Solar in Chile

Solar power in 2015 accounted for less than 5% of Chile’s total electricity generation, but because of decreased demand and inadequate grid connections it’s already generating surpluses that have to be curtailed or which result in the power being sold at zero cost . Yet to meet its target of 20% renewable energy from non-hydro sources by 2025 Chile plans to install yet more intermittent solar and wind energy by 2020/21. Development of untapped dispatchable renewables such as hydro – Chile’s cheapest source of renewable energy – and geothermal, both of which Chile has in abundance, is hindered by lack of grid connections and environmental opposition.

First some hard-to-find facts on Chile’s electrical sector. Installed capacity is on the order of 20,600MW, with approximately 33% diesel /gas, 32% hydro, 22% coal, 6% solar 5% wind and 2% biomass/biogas. Peak demand is pn the order of 7,000MW, or little more than a third of total installed capacity.

So far so good, but even at the current low level of solar penetration there are problems. Figure 1, reproduced from the Chilean Ministry of Energy report Chilean Experience in Developing Electric Power Infrastructure , illustrates some of them:

Figure 1: Chile’s four “grid regions”

Chile is divided into four electrical regions based on existing grid networks (SING = Sector Interconectado del Norte Grande, SIC = Sistema Interconectado Central). The SING and SIC grids are presently not interconnected, and plans to connect them are apparently suffering delays. But while they remain unconnected the superior solar resources in SING – reportedly the best in the world – will not have access to major centers of consumption in the SIC, where all the existing solar installations are. And when they do their output will have to be transmitted up to 1,500km to reach them.

The Aysén and Magallanes regions have no grids worth speaking of but contain large amounts of untapped hydro potential and also Chile’s best wind resources. Given the nature of the terrrain in these regions it seems unlikely that major grid connections with Central Chile will ever be feasible. (How bad is the terrain? The developers of the on again/off again 640MW Hidroaysén project at the northern end of the Aysén region are/were considering shipping the power out through a submarine cable.) The best geothermal resources tend to be located in remote and inaccessible areas high in the Andes.

And at present Chile has no grid connections with its neighbors Argentina, Bolivia and Peru. To all intents and purposes it’s an island.

Chile also has an interesting electric generation history, which is summarized in Figures 2 and 3 (from the same source as Figure 1). If nothing else it highlights the dangers of relying on imported fuels:

Figure 2: Chile’s capacity growth since 1990. (Capacities are lower than those quoted elsewhere, which makes me suspect that gross capacity=derated capacity.)

The seminal event here was the progressive cutoff in gas imports from Argentina, which began in 2004 and prompted the increase in diesel and coal capacity. The impacts on price are shown in Figure 3:

Figure 3: Chile’s generation by source and generation prices since 2000.

After years of cheap hydro generation Argentinian gas imports finally decreased to zero in 2007 and the gap was filed with diesel generation. As a result electricity generation prices abruptly rose from historic levels of around $20/MWh to $260/MWh in 2007 and to $350/MWh in 2008. Since then prices have gradually declined to an average of around $100/MWh as diesel generation has been replaced with gas and coal. (Where does Chile get its gas from now? From Equatorial Guinea, Trinidad and Tobago and Qatar, not Argentina.)

Now we will take a look at solar.

Five years ago there was no solar in Chile. But since then Chile has jumped on the renewables bandwagon and passed laws to encourage renewables development which are summarized by IRENA as follows:

Chile has a target to generate 20% of its electricity from renewable sources by 2025 (excluding hydro plants larger than 20MW). This target was established in 2013 by Law 20698, better known as “Law 20/25,” and updated a previous target of 10% renewable electricity by 2024 …… The 2014-2018 Energy Programme aims at achieving a 45% renewable energy share for new electric capacity installed between 2014 and 2025.

(Why are hydro plants larger than 20MW excluded? Maybe for the same reason that California excludes pumped hydro, which it believes would limit the growth of more innovative storage technologies, or maybe because of anticipated public opposition to anything much larger than a water wheel.)


A quota obligation is Chile’s main support scheme for renewable electricity.

Under the quota obligation system Chile has put megawatts out for bid, and in the most recent 2015 auction (for 1,200MW) solar and wind swept the field :

Renewable energy technologies have made a clean sweep at Chile’s latest energy auction, with five global wind and solar companies awarded 20-year contracts to supply 1,200GWh to the Chilean electricity market from 2017. The five companies – including Abengoa, Aela Generacion, Ibereolica Cabo Leones, Amunche Solar SpA (a subsidiary of Spanish Solarpack) and First Solar (via SCB II) – offered an average bid price of $US79.3/MWh, the lowest average price per MWh for a local energy auction since 2007. The most remarkable result came in with a bid of $US97/MWh for “overnight” solar. This came from Spanish group Abengoa. It still beat all the coal and gas alternatives. All of the bids for wind projects came in at $78-$95 per MWh, and solar PV came in as low as $US65/MWh.

And now the energy ministry plans more and larger auctions:

The energy ministry said it expected similar or lower prices for the next auction in April 2016 (now delayed for 90 days), for annual supplies of 12,500GWh, (which translates roughly into 3,000-4,000MW installed) for delivery 2020 and 2021. It is hoped that this trend towards cheaper renewables could halt Chile’s increase in electricity bills within three to four years and bring a reduction thereafter.

This of course is a forlorn hope. Solar and wind auction prices are not that much lower than current generation prices and as always they make no allowance for energy storage or grid upgrades. Figure 3 clearly shows that Chile’s cheapest source of electricity is hydro. Nevertheless, the impact of the auctions on wind and especially solar growth has been spectacular. Figure 4 shows the status of renewable capacity as of May 2016, starting from zero MW in 2011:

Figure 4: Status of renewables projects in Chile

Chile now has 1,113MW of solar installed and another 2,121MW under construction. Added together these will account for 14% of Chile’s installed capacity. And then there are another 11,549MW of RCA (Resolucion de Calificacion Ambiental, or environmentally approved) projects hanging fire. What if these projects win the lion’s share of the 2016 auction, say 3,000MW? Then solar will make up over 25% of Chile’s total installed capacity.

But solar power is already being curtailed at the current low level of solar penetration:

Carlos Finat, executive director of the Chilean Renewable Energy Association (ACERA), told PV Tech that there are two major reasons for the curtailments. Firstly, the Chilean power system is weak and attempts to resolve the requirements of the transmission system have experienced major delays. The second major reason for curtailment is that for security and flexibility reasons, CDEC has to constantly run at least three units (around 75MW) from the 152MW Guacolda coal-fired power station operated by Chilean power firm AES. Finat claimed that the possibility of curtailment was very clear and public to companies looking to invest in Chile at an early stage. He added: “They were able to foresee that this would occur. The capacities are already known with the studies that they must have run. It will probably take a bit longer, but my feeling is that many companies have already considered this effect in their business plan. Of course it is not a good situation. Of course probably it will take longer than expected, but it is a manageable situation.”

But will it be manageable if Chile’s solar capacity expands by a factor of five by 2020/21? I doubt it. Chile should be pursuing hydro if it wants the lower electricity rates and energy security its consumers are clamoring for. On the other hand there is considerable environmental opposition to the construction of new hydro plants in pristine areas, which explains the on again/off again history of the Hidroaisén project. Chile clearly has to make up its mind whether it wants pristine areas or cheaper electricity, or at least strike a balance between the two.

But for the time being solar leads the field. Yet despite its low level of penetration it has already reached the point where El Hierro-type solutions to smooth out solar intermittency are being considered:

Valhalla Energia plans to build a hydroelectric power generation plant in Chile’s parch ed Atacama desert, the world’s driest, by using solar energy to pump sea water up the side of a 600 meter cliff and then have it rush back down to the Pacific Ocean below. The $400 million project will use solar energy to pump seawater to the top of a coastal cliff, where it will be stored in natural depressions. At night, electricity will be generated by releasing the water and letting gravity do the rest, the company said. Valhalla is still awaiting regulatory approval though for its 600-megawatt, $500 million Cielos de Tarapaca solar panel energy project, which will provide the energy needed to pump the seawater up the cliff wall. “We found these natural depressions that we believe were very ancient lakes, but obviously there is nothing there now, it is a desert, that will allow us to store water,” the company’s co-founder and chief executive Juan Andres Camus told Reuters. If the natural depressions are filled to capacity with seawater, the project can continuously provide hydroelectric energy for a little over nine days, Camus said. “That adds a lot of value from the point of view of system’s security.” Construction on Espejo de Tarapaca is expected to begin in the second half of 2016 and commercial operation is slated for 2020.

Figure 5: The Valhalla pumped hydro project

Hopefully it will work better than Gorona del Viento.

This entry was posted in Energy and tagged , , , . Bookmark the permalink.

34 Responses to Solar in Chile

  1. Greg Kaan says:

    Further from the Reuters article:

    Chilean regulators approved the environmental impact study for its 300-megawatt Espejo de Tarapaca project, located 100 kilometers south of Iquique in northern Chile

    So they don’t consider the Atacama desert a pristine area?

    A little over nine days at 300 MW is only 65 GWh – it doesn’t seem much for $400 million (presumably US$)

    • Peter Lang says:

      Hi Greg,

      How did you calculate the 65 GWh energy storage capacity. To calculate the energy storage requires a figure for the value of water stored in the smaller of the two reservoirs (in this case the upper reservoir) and flow rate at maximum power (or volume, hydraulic head and energy efficiency).

      • Greg Kaan says:

        Hi Peter

        I made the calculation simply by assuming that the figures provided for the system in the Reuters article were correct – 300 MW x 9 (days) x 24 (hours) = 64.8 GWh so rounding up for “a little over nine days” gave 65 GWh.

        I am assuming that the system can “continuously provide hydroelectric energy” at maximum stated output which could well be wrong be I was just trying to ballpark the scale of the solution.

        • Peter Lang says:

          Hi Greg,

          I see how you calculated it. However, I think that is not correct. The energy storage capacity is as I explained, is the volume of the smaller reservoir x the hydraulic head less efficiency losses. The capacity factor takes into account how may hours per year you can pump and hours per year x power generated per year.

          • Greg Kaan says:

            It wasn’t meant to be a definitive analysis of the proposal – just a ballpark (maximum) estimate to get a feel of how much energy would stored.
            It would be seem to be of the order of Lake Mackenzie in Tasmania which is pretty small in the scheme of Hydro Tasmania’s reservoirs

        • Peter Lang says:


          To clarify my point. There is no relationship between generating capacity and energy storage capacity. They are totally independent of each other.

        • robertok06 says:

          “300 MW x 9 (days) x 24 (hours) = 64.8 GWh so rounding up for “a little over nine days” gave 65 GWh.”

          Hello: I would naively cut in half your estimate… you cannot use pumped-hydro 24h/24… they use the same reversible pumps up and down-hill, so at most 12h pumping and 12 hours generating… or maybe I miss something here?

          • OpenSourceElectricity says:

            Capacity is 55,000,000m³*600m =90GWh according to available DAta of the system, much bigger than neccesary for the diurnal storage. Maybe they plan extensions in the future.

    • Alex says:

      That’s a good price. Imagine scaling it up 50 fold to $20 billion, to provide 3.2TWh of storage. The UK would love to be able to get that.

      It shows perhaps what you can do if you have the right terrain and solar resource.

    • So they don’t consider the Atacama desert a pristine area?

      No they don’t. Since the nitrate boom of the 1800s the Atacama has historically been considered a place one goes to get resources. Most of Chile’s major copper mines are located in or next to the Atacama and most of the power demand comes from these mines. Some solar plants have reportedly signed contracts to sell power to the mines, but there’s no way a mine is going to work 24/7/365 on solar power alone.

  2. Peter Lang says:

    here’s a rough estimate of the capital cost per watt average power supplied from the PV installation plus pumped hydro storage:

    say $3/W for pumped hydro
    Say $1/Wp for 600 MW solar PV installation (their claim)
    Need say 5 times peak capacity over build of PV to achieve say 40% capacity factor from the system (PV + pumped hydro). [Note: must be capable of supplying reliable 40% capacity factor through worst periods of winter].

    Total cost for 40% capacity factor = $3/W + 5 x $1/W = $8/W
    Cost per average W supplied = $8/ 40% = $20/W average

    c.f. nuclear at say $6/W and 80% capacity factor = $6/80% = $7.5/W

  3. Euan Mearns says:

    One thing I’m increasingly mystified about is how Green Thinkers think. It seems like renewables means wind and solar and that’s it. Why would you consider a salt water reservoir in the Atacama desert that is presumably a million miles away from market while opposing traditional hydro schemes. And why is it OK to pave wilderness with PV panels?


    $400 million
    300 MW

    Coire Glas (Scotland)

    £800 million
    600 MW

    Its interesting to note that this proposal is to balance 600 MW of solar that costs $500 million. So one would need 2 Atacamas for the balancing act. We can conclude that converting solar to dispatchable power will treble the cost assuming the Atacama has storage capacity to cover the diurnal cycle.

    • Helmut says:

      Cometition for power production in this region are Diesel, open cycle gas turbines, and dry cooled coal power or CCGT power stations, all of them expensive per kWh (>10ct/kWh approximately)
      Hydropower would be located further away so would also need big grid extensions to connect the north grid with the central grid.
      Economic reasonable would then be to build the hydro power stations with double turbine and generator sets, to provide power to the north during the night, and to close hydropower down and import cheap solar power from the north. Solar power with a average capacity factor somewehere around 30% in the atacama desert can become quite cheap per kWh.

  4. gweberbv says:

    Does anyone has a deeper knowledge how these auctions in Chile work? For how long do the contracts run? And are the auctions split into peaking time and baseload and/or summer and winter?
    I expect with electricity prices reaching up to 200 $/MWh a lot should be possible for PV and wind.

    I also wonder what happened in 2006 that gave rise to an increase in electricity prices by roughly a factor of 15 two years later (Fig. 3). Must have been interesting times.

  5. Thinkstoomuch says:

    Roger thank you for more things to think about.

    An Appreciative,

  6. For those interested in capacity factors. If I interpret the last table in the “Chilean Experience in Developing Electric Power Infrastructure” document correctly the numbers assume for planning purposes are:

    Small hydro: 63%
    Wind: 34%
    Large hydro: 61%
    Solar PV: 33%
    Solar CSP: 52%
    Geothermal: 85%

    • Alex says:

      Those figures make sense for Chile but for large hydro and perhaps Geothermal, the capacity factor is “elective”. They can choose the capacity factor from hour to hour to help out variable demand and supply. A low capacity factor can be a good thing in this instance.

      The solar CSP capacity factor is a function of the storage size relative to the collector size. It’s a design choice, and I’m not sure how useful it is as a measure.

      Quite a few wind advocates dismiss capacity factor as being meaningful, as it can be manipulated by reducing generator size. However, that tends to have a cost and the people doing the dismissal are usually trying to defend German onshore wind (CF ~17%).

      That solar PV figure of 33% is probably the highest in the world. The Atacama is often cited as the area with the least rainfall.

  7. Some 500+ miles to the north of the Atacama Desert (Chile) is Lake Titicaca, high in the Andes, bordering Peru and Bolivia, which is the largest freshwater lake in South America and the highest of the world’s large lakes.

    I suggest that Lake Titicaca could be used as a natural upper reservoir of the mother of all pumped-storage hydro-schemes.

    Presumably the South Americans would always wish to retain the freshwater natural resource of Lake Titicaca, which would rule out using the sea as the lower reservoir and necessitate the construction of artificial lower reservoirs and rule out ever draining all the water out of Lake Titicaca, but only some fraction of it.

    Nevertheless to illustrate the hydro-electric energy storage capacity of Lake Titicaca for pumped-storage schemes, one can calculate the gravitational potential energy of all the water in Lake Titicaca with respect to sea level, as follows.

    Water volume 893 km3
    = 893 x 10^9 m3
    = 893 x 10^12 litres
    Water mass = 893 x 10^12 Kg
    Surface elevation 3812 m
    Average depth 107 m
    Average elevation of water = 3812 – 107 = 3705 metres

    Gravitational potential energy of water in Lake Titicaca with respect to sea-level
    P.E. = m x g x h
    = 893 x 10^12 Kg x 9.81 m/s^2 x 3705 metres
    = 3.25 x 10^19 Kg m^2/s^2 (Joules) (Watt-seconds)
    divide by 3600 to convert to Watt-hours
    divide by 1,000,000 to convert to MW-h
    9,030,000,000 MWh
    divide by 1000 to convert to GW-h
    9,030,000 GWh
    divide by 1000 to convert to TW-h
    9,030 TWh
    divide by 1000 to convert to Peta Watt – Hours (PWh)
    9 PWh

    – which is greater than the entire electrical energy generated in the United States in a whole year!

    So pairing up the solar power potential of the Atacama Desert with the pumped-storage hydro scheme potential of Lake Titicaca, via a 500+ mile grid connection, should be feasible and a uniquely favourable combination of natural renewable energy resources!

    Scottish Scientist
    Independent Scientific Adviser for Scotland

    • Peter Lang says:

      I suggest that Lake Titicaca could be used as a natural upper reservoir of the mother of all pumped-storage hydro-schemes.

      A few minor issues to deal with:

      1. If its pumped hydro with sea as lower reservoir, sea water would be pumped into the upper reservoir. That is unlikely to be acceptable for environmental reasons.

      2. If it’s not going to use the sea for the lower reservoir, where is the lower reservoir to be located? What’s it’s active volume? What’s the likely rate of water loss from leakage and evaporation?

      3. The hydraulic head is around five times greater than large conventional turbines can handle. You’d need around four pumped storage dams and generating/pump stations in between. And one of them goes out of action and the whole system is out of action.

      4. You haven’t defined how many tunnels, tunnel route and the steel lining requirements. The cost of that would be prohibitive. If you try to use surface pipes what diameter and what steel thickness do you calculate you’d need to hold the pressure and provide the flow rate required?

      5. What’s the capital cost of generating capacity ($/kW) and energy storage capacity ($/kWh)?

      Here’s some examples of some of what’s involved:

      • 1. In general, it might be possible to use fresh-water bags floating in the sea as the lower reservoir. See
        In the case of Lake Titicaca, the sea is about 160 miles away and a lower reservoir on land could be much closer.

        2. For a lower reservoir on land, the most useful low-land lies to the North-East of Lake Titicaca, is about 3 times nearer and gets much more rainfall compared to the less useful three times further and dry low-land to the South-West of Lake Titicaca.

        So for a lower reservoir on land, look to the north east of Lake Titicaca would be my suggestion.

        The active volume and energy storage could be greater than any other pumped-hydroelectric scheme in the world, by orders of magnitude and certainly big enough to meet all the energy storage needs of South America for the foreseeable future.

        I’m not proposing a specific size of scheme, not today anyway, so I’m not even going to attempt to answer here any of Peter Lang’s questions asking about the specific engineering design details.

        3. Whilst I readily and happily concede the “need” for “around four pumped storage” “generating/pump stations in between” the upper and lower reservoirs I reject the absolute need for intermediate dams and reservoirs.

        All that is really needed are intermediate surge tanks to cope with small transient flow mismatches between the generator/pumps operating in series.

        4. I haven’t defined such details because those are scheme-specific and so not for me to define here.

        5. See 4.

        I very much appreciate Peter Lang’s experience and leadership as to what is required to flesh out a engineering feasibility study and costing for pumped-hydro energy storage schemes.

        However, I have not even got the time and resources to answer Peter’s interesting questions about such details for my Strathdearn Pumped-storage Hydro Scheme proposal.

        So my thanks again to Peter Lang. I do welcome anyone posing such valid questions, “to be answered” (by someone, sometime, only not by me, not at this time anyway).

    • JohnF says:

      Frankly, this idea, whatever its technical merits,is completely unfeasible given Bolivia’s historic and ongoing antagonism towards Chile.
      As an example, Chile has just gone to the International Court of Justice in The Hague to counter Bolivian claims over Rio Silala, a river only by name, in fact little more than a stream. Just do a Google image search to see what I mean. Unfortunately in S. America its never technical problems that stop progress.

  8. Andy Dawson says:

    If you’ve not run across “Scottish Scientist” before, he’s prone to some (shall we say) “unusual” enthusiasms, including a vast pumped storage scheme based on Peynn canals using sea water above Loch Ness.

    It’s also worth asking to what bodies he’s actually an ” independent advisor”.

    • Andy Dawson says:

      sorry, that should have read “open canals”

      • “Vast” is right. Up to 6,800 GWh energy storage.

        No, actually, I don’t think it is worth asking ad hominem personal questions about me.

        I advise all readers of my Scottish Scientist blog as an independent and anonymous scientist blogger.

        If I was appointed to advise a particular body I wouldn’t be “independent” of the body that had appointed me, now would I? I’d be an employee of that body, giving them the dependent advice they wanted to get, right?

        I have written “2-way power canal”, possibly the biggest ever canal construction in the world.

        Full details here

        World’s biggest-ever pumped-storage hydro-scheme, for Scotland?

        If anyone has a question about my science I will answer them. Questions about me, I don’t invite and I won’t answer.

  9. jacobress says:

    Since Chile already has installed capacity twice as big as peak demand, they need the solar capacity (with or without pumped storage) like a hole in the head. It’s pure ideology.

    Let the big mining companies in the North supply their own needs in electricity the best they can. If solar makes sense for then – let them do it.

    The Government holding “capacity auctions” while there already is oversupply is pure ideological nonsense.

    • gweberbv says:


      if there is oversupply, how do we explain electricity prices of 150 US Dollar per MWh? In Central Europe, we have also oversupply and as a result wholesale prices dropped to something like 30 bucks per MWh.

      • JohnF says:

        Until the SIC – SING interconnector is built, solar energy will be underutilized as its main market would be Chile’s Central Region, where most of the people are. However there has been a strong and vocal opposition to most of the energy initiatives proposed over the last few years, motivated as much for political and monetary considerations as for the environmental reasons claimed, that’s my opinion anyway. So even this vital interconnection may not happen any time soon. It doesn’t help either that every governmental energy plan is tossed out of the window every four years when a new government takes over.

        Incidentally one of the reasons an undersea power transmission system was mooted for the Hydroaysén project was not so much the terrain; its rather that the terrestrial transmission line ROW would have run through a very large private nature reserve, Parque Pumalín. This had been set up by a wealthy American philanthropist, Douglas Tompkins, to preserve the area in its virgin state. He didn’t like the power line idea, and gathered a lot of local and international support for his cause. The dam project itself had its RCA (environmental approval) rescinded by president Bachelet shortly after coming to power in 2014.

Comments are closed.