This study doesn't correct for baseline exponential decay due to inflation, to better highlight the meaningful variations. By comparing based on 1914 dollars it also causes old variations to be relatively more extreme and newer inflationary events to look less extreme. You must compare apples to apples.
Finally the events are quite cherry-picked. It is a conclusion looking for a result, when the statistical reason for choosing those 4 events simply isn't evident when you look at the data itself. There is no mathematical rule you could apply to your dataset that would distinctly highlight those 4 periods.
Batteries are an expensive solution that doesn't scale well at the grid level. It is useful for grid stability (fast frequency response) but simply a non-starter when you're dealing with national grids.
Batteries are an added cost to the system, without producing more electricity, and as a result prices will go up.
A far cheaper source of flexibility is Demand Side Response. Particularly data centres that are willing to be market actors. Compute can happen anywhere, so it should happen where the wind blows and the sun shines. It is cheaper to transmit bits than Megawatts.
Gas power generation is a necessary evil to balance out the variability of intermittent energy generation (i.e. wind and solar).
Hydropower isn't a feasible alternative because the easy resources have been developed.
The only alternative source of flexibility available today is demand side response.
Edit: I appreciate the down votes, as I've not explained in detail. It is a complex issue. My opinions are based on having a phd in the topic, 10+ years in control rooms, years of market operations and design, and years contributing to europe-wide risk assessment methodologies.
When you're in a transactional environment, it is expected. For example, if you are a car salesman and someone comes to you with a problem like "what car should I buy?", and your response is, "you should buy X because Y", that is helpful.
Most communications are not contained within a transactional context. If your mother asks "what car should I buy?", responding with "you should buy X because Y" might be 'correct' but what if she was actually reaching out, and that she doesn't actually need a car but it just 'feels' like it'd help?
In such a case you'd be rejecting her attempt at creating belongingness and a chance for empathy ("why do you need a new car mum? what happened?") and instead hit her with a status ("I know more about this than you") and confrontation ("I have resolved what I thought the situation was so now this conversation is over").
This is at least my interpretation of the terminology. If your question wasn't searching for a solution but instead you were reaching out to feel like your perspective belongs, then I apologise if I come across as confrontational, and I do not mean to imply I know more or less than you about the topic.