a free sample for any of 49 US cities.
A 60cm raster of tree-canopy presence and absence.
96.6% overall accuracy, up to 2,500x finer than the NLCD tree-canopy layer.
Urban tree-canopy (UTC) assessments, wildfire defensible-space analysis, carbon storage and sequestration, and utility vegetation risk.
Annually, as new high-resolution imagery becomes available.
Talk to our team about Tree Map coverage for your region, or download a free sample to evaluate the data yourself.