🗨 About Me

I am Michael Edidem, a PhD Researcher in Geographic Information Systems and Environmental Modeling at Southern Illinois University Carbondale. I work at the intersection of Geospatial AI, hydrologic and environmental modeling, and physics-guided machine learning — developing scalable, data-driven systems for analyzing groundwater–surface water interactions, elevation-derived hydrography, and environmental decision-support workflows.

My work bridges research and engineering, translating advanced machine learning methods into reproducible, high-performance geospatial pipelines.

🔭 Vision: Turning elevation data and hydrologic signal into decisions that protect the ground and water beneath us — where physics, geospatial AI, and environmental stewardship meet.

📄 Curriculum Vitae: View CV

🌍 Research Interests

📖 Education

• PhD, Geographic Information Systems and Environmental Modeling — Southern Illinois University Carbondale, Expected 2027
• M.S., Geography and Environmental Resources (Spec: Geospatial AI) — Southern Illinois University Carbondale, 2024
• B.S., Geoinformatics and Surveying — University of Uyo, Nigeria, 2019

💼 Experience

Research Assistant / Teaching Assistant, Southern Illinois University Carbondale  08/2022 – Present

• Developing large-scale geospatial ML pipelines for mapping and environmental modeling using Python and GIS tools.
• Built Theory-Guided Neural Network (TgNN) surrogate models for scalable spatial simulation, reducing computational cost while preserving physical consistency.
• Implemented deep learning architectures (U-Net, CM-UNet, Faster R-CNN, YOLOv5, Transformer-based models) for elevation-derived hydrographic feature extraction and classification.
• Conducted TauDEM-based topographic and stream network analysis for hydrologic connectivity studies in agricultural watersheds.
• Teaching introduction to GIS (GEOG 401/502) and mentored over 500 students in geospatial analysis, research methods, and capstone project development.

GIS Specialist Intern, CDM Smith, Chicago, IL  05/2025 – 08/2025

• Evaluated and optimized an AI-based wetland delineation model to improve scalability and accuracy.
• Implemented and evaluated a peer-reviewed spectral diversity method for biodiversity monitoring and presented findings to stakeholders.
• Expanded a Bayesian model to correct additive bias in PFAS concentration estimates and automated extractable organofluorine soil data processing for reproducible analysis.
• Co-led a GitHub training session for engineering teams to enhance version control and collaborative workflows.

Geographic Information Systems Officer, Ikeja Electricity Distribution PLC, Lagos  06/2021 – 07/2022

• Developed and managed a customer enumeration application with real-time tracking, improving customer data accuracy by 30%.
• Coordinated UAV deployment for powerline inspection and feeder fault detection, reducing inspection time by 40%.
• Updated enterprise geospatial databases and developed analytical maps to support data-driven decision-making.

Project Control | GIS Analyst, Subsea100 Global, Port Harcourt  08/2020 – 07/2021

• Oversaw GIS project planning and spatial data workflows, preparing technical reports and bid documents for engineering stakeholders.

Early Geospatial & Surveying Experience  2017 – 2020

• Conducted cadastral, engineering, and maritime surveys using GNSS and total stations; produced CAD-based maps and managed spatial datasets for infrastructure and land-use projects.

🔬 Research Experience

Research Fellow — Optimizing Deep Learning Models for Geospatial Applications, University of North Texas  05/2024 – 08/2024

• Designed and optimized DL models for drought forecasting and drainage detection, improving computational efficiency through TPU optimization.
• Developed automated pipelines for processing large-scale geospatial datasets, reducing data preparation time by 50%.

Research Fellow — I-GUIDE Summer School: Convergence Science in Action, Boulder, Colorado  08/2023

• Problem set: Improving the 3D Representation of Rivers in Digital Elevation Models (DEM).
• Deployed deep learning models for river delineation and applied Random Forest regression for max river depth estimation.

🚀 Selected Projects

GeoFewLab

Soil Swin-UNet

Deep learning–based soil and land-surface segmentation using Swin-UNet architectures, applied to drought forecasting across remote-sensing drought indices. Primary developer within GeoFewLab.

Code  ·  PyTorch Swin-UNet Remote Sensing

Nebraska

Groundwater Simulation & Prediction

Large-scale groundwater modeling and scenario-based prediction for Nebraska, integrating MODFLOW with ML surrogate models for water-resource management decisions.

MODFLOW-2005 FloPy PINNs Transformers

CONUS

GeoAI-Based Drainage Crossing Detection

End-to-end deep learning pipelines detecting drainage crossings from elevation-derived hydrographic data, supporting infrastructure monitoring and hydrographic mapping across the contiguous USA.

Code  ·  U-Net CM-UNet Faster R-CNN YOLOv5

📝 Publications

• Identification of Drainage Crossings on High-Resolution Digital Elevation Models Using Explanatory Deep Learning Approaches, Edidem, M., Xu, B., Li, R., Wu, D., Rekabdar, B., Wang, G., Frontiers in Artificial Intelligence, 2025
• GeoAI-based Drainage Crossing Detection for Elevation-derived Hydrographic Mapping, Edidem, M., Li, R., Wu, D., Rekabdar, B., Wang, G., Environmental Modelling & Software, 2025
• Exploration of TPU Architectures for the Optimized Transformer in Drainage Crossing Detection, Nazeri, A., Godwin, D.W., Panteleaki, A.M., Anagnostopoulos, I., Edidem, M., Li, R., Shu, T., Proceedings of 2024 IEEE International Conference on Big Data, 2025
• Enhancing Hydrologic LiDAR Digital Elevation Models: Bridging Hydrographic Gaps at Fine Scales, Wu, D., Li, R., Edidem, M., Wang, G., JAWRA – Journal of the American Water Resources Association, 2024
• Classification of Drainage Crossings on High-resolution Digital Elevation Models: A Deep Learning Approach, Wu, D., Li, R., Talbert, C., Edidem, M., Rekabdar, B., Wang, G., GIScience & Remote Sensing, 60:1, 2023
• Networking Public Health Emergency Facilities with Incidence of Road Accident in Uyo, Udoh, I.B., Edidem, M.I., Eyofen, I.E., 2018

🎤 Conference Presentations

• 2026
  • AWRA Geospatial Water Technology Conference, Niagara Falls, NY, May 18–20, 2026.
    • Accuracy Is Not Enough: Evaluating Physical Consistency in Deep Learning-Based Groundwater Model Surrogates — Edidem, M., Li, R., Kharazi, P.
• 2025
  • AAG Annual Meeting, Detroit, MI, March 24–28, 2025.
    • Towards a Comprehensive Culvert Detection and Classification Dataset for Enhanced Infrastructure Monitoring — Edidem, M., Li, R., Wang, G., Rekabdar, B.
• 2024
  • IEEE International Conference on Big Data, Washington, DC, December 15–18, 2024.
    • Exploration of TPU Architectures for the Optimized Transformer in Drainage Crossing Detection — Nazeri, A. et al., Edidem, M., Li, R., Shu, T.
  • AGU Fall Meeting, Washington, DC, December 9–13, 2024.
    • Classification of Drainage Crossings Based on Advanced Deep Learning Models and High-Resolution Digital Elevation Models — Edidem, M., Li, R., Wang, G., Rekabdar, B., Xu, B.
  • AAG Annual Meeting, Honolulu, HI, April 16–20, 2024.
    • Geospatial AI Solutions for Locating Drainage Barriers in Elevation-derived Hydrographic Mapping — Edidem, M., Li, R., Wu, D., Wang, G.

🎖 Honors and Awards

• SIUC Outstanding Thesis Award (University-wide), 2025–2026
• SIUC Student Creative Activities and Research Award, 2025
• SIUC David G. Arey Memorial Award (Best MSc Thesis, Department), 2025
• NSF Summer School Award for Deep Learning Systems in Advanced GPU Cyberinfrastructure, 2024
• NSF Travel Award for AAG Geospatial Cyberinfrastructure Workshop, 2024
• NSF I-GUIDE Summer School Award, 2023

🤝 Volunteering

GIS Analyst, Geohazards Risk Mapping Initiative – Nigeria  08/2020 – 08/2022

• Developed a GIS flood risk model integrating socio-economic data with hazard maps to assess vulnerability, risk, and damage.
• Created high-resolution flood hazard inundation maps for land-use planning, flood risk mitigation, and emergency planning.

Geospatial Data Scientist, Omdena – Improving Food Security and Agriculture in Senegal  11/2020 – 02/2021

• Developed software combining satellite imagery, meteorology, and crop/disease data to improve food security and agriculture in Senegal.

🏛 Professional Affiliations

• President, MANRRS (Minorities in Agriculture, Natural Resources and Related Sciences), SIUC Chapter — 2026 – Present
• American Water Resources Association — 2025 – Present
• American Association of Geographers — 2023 – Present
• American Geophysical Union — 2023 – Present

🛠 Technical Expertise

• Programming & ML: Python, PyTorch, TensorFlow, deep learning model optimization, automated ML workflows
• Geospatial & GIS: GeoPandas, Rasterio, GDAL, ArcGIS Pro, ERDAS Imagine, Google Earth Engine, PostGIS
• Hydrologic Modeling: MODFLOW, FloPy, TauDEM
• Computer Vision: U-Net, CM-UNet, YOLOv5, Faster R-CNN, Transformer-based models
• Remote Sensing: LiDAR/DEM analysis, deep learning for geospatial imagery
• Infrastructure: Git, large-scale geospatial data pipelines, TPU/HPC computing, AWS, GCP