Event-Based Climatology of Tropical Cyclone Rainfall in Houston, Texas and Miami, Florida

Abstract

Tropical cyclone (TC) rainfall amounts are compared from 1950–2017 for Houston, Texas and Miami, Florida to estimate the risk of TC rain in both cities. Following the wake of Hurricanes Harvey and Irma in 2017, concern has risen over the future of raininess in these locations. Per-event rainfall amounts are aggregated using tracks taken from HURDAT, time-of-rain gathered from National Weather Service daily weather maps, and rainfall totals taken from airport monitoring stations. Risk analysis tools include descriptive statistics, time series, and return frequencies for Houston and Miami, and spatially interpolated surfaces for Hurricanes Harvey and Irma. The season duration is longer in Miami than in Houston. The uppermost rainfall events in the distribution for Houston show a significant increase through time, suggesting the most intense rainfall events are becoming worse for Houston. The expected return frequency for a Harvey-like event (940 mm) in Houston is every 230 years, on average, and the 90th percentile rain of 286 mm is expected once every 17 years (11–29; 90% significance). The expected return frequency for an Irene-like event (261 mm—maximum for location) in Miami is every 173 years, on average, and the 90th percentile rain of 167 mm is expected once every 11 years (7–17; 90% significance). Results show a substantial difference between Houston and Miami TC rainfall climatologies similar to the differences of Hurricanes Harvey and Irma. Though emergency management must be tailored for each TC, management for inland TC rainfall may be more applicable in Houston than in Miami.

Keywords

• hurricane
• daily weather maps
• extreme value theory
• gulf of Mexico
• risk
• harvey
• irma
• 2017