Parts | Days | Selection | Search | Updates | Downloads | Help

DY: Fachverband Dynamik und Statistische Physik

DY 10: Posters I

DY 10.47: Poster

Monday, March 14, 2011, 17:00–19:00, P4

Towards first-passage-time prediction for temperature data — •Anja Garber, Nicholas Moloney, and Holger Kantz — Max-Planck-Institut für Physik komplexer Systeme, Dresden

In the operational short term weather prognoses, detailed models are run only for a lead time of ten days, preventing predictions further into the future. However, in the more medium range exceeding the lead time of model runs where detailed predictions are currently unavailable, already the first passage time until a threshold crossing can be of significant interest, e.g. the expected time until first frost.

In order to develop methods towards a suitable approximation and prediction of these first passage times from the existing model runs, we study the first passage time distribution both for an autoregressive model process and actual temperature data.