Land & Carbon Lab
Annual DIST-Alert Drivers
Identifies the potential conversion of natural land in Latin America on an annual basis (2023-2025) by applying the rule-based Land Disturbance Alert Classification System (LDACS) to DIST-ANN alerts.
Caution: The accuracy of the conversion data is dependent on the accuracy of the input datasets, thus any omissions and commissions in the DIST-ANN data are inherent. For example cloudy conditions can mean disturbances are temporarily missed, until clear satellite images are available, meaning cloudy conditions at the end of the year can mean disturbances are identified in the following year.
The method indicates potential conversion, and many alerts remain unclassified where a driver cannot be established from the method. Conversions represent "potential" drivers, and require ground-truthing or high-resolution imagery for definitive confirmation of the change and cause of change.
The conversion class contains several drivers including conversion to agriculture and urban use.
The potential conversion class has a relatively low producer accuracy, meaning not all examples of conversion are captured in the dataset.
Known limitations including confusion between conversion and fire occur. This was particularly prevalent since 2024 was an extreme fire year in the region.
Didn't find what you were looking for?
