PIKART: A Comprehensive Global Catalog of Atmospheric Rivers

Catalogs of atmospheric rivers (ARs) are vital resources to advance AR science. Many available AR catalogs have regional extent, although only global AR catalogs can record large-scale heterogeneities in AR transport. However, identification of ARs at a global scale presents substantial challenges caused by regionally and temporally varying atmospheric conditions, for example, (high) low atmospheric moisture content in (low-) high-latitudes. Here, we introduce the PIK Atmospheric River Trajectories (PIKART) catalog, a global and comprehensive compilation of AR activity covering 84 years (1940–2023) at a high spatiotemporal resolution of 0.5° and 6 hr. Building on a previously developed image processing technique, PIKART identifies ARs in an unsupervised fashion that exploits characteristics of endogenous synoptic-scale IVT variability. It tracks ARs prioritizing large and strong features, allowing for physically sound temporal gaps which ultimately improves the representation of long-lived ARs. PIKART identifies a considerable number of high-latitude ARs, matching current best estimates on polar AR frequencies while still maintaining a physically sound frequency of tropical ARs. It furthermore substantially extends the scope of previous catalogs by providing secondary AR properties such as a novel index of inland penetration, land-intersecting locations, and AR ranks. Available as a compilation of AR conditions and AR trajectories, PIKART facilitates the study of ARs from both the Eulerian and Lagrangian perspective. As a first overview of the catalog's scope, we use PIKART to reveal (a) additional hotspots of AR activity, particularly at the poles and in the tropics, (b) exposure to considerable AR impacts in less-studied continents (e.g., South and East Asia, Antarctica and Oceania), (c) inland penetration of ARs into less-studied regions (e.g., Northwest Africa), and (d) a significant intensification of AR-related moisture transport. The PIKART catalog constitutes a valuable resource for future studies in AR science.