Austrian Hydro Outlook 2026/27: Reservoir Levels, Weather Risks and the Market Balance for Next Winter | Xiulan He, LSEG

Good afternoon everyone, my name is Xiu Lan, I come from London Stock Exchange Group, I'm a senior power research analyst, and in our power research group we does fundamental modeling like solar, wind and hydro, as well as we do short term and long term price forecast. And I started my career as a hydrologist, so I'm very happy to say today that finally hydro is on spot. So today I'm going to talk about hydro risks and the market balance for next winter. We will focus on Austria. If you notice in the agenda, my topic, there is a reservoir level, but I removed it from my topic and we will find out why. For the content, I will cover Austria's hydro starting points and the winter inflow situation system balance risks. I will finish with the Austria power price outlook. So here's the map that you already see in Mark's presentation. I just zoomed in to Central Europe. It shows the top two centimeters soil moisture. It was updated this Tuesday from NASA satellite product. We can see that across Central Europe is all dark red. And this color means percentile zero for period of 1948 to 2012. So across Central Europe is severe droughts, extended droughts. That's on the historically low level. So now let's look at the other hydro component. To understand this, I will quickly introduce our hydro energy model. How it works? It takes precipitation. It takes temperature. And then it puts this amount of water into different boxes. When the temperature is lower than a certain threshold, it will move the water to the snow box. Otherwise, it will go to this groundwater part. There are two boxes to fill in first. And then whatever left will be the water. So now the inflow. And it is snow water content and soil water content that defines the hydro status. And in a situation right now where we are in severe deficit, whatever precipitation comes in, we need to fill in those boxes first. That means we will not see an immediate increase of inflow with rainfall event. And here I listed the current status of all the Alpine countries. What we can see is Austria is all red. That means everything is on percentile zero for the past decade. It is Austria that is the most hydro stressed market around the Alpine region. For other regions like France, this spring, it got some exceptional rainfall. The inflow is high at that time. And that legacy lasts until today. So the soil water content is not at lowest yet. And the inflow is also drought, but it's not severe. For Switzerland and Italy in early August. The first time it's not at lowest yet. However, for Austria, it's just dry all the way. And now let's zoom into Austria. How did it arrive to this situation? It all starts from the input, right? There's no input. There's no inflow. So here is the effective precipitation. 2026 in red, last year in blue, and 2018 in gray. 2018 is the second driest year in the last decade. So what we can see is since last year, the effective precipitation has been below normal for the whole year. And this year since July is getting even more extreme. And with that, we had a monthly inflow below normal for entire 2025 and 2026. Actually until today, Austria has had 22 consecutive months of below normal inflow. The last time the inflow was higher than normal was in October 2024. And that leads to the low reservoir level. One thing to mention is that temporary drought doesn't lead to a low reservoir level. Like in 2018, the first half of the year was actually quite OK for inflow. So the second half of the year, although the inflow was low, the reservoir level was not low. But last year at this time, at winter, it's already had one year of drought. So last year's reservoir level was quite low. And this year, right now it's at the minimum. Because of this extended drought, it will take much more effort for the reservoir level to recover. Now let's look at the Austria's hydropower system. Here I plotted the Renau River generation and the reservoir generation for the period of past decade weekly data against each of the hydro components from our model. What obvious is that both Renau River generation and reservoir generation are highly correlated with inflow. So Austria's hydropower system is an inflow-driven system. What matters is the water comes into the system, the water left of the system. That matters more than what is stored in the system. So we don't see a clear correlation even with reservoir generation with the reservoir content. There's no clear correlation. And that is why I removed reservoir from my presentation. I think from inflow itself, we can get a very good estimate on the hydropower generation as well as we could use this inflow to quantify the hydro risk. And how will the inflow look like in the next two quarters? We have two types of forecasts. Base case forecasts, what I call it's a long-term forecast. We use our model first to run with ECMWF's monthly forecast. That one covers next six weeks. This forecast has very good forecast scale, but it's only six weeks. And after that, we will plug in historical weather years, run it for up to one year. And another forecast is EC seasonal. It's also from ECMWF. It published only once per month. It covers seven months horizon. So it just covers negative quarters. Very good. However, this EC seasonal forecast scale is much worse than the monthly. We shouldn't trust it that much. But it's good to give a regional climate anomalies. And we can see here, this is the pink one is from the EC seasonal forecast. It looks like next year, Q1, the inflow could be higher than normal. I keep it just as a benchmark for a VAT scenario. But we should really remember that it's very high uncertainty around here. I'd like to stick to the long-term forecast. So for Q4, the inflow will be below normal and will be lower than 2018 situation. And for Q1, it will still be below normal, more or less in line with 2026 level. Here we can see a comparison of inflow and the hydro generation. On top here is for Q4. We see historically the light blue one is the share of Reynolds River generation. The dark blue one is the share of Red River generation. And the light, the green line is the inflow through the years. So it's quite clear that there is a correlation there. And if we look at the long-term forecast for Q4, the level will be around here, a historical low. The hydro generation share will probably lower than this year, lower than 40%. And when it comes to Q1, according to the long-term forecast, base case forecast, the inflow will be around here. So hopefully the generation will come back to 2025 level, which is still below 40%. For the history monthly generation mix, I highlight the last winter here. It was already a winter with low hydro production. We can see Austria's system is heavily depends on the thermal generation and import. And this year already from June to July, the hydro generation, the blue one had a huge deduction from 73% to 48%. And already from July, we see a big part, it depends on the import. Well, in normal wet summer, June, July, August is normally the, we see Austria have larger amount of export. Now let's look at the price. Here I plotted the hourly price profile, monthly average. An Australian red and the German blue on top is 26, below here is 25. What we see clearly is in the summer season, there is a price coupling. Both German price and the Austrian price follow each other quite well. And in the winter season, January, February and October onward, we see a lift of the price in both countries. But Austria increased much more. They are the premium of Austria price against German price. And the market behind it is probably during winter season. There is low hydro production, the water value get high. When there is low hydro production, the system depends more on the thermal generation, on the import from Germany many. And when it comes to import, there is congestion problems. All this lifted the Austria price higher. And this year already from August, we see that the Austria price taking off. So it's really drought driven this premium. For this winter's Austria price forecast, I will implement a method that uses our forecasted German price plus our forecasted Austria premium. And this premium, I will use inflow to forecast this premium. And that is because we see a very good correlation of inflow and the Austria premium. Here is the monthly price spread of Austria and Germany through the years. And here the green line is the inflow, monthly inflow, and the black line is the gas SMC price. We see clearly that in past two winters, when the inflow is low, there is a high premium. And in summer 24, when the inflow is high, Austria is trading at the discount price against Germany. So I took the winter months, a plot inflow against the price spread. We get a very good fitting. And with this fitting, together with our forecast inflow, we could get an estimate on the price premium. I tried to do the same for the black one, the gas RMC, but we don't see any good correlation. So it's clear that this premium is driven by inflow. Here are the forecasted premium. This one is using the base case inflow, which is the green one, base case inflow, for the next two quarters. And the blue bar is the market price premium for the next two quarters. The pink one is the forecasted one. Similar here, we are using a slightly wetter scenario of the inflow. What we see here is, for short term, October to December, our base case forecasted premium is around 21 to 24, and the market is around 24 to 25. So the market premium already largely reflected this hydro risk. What difference is when it comes to January, February, the market is around 21. Market premium is around 21. Where our base case shows that the premium could go as high as 26, 29. So this is where the market has underestimated the premium. Or let's say the market is expecting a faster recovery of the inflow. Now let's look at the German winter outlook, the German price. Here's the German generation mix output from our forward model. We see clearly that this winter, German will heavily depend on the renewable generation, especially this dark green one, which is the wind generation. So when the system is so heavily depends on the renewables, the weather becomes the main driver for the price. Especially the wind generation, we know that wind is very difficult to forecast. There is high uncertainty on the forecast. For example, this is the dark line is average forecast to wind generation share. And with the uncertainty spread is very high. Take January, for example. Average is around 45% wind production. But it could be as low as 25% or as high as 75%. So there's a large difference. And if the wind doesn't deliver as expected, this whole 20% have to be covered by thermal generation or by import. That will push the price high. Or if the wind is extremely strong, that will just push the price materially low. That's what we see here when it comes to December, January, February. We have a large spread of the price forecast. On average, our forecast price is the purple one. And the market price is the blue one. The market price follows the gas SMC closers, usually like that. And we are bearish on October, November and February onwards. Generally, we are neutral for the German price. And December, the market price is a little bit bearish. That's because December is a holiday month. During this Christmas season, the demand can be really low. And we saw that in past two winters when the demand is low and some kind of wind change can easily change the direction of the price. So there's a high volatility there. Now we arrive at the Austria price outlook. Like we said, we use the forecasted German price, the average price, plus the forecasted Austria premium, which reflects the hydro risk. Here, the bars are the market price. The gray one is Germany, and then the green one is for Austria. The whisker, the gray whisker is our forecasted German price. And the purple whisker is the forecasted Austria price using the base case inflow. What we see here is it's clear that Germany is the main market driver. Austria price direction follows quite closely with the German price. October, November, bearish for both, and also February, March. December, bullish. We have bullish signals. It's only in January here when the German market is neutral. When we add on that bullish Austria premium, we see a bullish signal here on the Austria price. So key takeaways, Austria enters winter super hydro stressed. And our base case inflow shows that the inflow will stay below normal for next two quarters. The hydro risk is likely underpriced when it's deep into the winter season, January, February. And lastly, Germany is the main market driver. German fundamentals pose the main outright price risk. And thank you. We see that you, you, it seems to be more important than to Austria in the time to come due to the very poor hydro situation. you mentioned also that there could be congestion issues, can you say a little bit more about what kind of congestion issues we're talking about here because German north, German south is different, most renewables come from German north wind and those things also the line itself has this capacity issue and when the German market itself is saturated or it's tight, there could be like not this much that can be exported to Austria and that same problem is also like we saw in Italy north actually because Austria is also important from Italy north but that price there is much higher So how nervous should this room be when you are showing these results? Good question The situation probably slightly worse than last winter and I don't want to scare you guys but there is weather forecast showing that in January, February the temperature could be colder than normal That's also a bad signal, right? Because when the temperature is cold whatever precipitation comes in will go to the snowbox It will not be inflowed immediately Meanwhile there is a high heating demand So that will create some kind of crisis Christian? Do we have any questions from the floor? If that's not the case then I'll take over Well you said that it looks like the market is currently still underestimating the risk in January and February Is that because of those forecasts or are they just waiting out hoping it will get better? Yeah It depends on our based on our inflow forecast we see that if the inflow really keeps it that low level there should be a higher premium Alright Thank you Shalane Thank you Very good We give her a big hand and here is a small present from a suspect Thank you so much Christian Alright