BEGIN:VCALENDAR VERSION:2.0 PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4// BEGIN:VEVENT UID:20260710T061045EDT-0001Kcm4Lu@132.216.98.100 DTSTAMP:20260710T101044Z DESCRIPTION:Please join us as we welcome Dr. S. Lovejoy and Dr.  L. Del Rio Amador from the department of Physics at 91 for their semi nar titled 'Harnessing butterflies for improved monthly\, seasonal\, and i nterannual forecasts'. Coffee will be served.\n\nAbstract\n\nOver the past ten years\, a key advance in our understanding of atmospheric variability is the discovery that between the weather and climate regime lies an inte rmediate “macroweather” regime\, spanning the range of scales from ≈ 10 da ys to ≈30 years (in the anthropocene\; it is longer in the pre-anthropocen e).   Macroweather statistics are characterized by two fundamental symmetr ies: scaling and the factorization of the joint space-time statistics.  In the time domain\, the scaling has low intermittency with the additional p roperty that successive fluctuations tend to cancel.  In space\, on the co ntrary the scaling has high intermittency corresponding to the existence o f different climate zones.  \n\nThese properties are fundamental for macro weather forecasting.  For example:\n\nThe temporal scaling implies that th e system has a long range - indeed elephantine - memory that can be exploi ted for forecasting.\n The low temporal intermittency implies that mathemat ically well-established (Gaussian) forecasting techniques can be used.\n Th e statistical factorization property implies that although spatial correla tions are large\, that they are not useful in making forecasts.\nThese pro perties can be directly exploited by the Stochastic Seasonal and Interannu al Prediction System (StocSIPS).  StocSIPS is a straightforward\, highly e fficient forecasting system that makes global\, monthly\, seasonal and int erannual forecasts.  Using hindcasts\, we compare StocSIPS with Environmen t Canada’s CanSIPS model\, finding that most of the earth\, for horizons b eyond about one month\, that StocSIPS is significantly more accurate. \n\n StocSIPS’ advantages include:\n\nConvergence to the real – not model - cli mate: The key to StocSIPS skill is the forecasting module that uses past d ata – and the huge memory in the system - to ensure that the forecast conv erges to the real world climate.\n Speed: In order to get good statistics\, conventional seasonal to annual forecasts typically re-forecast over ten to twenty realizations\, each time using slightly different initial data o ften taking the equivalent of hundreds of thousands of CPU hours on the wo rld’s fastest computers.  In comparison\, StocSIPS uses only a few minutes of CPU time to directly calculate the statistics of an infinite number of realizations.\n No data assimilation: StocSIPS can directly forecast eithe r gridded or individual station data\, there is no need to transform the i nput data to make it digestible by the numerical model\; StocSIPS avoids c omplex data “assimilation” techniques.  \n No ad hoc post processing: The r aw temperatures and precipitation rates forecast by conventional models ha ve unrealistic variability.  This is usually “corrected” using complex ad hoc post processing algorithms that use hindcasts to incorporate past info rmation in order to make the forecasts more realistic.  StocSIPS uses only past information with a theoretically justified forecast procedure.\n No n eed for downscaling: Conventional models have pixels of 100\,000 km2 or mo re in size and must be “downscaled” to adapt them to local conditions.   W henever long station temperature series are available\, StocSIPS can forec ast them directly.\n DTSTART:20160411T193000Z DTEND:20160411T203000Z LOCATION:Room 934\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue Sherbrooke Ouest SUMMARY:Seminar: Dr. S. Lovejoy and Dr. L. Del Rio Amador URL:/channels/event/seminar-dr-s-lovejoy-and-dr-l-del- rio-amador-260066 END:VEVENT END:VCALENDAR