Hydroelectricity has long been assumed to be the cornerstone of any future low-carbon economy. It is the single largest source of renewable energy in the world, representing 16 percent of global electricity generation—a figure that is set to grow by about three percent a year over the next quarter century. Here in Canada, hydro delivers 60 percent of our electricity, a figure that climbs to 90 percent in British Columbia and 96 in Quebec.
It’s cheap. It’s clean—emitting just six grams of carbon dioxide CO2 per kilowatt hour generated compared to the 1024 grams emitted by coal, the 45 grams emitted by photovoltaic solar and the 16 grams emitted by wind. And, crucially, it is not intermittent like other renewable energy sources such as wind and solar. That is to say, while the sun does not always shine and the wind does not always blow, hydro is ‘always on’, allowing everything in a modern economy from hospitals to factories to keep running without fear of regular black-outs.
It’s also flexible, offering the ability to quickly ramp up or wind down the amount of electricity dispatched depending on need. As a result of all these favourable attributes, in most scenarios for the transition to a clean-energy economy that we need in order to prevent catastrophic climate change, hydroelectric does a lot of the heavy lifting. When Ontario switched off its coal plants, it was hydroelectric, together with nuclear, that filled the gap.
But what if climate change itself started to make hydroelectricity less dependable?
This appears to be the worrying conclusion of some PICS supported researchers who have looked at projections for the rest of the century for British Columbia, but which have worldwide implications.
The story begins at the top of the province’s mountain ranges. As is happening around the globe as temperatures increase, glaciers here are melting rapidly. A bit of annual melt isn’t a problem; indeed, billions of people depend upon this meltwater. But the snow that falls upon the glaciers each year is supposed to make up for the losses, and that’s not happening any more.
Historically, researchers have kept what amounts to a sort of ledger that tracks this process, accounting for the water accumulating as ice and the water exiting as runoff. This has been useful enough in the past when glaciers changed shape slowly and their motion could be measured directly. But projections of future glacier change need more than this ledger system. They also need to account for the underlying physics of glaciers—how ice moves—and earlier studies came in for criticism for their failure to do so.
However, last April, a group of researchers from the Universities of Victoria, British Columbia, Northern British Columbia and Iceland with some support from PICS published a paper in the journal Nature Geoscience that took into account these ‘ice dynamics’ for the first time in a series of simulations of what is likely to happen from now until the end of the century. They focussed their gaze on glaciers of the Coastal, Interior and Rocky Mountain ranges of BC and Alberta, and also upped the spatial resolution—basically like having a better camera—compared to most other studies.
They ran simulations for each of the four main scenarios of likely global temperature increase developed by scientists with the United Nation’s Intergovernmental Panel on Climate Change, ranging from 1.5°C of warming up to just shy of 5°C. And for all but the lowest temperature increase, by the year 2100, some 75 percent of the ice area and 70 percent of the ice volume of Coastal mountain glaciers will have been lost compared to 2005. The Interior and Rocky mountain glaciers meanwhile will essentially have disappeared, losing 90 percent of their ice area and volume.
The results were shocking and made headlines throughout western Canada and across the country, due to the profound impacts this will have on alpine tourism, ecosystems, and agriculture, which depends on the water for irrigation. The Columbia Icefield between Banff and Jasper National Parks is one of Canada’s major tourist attractions. The paper starkly showed computer-composited ‘aerial’ images of the grand icefield diminishing from tens of kilometres wide in 2010 to essentially nothing by the end of the century.
The researchers hinted at how glacier melt might affect hydro power in the province, reminding readers that the Columbia River, which flows from its headwaters in the interior of BC out to the Washington and Oregon coast, delivers the largest hydroelectric production of any river in North America. In 1964, Canada and the US signed a treaty sharing the hydroelectric spoils of the Columbia between them, a treaty that is up for renegotiation in 2024.
But the paper only spoke briefly about the effects on hydroelectricity, with no hard numbers attached. Nevertheless, Gerry Clarke, the lead researcher on the team, is now working with BC Hydro, the province’s energy company, to develop better power production planning in the future.
Meanwhile, a second paper published around the same time by a pair of mechanical engineers, Ned Djilali from the University of Victoria and Simon Parkinson from the International Institute for Applied Systems Analysis in Austria, did go into precisely those sort of details. Supported by the Pacific Institute for Climate Solutions via its 2060 Project looking at the future of Canadian energy systems, the researchers’ findings have both good and bad news for how much we can continue to depend on hydroelectric as the foundation of a clean-energy economy.
Parkinson and Djilali combined three sets of data: results from models of hydrologic (water) cycles at the scale of a river basin, together with global climate model simulations downscaled for BC, and projections of electricity requirements, including how a warming climate may increase demand for electricity for more air-conditioning and reduce it for less heating.
First the good news from their modelling. On the supply side, the likely increase in streamflow from melting glaciers is significant, delivering an increase in the province’s potential annual hydropower of 11 percent. Meanwhile on the demand side, the reductions in heating as a result of warmer temperatures drown out the modest increases in cooling in the summer, delivering a reduction in average and peak demand of two percent. Together, this translates to an increase of roughly 11 terawatt hours of available energy by 2050. That is, climate change will give BC much more energy supply that we can tap into.
For a while.
Clarke’s paper notes that the high point of glacier volume loss, and thus of meltwaters running into streams and rivers will be from 2020-2040.
So the moral of the story from both papers is that in a warming world, glacier-fed BC hydropower stations will be capable of offering us more electricity as we approach mid century. This largely dovetails with BC Hydro’s own findings that annual precipitation in BC has increased by about 20 percent over the last century as temperatures increased matched by a modest increase in inflows into BC Hydro reservoirs (although the trends are small), and across Canada the increases have ranged from 5 to 35 per cent. As a result, the company expects to see a modest increase in annual water supply for hydroelectric generation up to mid-century.
Parkinson and Djilali do not conclude that this beneficial effect of climate change is any reason to sit back and relax. They stress the uncertainties in projected climate impacts, and so how BC Hydro configures its technological and infrastructural response over the next few decades will require significant operational flexibility over the long term in order to ensure reliability. This requirement for infrastructural flexibility will actually increase costs.
And what happens after 2050 should give pause. If the high point of run-off from glaciers comes around that time, or a decade or so before, what are the consequences from reduced run-off after that point? Furthermore, the majority of climate change impacts are actually likely to happen in the second half of the century, but even more uncertainly surrounds climate conditions after this point.
“These papers are very broad and coarse,” says Markus Schnorbus, lead hydrologist with the Pacific Climate Impacts Consortium, explaining that much work now needs to be done to scale these findings down, as what happens more locally varies considerably. In BC, the role of glacier runoff may be important for the smaller reservoirs, and less so for the bigger reservoirs. “The Mica and Peace dams, the main sources of our hydropower in BC, don’t depend upon glacier run-off, but from snowmelt and rainfall.”
So this is something Schnorbus and other researchers are now trying to answer: How important are glaciers for hydroelectricity compared to the role of rainfall and snowmelt? Will the increase in rainfall be enough to make up for the glacier losses? Also, With increasing temperatures, the proportion of precipitation that falls as snowfall can be expected to decrease over the long-term.
The increased precipitation as a result of climate change will be seen in most seasons and on an annual basis, although there will be sharply drier conditions in the summer. Warming will also trigger earlier spring snowmelt, which when combined with warmer and drier summers, means that run-off in the summer will be lower in many provincial basins.
“What does happen when the glaciers are gone?” Schnorbus continues. “There’s still a lot of speculation going on. How significant, it’s still hard to say. We haven’t looked too hard at its effects on coastal regions. While the Campbell will be hit by a substantial loss of snowpack in the future, it has negligible glacier runoff. Meanwhile, some of the smaller reservoirs may be more affected – Daisy Lake and Allouette.”
The most vulnerable, he says, are likely run-of-river operations. These are smaller hydroelectric dams that involve no or very little storage capacity, no reservoirs. Their viability during the late summer and fall is heavily dependent on glacier run-off. Some environmental groups in the province have suggested run-of-river hydroelectricity as an alternative to large-scale hydro due to the significant land footprint required for the latter. “But in terms of reliability, they are already very much at the mercy of the whims of nature.”
Schnorbus also notes that while things don’t look too bad on an annual basis, what can be just as important is the changes from season to season. “More water annually does not necessarily mean more power.”
And substantial changes to the seasonality of energy supply was precisely the problem that researchers from Washington state found when they modelled the effects of projected climate change over the rest of the century on the hydroelectric system that, like BC, supplies much of the US Pacific Northwest with its power (albeit with more fossil fuels and nuclear in the mix).
In the 2020s, regional hydropower production increases by 0.5-4 percent in winter, but then decreases by 9-11 percent in the summer, for total annual reductions of 1-4 percent. Slightly larger increases in the winter and slightly larger decreases in the summer, are projected at mid-century and by the 2080s.
One question that immediately comes to mind is whether similar effects will be seen elsewhere in Canada and indeed the world. The BC researchers note that the impact of climate change on demand seen in this province will be similar in the rest of the country, and the impacts on hydropower compare well with Nordic Europe, which also is highly dependent on hydroelectricity. A 2011 paper looking at the Upper Danube basin, Europe’s second largest and running through the territories of 19 countries, suggested a slight to severe decline in hydroelectric power generation up to the 2060s, with increases in the winter and decreases in the summer.
“In other regions, the situation is very different again,” notes Schnorbus, “such as the altiplano in South America. There, glaciers are far more important.”
In January this year, researchers from Vienna’s International Institute for Applied Systems Analysis and Wageningen University in the Netherlands answered that question, and their findings are not encouraging.
They used data from over 24,000 hydroelectric plants around the world, equivalent to almost 80 percent of the planet’s hydro dams, to build a model of the interplay between climate change and hydropower worldwide over the course of the rest of the century. They explored the outcomes that are likely to occur from 1.5°C of average global warming up to about 5°C. And they found that most places will suffer from reductions in usable capacity at this mid-century mark, but these effects vary wildly depending on the region.
Canada, northern Europe, central Africa, India and northeastern China will be blessed with an increase in hydro capacity. Indeed, most of the world’s land surface will experience boosts in streamflow. The trick is that most hydropower plants are located in regions where “considerable declines” are projected, such as the United States, central and southern Europe, southeast Asia, and the southern parts of South America, Africa and Australia. And at 1.5°C of warming, some 61 percent of hydro plants are hit by a diminishment of hydro capacity from 2040-2069, climbing to 74 percent of plants at 5°C of warming. And as many as a fifth of plants will experience what the researchers describe as “strong” reductions in usable monthly capacity (more than 30 percent decreases) throughout the 2050s.
The researchers also modelled whether certain adaptation options could mitigate this decrease, notably increases in the efficiency of hydropower plants. For most regions, a modest increase in efficiency was enough to completely offset the impacts decreased streamflow over the course of the year, so long as these shifts began to happen by the start of the next decade. Australia and South America however are still hit by decreases in usable hydro capacity no matter what is done.
The take-away message from all the research is certainly not panic, but that adaptation options including greater operational flexibility need to start being included in planning, whether due to projections of increases or decreases in hydroelectricity, and due to the increased seasonality of production. And this needs to be incorporated today.
Nevertheless, as the planet warms, any hydro-power advantage gained from an increase in glacier run-off will be short-lived, as this ancient, icy resource, it now turns out, has a finite life expectancy.
Researchers calculate that Canada’s northern latitudes require astonishing number of batteries or solar panels