Geography Professor & GIS Educator | GeoAI · Python · Google Earth Engine | Climate & environmental applications

I'm a geography professor at Texas A&M University-Kingsville with 20+ years of experience in GIS, remote sensing, and climate science. I teach in our GIS certificate program and in geography and geoscience courses. Alongside that work, I'm building GeoAI skills (Python, machine learning, Google Earth Engine) to bring remote sensing foundation models and applied ML into my research and teaching.

Free introductory resources for geospatial students and professionals moving from desktop GIS toward the modern Python and SQL stack.

Python: Start with Geospatial Python Tutorials (repo) from Spatial Thoughts. Each tutorial is a self-contained Google Colab notebook with datasets and, for many, a video walkthrough, covering GeoPandas, Xarray, satellite imagery, and more. They're written for learners who want to advance their skills, so if you're brand new to Python, take the free Python Foundation for Spatial Analysis course first. When you're ready for a full textbook, continue with pyGIS (pinned above; original by Michael Mann, read it online), which covers vector and raster data, remote sensing, and machine learning with GeoPandas, Rasterio, and scikit-learn. It is still under development, and it expects some Python experience.

SQL: Spatial Data Management with DuckDB (book site) by Qiusheng Wu starts at SQL basics and builds to spatial queries, joins, and real datasets. DuckDB installs like a Python package, so there's no database server to set up, and the examples run in Colab or Binder, with a companion video playlist. It can also query cloud-hosted GeoParquet files directly, without downloading them.

Next steps (cloud-native and big data):

• Cloud-native formats: the Cloud-Optimized Geospatial Formats Guide explains COG, GeoParquet, and Zarr, which let software read just the part of a file it needs over the web instead of downloading the whole thing.
• Data too big for memory: Xarray Tutorial for rasters and climate data, then the Dask tutorials in Geospatial Python Tutorials for parallel processing.
• Spatial SQL at cluster scale (awareness only): Apache Sedona runs spatial SQL across clusters; see its Spatial SQL tutorial.

• 🔭 Building: a multi-temporal El Niño climate risk pipeline. Google Earth Engine (MODIS LST, ERA5-Land soil moisture, NOAA ONI) → Cloud Storage → partitioned, clustered BigQuery tables with spatial joins. Next: matrix construction and ConvLSTM prototyping.
• 💧 Applied focus: ongoing research on the South Texas Sand Sheets (geomorphology and spatio-temporal change, using remote sensing and GeoAI), and Southern High Plains / Ogallala Aquifer depletion and long-term drought, using GRACE/GRACE-FO gravimetry and USGS groundwater data.
• 🌱 Learning: intermediate Python and SQL, BigQuery GIS, ML fundamentals, and turning Colab notebooks into modular, tested Python.
• 🧭 Exploring: geospatial foundation models (Prithvi, Clay), SAM/SamGeo, and Hugging Face for use in my courses and research.

Languages & data:

Geospatial & remote sensing:

Cloud & data engineering:

Dev workflow:

Building toward:

I keep curated GitHub star lists for learning the modern geospatial data science, remote sensing, climate, and GeoAI stack. They are meant for students and professionals who want a short, vetted path through the tools instead of searching millions of repositories. Suggested order for newcomers is top to bottom within each group:

Method lists (domain-agnostic tools and techniques):

• education-training-resources: Start here: courses, textbooks, and tutorials for learning geospatial Python, SQL, ML, and data science.
• raster-vector-core: Core raster and vector libraries: GeoPandas, Shapely, Rasterio, GDAL, pyproj, and related readers and utilities.
• gridded-multidimensional-data: Gridded and multidimensional data: Xarray and its ecosystem, Zarr, Dask, NetCDF, and CF conventions.
• cloud-native-stac: Cloud-native geospatial: STAC clients and catalogs, COG, GeoParquet, GeoArrow, tile servers, and cloud data access.
• spatial-sql-data-wrangling: Spatial SQL and data wrangling: DuckDB, PostGIS, BigQuery, Apache Sedona, and cloud-scale spatial SQL tools.
• google-earth-engine: Google Earth Engine: geemap, the Python API, Xee, community tools, courses, and the data catalog.
• earth-observation-remote-sensing: Remote sensing and Earth observation: NASA data tutorials, spectral indices, change detection, and satellite imagery notebooks.
• geoai-remote-sensing-ml: Machine learning for geospatial data: TorchGeo, GeoAI, segmentation, deep learning for remote sensing, and agents.
• geoai-foundation-models-hf: Geospatial foundation models (Prithvi, Clay, SAM and relatives) with Hugging Face weights and fine-tuning toolkits.
• spatial-statistics-networks: Spatial statistics and analysis: PySAL, OSMnx, momepy, network and urban analysis tools.
• visualization-communication: Mapping, interactive maps, dashboards, and notebooks for communicating geospatial results.
• python-engineering-reproducibility: Python engineering patterns: notebooks, project templates, and reproducible environments for geospatial work.

Applications and tools:

• climate-weather-data: Climate and weather data science: MetPy, xclim, climpred, ERA5 and CMIP tools, and tutorials.
• open-source-gis: Open-source desktop and server GIS: QGIS, GRASS, GeoNode, pygeoapi, and curated awesome lists.
• ai-assisted-coding-tools: AI-assisted coding tools for research and teaching workflows.

• Collaborations across STEM fields, as well as in the Humanities; field research design and data collection, including ground and aerial remote sensing, weather and climate, and environmental sensors; grant and proposal writing.

• LinkedIn: in/brent-hedquist-94267612
• Hugging Face: huggingface.co/bhedquist
• ORCID: 0000-0003-1840-8779