Toward a Comprehensive Detection Algorithm for Atmospheric Blocking Events

Abstract

Atmospheric blocking is a phenomenon in which upper-level atmospheric flow becomes quasi-stationary for an extended period (i.e., several days to weeks), leading to some of the most impactful large-scale extreme weather events in the mid-latitudes including droughts, heatwaves, flooding, and cold air outbreaks. Although blocking has long been recognized, the physical mechanisms responsible for blocking formation, maintenance, and decay remain poorly understood, posing a challenge for improving long-range weather forecasts. A key obstacle in this effort is the ambiguity in how blocked flow patterns should be defined, resulting in the absence of a unified objective blocking detection algorithm. Using ERA5 reanalysis data, this thesis develops a novel objective algorithm to comprehensively detect and characterize blocking events in the Northern Hemisphere. Building upon established blocking identification metrics, the algorithm incorporates three dimensional tracking to approximate the dynamic spatiotemporal evolution of blocked flow patterns. It also accounts for blocking diversity by classifying events into ridge and dipole flow patterns. Applying this algorithm, a data set of Northern Hemisphere blocking events from 1950–2024 is established. The resulting climatology and trends of blocking characteristics are broadly consistent with previous studies that use alternative detection methods. The relationships between blocking and latent heating, as well as the influence of the El Niño Southern Oscillation and the North Atlantic Oscillation, are examined. These relationships exhibit notable differences from previous findings. Overall, the results suggest that a comprehensive blocking detection approach offers meaningful improvement over more simplified algorithms.