A high-resolution inland surface water body dataset for the tundra and boreal forests of North America

Abstract

Inland surface waters are abundant in the tundra and boreal forests of North America, essential to environments and human societies but vulnerable to climate changes. These high-latitude water bodies differ greatly in their morphological and topological characteristics related to the formation, type, and vulnerability. In this paper, we present a water body dataset for the North American high latitudes (WBD-NAHL). Nearly 6.5 million water bodies were identified, with approximately 6 million (∼90 %) of them smaller than 0.1 km2. The dataset provides area and morphological attributes for every water body. During this study, we developed an automated approach for detecting surface water extent and identifying water bodies in the 10 m resolution Sentinel-2 multispectral satellite data to enhance the capability of delineating small water bodies and their morphological attributes. The approach was applied to the Sentinel-2 data acquired in 2019 to produce the water body dataset for the entire tundra and boreal forests in North America. The dataset provided a more complete representation of the region than existing regional datasets for North America, e.g., Permafrost Region Pond and Lake (PeRL). The total accuracy of the detected water extent by the WBD-NAHL dataset was 96.36 % through comparison to interpreted data for locations randomly sampled across the region. Compared to the 30 m or coarser-resolution water datasets, e.g., JRC GSW yearly water history, HydroLakes, and Global Lakes and Wetlands Database (GLWD), the WBD-NAHL provided an improved ability on delineating water bodies and reported higher accuracies in the size, number, and perimeter attributes of water body by comparing to PeRL and interpreted regional dataset. This dataset is available from the National Tibetan Plateau/Third Pole Environment Data Center (TPDC; http://data.tpdc.ac.cn, last access: 6 June 2022): https://doi.org/10.11888/Hydro.tpdc.271021 (Feng and Sui, 2020).