W. Joseph Durkin, Ph.D.

Professional Summary

Geophysical remote sensing and geodesy specialist with a Ph.D. from Cornell, postdoctoral fellowship at the Byrd Polar and Climate Research Center (OSU), and 5 years as a staff scientist at MITRE. Research has centered on measuring surface deformation from noisy, irregularly sampled, and opportunistic satellite observations (e.g., InSAR, SAR, GNSS, stereo photogrammetry, and altimetry) and using those measurements to estimate the parameters of geophysical models of glacier dynamics, mantle rheology, and seismic sources. At MITRE, applied the same black-box problem framework to navigation and PNT integrity for govt. customers, including GNSS spoofing detection and geolocation, inertial navigation, and lunar tidal deformation modeling. Published in Journal of Geophysical Research, The Cryosphere, Frontiers, ICRA, ION GNSS+, and IEEE MILCOM.

Professional Experience

• Co-developed a GNSS spoofing detection and geolocation algorithm for a single moving UAV platform; led HPC-based simulation and benchmarking, and currently leading payload design and hardware integration for a field demonstration program.
• Served as subject matter expert on lunar geodesy and geophysics for an advisory effort on improving the lunar reference frame, authoring a whitepaper surveying the state of the art in lunar geodesy, geophysical processes, and their implications for reference frame accuracy.
• Designed and implemented SBTides, an open-source Python toolkit for modeling viscoelastic solid body tides on planetary surfaces using spherical harmonic expansions and complex Love numbers, filling a gap in available software for non-terrestrial geodetic reference frame modeling. Published in ION GNSS+ 2023.
• Designed and benchmarked algorithms for incorporating Earth gravity models into inertial navigation solutions, using HPC simulation to characterize globally the conditions under which higher-order gravity anomalies meaningfully degrade INS coasting performance.
• Developed spoofing detection monitors within the PNT Trust Inference Engine (PNTTING) reference architecture, including a Wiener process anomaly detector, and used HPC-based optimization to probe monitor evasion strategies. Published in ION GNSS+ 2024.
• Performed geospatial image analysis of terrain and land cover rasters (DSM, DTM, clutter maps, NLCD) to develop physics-based RF clutter loss features for a convolutional neural network predicting clutter loss statistics across frequency bands. Published in IEEE MILCOM 2024.

• Developed a simulation environment in Unreal Engine 4 and AirSim for testing multi-agent search and rescue algorithms, including sensor simulation, object detection, viewshed computation, and coordinate transformations between agent and world frames. Published in IEEE ICRA 2021.
• Detected and filtered jitter artifacts in pushbroom satellite DEM time series by fitting a correction model to stable off-target surfaces.
• Applied Gaussian Process Regression to denoise Antarctic GNSS time series contaminated by snow infiltration artifacts, recovering clean displacement estimates.
• Developed a least squares network adjustment of DEM stacks to estimate Alaska-wide glacier thinning rates. Processed tens of terabytes of stereo DEMs on HPC, achieving decimeter precision.
• Developed a Jacobian-based inversion demonstrating that weighted combinations of ground deformation observations from the existing Antarctic geodetic GNSS network can isolate and monitor mass changes at individual glaciers with major sea level rise implications.
• Co-advised a Ph.D. student on identifying stable coregistration references within the Antarctic Ice Sheet, enabling HPC-scale DEM time series analysis where traditional rock outcrop references are sparse.

Key Technical Skills

• Remote Sensing: InSAR processing, satellite altimetry, optical/multispectral imagery (Landsat, Sentinel), stereo photogrammetry, pixel tracking
• LiDAR: Point cloud processing (PDAL), DTM/DSM construction, vegetation masking, feature extraction
• Geospatial Analysis: Python (GDAL, rasterio, shapely, NumPy, Pandas), MATLAB, QGIS, coordinate transformations, time series analysis
• Simulation & Robotics: Unreal Engine 4, AirSim, YOLO object detection, viewshed analysis
• Statistical Methods: Gaussian Process Regression, parameter estimation, optimization (simulated annealing, genetic algorithms, surrogate optimization)
• HPC: On-premises cluster computing (NASA, OSU, MITRE); batch processing and workflow automation
• Software Development: Python, MATLAB, shell scripting, Git

Education

Dissertation: Investigating Dynamic Glacier Processes, Mass Loss, and Coupled Interactions with the Solid Earth Using Satellite Geodesy

Used pixel tracking, DEM time series analysis, and InSAR to measure glacier dynamics and mass loss in Alaska and Iceland. Key results include characterization of the geometric controls driving advance of Yahtse Glacier, estimation of icefield mass loss rates in southeast Alaska and their solid Earth deformation signature, and a feasibility analysis for detecting long-wavelength glacial isostatic adjustment with InSAR in a high-atmospheric-noise environment.

Relevant coursework included geophysical signal processing, computational geophysics, near-surface geophysics, complex theory, differential equations, physical oceanography, and electromagnetic and optical physics. Participated in research cruise MGL1206 to survey Tamu Massif and Shatsky Rise, oceanic plateaus in the northwest Pacific, contributing to multibeam bathymetry data collection, processing, and publication.

Honors & Awards

• Byrd Postdoctoral Scholarship, Byrd Polar and Climate Research Center, Ohio State University
• NASA Earth and Space Science Fellowship, 2017--2019
• Cornell Earth and Atmospheric Science Department's Award for Excellence in Research
• European Geosciences Union Outstanding Student Presentation Award

Selected Publications

1. D. Dominic, W. J. Durkin, et al., "Machine Learning Prediction of Multi-Band Clutter Loss Using Deep Neural Networks," IEEE MILCOM, 2024.
2. W. J. Durkin, P. Larkoski, and J. J. Rushanan, "PNT Trust Inference Engine Reference Architecture," Proceedings of ION GNSS+, 2024.
3. W. J. Durkin and C. Davis, "SBTides: A Modular Toolkit for Modeling Viscoelastic Solid Body Tides in Python," Proceedings of ION GNSS+, 2023.
4. W. J. Durkin, T. Wilson, and M. Bevis, "A novel approach to exploit elastic deformation to constrain regional ice mass change in Antarctica," arXiv preprint arXiv:2212.06577, 2022.
5. R. Ghods, W. J. Durkin, and J. Schneider, "Multi-Agent Active Search Using a Realistic Depth-Aware Noise Model," IEEE ICRA, 2021.
6. W. J. Durkin, S. Kachuck, M. Pritchard, et al., "The importance of the inelastic and elastic structures of the crust in constraining glacial density, mass change, and isostatic adjustment from geodetic observations in Southeast Alaska," Journal of Geophysical Research: Solid Earth, 2019.
7. E. Berthier, C. Larsen, W. J. Durkin, et al., "Unabated wastage of the Juneau and Stikine icefields (Southeast Alaska) in the early 21st century," The Cryosphere, 2018.
8. M. Willis, W. Zheng, W. J. Durkin, et al., "Massive destabilization of an Arctic ice cap," Earth and Planetary Science Letters, 2018.
9. W. J. Durkin, T. Bartholomaus, M. Willis, and M. Pritchard, "Dynamic changes at Yahtse Glacier, the most rapidly advancing tidewater glacier in Alaska," Frontiers in Earth Science, 2017.
10. J. Zhang, W. W. Sager, and W. J. Durkin, "Morphology of Shatsky Rise oceanic plateau from high resolution bathymetry," Marine Geophysical Research, 2017.