Ed Zaron

Research Assistant Professor

Dept. of Civil and Environmental Engineering

Portland State University

P.O. Box 751

Portland, OR 97207-0751

PI: Zaron; Co-I: Marie-Aude Pradal, Johns Hopkins University.

We are developing techniques to use knowledge of the ocean surface tide to improve maps of seafloor bathymetry. Our approach is to use inverse methods with a data assimilative hydrodynamic model to simultaneously infer both tides and bathymetry.

PI: David T. Sandwell, SIO/UCSD; Co-I: Zaron.

Uncertainties in tides near the coastline contribute errors to marine gravity models derived from satellite altimetry. I am developing new techniques for high-resolution empirical mapping of tides that utilize data from non-repeat altimeter missions.

PI: Richard D. Ray, NASA/GSFC; Co-I's: Gary D. Egbert and Zaron.

The objective of this project is an improved understanding of surface and internal tides, and their relationship to non-tidal phenomena.

PI: Christopher N. K. Mooers.

This project is evaluating prototype operational numerical ocean models of the Gulf of Mexico. My role is the development of software and evaluation of metrics for analysis of models.

PI: Melinda Peng, NRL-Monterey

David Jay of PSU recently discovered that tides are changing in the North Pacific. This project aims to follow up on his initial data analysis and identify the causes of the change. Although the observed rate of change is small, approximately 1% per century, it is a puzzle when we consider the processes which could interact with tides on large, ocean-basin, scales. Using numerical models of the ocean we are assessing a number of possible dynamical causes, and also developing new approaches to tidal analysis which make better use of long, sparse, time series.

Together with collaborators at the Naval Research Lab, Stennis, MS, I am building an improved and automated version of the OTIS software for data-assimilative tidal modeling. Emphasis is on shallow-water modeling, with the goal of automatically tuning the representation of the model forcing error covariance.

I have been working with student volunteers and a graduate student to develop low-cost equipment for studying turbulence in flowing soap-water films. Present efforts are aimed at using modified phonograph cartidges as transducers to measure the dynamic pressure of the turbulent flow.

I am collaborating with Alan Blumberg's group at the Stevens Institute of Technology to develop an efficient means of inverting surface velocity for bottom depth. This is an interesting and strongly nonlinear estimation problem with many practical applications.

Gary Egbert (Oregon State University) and I have a project to assimilate satellite altimetry data into a three-dimensional primitive equations model in order to better quantify the energy flux of tidal internal waves, and infer the mechanisms responsible for dissipating these waves. The ultimate objective of this project is to test the hypothesis that tidal energy drives much of the global thermohaline overturning circulation, a quantity of central importance to climate change studies.

Together with Jim Moum at Oregon State University, I analyzed a widely-used parameterization of shear-driven turbulence in stratified flows. We found a new empirical relationship between turbulence diffusivity and gradient Richardson number using basic principles of dimensional analysis. The relationship collapses the turbulence data within a factor of 2, which is much improved over previous Richardson-number-based schemes, but we also demonstrate the lack of universality of these approaches. Essentially, the nonlinear dependence of diffusivity on Richardson number causes the relationship between mean quantities to deviate strongly from the relationship between the instantaneous quantities, leading to a Landau-type argument against universality.

I am involved in a project with David Jay (Portland State University) looking at the Columbia River Estuary and Plume (River Influences on Coastal Ecosystems, RISE). We are looking at the currents and mixing associated with the tidal plume, and the larger sub-tidal plume, using ADCP and CTD data from a towed, undulating platform.

PEZ-HAT is the data-assimilating primitive equation model I have been developing and using for our analyses of satellite altimetry. The model and its adjoint are designed to execute efficiently on computer clusters using domain decomposition via MPI. The model equations and finite-difference methods are based on MOM3, with some minor changes specific to tidal applications.

The Inverse Ocean Model (IOM) is an advanced software system for data assimilative modeling, developed by Andrew Bennett (Oregon State University) and colleagues under NSF-ITR funding. It uses advanced program generation tools to create a custom data assimilation system for any numerical model and observing system. I participated as a post-doc on this project and developed the tidal applications with PEZ-HAT.