GeoAI | Earth Observation ML | Environmental Intelligence | Geoscience × Machine Learning I am a Geo-Data Scientist bridging the gap between foundational Geology and state-of-the-art Machine Learning. With over 5 years of industry experience at Fugro and Hindalco, combined with an MSc (4.0 GPA) from Purdue, my research focuses on developing robust, explainable, and physics-informed AI models to transform complex Earth Observation data into actionable climate resilience knowledge.
I am especially interested in using AI to solve real-world problems related to:
- natural hazards
- environmental change
- remote sensing
- geospatial intelligence
- sustainability
- explainable AI for Earth systems
- MSc in GeoData Science
- Background in geology, geophysics, and geospatial data
- Interested in GeoAI, Earth Observation, environmental ML, and interpretable AI
- Building a portfolio of projects in remote sensing, hazard intelligence, and environmental analytics
Explainable GeoAI: Leveraging XAI (SHAP/LIME) to understand causal drivers in environmental hazards.
Physics-Aware Modeling: Integrating physical constraints into Deep Learning for subsurface and atmospheric systems.
Climate Resilience: Using Earth Observation (Landsat/Sentinel) to map and mitigate urban heat and wildfire risk.
Languages & Tools
- Python
- SQL
- Git & GitHub
- Jupyter Notebook
- VS Code
Geospatial & EO
- Google Earth Engine
- GeoPandas
- Rasterio
- rioxarray
- xarray
- GDAL
- geemap
- QGIS
Machine Learning & Data Science
- scikit-learn
- XGBoost
- LightGBM
- PyTorch
- pandas
- NumPy
- Matplotlib
- Plotly
Focus Areas
- GeoAI
- Earth Observation
- Environmental Intelligence
- Spatial ML
- Explainable AI
- Geoscience Data Science
I am currently building and publishing projects in:
- Wildfire Risk Mapping with Explainable GeoAI
- Urban Heat Island and Vulnerability Analysis
- Flood Mapping using SAR + Optical Data
- Drought / Crop Stress Monitoring from EO Time Series
More projects will be added here as they are completed.
Right now, I am focused on: Advanced GeoAI Research: Refining a workflow for Urban Heat Island Analysis using Landsat-9 TIRS data, focusing on Explainable AI (SHAP) to identify causal environmental drivers. Reproducible Science: Structuring all projects as "Research Repositories" with Conda environments and documented physics-based assumptions. Intelligent Earth CDT Preparation: Actively bridging the gap between industry experience at Fugro and academic research in AI for the Environment.
Develop robust, interpretable AI frameworks that respects the physical constraints of Earth systems. Contribute to climate resilience by providing decision-ready knowledge through automated hazard intelligence pipelines. Pursue doctoral research at the intersection of Geoscience and AI to solve complex, multi-scale environmental problems.
My long-term goal is to contribute to AI for Earth, environmental sustainability, and interpretable geospatial intelligence.
- LinkedIn: Shreya Jariwala
- GitHub: github.com/ShreyaJari
I’m currently developing open projects around:
- Earth Observation machine learning
- GeoAI research workflows
- environmental risk mapping
- geospatial explainable AI
Stay tuned.