Dissertation: Mesoscale Ensemble-Based Data Assimilation and Parameter Estimation
M.B.A., Finance and Investments (1994)
The George Washington University, Washington, DC
B.S. in Mechanical Engineering (1991)
Boğaziçi University, Istanbul, Turkey
Professional Experience
Scientist (June 2017 – Present)
CIMAS, University of Miami & NOAA/AOML/HRD, Miami, FL
Associate Scientist (2013 – 2017)
CIMAS, University of Miami & NOAA/AOML/HRD
Assistant Scientist (2009 – 2013)
CIMAS, University of Miami & NOAA/AOML/HRD
Postdoctoral Associate (2008 – 2009)
CIMAS, University of Miami & NOAA/AOML/HRD
Postdoctoral Fellow (2005 – 2008)
NCAR (MMM and IMAGe Divisions), Boulder, CO
Research Interests
Ensemble-based data assimilation (EnKF)
Hurricane data assimilation and modeling
Radar data assimilation
Simultaneous state and parameter estimation
Chaos theory and predictability
Research Products & Software Development
Hurricane Ensemble Data Assimilation System (HEDAS) (2008–Present)
NOAA/AOML
A state-of-the-art, high-performance computing system to incorporate, in real time, high-resolution aircraft observations of tropical cyclones into numerical weather prediction models. It improves forecast skill and provides a complete picture of 3D tropical cyclone structure.
NOAA/AOML/HRD HRDOBS Dataset (2024–Present)
A comprehensive dataset that includes all tropical cyclone inner-core observations in a uniform format with appropriate error estimations, bringing together observations from various platforms and disparate online locations.
An interactive Python/Streamlit application designed to visualize the limits of predictability in chaotic systems. It serves as the digital companion to the research published in Chaos (2024), allowing users to run real-time ensemble simulations and visualize error growth.
Online Quality Control Technique for Ensemble-Based Data Assimilation Applications (2022–Present)
An advanced technique based on observation innovations normalized by total uncertainty. Combined with a nonparametric outlier identification mechanism, it effectively identifies outlier observations in highly non-Gaussian distributions typical of the tropical cyclone inner-core region.
An advanced observation processing technique that accounts for storm motion within a given processing window, allowing for homogeneous distribution of observations throughout assimilation cycles. Routinely used in HEDAS.
Atmospheric Parameter Estimation Pioneer
Since Ph.D. work at Texas A&M University, demonstrated and advocated the utilization of meteorological observations in data assimilation systems to estimate and improve key internal features of numerical weather prediction models. Author of the “Parameter Estimation” chapter in the Encyclopedia of Atmospheric Sciences.
Idealized, Non-Linear Numerical Sea Breeze Model (2003–2005)
Texas A&M University
Used to demonstrate the value of ensemble-based data assimilation in thermally forced circulations and later coupled with a chemical tracer model to evaluate how meteorological observations can be utilized to better model ozone transport.
Leadership & Service
Principal Investigator, NOAA Hurricane Field Program (2014–Present)
