Interdisciplinary Multi-adaptive Oceanic Data-Driven Forecasting - - PowerPoint PPT Presentation

DDDAS Panel at ICS'02

Interdisciplinary Multi-adaptive Oceanic Data-Driven Forecasting

PIs: N.M. Patrikalakis, J.J. McCarthy, A.R. Robinson, H. Schmidt Staff: C. Evangelinos, P.F.J. Lermusieux, P.J. Haley Jr., S. Lalis Students: B. Renard, D. Guo, P. Moreno, O. Logoutov, R. Chang http://czms.mit.edu/poseidon

Ocean Science and Data Assimilation

• Field and remote
• bservations
• Models:

– Dynamical – Measurement – Error

• Assimilation schemes
• Sampling strategies
• State and parameter

estimates

Applications

• Pollution control

– Outfalls – Spills – Harmful algal blooms

• Resource exploitation and management

– Fisheries – Oil platforms

• Maritime and naval operations, law enforcement

Pollution Control: Harmful Algal Blooms

Resource Exploitation: Fisheries

Maritime and Naval Operations: Egypt Air crash (10/31/1999)

Egypt Air Flight 990 - Floating Debris Dispersion southern (left) vs. northern (right) impact point

Background: Physical-Biological Oceanography in HOPS

• Primitive Equation

(PE) dynamical model

• Multiple biological

models

• Optical dynamical

model

• Interface with ocean

acoustics

• Assimilation for

physics, acoustics and biology

Background: Ocean Acoustics

• Study of reflection, scattering

and transmission loss

• Interface with physical
• ceanography (sound speed)
• Parabolic equation, ray

propagation, normal mode and wavenumber integration based dynamical models

• Propagation characteristics

help estimate physical vars.

• Scattering characteristics help

estimate biological vars.

Current Work and Plans

• Data Assimilation and Error Estimation:

– Operational real-time Error Subspace Statistical Estimation

(ESSE): physics-biology-acoustics

– Error metrics for quality control, adaptive modeling and

sampling

• Advanced Models:

– Adaptivity:

• Biological Oceanography: Dynamical Models
• Physical Oceanography: Adaptive Primitive Equation terms

– Interactions:

• Physical Oceanography with Acoustics
• Acoustics with Biological Oceanography
• Distributed Computing Infrastructure and Web User Interface

Error Subspace Statistical Estimation

• Improved forecasts (with a

reliable model)

• Dynamically forecasted

error

• Ensemble forecasts (Monte-

Carlo approach with full nonlinear model)

• In addition ESSE provides:

– Most significant

variability modes

– Statistics for variability,

predictability, model error

– Multivariate, multiscale

correlations

Use of ESSE Assimilation in Coupled Simulations

Aug.-Sep. 1998 Mass. Bay experiment: ESSE assimilated simulation results compared to unassimilated forecast and persistence results.

Next Generation Oceanic Data Assimilation

• Issues arising from the use of ESSE in an
• perational real-time interdisciplinary

nowcasting/forecasting:

– Large amounts of compute power – Quality of service – Transparent data management

Solution through distributed computing infrastructure

• Error metrics for automated quality control

– Guided and automatic adaptive modeling and

sampling

• Visualization of uncertainties (feature extraction)

Interdisciplinary Interactions

Velocity, Temperature, Salinity Sound Speed Sound Speed Species concentration

Physical Oceanography Biological Oceanography Ocean Acoustics

Next Generation Interactions

• Physical

Oceanography to Ocean Acoustics

• Ocean Acoustics

to Physical Oceanography

• Ocean Acoustics

to Biological Oceanography

Distributed Computing Framework

Summary

• Data driven simulations via data assimilation
• Simulation driven adaptive sampling of the ocean
• Interdisciplinary character of ocean science

involving interactions of physical, biological and acoustical phenomena

• Extend state-of-the-art by providing feedback from

acoustics to physical and biological oceanography

• Use of distributed computing infrastructure to

integrate the disciplines and enable the new science.