Assessing Tropical Cyclones’ Contribution to Precipitation over the Eastern United States and Sensitivity to the Variable-Resolution Domain Extent

Abstract

Tropical cyclones (TCs) can subject an area to heavy precipitation for many hours, or even days, worsening the risk of flooding, which creates dangerous conditions for residents of the U.S. East and Gulf Coasts. To study the representation of TC-related precipitation over the eastern United States in current-generation global climate models, a novel analysis methodology is developed to track TCs and extract their associated precipitation using an estimate of their dynamical outer size. This methodology is applied to three variable-resolution (VR) configurations of the Community Atmosphere Model, version 5 (CAM5), with high-resolution domains over the North Atlantic and one low-resolution conventional configuration, as well as to a combination of reanalysis and observational precipitation data. Metrics and diagnostics such as TC counts, intensities, outer storm sizes, and annual mean total and extreme precipitation are compared between the CAM5 simulations and reanalysis/observations. The high-resolution VR configurations outperform the global low-resolution configuration for all variables in the North Atlantic. Realistic TC intensities are produced by the VR configurations. The total North Atlantic TC counts are lower than observations but better than reanalysis.

Anthropogenic Influence on Hurricane Dorian’s Extreme Rainfall

“…This work makes use of the variable-resolution configuration of the Community Atmosphere Model version 5 (CAM5; Neale et al 2012) with a 28-km nest over the North Atlantic [as in Reed et al (2020)]. CAM5, at grid spacings of 28 km, has been used previously to explore tropical cyclones and rainfall at both climate (e.g., Zarzycki and Jablonowski 2014;Wehner et al 2015;Bacmeister et al 2018;Stansfield et al 2020a) and weather time scales (e.g., Zarzycki and Jablonowski 2015;Wehner et al 2019;Reed et al 2020). Following the methodology of Zarzycki and Jablonowski (2015), short 7-day ensemble hindcasts are initialized both in advance of and after Hurricane Dorian's landfall in the Bahamas at 12-h increments starting at 1200 UTC 30 August and ending at 0000 UTC 4 September for a total of 10 initialization times.…”

“…For both the actual and counterfactual ensembles, the TempestExtremes software package (Ullrich and Zarzycki 2017) is used to detect and track the simulated storms and extract storm-related rainfall using the approach of Stansfield et al (2020b), which specifies storm rainfall to be within an outer radius defined by a 8 m s -1 threshold of the azimuthally averaged azimuthal wind speed. Hurricane Dorian's observed rainfall estimates from NASA's Integrated Multi-satellitE Retrievals for GPM (IMERG; https:// pmm.nasa.gov/data-access/downloads/gpm) are used to calculate the maximum 3-hourly S11 JANUARY 2021 AMERICAN METEOROLOGICAL SOCIETY | rainfall amount, while recognizing that there are substantial uncertainties associated with heavy tropical cyclone rainfall in satellite estimates (e.g., underestimating storm rainfall over land; Chen et al 2013).…”

TempestExtremes v2.1: A Community Framework for Feature Detection, Tracking and Analysis in Large Datasets

“…TE has also been used for tracking storms in aquaplanet simulations (Chavas and Reed, 2019) to better understand how thermal forcing impacts TC genesis and size. Recent work by Stansfield et al (2020) has also leveraged some of the more advanced capabilities in TE to filter fields (e.g., precipitation) in the vicinity of tracked features to evaluate model performance.…”

“…As argued by Schenkel et al (2017), the largest radius outside of the eyewall where the azimuthally-averaged wind speed exceeds 8 m s −1 (r8) tends to be a good measure of the outer size of a TC. In Stansfield et al (2020), TE was used to examine the distribution of r8 among TCs in reanalysis data in ERA5 and a series of runs from the Community Earth System Model (CESM). This paper further compared and contrasted TC-related precipitation within r8 against precipitation within a fixed distance of 500 km.…”

“…For many atmospheric and oceanic features, automated object identification and tracking in large datasets has enabled targeted scientific exploration of feature-specific processes. Software tools for feature tracking, colloquially referred to as "trackers", are valuable for evaluating model performance (Davini and D'Andrea, 2016;Stansfield et al, 2020), understanding upstream process drivers (such as large-scale meteorological patterns; e.g., Grotjahn et al, 2016), and projecting future changes in feature characteristics and climatology (Roberts et al, 2020a). When well-engineered, these automated tools provide a means by which we can analyze the multiple petabytes of climate data now available (and anticipated in the next decade) (Schnase et al, 2016;Hassani et al, 2019).…”