Even as fossil-fuel power stations emit CO2 and drive climate change, they will likely generate less electricity in a warmer world. Both fossil fuel and nuclear electricity generation plants have already been affected by higher water temperatures and reduced river flows. And in the future the same problems could mean that summer average generation capacity could fall by one-twentieth to one-fifth from past levels. So says the largest study yet of climate change’s effect on cooling water used in power generation, by Michelle van Vliet from Wageningen University in the Netherlands and her colleagues.
“Our results show overall increases in river water temperature under climate change, with highest increases in the southern part of Europe and south-eastern part of the US,” Michelle told Simple Climate. “For large parts of the US and Europe the results show an overall increase in high river flow and decreases in low river flow. The combination of higher water temperatures and lower river flows during summer periods can be especially critical for cooling water use.”
Nuclear or fossil fuel plants with water cooling currently produce over three out of every four units of electricity consumed in Europe and over nine out of every ten in the US. They use more river water than farming and more than twice as much as the demand from homes in each region. That thirst poses these “thermoelectric” power stations problems in a warming world, Michelle explained. “Due to climate change, periods with low river flows in combination with high water temperatures are expected to occur more often, and this could cause significant reductions in cooling water availability during summer,” she said.
Following three streams
To understand effects on power production, so that we can adapt to them, Michelle and her fellow scientists wanted to produce a forecast with trustworthy numbers. The key data needed included river flow, water temperatures and how they affect rivers’ cooling ability and power plants’ usable capacity. That meant hard work was needed to create a simulation that included all these factors, by combining three different types of model. Michelle used large-scale river water temperature and a physical model of water flows over the land surface developed by Dennis Lettenmaier and John Yearsley from the University of Washington. Then, to simulate how these affected usable power plant capacity, Michelle worked with Stefan Vögele of the Institute of Energy and Climate Research in Jülich, Germany
In a paper published in scientific journal Nature Climate Change on Sunday, the researchers first looked at how well their simulations modelled real data from the period 1970-2000. “We evaluated the performance of the hydrological and water temperature modelling framework using observed daily river flow and water temperature records of a high number of river stations,” Michelle explained. “This showed an overall realistic representation of the observed conditions.”
Then, to create forecasts for the EU and US, Michelle examined various ways that society, the economy and climate will change for the period from 2031-2060. That timeframe is needed because thermoelectric power plants are expensive to build, but are supposed to run for 50-60 years to pay back those up-front costs. In particular she looked at two climate scenarios from the Intergovernmental Panel on Climate Change. In scenario A2 the world’s technological change is fragmented and slow, while scenario B1 assumes environmental sustainability and a much more rapid introduction of renewable energy generation.
Low on juice
Michelle’s team used their models to look at daily water temperature and river flow at 61 power plant sites in the US and 35 in Europe during these periods, two-thirds of which will still be running in 2030. Averaged across both the A2 and B1 scenario, water temperatures were more than 0.7°C higher, and the lowest flows that rivers reached got even lower. That did reduce how much electricity the power stations could generate. “Our results show an average decrease in the usable capacity of thermoelectric power plants by 4-16 per cent for the U.S. and 6-19 per cent for Europe for summers in the period 2031-2060 relative to 1971-2000,” Michelle said.
How severely a power plant is affected depends upon its design. Those with cooling towers that recirculate water will be less affected than ‘once-through’ plants that pump used, warmer, water directly back to its source. For example, the probability of power production falling to one-tenth of its normal value more than doubled in the A2 scenario for both types of plant. But in once-through plants that probability works out as more than one day a year, whereas in recirculating plants it only increases to less than three hours per year.
Electricity producers could adapt to climate changes by switching to natural gas powered stations, which use less water, or by cooling thermoelectric plants with seawater. But with these plants designed to last so long and relying on local water sources, such changes wouldn’t be quick to do and therefore planning for them should start today, Michelle’s team wrote. But perhaps even more importantly, which scenario the world follows in future will affect how easy it is to adapt. “It is expected that the vulnerability of the energy sector to climate change will be less high under a scenario with a rapid introduction to renewables than a scenario that considers slow technological change with many fossil fuelled power plants in need of cooling water,” Michelle noted.