> For the complete documentation index, see [llms.txt](https://docs.carto.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.carto.com/carto-user-manual/workflows/components/google-pdfm-embeddings.md). # Geospatial Foundation Models An extension for accessing open source Geospatial Foundation Models: large-scale AI models trained on diverse geospatial data like satellite imagery, topographical maps, and other location-specific datasets, that encapsulate spatial information and relationships into dense representations in a multi-dimensional space (embeddings). ## Google PDFM County Embeddings **Description** This component loads Google's PDFM Embeddings at the county level. This component loads [Google's Population Dynamics Foundation Model (PDFM) Embeddings](https://research.google/blog/insights-into-population-dynamics-a-foundation-model-for-geospatial-inference/) at the county level. Google's (PDFM) Embeddings are condensed vector representations designed to encapsulate the complex, multidimensional interactions among human behaviors, environmental factors, and local contexts at specific locations. These embeddings capture patterns in aggregated data such as search trends, busyness trends, and environmental conditions (maps, air quality, temperature), providing a rich, location-specific snapshot of how populations engage with their surroundings. **Output** * **Output table:** This component generates an output table with the PDFM embeddings at the county level. ## Google PDFM ZCTA Embeddings **Description** This component loads [Google's Population Dynamics Foundation Model (PDFM) Embeddings](https://research.google/blog/insights-into-population-dynamics-a-foundation-model-for-geospatial-inference/) at the ZIP Code Tabulation Area (ZCTA) level. Google's (PDFM) Embeddings are condensed vector representations designed to encapsulate the complex, multidimensional interactions among human behaviors, environmental factors, and local contexts at specific locations. These embeddings capture patterns in aggregated data such as search trends, busyness trends, and environmental conditions (maps, air quality, temperature), providing a rich, location-specific snapshot of how populations engage with their surroundings. **Output** * **Output table:** This component generates an output table with the PDFM embeddings at the ZCTA level. ## Google Earth Engine Satellite Embeddings This component loads [Google Earth Engine's Satellite Embeddings](https://developers.google.com/earth-engine/tutorials/community/satellite-embedding-01-introduction) summarized for input geometries. The Satellite Embeddings dataset provides global, analysis-ready 64-dimensional vectors for each 10-meter land pixel, capturing annual temporal patterns of surface conditions using data from multiple Earth observation sources. Unlike traditional spectral bands, these vectors summarize complex, multi-source relationships in a more abstract but powerful format. **Input** * **Input table:** This component expects an input table with a geography column **Settings** * **Geography column:** A column that contains the geography data. * **From:** The starting year of data to extract. * **To:** The end year of data to extract. * **Reducer function:** Function to summarize the raster data over each polygon (e.g., mean, min, max). * **Scale in meters:** The scale of the source data in meters. **Output** * **Output table:** This component generates an output table with the Satellite Embeddings. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.carto.com/carto-user-manual/workflows/components/google-pdfm-embeddings.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.