With the support of the Swiss Federal Office of Energy (SFOE), seasonal forecasts were produced for 12 different catchment areas allowing for a wide range of hydrological regimes to be evaluated with cutting-edge technology. Following a successful calibration process, the data was validated based on Ultracam photogrammetry Snow Height (HS) data that was collected by the Institute for Snow and Avalanche Research (SLF) during four drone flights.

The geospatial data consisted of snow cover extent (SCE), snow height (HS), and Snow Water Equivalent (SWE) was fused with AI and then integrated into rainfall-runoff models, proving to be especially effective in accurately estimating water supply in spring and summer, and resulting in improved input predictability and tangible ROI for long-term trading (from 60, 90, 120, 150 to 180 days periods).

Case in point: By integrating satellite snow data for forecasting and seasonal prediction of water supply in alpine dams, project participant SIG (Services Industriels de Genève) wanted to better manage renewable electricity supply.  The above visualization demonstrates how a fusion of geospatial data and AI can be applied to improve long-term water availability forecasts for hydropower assets that are subject to snow melt inflows from surrounding mountain ranges. 

Overall, the project results demonstrated that SWE data specifically can be successfully produced with a really high correlation (R-value of 0.86), especially for high altitude catchments, when compared to validation data from a high-resolution sensor. Furthermore, when integrated into rainfall-runoff modelling, forecasting errors could be greatly reduced by 20% to 50% and some hydropower energy companies were able to envisage annual income increases of up to 1.2%.

“We discovered that the SWE values are a determining factor for the quality of long-term seasonal inflow forecasts and can also improve forecasting performance by 5%-10%, especially for large catchments and time frames between 60 to 180 days,” said Daria Lüdtke, CTO & Geospatial Technologist.

By having a long-term forecast that is as unbiased as possible and resilient to climate anomalies, companies can avoid volatile market situations, reduce downside risk and be less vulnerable to today’s more frequent climate anomalies. Traders can use these optimal long-term forecasting capabilities to sell energy in long-term contracts several months in advance, ensuring income as early as January/February.

To find out more, register for the forthcoming webinar in which the company delve's into the highlights and key take-aways of this project with integration partner Hydrique, and long-standing client, Alpiq.