|
|
| Fig. 1: A single-phase, three-winding converter transformer used for HVDC conversion from DC to AC and back in Baie-James, Qubec. (Source: Wikimedia Commons) |
In the early 2000s, the Trans-Mediterranean Renewable Energy Corporation (TREC) proposed the Desertec concept: building large-scale solar power plants in the deserts of North Africa and transmitting the electricity to Europe via long-distance high-voltage power lines. The idea was to harness the abundant solar resource in the desert to supply clean energy to population centres thousands of kilometers away.
On a map the idea looks straightforward: draw a few thick transmission lines from the desert to the load centers and collect clean power. In truth, the reality is more complex. Political instability, colonialism concerns and financing challenges all play a role in determining whether such projects can be realized. However, we will focus here only on the technical aspect of power transmission.
In particular, we focus on distances of roughly 1000 to 4000 km at multi-gigawatt power levels, where the choice of transmission technology is no longer obvious. At these scales both resistive losses in the conductors and the capital cost of the lines become important, and the traditional solution high-voltage alternating-current (HVAC) overhead lines competes with high-voltage direct-current (HVDC) technology and long submarine cables.
Modern power systems move bulk electricity with high-voltage transmission lines. The key idea is that raising the voltage reduces the current for a given power flow, and resistive losses scale as Ploss = I2R, where I is the current and R is the wire resistance. Doubling the voltage halves the current I and cuts resistive losses by a factor of four. Today most long-distance lines operate at hundreds of kilovolts.
Historically the grid was built around high-voltage alternating current , mainly because transformers make it easy and efficient to step AC voltages up and down. Over the last few decades HVDC has become commercially mature. HVDC lines are attractive for very long distances, for undersea cables, and for interconnecting asynchronous AC grids. In practice HVDC lines are still rare compared with HVAC, but they are used increasingly for specific applications.
HVAC lines are the workhorse of today's transmission system. Typical overhead HVAC lines operate at 220 to 765 kV and carry hundreds to a few thousand megawatts per circuit.
A key advantage of HVAC is that it can directly tie into existing AC grids using regular transformers. No converter stations are required as each end of the line connects through transformers and switchgear. However, over long distances the line's inductance and capacitance lead to phase shifts between voltage and current, resulting in reactive power flow. This effect increases with distance, forcing utility operators to install compensation equipment such as shunt reactors and capacitors.
HVDC transmission uses converter stations at each end of the line to turn AC power into DC and back again. The line between them carries constant-polarity current at a fixed high voltage, often ~320 to 800 kV in modern projects. HVDC avoids the phase shift issues of HVAC. Furthermore the entire conductor cross-section carries current, due to the abscence of the skin effect. This combination leads to more efficient transmission.
The drawback is that converter stations, as seen in Fig. 1, are expensive and introduce additional losses. Typical estimates place each converter at roughly 350 million dollars for a ~500 kV, multi-gigawatt HVDC line, although costs vary with technology and scale. [1] When the line is short, fixed costs and losses dominate making HVAC cheaper and more efficient. Beyond some break-even distance, the lower losses per kilometer and higher capacity of HVDC outweigh the additional cost and losses of the converter.
Submarine cables are essential for interconnecting countries separated by water and for bringing offshore wind power to shore. A survey by the European Commission's Joint Research Centre of existing projects finds that for long HVDC submarine cables the total losses, including converter stations, are on the order of 3.5 to 5 percent for 1000 to 2000 km links, while comparable HVAC cables would lose roughly 6.7 to 10 percent over the same distance. [2] A concrete example is the NordLink project, a ~525 kV HVDC submarine and underground cable connecting Norway and Germany. NordLink is about 600 km long, rated to transmit 1.4 GW, and cost roughly 2 billion euros to build. Measured operating data indicate total losses of about 4 to 5 percent at full power, consistent with the 3.5 to 5 percent per thousand kilometer estimate, not including converter losses. [2,3]
HVAC is still used for shorter submarine links when both ends belong to the same AC grid and the converter-station cost of HVDC cannot be justified. The Crete-Peloponnese interconnection built in 2021 consists of two parallel HVAC submarine cables linking the island of Crete to mainland Greece. Each circuit is rated for about 140 MW, for a total of 280 MW, over a length of roughly 174 km, and measurements over a year of operation show active power losses of about 6.75 percent over the link. [4] If we use simple resistive scaling and apply linear extrapolation, these losses become prohibitive at longer distances. A 1000 km HVAC submarine cable of similar design would lose nearly 40 percent of its power due to resistance alone, and the increasing reactive current would saturate its capacity much earlier. For the trans-continental distances relevant to a Desertec-style link from North Africa to Europe, the physics therefore strongly favours HVDC submarine cables.
On land the comparison between HVAC and HVDC is more balanced. Overhead lines have much lower capacitance per kilometer than submarine cables because the conductors are surrounded by air rather than water, so the reactive current drawn by an HVAC overhead line is much smaller. HVAC can therefore be used over much longer distances without excessive reactive effects.
For overhead HVAC lines, cost estimates in recent transmission planning studies fall in the range of roughly 1 to 4 million dollars per kilometer, depending on voltage level, terrain, and permitting difficulty. [5] Published cost ranges for overhead HVDC lines in North America span roughly 0.7 to 5.3 million dollars per kilometer, again depending on voltage level and corridor difficulty. [1] For a long (greater than 1000 km), multi-gigawatt corridor, that implies capital costs of order 1 to 5 billion dollars regardless of whether the line is AC or DC.
For very long distances and high power levels, HVDC becomes increasingly attractive due to its lower line losses and higher capacity. Trieb and Knies estimated that an HVDC line transmitting 5 GW over 3500 to 4500 km would experience total losses of about 10 to 15 percent while costing only 2 to 2.5 billion dollars (in 2004 USD), making it competitive with building multiple large power plants closer to load centres. [6]
From a physics perspective, losing 10 to 15 percent of the power after 4000 km may sound surprisingly modest. The economic impact depends on the value of the electricity. Current wholesale electricity prices in Europe are roughly 60 to 90 dollars per megawatt-hour. [7] For a 5 GW line with 10 percent losses, the lost power can be estimated as
With electricity valued at approximately 60 to 90 dollars per megawatt- hour, the cost of lost power is given by
| Lost Value | = | 500 MW × $90 MWh-1 × 8760 h year-1 | = | $394.2 million per year |
For lifetimes of 30 to 40 years, the total lost value sums to about 10 to 15 billion dollars over the line's life. This alone would be around five times the capital cost of building the line itself.
However, this back-of-the-envelope estimate does not include other costs and hurdles. Several billion dollars of upfront investment is needed to build the line and converter stations. The line would need to be kept full or nearly full for decades to realize the low cost per megawatt-hour. Political risk, permitting delays, and public opposition can easily derail such projects.
In practice, the barrier to desert-to-Europe transmission is therefore less the unavoidable 10 to 15 percent physical loss and more the difficulty of financing and siting multi-billion-euro corridors across multiple countries.
Long-distance electric power transmission is often perceived as a major obstacle to harnessing remote renewable resources such as desert solar power. A quantitative look at existing projects and engineering studies paints a more nuanced picture.
For submarine cables, the physics strongly favours HVDC. Real-world projects like NordLink demonstrate that 600 km scale HVDC links can deliver gigawatt-scale power with only about 4 to 5 percent losses, whereas HVAC submarine cables of comparable length would suffer much larger losses and practical limitations. For overhead lines, both HVAC and HVDC are technically viable over 1000 km and beyond, with losses of only a few percent per 1000 km, but HVDC becomes increasingly attractive at the multi-thousand-kilometer scale and for interconnecting different grids.
In a Desertec-style scenario transmitting several gigawatts over 3000 to 4000 km, the unavoidable physical losses are on the order of 10 to 15 percent, corresponding to hundreds of megawatts and hundreds of millions of dollars per year at current European electricity prices. These are not negligible numbers, but when spread over the enormous amount of energy delivered over a 30 to 40 year lifetime, they translate into a transmission cost of only a few dollars per megawatt-hour.
The real challenges of long-distance transmission are therefore less about pure efficiency and more about cost, risk, and governance: raising billions of euros of capital, securing rights-of-way across multiple jurisdictions, ensuring long-term political stability, and managing public perceptions of energy dependence. From a physics point of view, moving desert solar power to Europe is entirely feasible. Whether it is the best use of limited financial and political capital remains an open question, but one that should be debated on economic and geopolitical grounds rather than on misconceptions about transmission losses alone.
© Jan-Lucas Uslu. The author warrants that the work is the author's own and that Stanford University provided no input other than typesetting and referencing guidelines. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
[1] "Assessing HVDC Transmission for Impacts of Non-Dispatchable Generation," U.S. Energy Information Administration, June 2018.
[2] M. Adrelean and P. Minnebo, "HVDC Submarine Power Cables in the World," European Commission, 2015.
[3] S. E. Schradder and F. E. Benth, "A Stochastic Study of Carbon Emission Reduction From Electrification and Interconnecting Cable Utilization. The Norway and Germany Case," Energy Econ. 114, 106300 (2022).
[4] G. A. Barzegkar-Ntovom et al., "Active Power Losses in the Crete-Peloponnese Subsea HVAC Interconnection: Insights from One-Year Measurements," IEEE 11169230, IEEE Intl. Conf. on Environment and Electrical Engineering and IEEE Industrial and Commercial Power Systems Europe, 15 Jul 25.
[5] "Transmission Cost Estimation Guide for MTEP24," Midcontinent Independent System Operator. May 2024.
[6] D. C. KacKay, Sustainable Energy - Without the Hot Air (UIT Cambridge, 2009).
[7] "Electricity 2025 Mid-Year Update," International Energy Agency, July 2025.
