An opportunity index to anticipate when subseasonal predictions are useful

Abstract The operational use of subseasonal atmospheric predictions remains a major challenge due to the intermittency of skill on these timescales. Often only specifically trained users can keep track of so‐called “windows of forecast opportunity” (WFOs). WFOs are periods during which prediction skill is enhanced because specific states of the atmosphere, ocean, or land surface temporarily enhance predictability. Here, we propose a novel method to combine simultaneously active WFOs into a single opportunity index, which can be used operationally like a traffic‐light system and without expert knowledge to anticipate enhanced or reduced subseasonal prediction skill. For predictions of two‐weekly and monthly mean temperature anomalies in Switzerland during summer—a region and season with particularly low predictability—we demonstrate that skill can nearly double for a high opportunity index compared with a low one. The use of such an index could thus advance the year‐round operational usability of subseasonal predictions in many other regions of the world.