Opinion Pieces  
21 February 2017

Energy forecasts and planning vs. reality. The rational debate

Anthonie Cilliers, nuclear engineering specialist, discusses the predictions and modelling required to consider the proven track record of renewables in energy forecasts.

In the past week, I read with interest how Anton van Dalsen asks if a rational debate is possible on South Africa's nuclear plans (www.politicsweb.co.za).

Anton claims that, “without going into a detailed technical discussion, we can refer to the situation in Ireland, where the grid operator controls a system with 60% renewable energy (i.e. non-synchronous) today, with plans to go to 75% by 2020. South Africa will not be at these levels for 20 years. There seems to be a fixation on ESKOM’s (sic) side with nuclear power, coupled with an unwillingness to consider alternatives, which have a proven track record and which are not expensive.”

Maybe I am harsh on Anton, his opinion piece is just one of many I have read over the last couple of months. All of them claiming that renewable energy has a proven track record. Where a track record lacks, the CSIR has done statistical modelling aiming to prove that a bulk of wind and solar, augmented by limited gas turbines would be enough to supply the needs of the entire electrical grid in South Africa, all be it many years from now.

The models claim that a combination of wind, solar PV and natural gas would be the cheapest option. Of course, for this model to work, you need to limit the installed capacity of natural gas turbines to a minimum and prove that they will only be used in very special occasions, wind and solar need to be available all the time. Otherwise the combined cost of gas and renewable becomes your new base price for installed capacity.

The CSIR claims that spreading these sources geographically, limits the intermittency and unpredictability. I know something about statistics; in the nuclear industry we use extensive Monte Carlo modelling when calculating all nuclear reactions. When billions of interactions happen in one second, the average of a second can accurately be modelled, still you can never ever predict the outcome of the next interaction. This is also the case when predicting weather patterns - the driving force behind wind and solar (unfortunately, the weather doesn’t allow billions of interactions to average out, so accuracy would start off being questionable at best).

In such a model, the average wind over a year can be predicted (when it comes to rainfall we know this is rather inaccurate). It is even possible to predict the wind for certain months to an acceptable degree of accuracy. Predicting it for a week, or a day, becomes close to impossible. When spreading these sources geographically it may be said, the wind will at least blow somewhere, and the whole country will not be cloudy at once. This could be right, statistically it could even be shown that this could be true most of the time.

However, these statistics do not tell you anything about any instance in time and for this reason I would venture as far as to say, that wind and solar energy sources should in all cases be matched, kw for kw, with a reliable backup energy source, or energy storage system up to the maximum of the grid demand. If such a grid cannot supply the required demand for a single hour in a year, this would result in more than simply load shedding. It is impossible to statistically predict a minimum (above 0kW) wind or sun availability at any time although weather statistics may allow averages over time.

The biggest drive for renewable energy is climate change. Every article I read on the topic asks us to think about the environment.

I digress, having discussed predictions and modelling we need to consider the  proven track record Anton van Dalsen is talking about:

My question would be why is a debate necessary, when we can simply look at the facts of reality? Nothing provides more rationality than facts, right? If Anton would like to use Ireland as a prime example, let us check those facts.

In this regard, I found a wonderful website: www.electricitymap.org. I would urge anyone to access it as it provides a wealth of information on the European electricity grid in an amazing visual way. This article cannot do justice to all the comparisons and information to be extracted from it. The site provides a real-time map of Europe indicating how energy is being produced in each country, who is importing and exporting electricity, from whom and who is producing the most CO2 at any given time. The European grid is probably the most interconnected grid in the world, so it provides a very relevant benchmark.

From this information, I listed several interesting European countries from the lowest carbon emitting, to the highest. Europe has many more countries, and all of them are interesting, so for those not listed, the website is free to access. Below, the energy sources for each country are listed, along with the installed capacity as well as the maximum produced during a 24-hour period by the various sources.


Norway is a Scandinavian country with an enormous potential for hydro power generation. For this reason it is possible for Norway to produce clean energy while being the 11th cheapest producer of electricity. When it comes to wind, Norway has limited installed capacity, but manages to utilise up to 62% of the installed wind capacity when running at its peak.

  • Wind – Installed: 856MW, max: 218MW
  • Hydro – Installed: 36,5GW, max: 19,2GW
  • Gas – Installed: 1,78GW, max:383MW
  • CO2 – 34g/kWh
  • Price – R1,30/kWh


Like Norway, Sweden has large amounts of hydro potential. Sweden also opted for nuclear power, utilising 100% of its installed capacity (50% of entire grid) continually. Sweden also has wind capacity, supplying up to 57% of the installed wind capacity at peak. An unknown energy source on the grid pushes the carbon emissions up to an average of 60g/kWh. Electricity price in Sweden is the 2nd cheapest in Europe behind Bosnia and Herzegovina.

  • Wind – Installed: 5,96GW, max: 4,03GW
  • Hydro – Installed: 15,9GW, max: 7,67GW
  • Nuclear – Installed: 9,1GW, max: 9,1GW
  • Unknown source – Installed: 1,41GW, max: 1,4GW
  • CO2 – 60g/kWh
  • Price – R1,08/kWh


France opted to install a rather large fleet of nuclear power plants in the 1970s to ensure energy security when prices of oil and gas started rising. Today 70% of electricity in France comes from nuclear. Whereas, wind, solar and hydro supplies limited electricity, but when these sources are supplying, gas and oil are utilised to make up for intermittencies. For this reason, carbon emissions rise during these times. France comes in at number 22 for cheapest electricity in Europe.

  • Wind – Installed: 10,4GW, max: 1,76GW
  • Solar – Installed: 6,58GW, max: 3,63GW
  • Hydro – Installed: 23,5GW, max: 7,2GW
  • Nuclear – Installed: 63,1GW, max: 47,1GW
  • Gas – Installed: 10,9GW, max: 8,51GW
  • Coal – Installed: 2,93GW, max: 2,03GW
  • Oil – Installed: 6,67GW, max: 295MW
  • CO2 – 101g/kWh
  • Price R1,67/kWh


Another Scandinavian country, Finland opted for more of an energy mix, but might have limited hydro capacity compared to their neighbours. Finland relies on wind, hydro, biomass, nuclear, gas, coal and oil for electricity. Nuclear power is the only source that runs constantly at maximum capacity. Again, gas and renewable sources are interchanged as necessary.

  • Wind – Installed: 1,43GW, max: 507MW
  • Hydro – Installed 3,21GW, max: 1,25GW
  • Biomass – Installed: 2,87GW, max: 1,43GW
  • Nuclear – Installed: 2,78GW, max: 2,78GW
  • Gas – Installed: 1,8GW, max: 822MW
  • Coal – Installed: 2,85GW, max: 2,78GW
  • Oil – Installed: 1,43GW, max: 0GW
  • CO2 – 190g/kWh
  • Price – R1,18/kWh

Northern Ireland

Northern Ireland is rather small with an entire grid capacity of 863MW. 38% of the demand is supplied by the rest of Great Britain. It is, however, interesting to note that the carbon emissions from this country is more than double that of France, and it is only supplied by wind and gas. Electricity prices for Northern Ireland comes in at 37 out of 41 countries.

  • 38% of power from Great Britain
  • Wind – Installed: 666MW, max: 270MW
  • Gas – Installed: 659MW, max: 197MW
  • CO2 – 226g/kWh
  • Price R2,25/kWh


Now we get to the example Anton used. I might add that many other rational debate supporters like to use the example of Ireland. Ireland utilises wind, hydro, biomass, gas, coal and oil to supply its electricity requirements. It is important to note that 66% of the Irish grid is supplied by fossil fuels (not the other way around as Anton claims), which shows up in the carbon emissions (more than 3 times that of France per kWh). The reason for it being the 34th most expensive electricity supplier in Europe is open for a rational debate.

  • Wind – Installed: 3,03GW, max: 1,34GW
  • Hydro – Installed: 508MW, max: 217MW
  • Biomass – Installed: 344MW, max: 327MW
  • Gas – Installed 3,8GW, max: 1GW
  • Coal – Installed: 855MW, max 782GW
  • Oil – Installed: 811MW, max: 421GW
  • CO2 – 365g/kWh
  • Price R2,10/kWh


When it comes to Germany, I find the situation comical at best, ironic on any given day, and truly heartbreaking the rest of the time. Germany will phase out all nuclear power by 2022. This was a political decision based on irrational fears of nuclear power. When facts state that nuclear power has the lowest mortality rate (World Health Organisation) of any energy industry, with all nuclear accidents included.Nuclear contributes insignificant portion of the “non-natural” radioactivity to the environment (0,1% compared to the phosphate industry – 55%) and a probability of a nuclear accident currently lower than that of a meteorite striking the earth. I wonder how the plans are doing building that meteor shield?

In any case. Germany has embarked on a massive rollout of renewable energy over the last couple of years. To the extent that they now have a total installed capacity of more than 80GW of solar and wind sources. On a given day they could supply up to a combined maximum of 32GW of electricity from these sources. Fossil fuels still provide 33% of the supply, with coal supplying most. Why they don’t supply most of this electricity from gas (being the slightly less carbon intensive source) is uncertain. Costs of electricity seem to be the only plausible reason, with Germany supplying the second most expensive electricity in Europe behind Denmark.

Denmark utilised only wind, gas and coal with carbon emissions of 328g/kWh. It is important to note that still more than half of the demand is supplied by reliable energy sources – even with a massive 80GW installation of wind and solar. Germany produces a massive 427g of CO2 per kWh, more than 4 times that of France and more than 7 times that of Sweden – a nuclear country. Still kw for kw, the entire grid demand can be supplied without any renewable energy sources.

  • Wind – Installed: 43,2GW, max 18,7GW
  • Solar – Installed: 39GW, max: 13,3GW
  • Hydro – Installed: 12,9GW, max: 4,3GW
  • Biomass – Installed: 8,19GW, max: 5,1GW
  • Nuclear – Installed: 11GW, max: 6,52GW
  • Gas – Installed: 26,8GW, max: 1,58GW
  • Coal – Installed: 50,7GW, max: 24,6GW
  • Unknown source – Installed: 5,42GW, max: 5,42
  • CO2 – 427g/kWh
  • Price – R2,76/kWh second most expensive after Denmark

So, looking at the facts, I believe it is time we become brutally honest with ourselves. South Africa can follow the route of Germany, we could install 80GW of renewables on our grid, we could even double or triple it as the CSIR recommends. We will still not make a dent in what we are trying to achieve.

I probably don’t need to summarise these facts, but for clarity: the cleanest energy sources are demonstrated to be hydro and nuclear, not anything else. If one cares about the environment, this is what should be supported.

When augmenting energy sources with natural gas, it is important to note that it poses a very large risk to energy security and price fluctuations – gas is cheap now, how long will it last?

The most expensive electricity demonstrated come from countries with large wind and solar installations. If one cares about one’s pocket, stay away from these. For more rational debate and thinking it should be guarded against quoting other people, rather visit the electricitymap.org website (or any other such resource), look at the information and make rational arguments. Information is often more reliable than personal opinions of the information.


Anthonie Cilliers Africa Nuclear Africa webinar ESI AfricaAnthonie is a Nuclear Engineering specialist with a PhD Nuclear Engineering and M.Eng (Computer & Electronic) degrees. He works at the North-West University as Programme Manager: Nuclear Engineering and as research lead on numerous projects. He also specialises in Nuclear Knowledge Management where he is the initiator and coordinator of the national academic network, “South African Network for Nuclear Education, Science and Technology” (SAN-NEST).

Share This Article
Join the Discussion

Leave a Reply

Notify of
Sort by:   newest | oldest | most voted
Matthew Turner
Hi Anthonie, Not to be too harsh on you, but did your information on cost of energy generation consider when the plants were installed? It’s pretty rational that nuclear and hydro plants that were commissioned in the 1980’s will be largely depreciated by now, and thus will be “less expensive” than renewables (wind or solar) commissioned in the past 5 years. This seems to be something that nuclear proponents like you conveniently ignore. My questions to you: 1. If nuclear really is the solution (answer: it isn’t!), why is private sector capital not being invested in new nuclear builds? 2.… Read more »
Ashley Theron
Dear Matthew, thank you for not being too harsh on me, you pose very valid questions and I am sure other readers may also be interested in the answers. 1. As we all know, nuclear power in its current form has a large capital cost attached to it, not because it is expensive, but because they are built in large power plants with 1,000s of MWs that it produces, the plants also run for 60 to 80 years, so the return on investment in financing terms could be set up come after 20 years with 40 to 60 years left… Read more »