Severe weather events (high winds, icing, and lightning strikes) that threaten an electric utility's above-ground power grid are among the most difficult to predict and assess because of their highly time-dependent geospatial distribution. However, recent advances in meteorological sensor systems, mesoscale numerical forecast models, communications technology, and geographical information systems (GIS) have enabled regional severe weather monitoring, forecasting and assessment at unprecedented spatial and temporal resolutions.

Following several major power outages in the St. Louis metropolitan area in 2006, the St. Louis-based utility AmerenMissouri made significant changes to how they responded to high-impact weather events. To provide up-to-the-minute highly detailed weather forecasts designed to support time-critical decisions, the Saint Louis University Department of Earth and Atmospheric Sciences and AmerenMissouri partnered to create QuantumWeather©. Forecasts provided by the QuantumWeather© team have been used by Ameren Missouri to reduce total power outage duration, preposition restoration supplies and crews, mitigate the risk of transporting resources through inclement weather, and better manage personnel by deciding whether to keep restoration crews working past normal shift hours. Thus, the benefit of QuantumWeather™ is realized both in better service to electric consumers and financial savings to the utility company.

To create the required neighborhood-scale short-term forecasts and diagnoses, QuantumWeather© has created a mesonet composed of 75 surface sensor suites across and surrounding AmerenMissouri's service territory (roughly the Eastern half of Missouri). Each surface sensor reports temperature, relative humidity, pressure, wind direction, wind speed, wind gust, rainfall and precipitation rate every minute; data from each solar-powered sensor is communicated wirelessly to a local receiver and then sent over the AmerenMissouri SCADA network to a central repository where it is quality controlled in real-time. 22 surface stations have been installed in the immediate St. Louis metropolitan area; creating a grid with average station spacing of 5-10km. These closely spaced stations allow optimal coverage of ongoing meteorological events within AmerenMissouri's most critical and dense service areas. The remaining stations are distributed throughout much of Missouri. In addition to the surface sensor suites, two mobile GPS-based rawinsondes augment the surface network providing detailed information about the vertical structure of the pre-storm environment. Complementing the surface and upper air mesonet are two X-band Doppler radars. These radars are designed to provide the three-dimensional wind, pressure and buoyancy fields from a multi-Doppler analysis. This detailed mesoscale temperature, pressure, wind, and humidity data are used both as a diagnostic and as a tool to dynamically initialize the Advanced Weather Research and Forecasting (WRF-ARW) model. The resulting high-resolution meteorological data fields are being combined with information on the power grid configuration, local ground and tree cover information, and historical power outage information to pinpoint areas at highest risk of power interruption.