Images of an infrared sky camera

Problem The Numerical Weather Forecast (NWV) is primarily dependent on the principle of data assimilation. Two sources of information are combined: a model that makes a prediction from an initial state and observations/measures to improve this prediction. This is necessary because the predictions of the model are flawed by insufficient knowledge of the exact initial state, discretisation and model errors. The project ICamCloudOps The idea of the project ICamCloudOps (Intelligent Camera Cloud Operators for NWP) is to derive observations of clouds from sky cameras to use them for data assimilation. The aim of the ICamCloudOps project is to use artificial intelligence methods to improve wind and cloud analysis and cloud predictions. Wind and clouds have a significant and direct impact on all levels of traffic: Air, rail, ship and individual traffic is largely influenced by strong winds, gusts, thunderstorms and precipitation. In detail, it’s about: — the development of new AI-based techniques as components for the simulation of camera observations (so-called forward operators) of clouds in the infrared and visible range from the atmospheric variables of the numerical weather forecast — the use of intelligent forward operators for camera observations of clouds and their use in the prediction systems of numerical weather forecasting — improving the prediction of high-impact weather in the form of thunderstorms, clouds, wind and heavy rains and predicting yield fluctuations in renewable energies, especially for wind and photovoltaic systems. About this dataset Images of the all-sky infrared camera Sky InSight developed by Reuniwatt. The camera was installed as part of the mFUND project ICamCloudOps at the meteorological observatory Lindenberg — Richard-Aßmann-Observatorium (MOL-RAO) in Brandenburg and provides infrared images of the sky. The photos show the brightness temperature of the clouds and the sky observed by the Sky InSight.