By Justin Sharp, EPRI
Note: This is a guest blog post; it represents the views of the author alone and not the American Meteorological Society.
The “Transition to Carbon-Free Energy Generation” Presidential Session at the AMS 104th Annual Meeting discussed the status of–and barriers to–the U.S. transition to renewable energy. During that panel, I and several other speakers discussed how the weather enterprise will be key to this effort. Meteorological expertise is a keystone of power systems with large shares of renewable energy.
Weather Drives a Vastly More Complex Electric System
Existing electric systems are some of the largest and most complex machines humankind has ever built, with every component linked and synchronized. Electricity consumption is increasing rapidly as sectors currently powered by fossil fuels switch to electric power, and demands from data centers, AI, and cypto-mining escalate. Many of these new loads, such as electric vehicles and indoor heating/cooling/ventilation systems, are affected by weather, especially temperature. At the same time, extreme weather events continue to cause infrastructure outages, a trend likely to increase with climate change.
Amping up solar and wind power means electricity generation is affected by additional weather variables: wind speed, clouds, and aerosols. Thus, weather-dependent generators, sited across broad geographic areas, produce complex interactions that can have large impacts that were never previously imagined.
Planning and operating such an electric system, day and night, through heat and cold, sun and cloud, wind and calm, with increasing amounts of weather dependent load, using large numbers of wind and solar generators and energy-limited storage devices, is an unprecedented challenge for the sector.
Better Data for Better Grids
Our ability to forecast renewable energy generation is improving rapidly, and better weather forecasts can reduce uncertainty in our estimates of future generation, easing the integration of renewable energy into grid operations. However, by themselves, even perfect forecasts cannot solve the problem of variability and shortfalls in renewable energy generation across the year. Better historical weather data (and best practices for their use) are vital to plan and build electric systems that can most effectively meet our highly variable energy demand, using diverse power sources and energy storage to ensure reliability across environmental and grid conditions.
Right now, the power sector is blind to a lot of key challenges, with power system planners often relying on weather data that is less certain and more limited than they believe it is. For example, gridded data from numerical weather prediction models are often utilized in planning tools without validation or uncertainty quantification and as if they have observational quality. This can result in important risks being missed. For instance, cold days lead to high electric demand and an increasing risk of infrastructure outages. Such critical days often occur in conjunction with strong inversions; frequently the combination of model resolution and/or parameterizations does not properly handle these inversions, resulting in over-estimates of wind and temperature, and under-estimates in clouds and fog. Issues like these could result in under- or over-building infrastructure, potentially leading to reliability concerns or incurring unnecessary costs.
Building a Weather-Data Infrastructure
Just as meteorologists employ models to diagnose and forecast atmospheric phenomena, electric system specialists utilize power system models to optimally plan and operate the grid. As electric grids evolve to include large amounts of renewable generation and energy storage, ensuring reliable, affordable electric power requires, a) improvement of these models to fully consider the uncertainty inherent in the weather and b) best-in-class, fit-for-purpose weather and climate information to inform the models.
Increasingly detailed records of past weather conditions for large regions and long time histories are needed — yet they typically do not exist as observations and thus must be synthesized. Comprehensive validation of such model data is also essential, along with user education and data curation to ensure that stakeholders appropriately apply weather intelligence in their downstream analyses.
Assessing, validating, and hopefully bias-correcting weather model estimates requires large quantities of ground-truth weather data. The rapid buildout of wind and solar facilities is producing such a data network, but unfortunately, there is often significant resistance from owners to sharing this data. There’s hope though; the Electric Reliability Council of Texas (ERCOT) now mandates that all renewable generators provide their power and meteorological data to the public. We need to see similar approaches elsewhere, as soon as possible.
You can learn more about all these issues in an ESIG report I co-authored. In summary, two incredibly complex fields — the electrical grid and atmospheric sciences — are becoming increasingly intertwined. There is a need to work together across sectors to define the requirements for optimal meteorological support for ongoing planning and operation of evolving power grids, and to develop an operational framework for producing, disseminating, and ensuring appropriate use of this intelligence. EPRI and other organizations are working to convene stakeholders to respond to this urgent need and I encourage interested parties (including data users, data producers, and observational data owners) to contact me at EPRI. Only by working together across sectors can we create the reliable and affordable carbon-free grids needed to power the economy while ensuring a livable future for our planet.
Header photo: Pixabay from Pexels
Further Reading
- OSTI.GOV: The Evolving Role of Extreme Weather Events in the U.S. Power System with High Levels of Variable Renewable Energy (Technical Report)
- ESIG Blog: The Crucial Need to Incorporate Meteorology into The Renewable Energy Transition (Part 1)
- ESIG Blog: The Crucial Need to Incorporate Meteorology into The Renewable Energy Transition (Part 2)