Permafrost Discovery Gateway Webinar: Transformation of Big Imagery into Arctic Science Ready Products
Speaking: Chandi Witharana, University of Connecticut
Event Type: Webinars and Virtual Events
When: 22 September 2022
Where: Online: 9:00 am AKDT, 1:00 pm EDT
The Permafrost Discovery Gateway hosts a monthly webinar series on the second Thursday of each month at 9:00 am Alaska time, raising topics of interest to the permafrost community. The webinar aims to 1) connect the international science community interested in big data remote sensing of permafrost landscapes, and 2) provide the Permafrost Discovery Gateway development team with end-user stories (by the presenter and webinar participants), such as exploring tools the community needs to create and explore big data.
Commercial satellite sensors of Maxar Technologies have imaged the entire Arctic multiple times at sub- meter resolution over the last few decades. The pan-Arctic Maxar data set (> 2 PB) is freely accessible to the NSF-funded Arctic research community via the Polar Geospatial Center (PGC). High-resolution multispectral images enable transformational opportunities to characterize different permafrost landforms and to monitor micro-topographic transitions/disturbances occurring in the Arctic permafrost landscapes at finer granularity without compromising the geographical extent. Despite the unprecedented opportunities, the image repositories at PGC are largely underutilized and imagery- derived pan-Arctic science products are yet rare. Traditional remote sensing image analysis methods fail to grapple the sheer data volume and semantic complexities of image scenes, thus, limit the scalability of mapping applications. Knowledge discovery through AI, big imagery, high-performance computing (HPC) is just starting to be realized in Arctic permafrost science. We have developed a novel high- performance image analysis framework – Mapping application for Arctic Permafrost Land Environment (MAPLE) that enables the integration of operational-scale GeoAI capabilities into Arctic permafrost modeling. Interoperability across heterogeneous HPC systems, optimal usage of computational resource, and extensibility of mapping workflows are some of the key design goals of MAPLE. We have recently deployed MAPLE across the Arctic tundra to map ice-wedge polygons from thousands of Maxar imagery. Our mapping exercise has produced the first pan-Arctic ice-wedge polygon map, which consists of more than one billion individual ice-wedge polygons.