AI & Environment Resource HubFeng, Zhengpeng; Atzberger, Clement; Jaffer, Sadiq; Knezevic, Jovana; Sormunen, Silja; Young, Robin; Lisaius, Madeline C.; Immitzer, Markus; Jackson, Toby; Ball, James; Coomes, David A.; Madhavapeddy, Anil; Blake, Andrew; Keshav, Srinivasan · 2025-07-29
This paper introduces TESSERA, an open global remote sensing foundation model that fuses optical and radar satellite time series using Transformer-based encoders to produce ready-to-use 10 m embeddings, outperforming or matching state-of-the-art models across diverse ecological and agricultural tasks.
Feng, Zhengpeng; Atzberger, Clement; Jaffer, Sadiq; Knezevic, Jovana; Sormunen, Silja; Young, Robin; Lisaius, Madeline C.; Immitzer, Markus; Jackson, Toby; Ball, James; Coomes, David A.; Madhavapeddy, Anil; Blake, Andrew; Keshav, Srinivasan. 2025-07-29. TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysis. https://arxiv.org/abs/2506.20380 (AI & Environment Resource Hub; record paper-358; collection snapshot 2026-09-15).
Included in the original Scientific Paper collection. Labels below are metadata based suggestions. They are not verified findings, claims of effectiveness, or endorsements.
Source sheet: Scientific Paper · Row 362 · Original ID: paper-358.
A paper listing is not a quality assessment. Peer review and findings require source-level confirmation. Source link reachable · checked 2026-09-15. Source identity and required metadata verified. The import date is not the original date added.
Suggest a correction for this record →Ordered by the number of shared provisional topic labels. This indicates a browsing connection, not agreement between sources.
Shared topics: Research tools and geospatial methods; Climate and greenhouse gases; Biodiversity and ecosystems; Weather, hazards, and adaptation
This study presents a new machine learning (Light Gradient Boosting Machine) approach to generate high-resolution, hourly maps of gross primary productivity (GPP) for East Asia (2020–2021). The paper highlights clear diurnal patterns across different land cover types and latitudes, offering valuable insights for ecosystem monitoring and carbon cycle modeling.
Shared topics: Research tools and geospatial methods; Climate and greenhouse gases; Biodiversity and ecosystems; Agriculture and food
Earth Map is an innovative, free and open-source tool developed by the Food and Agriculture Organization of the United Nations (FAO) in the framework of the FAO - Google partnership. It was created to support countries, research institutes, farmers and members of the general public with internet access to monitor their land in an easy, integrated and multi-temporal manner. It allows everyone to visualize, process and analyze satellite imagery and global datasets on climate, vegetation, fires, biodiversity, geo-social and other topics. Users need no prior knowledge of remote sensing or Geographical Information Systems (GIS).
Shared topics: Research tools and geospatial methods; Climate and greenhouse gases; Agriculture and food; Weather, hazards, and adaptation
Continuous and reliable Earth observation is critical for real-world systems like hurricane forecasting and wildfire detection. GAIA demonstrates promising capabilities, both in filling large gaps in satellite records and converting satellite imagery into precipitation estimates. GAIA represents a new class of geospatial AI models with the potential to create significant impact across sectors—from more accurate weather forecasting to applications in insurance, power utilities, aviation, and agriculture.
Shared topics: Research tools and geospatial methods; Climate and greenhouse gases; Biodiversity and ecosystems; Weather, hazards, and adaptation
Prithvi-EO-2.0, a geospatial foundation model trained on NASA satellite data, outperforms its predecessor and other models in remote sensing tasks, aiding applications like disaster response and ecosystem monitoring.