Solving the CCS cost conundrum
Changes to the energy mix and government-backed development of CCS would put the climate goal within reach, according to DNV GL
Standards agency DNV GL’s forecast that there will be significant changes to the world’s energy mix over the next three decades, set out in its Energy Transition Outlook, is hardly a consensus-breaking view. But its almost 300-page report, built on a cost-based model, is more analytical and provides more robust predictions than most.
Its model predicts that the share of renewables in the world energy mix will increase from 14pc to 40pc in 2050 while oil and gas will decline from 54pc to 46pc. For fossil fuels, it sees carbon capture and storage (CCS) as an essential and increasingly affordable technology—as long as governments provide support for its development.
“Our model forecasts that global demand for oil will have fallen into decline by the mid-2020s,” DNV GL Oil & Gas CEO Liv Hovem tells Petroleum Economist. “At that point, the stage will be set for gas to become the world’s largest energy source, accounting for nearly 30pc of the global energy mix.”
DNV GL forecasts that only gas and intermittent renewables will be in greater demand in 2050 than they are today. “This forecast gives us a very clear message—there is no single pathway to a decarbonised energy mix,” she says. “A combination of sources will be the quickest route to delivering a secure supply of affordable, decarbonised energy.”
Gas would be required to complement intermittent renewables during times of peak energy demand, such as winter in colder climates. “Neither gas nor renewables alone will make for a successful energy transition,” says Hovem.
Several climate change-mitigating technologies—such as technologies to increase energy efficiency and generate renewable power—have already become commercially viable without government incentives. Among the available technologies, “CCS is by far the most underutilised tool in the box”, says Hovem, and the barrier to its deployment is “quite simply…cost”.
It is “encouraging to see that uptake of the technology will come”, Hovem continues, but “concerning that our model predicts it will not be employed at-scale until the 2040s”.
The fitting or retrofitting CCS to power plants and industry has almost only ever happened with government intervention; enhanced oil recovery (EOR) is the only viable business case for its implementation. CCS struggles to gain traction because there is a cheaper option for industry, she says. “Emitting carbon into the atmosphere costs virtually nothing.”
Government-set carbon prices are the most cost-effective way to incentivise emissions reductions, says Hovem, “yet, 85pc of global emissions are currently untaxed.”
The remaining 15pc of CO2 emissions costs less than $10/t of, according to the World Bank’s State of Trends of Carbon Pricing 2018. If the cost of emitting carbon into the atmosphere were to increase, the speed at which industry would deploy CCS would logically also increase.
DNV GL’s model assumes that the carbon tax varies across regions between $5/t and $25/t but will increase slowly to $25-60/t in 2050.
“The reason cost [of CCS] is high is that the technology has never been installed at the scale we are now witnessing for renewables and storage,” says Hovem. “When deployed, we expect a cost learning curve similar to those industries, with costs reducing 15-20pc per doubling of capacity.”
Cost learning curve
The cost learning curve depicts a path where the cost of implementing a new technology will decrease as it begins to be introduced at scale, as has occurred for solar and wind over recent decades. Each time the capacity of an energy source doubles its cost decreases—depending on its maturity—by a double-digit percentage.
“It is impressive that solar power is now cost-competitive with traditional power production without any incentives, in certain geographies,” says Hovem.
“If there is one word to characterise the future grid it is complexity… and digitalisation is the key enabler” Eriksen, DNV GL
“We believe that the cost of implementing CCS at scale will follow this very same curve. Cost may be a barrier to deploying the technology, but it need not be in the future. Although CCS has been around for a long time, it has not started its journey down the cost learning curve. Once it starts, we expect to see a drop in its cost.”
The task of setting CCS on a cost learning curve is a “chicken-and-egg situation”, as rollout will not occur while costs are high and costs will not decrease until the industry has learned from implementation.
“The money used to stimulate uptake of solar power technology may have seemed crazy at the time, but it gave it the helping hand it needed to start its journey down the cost learning curve,” says Hovem, referring in particular to German government incentives.
A full-scale Norwegian CCS project—where CO2 will be transported by ship from a cement factory in Telemark and a waste-to-energy plant in Oslo to an onshore facility on the Norwegian west coast for temporary storage, before being transported via pipeline to a subsea formation in the North Sea for storage—could have the same effect, she says. “The project demonstrates how all pieces of the CCS puzzle can come together in a complete chain, easing investment decisions for future projects with the potential to start CCS technology on its journey down the cost learning curve.”
Existing technology would be enough to contain global warming to the COP21 pledge of 1.5°C but only if the energy transition is backed by strong enforcement of policies set out in the Paris Agreement, according to DNV GL. Crucially, it also predicts that the transition would be affordable.
DNV GL predicts that energy use will peak by 2030, when efficiency improvements start to outpace economic growth. However, although it foresees a rapid energy transition it does not expect emissions to fall fast enough to limit warming to 2°C. The forecast predicts 2.5°C, which "should set alarm bells ringing", says Remi Eriksen, the firm's CEO.
"We forecast a very rapid transition unfolding, within the timespan of one generation. Existing technology can deliver on the Paris Agreement, but only with broad and strong policy backing," he says.
The cost of the transition will be eased by another of its predictions—that the cost of energy will consume a lower share of global GDP. “Our forecast does not reach for a magical solution or a revolution. Rather, it is the reducing costs of existing technologies which make a big difference."
The world relies on fossil fuels for 80pc of its energy needs and the report predicts this will reduce to 56pc in 2050. It also forecasts that energy consumption growth will rapidly decouple from GDP and population growth, to which it has historically been correlated. "At the same time as we grow more prosperous, that growth will use less energy and make less CO2."
DNV GL forecasts the 2030 aggregate energy demand peak based on a combination of factors including efficiency gains from electrification, in residential and industrial uses, through technologies such as heat pumps. Further efficiency savings will come from 'smarter' buildings.
Adoption of EVs—which Eriksen says are three to four times more efficient than internal combustion engine cars—will mean that overall emissions from transportation will be marginally lower by 2050, despite the global fleet expanding by up to 75pc.
Global electricity grids are forecast to almost triple in size by 2050, measured in voltage kms, and be greatly more efficient. "Electrification is the main theme in our forecast. We expect electricity, as an energy carrier, to more than double over the next three decades and electrification is a big driver for energy efficiency," says Eriksen.
"By mid-century, 63pc of the world's electricity will be supplied by solar PV or wind. This large share of renewables requires more flexible systems and that means huge increases in storage—we expect storage capacity to expand nearly fifty-fold. If there is one word to characterise the future grid it is complexity… and digitalisation is the key enabler."