DID YOU KNOW? DTIC has over 3.5 million final reports on DoD funded research, development, test, and evaluation activities available to our registered users. Click HERE
to register or log in.
Comparison of Los Alamos National Laboratory (LANL) Parallel Ocean Program (POP) Model Fields with Pacific Surface Drifter Measurements
NAVAL POSTGRADUATE SCHOOL MONTEREY CA DEPT OF OCEANOGRAPHY
Pagination or Media Count:
Model fields from the Los Alamos National Laboratory LANL Parallel Ocean Program POP 16 degree global circulation model are compared to measurements from over 1300 satellite-tracked surface drifters that were deployed in the tropical Pacific 20N to 20S, between 1979 and 1994, during the TOGA Pan Pacific Current Study. Geographic averages of 5 days averaged drifter velocity estimates for 2 deg. latitude x 8 deg. longitude bins are compared to similarly binned 3 days model snapshots from September 1992 to October 1994. Eulerian comparisons of the model mean velocities and their observed counterparts show that the model U mean is slightly higher in the equatorial region, while the model V mean is 50 greater in this region. Model SST mean values are 20 less than observed values in the eastern equatorial Pacific. Model variability is about 20 less than the observed quantity in equatorial regions, and 50 less poleward of 10S and 10N. Both model and observed velocity and SST covariance fields imply a net heat convergence toward the equator with the largest values in the region of instability waves north of the equator. Model velocity fields are used to produce simulated Lagrangian trajectories for uniform and non-uniform deployment strategies. Autocorrelation, time and length scales, diffusivity, and polarization are calculated and ensemble averaged by 5 deg. latitude bands for comparison with drifter based Lagrangian statistics. Time and length scales are too long and diffusivities too low compared to observations, but data sampling in the simulated fields was biased by trajectories that overlap current regimes. These differences, in both Eulerian and Lagrangian comparisons, may be related to the lack of a surface mixed layer, inadequate representing of wind forcing, still too coarse grid resolution, and deficiencies in simulating the mean structure of the d
APPROVED FOR PUBLIC RELEASE