# Inferring Height from Earth Embeddings: First insights using Google AlphaEarth

arXiv: https://arxiv.org/abs/2602.17250 · HTML: https://mlsanity.com/paper/2602.17250/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 45 references that are in the map.
- First posted on arXiv 19 Feb 2026.

## Builds on (most cited first, 18 of 18)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [UNet++: A Nested U-Net Architecture for Medical Image Segmentation](https://mlsanity.com/paper/1807.10165/) (2018)
- [CROMA: Remote Sensing Representations with Contrastive Radar-Optical Masked Autoencoders](https://mlsanity.com/paper/2311.00566/) (2023)
- [AlphaEarth Foundations: An embedding field model for accurate and efficient global mapping from sparse label data](https://mlsanity.com/paper/2507.22291/) (2025)
- [Lightweight, Pre-trained Transformers for Remote Sensing Timeseries](https://mlsanity.com/paper/2304.14065/) (2023)
- [A Generalizable and Accessible Approach to Machine Learning with Global Satellite Imagery](https://mlsanity.com/paper/2010.08168/) (2020)
- [Towards a Unified Copernicus Foundation Model for Earth Vision](https://mlsanity.com/paper/2503.11849/) (2025)
- [Galileo: Learning Global & Local Features of Many Remote Sensing Modalities](https://mlsanity.com/paper/2502.09356/) (2025)
- [TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysis](https://mlsanity.com/paper/2506.20380/) (2025)
- [GEO-Bench-2: From Performance to Capability, Rethinking Evaluation in Geospatial AI](https://mlsanity.com/paper/2511.15658/) (2025)
- [Global and Dense Embeddings of Earth: Major TOM Floating in the Latent Space](https://mlsanity.com/paper/2412.05600/) (2024)
- [SMARTIES: Spectrum-Aware Multi-Sensor Auto-Encoder for Remote Sensing Images](https://mlsanity.com/paper/2506.19585/) (2025)
- [NeuCo-Bench: A Novel Benchmark Framework for Neural Embeddings in Earth Observation](https://mlsanity.com/paper/2510.17914/) (2025)
- [SRAI: Towards Standardization of Geospatial AI](https://mlsanity.com/paper/2310.13098/) (2023)
- [Cropland Mapping using Geospatial Embeddings](https://mlsanity.com/paper/2511.02923/) (2025)
- [S2Vec: Self-Supervised Geospatial Embeddings for the Built Environment](https://mlsanity.com/paper/2504.16942/) (2025)
- [Democratizing planetary-scale analysis: An ultra-lightweight Earth embedding database for accurate and flexible global land monitoring](https://mlsanity.com/paper/2601.11183/) (2026)
- [Utilizing Earth Foundation Models to Enhance the Simulation Performance of Hydrological Models with AlphaEarth Embeddings](https://mlsanity.com/paper/2601.01558/) (2026)

## Built on by (most cited first, 4 of 4)

- [Exploring the potential of AlphaEarth and TESSERA embeddings for Fine-scale Local Climate Zone Mapping: A case study across five cities in Switzerland](https://mlsanity.com/paper/2606.20034/) (2026)
- [Biomazon: A Multimodal Dataset for 3D Forest Structure and Biomass Modeling in the Amazon Basin](https://mlsanity.com/paper/2606.05368/) (2026)
- [Earth Embeddings](https://mlsanity.com/paper/2608.03410/) (2026)
- [Applying foundation model embeddings towards urban livability evaluation](https://mlsanity.com/paper/2609.09429/) (2026)

## Abstract (arXiv)

This study investigates whether the geospatial and multimodal features encoded in textitEarth Embeddings can effectively guide deep learning (DL) regression models for regional surface height mapping. In particular, we focused on AlphaEarth Embeddings at 10 m spatial resolution and evaluated their capability to support terrain height inference using a high-quality Digital Surface Model (DSM) as reference. U-Net and U-Net++ architectures were thus employed as lightweight convolutional decoders to assess how well the geospatial information distilled in the embeddings can be translated into accurate surface height estimates. Both architectures achieved strong training performance (both with R2 = 0.97), confirming that the embeddings encode informative and decodable height-related signals. On the test set, performance decreased due to distribution shifts in height frequency between training and testing areas. Nevertheless, U-Net++ shows better generalization (R2 = 0.84, median difference = -2.62 m) compared with the standard U-Net (R2 = 0.78, median difference = -7.22 m), suggesting enhanced robustness to distribution mismatch. While the testing RMSE (approximately 16 m for U-Net++) and residual bias highlight remaining challenges in generalization, strong correlations indicate that the embeddings capture transferable topographic patterns. Overall, the results demonstrate the promising potential of AlphaEarth Embeddings to guide DL-based height mapping workflows, particularly when combined with spatially aware convolutional architectures, while emphasizing the need to address bias for improved regional transferability.