Weather is a strong external driver of integrated energy systems that simultaneously impacts the demand-side, energy transmission capability, and the supply-side (Rübbelke and Vögele, 2011). Temperature and wind affect energy demand for heating/cooling. Cloud cover affects demand for lighting. The weather also drives Renewable Energy (RE) generation (wind and solar PV in particular). Numerical Weather Prediction (NWP) models are used with local conditions and turbine power curves to produce deterministic or probabilistic wind energy forecasts (Jung and Broadwater 2014, Courtney et al. 2013). PV power forecasting uses NWP, satellite imagery and irradiance sensors (Holmgren et al. 2014). On the transmission side, thermal cooling capability of overhead electricity transmission lines is subject to highly localised weather conditions. Extreme events can impact other systems. Extreme precipitation produces floodwater which can stress pumping stations, leading to combined sewer overflows. Conversely, drought conditions can reduce hydroelectric power generation. We are not aware of a single model that combines predictions of wind and PV generation with energy demand at a high spatio-temporal resolution. This WP will use high-resolution NWP models allowing better representation of important small scales in space (e.g. topography, site exposure, aspect) and time (e.g. fluctuations and extreme events). Output will be statistically post-processed (Sweeney et al., 2013) to identify links with energy system data. A model will be developed that systematically integrates NWP data with spatially resolved energy data to simultaneously forecast RE energy supply and weather-driven demand. The model will predict the supply and demand response of the energy system to atmospheric changes such as heat waves, cold spells, blocking highs and storms, with a prediction period from hours to days ahead. The model will additionally warn of weather-related impacts for energy systems infrastructure such as pumping stations.