eDaRT: the Ecosystem Disturbance and Recovery Tracking system prototype supporting ecosystem management in California

Alexander Koltunov, University of California, Davis, akoltunov@ucdavis.edu (Presenter)
Carlos Ramirez, USDA Forest Service, carlosramirez@fs.fed.us
Susan L. Ustin, University of California Davis, slustin@ucdavis.edu

In recent years, several Landsat-based land cover/use change monitoring algorithms have emerged, partly addressing the rapidly growing demand for various types of land cover and habitat change information at Landsat spatial resolution and various temporal scales. The research behind earlier algorithms based on annual comparisons is paving the way to more advanced methods, taking advantage of the entire time series and method ensembles.

The USDA Forest Service (USFS) Pacific Southwest Region has increasingly used outputs from the Ecosystem Disturbance and Recovery Tracker (eDaRT) system for disturbance event detection, tracking habitat changes, cumulative effect assessment, and higher-level derived product generation. The algorithm is based on a novel enhancement of a multitemporal anomaly detection method called the Dynamic Detection Model. It combines analyses across temporal and spatial domains and compensates the signal for phenological stage differences. The eDaRT system is automatically trained for a monitored region (e.g. a Landsat path-row tile), subject to user-defined baseline period guidelines. In the detection mode, eDaRT version 1.2.299 processes all available images in a fully automated way, including spectro-temporal cloud/snow mapping, and cumulative disturbance/regeneration effect detection. Furthermore, the disturbance event timing and relative magnitude statistics are calculated based on the time series of single-date detections. Other features of the system include: detects damage to understory vegetation, shrubland and grassland ecosystems, with stability after long data gaps.

With ongoing developments, eDaRT is expanding to become a regional platform for serving field manager community with customized user-level products and assessments that can be rapidly generated from its basic functionality. The eDaRT developers and user community are looking forward to data from Sentinel-2 mission and the LCLUC program sponsored research in Landsat-Sentinel blended products, which will maximize the contribution of NASA Earth observations to operational adaptive ecosystem management.

Presentation Type:  Poster

Session:  Theme 1: Tracking habitat change through new integrative approaches and products   (Mon 1:30 PM)

Associated Project(s): 

• Related Activity: Related Activity or Previously Funded CC&E Activity not listed ...details

Poster Location ID: 30