Hydropower storage may be the key to sustainable energy

Printable version Printable version

F. Lempérière, France
An analysis of future energy needs and sources should avoid utopias, clearly evaluate the needs, and quantify the potential, cost and impacts of possible solutions.



  • Nuclear energy is sometimes presented as the energy of the future, but present solutions are very limited, due to the scarcity of uranium resources, and new solutions which are using much less uranium, are far from industrial development. Worldwide, nuclear energy will be a rather marginal solution until at least mid century.
  • The storage of CO2 from fossil energies is possible, but may apply to a minor part of generation; it is also costly and increases energy needs. It should not prevent a huge reduction of fossil fuels utilisation.
  • Stabilising or reducing energy consumption through lifestyle changes has been suggested. This may be possible, to some extent, for the one billion people in industrialised countries who use presently two-thirds of the world energy. But for those eight billion people who will live mid century in other countries, an average per capita of at least half the energy used presently in industrialised countries will be needed, and this may be reached in China by 2020. Before mid century, world energy utilisation will thus be triple the present levels.As the share of renewable energies should increase from presently under 20 per cent to over 70 per cent, renewable energy need will be increased tenfold.


They are essentially for buildings, heavy industry and transport.

  • The world energy demand for buildings (housing, offices, small industry, schools, hospitals…) will include only a minor part for heating and hot water; this part, presently mostly supplied by oil and gas, could be provided directly by low-cost renewable energy (biomass, solar thermal, geothermy), and also through electricity. All other demands (including much air condi- tioning) will be essentially met through electricity.
  • In heavy industry, the direct utilisation of renewable energies will probably be low, especially for high-temperature heating. The main alternative to fossil fuels will be electricity.
  • Most energy for transport will be from three main sources: oil, bio-fuels and electricity. The share of electricity will be significant for rail, and also for road transport, either for electric or hybrid vehicles, or for supplying hydrogen for direct utilisation, or for association with bio-fuels.
  • Electric energy will be also favoured by new generation facilities (wind, photo- voltaic, solar, marine energies). The best places for low-cost electricity generation from wind, sun or hydro may be associ- ated with medium or long distance transport. Electricity has thus a great future, if its supply is adapted to needs in a timely manner, and if its generation includes a very low share of fossil fuels. It is necessary for using most renewable energies, and this justifies a clear specific analysis of potential, cost, and impact of possible sources of electricity generation. This specific analysis seems clearer and more realistic than traditional global studies of “primary energy needs”, which are poorly adapted to renewable energies.


  • In 2007, the global electricity generation (including losses from transport) was close to 20,000 TWh/year, increasing annually by 4 per cent; it will probably reach at least 50,000 TWh/year in 2040 (i.e. a world average of 6,000 KWh/year par capita as compared with 12,000 presently in industrialised countries).
  • Nuclear generation by present solutions, limited annually to 2 per cent of remaining uranium resources, will eventually supply 5,000 TWh/year (double what it provides now). The cost per KWh (excluding uranium supply) will be over 5 cents of U.S. $ in Europe, possibly less in China and India (all costs below are in cents of USD, actualised in 2009). The future cost of uranium supply is unknown, because its availability will be restricted because of decreasing supplier countries. New solutions requiring much less uranium may be implemented industrially after 2040, but their cost is unknown and probably over 5 cents.
  • Hydropower generating presently 3,000 TWh/year may reach a maximum of 6 or 8,000 TWh/year (including tidal and waves) in 2040 or 2050, at an average cost of 5 cents of USD/KWh. The utili- sation of biomass and geothermy for electricity generation will be probably limited to 1 or 2,000 TWh/year at similar costs. The conclusion is rather simple: In 2040, 35,000 TWh/year should be generated from fossil fuels (coal or gas), wind, or sun.

Capture d’écran 2013-07-16 à 14.07.55


  • The technical cost of gas generation may remain low (five cents/KWh) until mid century, but the commercial cost is unknown. The technical cost of coal power generation is presently low in countries using their own resources, such as China or India. It will be probably well over 5 cents in 2040 as the best mines will be exhausted (especially in Asia), labour cost will be higher, and CO2 should be stored or taxed. Few countries will export coal or gas.
  • Wind is now a well proven energy source. The direct cost per KWh varies considerably according to wind data. Three ways of generation are possible; close to needs as is presently practised in Western Europe; in very large farms in deserted areas with significant electricity transport; and offshore. A future average worldwide generation cost close to 5 cents/KWh and possibly lower appears likely. The theoretical world potential is very high, but many countries have few possibilities. It is thus unlikely that wind energy will generate more than 20 per cent of world needs, i.e. 10,000 TWh/year, requiring 3 to 400,000 km2 with 2 to 4 units per km2. The key problem is the intermittent generation (for instance, supplying half time 10 to 100 per cent of the capacity).
  • Solar electricity (photovoltaic, or by mirrors and steam) has unlimited potential, with the most increase avail- able for energy needs in sunny countries. Two utilisations appear likely; close to needs, for instance in build- ings, but probably for less than 5,000 TWh/year (1,000 KWh for five billion people), and very large plants using 10 km2 per TWh/year. Within ten years, it is likely that the generation cost per KWh for those plants will be well under 10 cents in sunny countries with low-cost labour, i.e. for most needs. As for wind, the main problem is the intermittent supply.


Without energy storage, intermittent wind or sun power would be replaced most time by fossil fuel power, and would thus generate much less than half of fossil fuel generation. Such a hypothesis is represented in Drawing A, below, with sun and wind power limited to 10,000 TWh/year.

Storing one or two days of sun or wind power, almost completely avoids the need for fossil fuel electricity limited to lack of sun and wind for a long time. This hypothesis is represented in Drawing B, below. In 2040 or 2050, wind and sun electricity generation may reach 30,000 TW/year, i.e. an average supply of 3.500 GW; a total storage need of 2,000 or 2,500 GW (0.25 KW worldwide per capita) will be advisable. At present, storage by pumping seems by far the most effective solution.

The existing Pumping Storage Plants (PSP) operating between two lakes at very different levels total 100 GW; their average investment cost has been in the range of 1.500 USD/KW. It is possible to find many other sites (Japan has imple- mented 25 GW, i.e. 0.2 KW per capita). It will be also possible to use, as part of PSP, many hydropower lakes presently used for energy generation. However, it is likely that onshore solution sites will meet only part of what is required, and offshore solutions may be necessary, especially in flat countries.

It is possible to use the sea as a low reservoir, and to create large, onshore high basins close to the sea (as, for instance, in Okinawa, Japan). This solution may be very cost-effective, and could have a great future in places with low population density (as in most of Africa and South West Asia). Such large onshore high basins with sea water may be less acceptable in countries such as Europe or China. Creating offshore large high basins that are 100 m over sea level and set within 10 or 20 km2 artificial atolls (called “emerald lakes”) may be also cost-effective.

All the solutions above are based on proven technologies, and their costs may be easily evaluated. It is thus possible to get the necessary storage associated with intermittent energies. The total correspon- ding investment for 2,000 GW would be 3 trillion USD for supporting 30,000 TWh/year of wind or sun electricity,i.e. an

investment of 0.1 $/KWh/year and a cost per KWh of 1 cent. There will be also a loss of 20 per cent of the stored energy linked with equipment output, i.e. 10 per cent of the total generated energy,i.e.. 0.5 to 1 cent per KWh. The total impact of storage on the cost of sun or wind energy seems thus under 2 cents. The total average cost of wind and sun energy generation trans- port and storage may then be close to 10 cents per KWh. This storage will be also very favourable to the grid operation, and to safety, and to peak capacity. The total worldwide areas (onshore and offshore) for 2,000 GW storage will be well under 50,000 km2, to be compared to 500,000 km2 for the reservoirs for existing dams.

Almost all countries have a large amount of wind or sun and/or hydropower. During rainy seasons, reduced sun energy will be balanced by much more hydropower.


World GDP, presently close to 50 trillion USD per year, may reach 150 trillion by mid century. The necessary investments for energy will be important during the next decades, especially if renewable energies are favoured. The extra invest- ments before 2050 for the solution B above, as compared with solution A, are very roughly evaluated below:

  • For solution B, a 20,000 TWh/y extra production of wind and sun energy will require 8,000 GW of generation plants, 2,000 GW of pumped storage plants and some transport facilities.

Wind and Sun
8,000 GW x 1.25 billion USD/GW = 10 trillion

2,000 GW x 1,5 billion USD/GW= 3 trillion

Grid = 2 trillion

Total = 15 trillion

  • But solution B reduces investments for thermal plants, CO2 storage, and for gas and coal mining and transport. The evaluation is complex, but it is likely that the extra investment of solution B will be under 10 trillion USD in 30 or 40 years, i.e. 300 billion USD/year, less than 0.5 per cent of the world income (or less than 50 USD/year per capita). For most countries, this investment will avoid or reduce the risk of a very high cost for all fossil fuels (including uranium) after 2030; at this time 80 per cent of the world population would need to import them from few countries.

Capture d’écran 2013-07-16 à 14.11.25



Over half of the energy supply will be through electricity, of which 80 per cent renewable, 10 per cent nuclear, 10 per cent thermal.

Most heating and hot water in buildings will be obtained directly by solar thermal and geothermy, or from biomass energy, which may also have a great future in road transport.

Oil will likely be essentially used for road transport and chemistry needs. The fossil fuels necessary for 10 per cent of electricity and for part of heavy industry needs may be gas, or possibly coal if associated with CO2 storage.

The overall share of fossil fuel may thus be reduced to 25 per cent. It would be 50 per cent without storage of electric energy.


Before mid century, the world will use three times the energy resources it presently requires. Most should be from renewable sources, and mainly through electricity. As the main renewable electricity sources are intermittent, some energy storage will be necessary: onshore or offshore pumped storage plants based on well proven technologies are a relevant cost-effective solution. Further needs that evolve as the century progresses might be met from solar energy or new nuclear solutions.

Beyond their impact on climate change, large investments in various renewable energies and relevant storage may well be economically justified in most countries. This would also reduce the huge risk of conflict for fossil fuel control.

Printable version Printable version

Back to top