Heatwaves are among Europe's most significant weather hazards, affecting millions of people as well as infrastructure, agriculture, and energy systems. But what actually causes a heatwave? Why do some persist for days or even weeks? And how do meteorologists forecast them?

Weather patterns are often connected across vast distances. A cold winter in Europe or an active Atlantic storm season can sometimes be traced back to oceanic and atmospheric anomalies occurring far beyond the region itself.

The level of detail available in weather forecasts has increased dramatically in recent years. What once focused on broad regional patterns is increasingly moving towards local, street-scale prediction. Modern forecasts can now identify which urban areas will retain heat overnight, where cold air may accumulate, or which locations face the highest risk of intense rainfall and flooding.

Few weather phenomena evolve as rapidly as a thunderstorm. A calm summer afternoon can quickly turn into heavy rain, lightning, hail, and strong winds within less than an hour. What begins as an ordinary cumulus cloud may grow into a towering storm reaching more than 10 kilometres in height. Despite their short lifespan, thunderstorms are among the atmosphere's most dynamic weather systems.

If your spring starts with itchy eyes and a runny nose, the meteoblue air quality tools can give you a fighting chance before you step outside.

During high-impact weather events, different forecast maps often show different outcomes. One model predicts stronger wind gusts, another shifts the peak a few hours later, and a third shows precipitation in a slightly different location. The natural question is: which one should we trust?

Terrain is one of the quiet architects of weather. Even when the larger weather pattern is set by fronts, pressure systems and the jet stream, hills, valleys and mountain ranges can decide where clouds form, where rain falls, where wind accelerates and where cold air lingers long after sunrise.

Short-term weather decisions often depend less on what might happen tomorrow and more on what is already unfolding now.

Every year on 23 March, the global meteorological community celebrates World Meteorological Day, recognising the vital role of weather and climate services in society.

In the first part of this guide, we explored how numerical weather prediction models simulate the atmosphere, from observational data and data assimilation to model grids and parameterisation. Modern forecasting goes further by combining global and regional models, ensemble systems and advanced computational techniques to better predict weather and its uncertainty.