Storm Analysis

Storm analyses form the foundational data sets needed for virtually any extreme precipitation or hydrologic risk study, including Probable Maximum Precipitation (PMP) and Precipitation Frequency (PF) studies. Systematic analysis of storm precipitation was routinely done by the U.S. Geological Survey (USGS), National Weather Bureau (now National Weather Service), U.S. Army Corps of Engineers and Bureau of Reclamation through the 1970s before the practice was discontinued. Since then occasional storms have been analyzed by the National Weather Service or USGS, but the majority of storms since 1972 have been analyzed with MetStat’s Storm Precipitation Analysis System (SPAS) and the next generation of storm analysis software, MetStorm.

Comparison of US Army Corps of Engineers hand analysis of the precipitation from the Sept. 1921 hurricane and MetStorm.

Isohyetal analysis of the 8-10 Sept 1921 Thrall, TX storm, as originally conducted by the U.S. Weather Bureau in 1946.

Storm Analysis Past and Present

Detailed storm precipitation analyses have been a dominant strength of MetStat since 1994, when the first storm analyses were conducted using Geographic Information Systems (GIS), then in 2002 with SPAS, and more recently with MetStorm. Storm analyses provide necessary input data for essentially every type of extreme precipitation study, including Probable Maximum Precipitation (PMP) and Precipitation Frequency (PF) studies. MetStat’s storm analyses have also been used in forensic investigations, hydrologic model calibrations/validations and storm severity assessments for the media. Given the importance of storm analysis, MetStat has invested in and pioneered tools for integrating radar, satellite, precipitation gauge data, climatology and expert judgement into high-resolution storm reconstructions. Our state-of-the-science methods combine the strengths of data from MetStat, our strategic partners and government agencies:

  • Weather Decision Technologies, Inc – Quality-controlled NEXRAD radar data
  • NOAA – Satellite-estimate precipitation data
  • Synoptic Data Corp – Quality-controlled precipitation gauge data
  • The PRISM Climate Group -Precipitation climatologies
  • MetStat – Integrating algorithms, software and expert storm analysts
Comparison of US Army Corps of Engineers hand analysis of the precipitation from the Sept. 1921 hurricane and MetStorm.

Storm total precipitation map for the MetStorm reanalysis of the 1921 Thrall, TX storm, with the core precipitation region based on the isohyetal map at right.

MetStorm: the Next Generation of Storm Analysis

MetStat originally conceptualized, developed, and implemented the Storm Precipitation Analysis System (SPAS) software in 2002. SPAS has been used/operated by MetStat and Applied Weather Associates to complete hundreds of storm analyses for extreme precipitation studies. In 2014-2016, MetStat independently developed a new and improved storm analysis software, MetStorm®, providing the next generation of software analysis. MetStorm is not a stand-alone software system, but a system used by expert storm analysts to produce consistent, quality-controlled storm analyses. MetStorm is designed to operate with a wide range of input data which makes it a flexible tool for producing gridded precipitation associated with very old storms (pre-1900) as well as storms that recently occurred. MetStorm is equipped to include bucket survey information that was collected in the Part I storm analyses by the Weather Bureau or U.S. Army Corps of Engineers (USACE) if available, and more recent datasets, including satellite information and Dual-Pol radar measurements.

To instill consistency among previous storm analyses conducted by others, MetStorm’s logic is built on similar techniques as those used in SPAS, by USACE and others, but includes numerous improvements. In comparison to the Storm Precipitation Analysis System (SPAS), MetStorm uses a more efficient data structure, integrates more comprehensive data sources, and includes refined logic for addressing shortcomings of SPAS. For example, MetStorm utilizes quality-controlled 1-hour precipitation gauge data, where each hourly precipitation measurement is accompanied by a quality control flag. Both are editable by the storm analyst, thereby providing powerful flexibility in using/excluding precipitation gauge data. Also, MetStorm reduces the high-degree of uncertainty with observation times of daily precipitation measurements by disaggregating the total daily precipitation into hourly precipitation based on nearby hourly gauges and/or radar data (if available). MetStorm also produces objective measures of uncertainty and a complete Average Recurrence Interval (ARI) analysis. Although the processing within MetStorm is completely different from SPAS, a comparison of pre-radar era storms indicates very similar results. However, MetStorm is markedly different and more far more advanced than SPAS for storms where radar data is available.

From the MetStorm software, each storm analysis results in a depth-area-duration (DAD) table and plot, uncertainty analysis, mass curves, a total storm isohyetal map, a color-shaded storm total precipitation map, and ARI analysis, and most importantly high-resolution gridded hourly (or 5-minute) precipitation. A storm analyst provides an interpretation of these plots in a storm report. MetStat has analyzed many large storm events; these are stored in an internal database for future studies to leverage.

MetStorm output has been trusted by the U.S. Army Corps of Engineers, the Flood Control District of Maricopa County, Santa Clara Valley Water District, BC Hydro and Power Authority, and numerous other clients.

Animation of rainfall accumulation at 5-minute intervals as analyzed by MetStorm in the Phoenix metro area.