A new satellite-based methodology for continental-scale disturbance detection.

Area of Science:

• Earth and Environmental Sciences
• Remote Sensing
• Ecology

Background:

• Major uncertainties in the global carbon cycle stem from the timing, location, and magnitude of disturbance events.
• Continental-scale data on disturbance location, extent, and duration are crucial for assessing ecosystem impacts of land cover changes.

Purpose of the Study:

• To develop an automated, economical, and systematic disturbance detection index for global application.
• To utilize Moderate Resolution Imaging Spectroradiometer (MODIS)/Aqua Land Surface Temperature (LST) and Terra/MODIS Enhanced Vegetation Index (EVI) data for disturbance detection.

Main Methods:

• Developed an algorithm based on the radiometric relationship between LST and EVI.
• Employed annual maximum composite LST data to identify changes in land-surface energy partitioning.
• Validated the algorithm using historical wildfire events and examined its sensitivity to irrigation and precipitation variability.

Main Results:

• The algorithm accurately detects the location and spatial extent of wildfires.
• It is sensitive to the recovery process of disturbed landscapes and irrigation.
• Demonstrated a close association with independently confirmed wildfire events.

Conclusions:

• The developed algorithm provides a precise and sensitive tool for detecting land disturbances globally.
• The method aids in understanding ecosystem impacts and improving global carbon cycle models.
• Further refinement with multiyear datasets is beneficial for areas with high precipitation variability.