GitLab
GitHub
LinkedIn
Map Gallery/Inland Empire Warehouses
(video)
I grew up in California's Inland Empire, where the landscape has changed dramatically over the past two decades. During my time in UCLA's MAGIST program, I had the opportunity to explore the shifting landscape with remote sensing imagery by employing Google Earth Engine, which I learned how to use in GEOG 415: Geospatial Data Science Futures. My first examination into the Inland Empire's warehouse explosion was rudimentary; I simply compared imagery from 1999 and 2019 (using LANDSAT 7 and 8 data, respectively). Below I've shared my first GEE script:
Google Earth Engine script
show script (GitHub)
Loading…A lot has changed in the Inland Empire in my lifetime; the rapid expansion of warehousing was the primary shift that I wanted to visualize.
The focus of GEOG 415 shifted towards more advanced geospatial analysis techniques, including the use of machine learning algorithms to identify and classify different types of land use changes over time, in particular for a collaborative project examining Eastern Washington state. The ML projects were quite involved, so my personal interest in the Inland Empire warehouse explosion moved to the backburner. I knew that I would want to revisit the project though, knowing that Google Earth Engine is an invaluable tool for creating timelapses and conducting complex analyses.
In Summer 2026 I employed LLMs (Claude Sonnet) to generate the code to create the timelapses. I was curious to test their capabilities in this domain.
With the models' help, we created 2 timelapses. The first shows the transformation from 1999 to 2019 using LANDSAT 7 and 8 data, and the second illustrates the changes from 2019 to 2024 using Sentinel-2 data. I am sharing the timelapses here because I am pleased with the results.
You can view the videos below, and use the links to download them or examine the code that generated them.
I would like to continue this thread in the future, though I am unsure to which extent I will employ LLMs. For example, I'm eager to analyze the warehouses' impact on the surrounding environment, especially in terms of heat, habitat fragmentation, and air quality changes. Seeing as these are more complex analyses, I am wary of relying too heavily on generative AI; but I was impressed with the Claude models' performance for the simple timelapse scripts.