Research has shown that an accurate representation of tropical weather can have a positive effect on Weeks 3-4 lead time extratropical forecasts. Retrospective forecasts, more commonly known as hindcasts, are performed with the tropics forced to closely match observational estimates. This allows for improved US West Coast precipitation forecasts in comparison to those without forcing. However, the origin of these improvements is not yet fully understood. In an attempt to do so, researchers from various institutions in Colorado (CSU, CU Boulder, NOAA/PSL) used a machine-learning method that subsets hindcasts by their initial conditions.
The results from the study, published in Geophysical Research Letters (GRL), yielded that, one subset in particular, characterized by an initially strong Aleutian Low, demonstrates larger improvements at Weeks 3–4 than the others. The greater improvements by forcing this subset can be attributed to model errors in simulating the interaction between the Aleutian Low and the teleconnection patterns associated with the Madden-Julian oscillation (MJO) and El Niño Southern Oscillation (ENSO). The authors emphasize that improving forecasts of tropical intraseasonal precipitation, especially during early MJO phases and under non-cold ENSO, is beneficial to producing better US West Coast precipitation forecasts.
The research was funded in part by CPO’s Climate Variability and Predictability (CVP) program and by NOAA/OAR’s Weather Program Office (WPO).
For more information, contact Jose Algarin.