Home | Meeting Announcement | Meeting Agenda | Call for Papers | Meeting Registration | Hotel Information | Travel Information | Attendees | Abstracts
2002 LAPCOD Meeting
Back to Abstracts Page | Next Abstract
Simple prediction algorithms for the Lagrangian motion
Leonid I. Piterbarg and Tamay M. Özgökmen
CAMS, University of Southern California and RSMAS/MPO, University of Miami
piter@math.usc.edu(Abstract received 09/27/2002 for session D)
ABSTRACT
A new algorithm is suggested for prediction of the position of an unobserved Lagrangian particle (predictor) in a stochastic flow given observations of other particles (predictors). The algorithm is similar to the classical center of mass (CM) procedure, but takes essentially into account the initial position of the predictand. It appears to be efficient for an initial tight cluster and small prediction time. A theoretical error analysis is given for the Brownian flow and a stochastic flow with memory. The asymptotic formulae are compared with simulation results to establish their applicability limits. Monte-Carlo simulations are carried out to compare the new algorithm with two others: CM and a Kalman filter type method. The algorithm is also tested on real data in the Tropical Pacific. We also present preliminary results concerning with the optimal initial predictor placement for CM algorithm.
Back to Abstracts Page | Next Abstract
2002 LAPCOD Meeting, Key Largo, Florida, December 12-16, 2000