A Predictive Geoelectric Field Model: Development, Results, and Challenges
Abstract
Historically, there are numerous documented examples of geomagnetic storms causing geomagnetically induced currents (GICs) in power grids resulting in power outages, service disruption, and impacts on routine operations. As a result, space weather impacts on the power grids are one of the top priority problems in today's society. To address this problem, NOAAs Space Weather Prediction Center (SWPC) has provided products and services that forecast, warn, and alert power grid operators of impending geomagnetic storms. However, during a meeting with grid operators in 2011, nowcasts and forecasts of the surface geoelectric field were identified as the key input needed for determining GICs since the regional geoelectric field can be used as input to power grid system models for determining the level of geomagnetically induced currents. SWPC, working with our partners has addressed this problem by: 1) working with the University of Michigan in 2016 to transition into operations a Geospace Model that was driven by solar wind observations at L1 to provide short-term regional predictions of magnetic variations at Earths surface that drive the geoelectric field; and 2) working with partners at the US Geological Survey (USGS) in 2020 to introduce a near real-time Geoelectric Model that used ground-based magnetometer data from the USGS and Natural Resources Canada (NRCan), along with a 3D empirical ground conductivity model, to estimate the regional geoelectric field in the US. Then, in June 2021, instead of using near-real time ground observed magnetic field data to drive the geoelectric model, we use predicted magnetic fields from the Geospace Model to drive the Geoelectric Model in a predictive mode. In this presentation, we show the first results from the coupled Geospace/Geoelectric model that can provide short-term predictions of the regional geoelectric field to support power grid operators. First results indicate that the model does surprisingly well at predicting the more slowly varying geoelectric field (many minutes); however, as expected, the model does not do as well with minute-by-minute variations. In addition to showing these results, we discuss model limitations, next steps towards producing useful geoelectric products, and the challenges for the research community to improve geoelectric forecasting.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021
- Bibcode:
- 2021AGUFMSM41A..02S