Queensland Woody Vegetation Age Since Disturbance from Satellite Time Series

A source of an estimated age in years since the last significant disturbance for woody vegetation in Queensland, derived from Landsat time series starting in 1988 and incorporating Sentinel-2 and Earth-i imagery. The method uses a sequential Conditional Random Fields classifier and heuristic rules to detect and track regrowing vegetation, combined with historic clearing data. The product represents different spatial scales, including 30-meter, 10-meter, and 1-meter resolutions.

Use Cases

• Train a model to predict woody vegetation age since disturbance using Landsat time series data and the Conditional Random Fields classifier method.
• Analyze the relationship between detected regrowth events and historic clearing data from the Statewide Land and Trees Study (SLATS).
• Map the spatial distribution of disturbance types (e.g., clearing, fire, flood) based on the estimated age since last disturbance for each woody pixel.
• Validate the regrowth detection heuristics by comparing the estimated start and end dates of regrowth events across different satellite resolutions (Landsat 30m, Sentinel-2 10m, Earth-i 1m).

Strengths

• Derived from a multi-decadal Landsat time series starting in 1988, providing a long-term perspective on vegetation change.
• Integrates data from multiple satellite sources at different resolutions (Landsat 30m, Sentinel-2 10m, Earth-i 1m).
• Combines automated regrowth detection with authoritative historic clearing data from the Statewide Land and Trees Study (SLATS).

Limitations

• The estimated age since disturbance does not always represent the true ecosystem age, as not all recorded disturbances result in complete woody vegetation loss.
• The product is a preliminary estimate, and its accuracy may vary across different disturbance types and landscape conditions.
• The dataset's specific spatial and temporal resolution details for the final age product are not explicitly quantified in the description.

Provenance

Source

Terrestrial Ecosystem Research Network's Data Discovery

Collection Method

Uses a sequential Conditional Random Fields classifier applied to Landsat time series, heuristic rules for regrowth detection, and integration with SLATS historic clearing data.

Time Range

Landsat time series starting from 1988.

Freshness

Last updated on March 20, 2026.

Geography

Queensland, Australia.