# Recoverable Geographic Location Information in Earth-Observation Embeddings

arXiv: https://arxiv.org/abs/2609.29151 · HTML: https://mlsanity.com/paper/2609.29151/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 17 references that are in the map.
- First posted on arXiv 24 Sep 2026.

## Builds on (most cited first, 12 of 12)

- [WILDS: A Benchmark of in-the-Wild Distribution Shifts](https://mlsanity.com/paper/2012.07421/) (2020)
- [Geography-Aware Self-Supervised Learning](https://mlsanity.com/paper/2011.09980/) (2020)
- [GeoCLIP: Clip-Inspired Alignment between Locations and Images for Effective Worldwide Geo-localization](https://mlsanity.com/paper/2309.16020/) (2023)
- [SatCLIP: Global, General-Purpose Location Embeddings with Satellite Imagery](https://mlsanity.com/paper/2311.17179/) (2023)
- [AlphaEarth Foundations: An embedding field model for accurate and efficient global mapping from sparse label data](https://mlsanity.com/paper/2507.22291/) (2025)
- [Tile2Vec: Unsupervised representation learning for spatially distributed data](https://mlsanity.com/paper/1805.02855/) (2018)
- [A Generalizable and Accessible Approach to Machine Learning with Global Satellite Imagery](https://mlsanity.com/paper/2010.08168/) (2020)
- [PANGAEA: A Global and Inclusive Benchmark for Geospatial Foundation Models](https://mlsanity.com/paper/2412.04204/) (2024)
- [TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysis](https://mlsanity.com/paper/2506.20380/) (2025)
- [Evaluating and Benchmarking Foundation Models for Earth Observation and Geospatial AI](https://mlsanity.com/paper/2406.18295/) (2024)
- [On the Generalizability of Foundation Models for Crop Type Mapping](https://mlsanity.com/paper/2409.09451/) (2024)
- [Global Renewables Watch: A Temporal Dataset of Solar and Wind Energy Derived from Satellite Imagery](https://mlsanity.com/paper/2503.14860/) (2025)

## Abstract (arXiv)

Earth-observation (EO) foundation models provide reusable embeddings, yet downstream task accuracy does not reveal whether these representations encode geographic information, which may be beneficial for location-aware applications but potentially detrimental when representations invariant to geographic location are desired. We therefore evaluate the geographic coordinate robustness of Tessera v1, Tessera v1.1, and AlphaEarth by testing whether coordinates can be predicted from the embedding representations using 284 quality-verified European solar farms from 2024. We assessed geographic information content information through the association between cosine and geodesic distances and through prediction of projected coordinates in EPSG:3035. Embeddings from all three EO foundation models contain recoverable geographic information. All prediction models significantly outperform training-range uniform random sampling baselines, with AlphaEarth exhibiting the strongest distance association and lowest mean geodesic error. Both Tessera variants also yielded higher geographic distance correlations than the Sentinel-2 controls. These findings motivate geographic information content as an additional criterion for auditing EO foundation models.