Northwestern Arctic Alaska surface water area vector files, Kotzebue study area, 2002-2019

Abstract

{Abstract:[Arctic landscapes are in a state of transition due to changes in climate occurring during both the summer and winter seasons. Scattered observations indicate that beavers (Castor canadensis) have moved from the forest into tundra areas during the last 20 years, likely in response to broader physical and ecosystem changes occurring in Arctic and Boreal regions. The implications of beaver inhabitation in the Arctic and Boreal are unique relative to other ecosystems due to the presence of permafrost and its vulnerability associated with beaver dams and inundation. Our study specifically examines the role of beavers in controlling surface water dynamics and related thermokarst development in low Arctic tundra regions. We mapped the number of beaver dams visible in sub-meter resolution satellite images acquired between 2002 and 2019 for a 100 square kilometer study area (12 years of imagery) near Kotzebue, Alaska and a 430 square kilometer study area (3 years of imagery) encompassing the entire northern Baldwin Peninsula, Alaska. We show that during the last two decades beaver-driven ecosystem engineering is responsible for the majority of surface water area changes and inferred thermokarst development in the study area. This has implications for interpreting surface water area changes and thermokarst dynamics in other Arctic and Boreal regions that may also result from beaver dam building activities. \n This geospatial dataset provides polygon vector files representing surface water area in a 100 square kilometer study area located near Kotzebue, Alaska. Surface water area maps were created using sub-meter resolution satellite imagery for the years 2002, 2007-2014, and 2017-2019. Image selection focused on cloud-free, ice-free, and calm surface water conditions with images being acquired between late-June and mid-August in a given year. All images were resampled to a spatial resolution of 70 centimeter to match the lowest resolution image in the time series prior to analysis. Within year image dates range from 25 June to 22 August with the average date of image acquisition being 17 July (table 1). Object-based image analysis was conducted in eCognition Essentials 1.3.]}