cs.CVNov 1, 2026

Forecasting Earth Surface Change in Embedding Space

Authors: Jingtao Li, Sadiq Jaffer, Anil Madhavapeddy, Pedro Sousa, Anthony Ortiz, Kyle Gao, Zhengpeng Feng, Yanfei Zhong, 

Organizations: Department of Computer Science, University of Cambridge · Microsoft AI for Good Research Lab · Aalto University · Wuhan University

Abstract

Repeated satellite imagery permits understanding of how our planet changes over time, letting us predict its future and plan mitigating actions. Most existing studies that rely on a time series of satellite observations forecast the spectral values at each pixel for a future target date. However, clouds, shadows, and sensor noise often corrupt these observations, so that spectral forecasters waste much of their capacity on low-level features or noise rather than real change. With the advent of Earth foundation models, we argue that forecasting in embedding space is a feasible and more effective alternative. To this end, we design a token space that handles irregular histories from different sensors and foundation models and supports forecasts at any query date. Tokens from this space are inputs to our proposed forecaster, E2^2Forecast, which is a transformer with two branches: a Gaussian branch that directly predicts a mean point forecast and per-dimension standard deviations, and a rectified-flow branch that generates forecast ensembles. On Sentinel-1 and Sentinel-2 benchmarks, forecasting embeddings instead of spectral values halves the cosine distance (from 0.20 to 0.10) and greatly reduces overconfidence, lowering the z-score standard deviation from 2.63 to 1.22. By comparing new observations with its forecasts, E2^2Forecast can also detect anomalous changes without labels or retraining, raising F1_1 from 41.58 to 48.70 on a global benchmark. This consistent superiority is validated on nine remote sensing foundation models. Beyond change detection, we believe our approach advances the field by permitting effective reuse of diverse pretrained task heads. Our code is publicly available at https://github.com/ucam-eo/e2forecast.

Figures & tables

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. SPEAR NeXT Causal Latent Forecasting Across Multiple Horizons for Spectral Temporal Earth Representation Learning

    Sep 15, 2026Rajiv Ranjan, Udaiveer Singh, Shashank Tamaskar +1Multivariate Time Series ForecastingGeospatial Foundation Models

  2. RS-WorldModel: a Unified Model for Remote Sensing Understanding and Future Sense Forecasting

    Mar 16, 2026Linrui Xu, Zhongan Wang, Fei Shen +4World Model LearningRemote Sensing VLMs