What will electricity demand look like tomorrow? Next week? What about prices two months from now? For the most part, we have a good idea, but such projections are certainly not infallible (see: Texas and New York) or immune to manipulation (remember: California). In truth, despite a fundamental need for answers, reliable electricity demand outlooks are harder to come by than one might assume. As the grid ages, generation inputs change, electrification intensifies, and climate change continues, that need only increases. No one place has all the answers, but the oceans may be able to lend some help.
According to a recent study, due to be published in the journal Nature Geoscience, ocean temperatures can predict U.S. heat waves up to 50 days out. More specifically, lead author Karen McKinnon, a postdoctoral researcher at the National Center for Atmospheric Research, has identified a distinct pattern of sea surface temperatures in the middle of the North Pacific Ocean.
The Pacific Extreme Pattern (PEP) is a juxtaposition of warmer-than-average water against cooler-than-average seas, and when it appears it can foretell an increased chance of summertime heat waves across much of the Midwest and East Coast on a particular day or week. “The stronger the PEP is (i.e. the larger the temperature anomalies), the more severe the upcoming heat is likely to be,” said McKinnon.
Using temperature data stretching back over 30 years, the research team was able to retrospectively predict the presence – or lack thereof – of extreme heat events during a particular year’s summer. At a lead time of 50 days, the researchers were able to predict a roughly 14 percent increase in the odds that an extreme heat event would strike the eastern U.S. during a given week. Closer in, at 20, 30, and 40 days, the scientists observed an increase in the odds to better than 1 in 2.
After successfully hindcasting – including the particularly damaging summer of 2012 – the team will begin making predictive forecasts for summer 2016 in early May. So, what are the implications for the utilities sector?
“Based on our understanding of the sector as well as conversations with some folks in the industry, we do believe that predictions could be used to better prepare for spikes in demand,” says McKinnon. “However, it is important to recall that our predictions are probabilistic, rather than deterministic. Thus, we are able to predict an increase in the odds that a hot day will occur by 3-4 times, but -- because these are extreme events -- the absolute probability that the event will occur is never predicted to be more than 35 percent.”
To be sure, in this industry, fluctuations in probability – and a greater ability to maximize operating reserves and/or adjust bidding strategy – however small, mean billions.
Take the aforementioned summer of 2012 for example; losses from the drought and largely unexpected associated heat wave are estimated to exceed $31 billion. More recently, in August of 2015, surging temperatures caused massive wholesale power price spikes in California, Texas, and New York, costing consumers and power retailers millions of dollars. Such examples are a dime a dozen, and are only becoming more regular.
Ignoring demand spikes, heat waves stifle generation capacity, and distribution and transmission systems, all of which are at critical junctures. By 2020, investments in electricity infrastructure will have fallen behind by $37 billion for distribution and $57 billion for transmission. By 2035, the power sector will require $2.1 trillion in investments. Moreover, the challenges of balancing supply and demand increase as variable renewable penetration grows.
Climate science doesn’t have to be a detriment to the energy industry; as an investment tool, its applications are nearly limitless. For their part, generators and power portfolio managers may discover that the oceans are more valuable than they ever could’ve imagined.
First published at Oilprice.com