Estimates of Extreme Precipitation Frequency Derived from Spatially Dense Rain Gauge Observations: A Case Study of Two Urban Areas in the Colorado Front Range Region. Issue 6 (2nd November 2017)

• Record Type: Journal Article
• Authors: Mattingly, Kyle S.; Seymour, Lynne; Miller, Paul W.
• Is Part Of: Annals of the American Association of Geographers. Volume 107:Issue 6(2017)
• Page Start: 1499
• Page End: 1518
• Publication Date: 2017-11-02
• Subjects: city planning -- extreme value statistics -- flooding -- precipitation design values -- precipitation return periods; 关键词: 城市规划 -- 极端值统计 -- 洪泛 -- 降水设计值 -- 降水重现期。; planificación urbana -- estadísticas de valores extremos -- inundación -- valores previstos de precipitación -- períodos de información de la precipitación; Geography -- Periodicals; Environmental sciences -- Periodicals; Geography; Electronic journals; Periodicals; 550
• DOI: 10.1080/24694452.2017.1309961
• Languages: English
• ISSNs: 2469-4452

Abstract

Abstract : Precipitation design values, which describe precipitation extremes expected within a specified time period, provide critical guidance for public policy and the design of hydrologic infrastructure. Unfortunately, conventional design value calculations are limited by a major assumption: They treat precipitation as a point-based phenomenon, measured at spatially isolated gauges. We argue that because precipitation occurs over areas, design value calculations should be based on an areal conception of precipitation. Using spatially dense Community Collaborative Rain, Hail and Snow (CoCoRaHS) Network data in two Colorado cities (Fort Collins and Boulder), we develop a "hyperlocal" design value calculation in which nearby observations within a representative precipitation region (RPR) "compete" to capture a single twenty-four-hour maximum. We find that design values for shorter return periods (one to fifty years) derived from just ten years of hyperlocal data, which we express as probabilistic distributions rather than single values, typically exceed those calculated using single-point records of 100 years or more. Hyperlocal design values for longer return periods (more than fifty years) are generally smaller than those calculated from the single-point data due to the temporally limited CoCoRaHS record. We also find that the dependence of design values on RPR size contrasts between Fort Collins and Boulder, as design values grow markedly larger as the RPR size increases in Fort Collins but not in Boulder. We attribute this behavior to the Boulder Global Historical Climatology Network station's location in a topographically favored area for precipitation and propose that future studies evaluate the hyperlocal design value method in a variety of geographic settings.