Toward a National Early Warning System for Forest Disturbances Using Remotely Sensed Canopy Phenology
- NASA Stennis Space Center
- ORNL
Imagine a national system with the ability to quickly identify forested areas under attack from insects or disease. Such an early warning system might minimize surprises such as the explosion of caterpillars referred to in the quotation to the left. Moderate resolution (ca. 500m) remote sensing repeated at frequent (ca. weekly) intervals could power such a monitoring system that would respond in near real-time. An ideal warning system would be national in scope, automated, able to improve its prognostic ability with experience, and would provide regular map updates online in familiar and accessible formats. Such a goal is quite ambitious - analyzing vegetation change weekly at a national scale with moderate resolution is a daunting task. The foremost challenge is discerning unusual or unexpected disturbances from the normal backdrop of seasonal and annual changes in vegetation conditions. A historical perspective is needed to define a 'baseline' for expected, normal behavior against which detected changes can be correctly interpreted. It would be necessary to combine temperature, precipitation, soils, and topographic information with the remotely sensed data to discriminate and interpret the changing vegetation conditions on the ground. Conterminous national coverage implies huge data volumes, even at a moderate resolution (250-500m), and likely requires a supercomputing capability. Finally, such a national warning system must carefully balance the rate of successful threat detection with false positives. Since 2005, the USDA Forest Service has partnered with the NASA Stennis Space Center and Oak Ridge National Laboratory to develop methods for monitoring environmental threats, including native insects and diseases, wildfire, invasive pests and pathogens, tornados, hurricanes, and hail. These tools will be instrumental in helping the Forest Service's two Environmental Threat Assessment Centers better meet their Congressional mandate to help track the health of the Nation's forests and rangelands. We envision two scales of forest monitoring: (1) a strategic, satellite-based monitoring of broad regions to identify particular locations where threats are suspected (i.e., early warning), and (2) a fine-scale, tactical tier consisting of airborne overflights and on-the-ground monitoring to check the validity of warnings from the upper tier. The tactical tier is already largely in place within the Forest Service and its State collaborators, consisting of aerial detection surveys (sketch mapping from aircraft), ground surveys, and trapping programs. However, these efforts are expensive and labor-intensive, can be dangerous, and may not provide sufficient broad-area coverage. Far from replacing the tactical tier, the national system will rely on the finer-scale efforts to confirm, validate, and attribute causes of detected forest disturbances. One important objective of the national warning system will be to help direct the focus of the tactical tier, making their efforts more cost efficient and effective.
- Research Organization:
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States). National Center for Computational Sciences (NCCS)
- Sponsoring Organization:
- Work for Others (WFO)
- DOE Contract Number:
- DE-AC05-00OR22725
- OSTI ID:
- 964361
- Journal Information:
- Photogrammetric Engineering & Remote Sensing, Vol. 75, Issue 10
- Country of Publication:
- United States
- Language:
- English
Similar Records
Toward A National Early Warning System for Forest Disturbances Using Remotely Sensed Land Surface Phenology
Improving Rangeland Monitoring and Assessment: Integrating Remote Sensing, GIS, and Unmanned Aerial Vehicle Systems