Subseasonal to Seasonal Forecasts

Beyond two weeks, forecasts of specific weather systems are not useful. However the statistics of weather patterns do have some predictability on timescales of weeks to months, including the statistics of high impact weather.  Using subseasonal and seasonal forecast information in your decision making process requires an assessment of the confidence you can place in our forecasts. Predictability on these timescales can vary depending on the time of the year, location, and climatic regime. CFAN's comprehensive predictability and uncertainty analysis provides unique and critical information for your decision making process.  


Forecasting for the sub-seasonal time range (2-6 weeks) has so far received much less attention than medium-range and seasonal prediction as it has long been considered as a “predictability desert”. However, recent research has indicated important potential sources of predictability for this time range which can be realized through better representation of atmospheric phenomena such as the Madden Julian Oscillation and improved coupling with, and initialisation of, the land-ocean- cryosphere and stratosphere.

CFAN’s scientists are at the forefront of research on subseasonal and seasonal predictions [link], and CFAN recently received a grant from NOAA to improve subseasonal forecasts for the energy and agriculture sectors [SBIR link].