Attributing extreme weather events to climate change is challenging because these events are by definition rare and therefore hard to evaluate reliably, and are affected by patterns of natural climate variability. For instance, the biggest cause of droughts and floods around the world is the shifting of climate patterns between El Niño and La Niña events. On land, El Niño events favour drought in many tropical and subtropical areas, while La Niña events promote wetter conditions in many places, as has happened in recent years. These short-term and regional variations are expected to become more extreme in a warming climate. There is considerable uncertainty about how hurricanes are changing because of the large natural variability and the incomplete observational record. The impact of climate change on hurricane frequency remains a subject of ongoing studies. While changes in hurricane frequency remain uncertain, basic physical understanding and model results suggest that the strongest hurricanes (when they occur) are likely to become more intense and possibly larger in a warmer, moister atmosphere over the oceans. This is supported by available observational evidence in the North Atlantic. Some conditions favourable for strong thunderstorms that spawn tornadoes are expected to increase with warming, but uncertainty exists in other factors that affect tornado formation, such as changes in the vertical and horizontal variations of winds.