The temperature outlooks until the first week of May are much lower than normal in the Nordic region. Read on to see the likely consequences for inflow- and snow conditions in SE1 and SE2.
Key findings
- We see delayed snowmelt in both Norway and Sweden, meaning the short-term inflow outlooks for hydro plants are significantly lower than normal.
- Looking to Sweden, the delayed melting process is significantly more pronounced in the SE2 price area than it is in SE1. Due to delayed SE2-inflows start of May, the inflow culmination is expected to be delayed this spring.
- The SE1 area would normally start its melting process later in May, which is confirmed by climatic scenario simulations.
Modelling the spring thaw
Energy Quantified has developed their hydrological models in cooperation with SMHI based on the SMHI Hydro GWh model from the HYPE-concept from SMHI. HYPE simulates water flow from precipitation through to its path to the sea via storage facilities and power generators.
The SMHI Hydro GWh model tool covers 40 years of climatic scenario simulations, with results proving to be of a very high standard.
To forecast the likely timing of the spring thaw and its impact on the hydrological situation, we need relevant data. These include:
- Snow levels: SMHI is updating their snow data on Snödjup (in Swedish) with information about measured snow levels (cm), see chart below. These data can be compared to the Energy Quantified/SMHI energy-modelled snow levels from the SMHI Hydro GWh model.
- From the SMHI webpages, it is possible to see the normal seasonal level (cm) as isocurves, making it possible to do a percentage comparison between the energy-models (TWh) and actual data (cm).
What the data tell us:
- For this year, we can see that the actual figures (cm) indicate somewhat lower snow levels in SE2 than initially modelled, whilst the SE1-data is closer to the expected amount. However, we do not see any reason to adjust the Energy-modelled snow levels for our inflow-outlook modelling.
- From a melting point of view we see that the SE2 snow would normally start melting the next few days, while the SE1-area has a later start-up. This means that most of the delayed melting will take place in SE2.
Scenario simulations – Sweden total
In the next chart we show daily simulated inflows for Sweden as a whole, from now until mid of August when the snow has melted. This shows:
- The 95th and 5th percentile of the climatic scenario simulations (weather scenarios 1980-2019) and latest 15-day EC ensemble simulation.
- Compared to average climatic scenario, the inflow the next 15 days are forecasted to come out rather flat, ending at about 50% of the climatic average level.
- It's also interesting to note that the average climatic scenario curve comes out close to the statistical long-term average (20 years). This is an important quality factor, showing that our modelling can be expected to be close to the eventual results.
- Both the 15 days and scenario simulations are based on the SMHI HYPE concept.
SE1 and SE2 inflows – culmination and scenario boundaries
The next chart models the inflow profiles and boundaries for SE1 and SE2 until the middle of August. For both SE1 and SE2 we see the daily 95th percentile accounting for 150%-200% of daily averages.
According to the delayed melting season and latest 15-day forecast, we believe that the SE2-culmination will be somewhat delayed too, but for the SE1-area we don't think the late start will influence the expected culminating date significantly.
SE2 is expected to culminate by May 20th, while SE1 is expected to culminate by about June 25th, more than 1 month later.
Snow scenarios for SE1 and SE2
The next chart shows all the 40 yrs snow-scenarios for SE1 until middle of August when the snow has melted. We see quite a symmetrical variation band around the average climate scenario (thick black). The 15 day forecast is slightly higher than the climatic average as the inflow level is moderate due to an early stage of the melting season.
For SE2 (below) we have not shown the 40 scenarios, instead focusing on the 95th and 5th percentile boundaries, alongside the simulated average and statistical long-term average (20 years). We see that the average climate and average long-term curve is quite close to each other this year as the current snowpack is close to normal.
The 15 days short-term forecast comes out at nearly unchanged level due to low temperatures and limited melting.
The climatic variation band is more or less symmetrical, as we saw from the SE1 curves too.
The variation band of the climatic snow curves would be strongly modified if we started a simulation from the end of the 15-day forecast, although we still expect that the snow curves would follow this simulation from the middle of June onwards.
Final words
This blog post has shown details of the SMHI Hydro GWh model output based on the SMHI HYPE concept in a situation with a 15-day forecast of low temperatures across Sweden, and by how long the melting process is expected to be delayed between SE1 and SE2 based on current forecasts.
The models conclude that cold temperatures in the short-term will lead to reduced snow melt in SE2. SE1 is not expected to be affected as much due to the fact the snow in this area only begins to melt after the forecast cold-snap.
These Hydro GWh models are established for daily updated scenario simulations at Energy Quantified for all countries covered in our hydrology section, and Energy Quantified will pay more attention to climatic scenarios in their hydrology modelling in the future.
