null Operational forecasting for early detection of changing extremes

Operational forecasting for early detection of changing extremes

The world has recently experienced a series of very severe and extensive weather events. Floods in Germany, Canada, and China; heat waves and wildfires in Greece, Turkey, Italy, Russia, and the United States; and heat waves in the Pacific NW and Siberia resulted in massive damage and many casualties. The magnitude and intensity of some of these events was such that they are rare, even under changing climate conditions. They can be attributed to climate change, as shown by The World Weather Attribution (WWA) network.

The average climate has already changed: mean global temperatures are currently 1.1 degrees C higher than in the early 20th century. Extreme weather will likely increase with climate change, and we urgently need to adapt. Preparing for more extreme events is not only needed for the future; it is needed now. 

Many countries are preparing for more extreme weather. How much the extremes will change, and in which direction, is less certain. As a society we are generally assuming that extreme events, like the average climate, are changing gradually. 

However, the recent series of very extreme events is causing us to revisit the question: what can we expect, now and in the future? The world will look very different if extreme events don't change gradually, but in spurts. While it may sound alarmist, maybe we should not interpret these recent events as “extreme,” but as the new normal in a world which is ~1.5 degrees C warmer.

If so, what can we do? This would mean that we can no longer estimate the probability of extreme events based on past conditions, which would have massive consequences for all engineering designs. 

While this possibility is not new to scientists, decision-makers are stirring, and these concepts are moving from the scientific realm into the arena of those who should act. We should not wait until climate models can provide us with more certain information about extreme events, or we may be too late. We should also take advantage of the tools we already have and our knowledge about acting in an environment of great uncertainty.

While some of the recent events do not appear in climate projections, we were able to forecast them operationally. Together the Delft-FEWS network covers a large part of the globe. This brings us to our proposal: Let’s bring all our knowledge/information together to explore whether the forecasts and warnings we produce daily could provide early detection of changes in extremes. 

Are you interested in exploring this topic together? Please get in touch.

Marc van Dijk, Jaap Kwadijk, and Albrecht Weerts