Kobuk Valley National Park Landscape Change Detection Using Remotely Sensed Data and Geomorphologic Assessments, 20-R8002Printer Friendly Version
Inclusive Dates: 10/17/08 02/17/09
Background - This project was undertaken to strengthen proposals submitted to the National Science Foundation (NSF) and National Aeronautics and Space Administration (NASA) to conduct geomorphologic and atmospheric field and modeling studies of the Great Kobuk Sand Dunes, Kobuk Valley National Park, Alaska. This aeolian dune system and its thermokarst environs at 67° north latitude are bellwethers for subarctic climate change; the dunes also serve as terrestrial geophysical analogs to Martian dunes potentially affected by movement-arresting permafrost. Scientific research in this park has lulled since the 1980s, so very little recent information about park resources was available at the time this project commenced.
Approach - To close this knowledge gap and support the needs of potential clients, this four-month project was funded to develop staff capabilities in the detection of change in remote subarctic landscapes. SwRI researchers developed a new method for processing multispectral imagery to estimate rates of landscape change. The multispectral data displacement analysis (MDDA) method builds upon the foundation of the co-registration of optically sensed images and correlation (COSI-Corr) technique. Using this method, staff estimated spatially distributed rates of sand dune migration at the Great Kobuk Sand Dunes.
Accomplishments - SwRI is the first to integrate multispectral data, the COSI-Corr technique, orientation filtering, and orientation projection to reliably measure slowly migrating sand dunes, landscape and vegetation changes, and other processes that affect optical data patterns. Adding the MDDA technology to SwRI's existing capabilities has expanded the range of ground movements researchers can detect using satellite data to those occurring horizontally. The Great Kobuk Sand Dunes migration rates are much lower than those of low-latitude aeolian dunes; this subarctic dune field may be arrested by the presence of permafrost deep within its core. Satellite imagery processed to quantify changes in thermokarst lake surface water area indicates a general decrease in area between 1985 and 2008; this decrease may result from a warming climate. Near infrared imagery processed to quantify changes in land surface temperature during spring thaw revealed relatively warm temperatures along a stream bisecting the sand dunes; warm stream temperatures may indicate areas of groundwater discharge.