Great Lakes Environmental Data Analysis

Python notebooks for fetching, processing, and visualizing satellite and buoy data over the Great Lakes, with a focus on southern Lake Michigan. The notebooks look at coastal upwelling, lake breezes, surface water temperature, and clouds — combining in-situ buoy observations (NDBC), gridded satellite SST (NOAA CoastWatch / GLERL via ERDDAP), and GOES-East satellite imagery.

Notebooks

• gl_sst.ipynb — Surface water temperature from the GLSEA gridded product. Builds ERDDAP requests, loads NetCDF into xarray, maps daily and averaged SST, and animates day-to-day SST anomalies to spot upwelling events.

• goes_cloud_animation.ipynb — A configurable pipeline that downloads GOES-East ABI imagery, renders each scan over a chosen region, and assembles the frames into an animation. Used here to look for lake-breeze cloud signatures, but the machinery works for any mesoscale cloud phenomenon.

• buoy.ipynb — Buoy diagnostics for southern Lake Michigan: wind, waves, and water-temperature drops associated with upwelling and lake breezes.

• nexrad_pyart.ipynb — Fetching, parsing, and visualizing NEXRAD Level 2 Doppler radar data from the KLOT (Chicago) station using the Python ARM Radar Toolkit (Py-ART).

Supporting files

• ndbc_io.py — Helper functions for fetching and parsing NDBC buoy text feeds.
• images/ — Saved figures and animations.
• .gitignore — Ignores cached downloads and local NetCDF slices (*.nc).

Data sources

• GLSEA / ACSPO — Daily gridded Great Lakes surface temperature (NOAA GLERL), derived from polar-orbiting satellite infrared via the ACSPO algorithm.
• NDBC — Real-time and historical buoy observations.
• GOES-East ABI — Geostationary satellite imagery, accessed with goes2go.

Getting started

Setup

Recommended (conda): this installs cartopy and the NetCDF libraries cleanly, which is the easiest path on Windows.

conda env create -f environment.yml
conda activate great_lakes

Alternative (pip):

pip install -r requirements.txt

Running the notebooks

Open any notebook and run the cells top to bottom. Each notebook sets its case date, region, and parameters in a configuration cell near the top — change those to retarget it to a different event or area.

New here? Start with buoy.ipynb — it begins by fetching and plotting buoy observations (a gentle introduction), then builds toward wind roses and statistical fits in later sections.