I’m traveling this week (see here), so my attentions are elsewhere. But with time to kill this morning, I was cruising some scientific back alleys and found this fascinating bit of business in GRL. It seems that ENSO, the best tool available for seasonal forecasting in some regions of the world, may not be working as well as it used to:

The interdecadal variation of the relationship between ENSO and summer precipitation in China has been examined based on observed monthly rainfall data and NOAA ERSST data from 1951 to 2003. Results show that the relation has weakened during the past two decades, and the significance of ENSO as a predictor has also decreased.