geopredict predictive knowledge factory and forecasting services


geopredict GmbH´s CLIMFOR technology has won the Accurate seasonal weather forecasts under Open Innovability Challenge from ENEL Green Power

09.12.2023, Greifswald, Germany: geopredict GmbH´s CLIMFOR technology has won the Accurate seasonal weather forecasts under Open Innovability Challenge from ENEL Green Power.

The challenge was “Precipitation and temperature forecast, on monthly and annual granularity, as accurate as possible 9-12 months ahead to be able to estimate the production of its hydropower assets, mainly Italy and Spain, where it is most difficult to accurately predict rainfall and temperature forecasts because they are less affected by oceanic phenomena that influence the climate globally” – Enel Green Power.

The weather parameters forecasted in this work uses CLIMFOR (Probabilistic Climate Forecasting System based on High-Resolution Space Based Atmospheric and Ocean Data) which is an AI system that includes all the patented technologies developed by geopredict GmbH for forecasting important climate parameters in short, medium, and long-term with high spatial resolution. CLIMFOR works autonomously and its approach is that of self-organising, spatiotemporal modelling of the near-surface atmosphere dynamics for local forecasting that implements an innovative, original knowledge extraction from data technology by inductive modelling of interdependent systems from high-dimensional noisy data.

Enel Green Power commented on the results submitted in this challenge from CLIMFOR technology: “Your solution is really interesting and we appreciated your work very much, therefore you are awarded as a winner of the challenge Accurate seasonal forecast.”

"Because unlike the current Global Circulation Model (GCM) or Atmosphere Ocean GCMs, which are domain-based, theory-based modelling approaches, CLIMFOR is the first climate forecasting engine utilising the data driven approach for climate/weather modelling. It uses the exclusively Earth Observation (EO) satellite data and optional ground data, thus increasing the accuracy, validity, and reliability of the information used for modelling in climate, energy, and agriculture applications by geopredict" says Frank Lemke, CEO, geopredict .

As part of this challenge geopredict GmbH has also demonstrated a proprietary 4D/5D Digital TWIN of the atmosphere dynamics which the company uses for it´s gridded, probabilistic forecasting of important atmospheric/weather parameters for any region on the globe. These short to long term forecasting (days to years ahead) of key atmospheric parameters such as solar radiation (GHI, DNI), wind speed, temperature, precipitation and others are used in several vertical applications in energy (renewable energy production forecasting, long-term resources forecasting a.o.), agriculture and Agri-PV (temperature, precipitation, soil moisture, radiation etc.), and climate (early warning systems).

For more info contact: Dr.-Ing. Sanjeev Kumar Gurram 

About geopredict GmbH: Geopredict GmbH develops high-resolution geo-forecasting services and products at the intersection of satellite earth observation, high-performance computing, and AI. The company addresses industrial and societal challenges of high-resolution spatiotemporal forecasting in energy, climate, oceanography, agriculture, and geology. It has received a number of competitive grants from the European Innovation Council, the European Space Agency, and from German national and regional funding programs to support the company's innovation. The company takes responsibility for current global challenges and sustainable development.

geopredict predictive knowledge factory and forecasting services

Location Pin


Brandteichstr. 20

17489 Greifswald



Scale Balanced

CO2 Neutral 

Certified by the EIC Greenhouse Gas Programme

CO2 reduction certified by EIC greenhouse gase programme. geo-forecasting, knowledge factory, energy forecast, climate forecast, renewable energy production, forecast,

© geopredict GmbH, All rights reserved.

This website makes use of cookies. Please see our privacy policy for details.