Main Page

A very simple guide to editing this wiki

Overview

TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysis

This text is suitable for sharing with third parties.

Satellite remote sensing enables a wide range of downstream applications, including habitat mapping, carbon accounting, and strategies for conservation and sustainable land use. However, satellite time series are voluminous and often corrupted, making them challenging to use: the scientific community's ability to extract actionable insights is often constrained by the scarcity of labelled training datasets and the computational burden of processing temporal data.

Our work introduces TESSERA, an open foundation model that preserves spectral-temporal signals in 128-dimensional latent representations at 10-meter resolution globally. It uses self-supervised learning to summarise petabytes of Earth observation data. We compare our work with state-of-the-art task-specific models and other foundation models in five diverse downstream tasks and find that TESSERA closely matches or outperforms these baselines. By preserving temporal phenological signals that are typically lost in conventional approaches, TESSERA enables new insights into ecosystem dynamics, agricultural food systems, and environmental change detection. Moreover, our open-source implementation supports reproducibility and extensibility, while the privacy-preserving design allows researchers to maintain data sovereignty.

To our knowledge, TESSERA is unprecedented in its ease of use, scale, and accuracy: no other foundation model provides analysis-ready outputs, is open, and provides global, annual coverage at 10m resolution using only spectral-temporal features at pixel level.

Preprint: https://arxiv.org/abs/2506.20380

Code: https://github.com/ucam-eo/tessera

15-minute AI-generated podcast: https://svr-sk818-web.cl.cam.ac.uk/keshav/wiki/images/9/9d/TESSERA.m4a

---

1. Background and motivation
2. Literature survey
3. Methods
   1. Computational cost estimation
   2. Hyperparameters
4. Global wall-to-wall map
5. Validation

Future Work

1. Improving representations
2. Downstream tasks

Datasets

1. Sources of environmental data
2. Austrian crop categories
3. California crop data
4. Earth Observation datasets for downstream tasks (spreadsheet)