Authors: David Keellings*, University of Alabama
Topics: Climatology and Meteorology, Physical Geography, Temporal GIS
Keywords: heat wave, climate change, spatial metrics
Session Type: Paper
Start / End Time: 8:00 AM / 9:40 AM
Room: Madison A, Marriott, Mezzanine Level
Presentation File: No File Uploaded
Heat waves are occurring more frequently across the globe and are projected to increase in frequency, intensity, and duration under climate change. Yet little is known about the spatiotemporal patterns of heat waves and seemingly simple questions remain unanswered: Are heat waves becoming larger or smaller in size through time? Is heat uniform within a heat wave or is there spatial variability? A new methodology is presented to quantify the spatial and temporal evolution of heat waves using shape metrics. The Parameter elevation Regression on Independent Slopes Model (PRISM) daily temperature and dewpoint temperature (4 km spatial resolution) are used to delineate heat wave areas. Each grid cell exceeding its own high percentile (85th, 90th, 95th) of the entire (1980-2017) summertime (June-September) daily mean (T) and equivalent temperature (TE) are identified. A hierarchical clustering algorithm is used to identify clusters of heat and delimit heat waves based on Euclidean distance between clusters. Heat waves are tracked daily from genesis through lysis based on spatial overlap with previously delineated heat waves up to 3 days prior. A suite of shape metrics is formulated to quantify the daily evolution of heat waves at the surface and aggregated statistics of each of these individual event metrics is calculated. Findings suggest that the number and total area of heat waves has increased across the U.S., but individual heat wave events are generally becoming more fragmented and complex in shape.