One third of electricity demand supplied by sun, wind and water (on average, that is)

Just completing a post that I started writing quite a while ago, but lay around unfinished. Central in this post is a tweet with this message (translated from Dutch):

An average of 1/3 of the total electricity demand in the Netherlands is now supplied by sun, wind and water. The record is from Sunday 24 April, with a nice 68%.
The low of last winter was on November 30, with only 4%.
#graphoftheday

It was accompanied by a graph showing the daily energy production by solar, wind and water as percent of total demand of the Netherlands:

Tweet BM_Visser 2023-02-11

The tick yellow line is the four weeks moving average and, indeed, it ends up at roughly one third of demand at the beginning of 2023. That is however only part of the story, as also hinted by the two values that are mentioned in the text of the tweet.

Although the average ends up around one third of demand, it is derived from a incredibly wide range. According to the tweet, solar and wind together with water produced between 4% and 68% of total demand in 2022 in the Netherlands.

There is also something in this graph that drew my attention, but it is not that clear from that graph. Unfortunately, I don’t have the data from the Netherlands. Luckily, this dynamic is not unique to the Netherlands, it is exactly the same in Belgium and the Belgian data is readily available.

This is a recreation of that graph using the Belgian data (only solar and wind, water power in Belgium is negligible):

Chart 0023a Belgium: solar and wind share of daily demand

The Belgian figures are close to that of the Netherlands, albeit a bit lower. The average share of Belgian solar and wind as percent of demand is roughly one fourth (compared to one third of the Netherlands). The range of the Belgian data is also somewhat smaller (between 1 and 58%) compared to the Netherlands (between 4 and 68%). The overall shape is however similar. There is the same funnel shape that is widening the more capacity is added.

Now it is easier to highlight a bit more the wide range that this average is derived from. These are the minimum and maximum values of the share of solar and wind in demand of each year:

Chart 0023a Belgium: solar and wind share of yearly demand

It is clear that the lower and upper boundaries don’t increase in the same way The lower boundary is hardly budging, it keeps close to the x-axis over the entire period. In 2022, the lowest daily share supplied by solar and wind was only about 1% of total demand. This didn’t change much over the years: it was roughly between 0.7% and 1.8% of demand between 2014 and 2022. This tells us that a lot of dispatchable capacity will still be needed at specific times of the year (in this case, pretty close to the expected demand and, looking at its shallow slope, that might be the case for quite a while).

The upper boundary behaves different. It shoots up exponentially. In 2022, the highest daily share supplied by solar and wind was 58% of total demand, coming from around 20% in 2014.

Meaning that the difference between the lower and upper boundary will keep increasing over time. Basically, electricity production by solar and wind will at times start to exceed demand, while the need for backup at specific times of the year will stay high.

Some people also seem to recognize this type of dynamic. Already the first comment below the tweet nails it (translated from Dutch):