TY - JOUR
AU - Szwarcman, Daniela
AU - Roy, Sujit
AU - Fraccaro, Paolo
AU - Gíslason, Orsteinn Elí
AU - Blumenstiel, Benedikt
AU - Ghosal, Rinki
AU - De Oliveira, Pedro Henrique
AU - Almeida, Joao Lucas de Sousa
AU - Sedona, Rocco
AU - Kang, Yanghui
AU - Chakraborty, Srija
AU - Wang, Sizhe
AU - Gomes, Carlos
AU - Kumar, Ankur
AU - Gaur, Vishal
AU - Truong, Myscon
AU - Godwin, Denys
AU - Khallaghi, Sam
AU - Lee, Hyunho
AU - Hsu, Chia-Yu
AU - Asanjan, Ata Akbari
AU - Mujeci, Besart
AU - Shidham, Disha
AU - Balogun, Rufai Omowunmi
AU - Kolluru, Venkatesh
AU - Keenan, Trevor
AU - Arevalo, Paulo
AU - Li, Wenwen
AU - Alemohammad, Hamed
AU - Olofsson, Pontus
AU - Mayer, Timothy
AU - Hain, Christopher
AU - Kennedy, Robert
AU - Zadrozny, Bianca
AU - Bell, David
AU - Cavallaro, Gabriele
AU - Watson, Campbell
AU - Maskey, Manil
AU - Ramachandran, Rahul
AU - Moreno, Juan Bernabe
TI - Prithvi-EO-2.0: A Versatile Multi-Temporal Foundation Model for Earth Observation Applications
JO - IEEE transactions on geoscience and remote sensing
VL - 64
SN - 0018-9413
CY - New York, NY
PB - IEEE
M1 - FZJ-2025-05777
SP - 4400120
PY - 2025
AB - This paper presents Prithvi-EO-2.0, a new geospatial foundation model that offers significant improvements over its predecessor, Prithvi-EO-1.0. Trained on 4.2 million global time series samples from NASA’s Harmonized Landsat and Sentinel-2 data archive at 30-m resolution, the new model incorporates temporal and location embeddings for enhanced performance across various geospatial tasks. Through extensive benchmarking with GEO-Bench, the model outperforms the previous Prithvi-EO model by 8% across a range of tasks. It also outperforms six other geospatial foundation models when benchmarked on remote sensing tasks from different domains and resolutions (i.e. from 0.1 m to 15 m). The results demonstrate the versatility of the model in both classical Earth observation and high-resolution applications. Early involvement of end-users and subject matter experts (SMEs) allowed constant feedback on model and dataset design, enabling customization across diverse SME-led applications in disaster response, land cover and crop mapping, and ecosystem dynamics monitoring. Prithvi-EO-2.0 is available as an open-source model on Hugging Face and IBM TerraTorch, with additional resources on GitHub. The project exemplifies the Trusted Open Science approach embraced by all involved organizations.
LB - PUB:(DE-HGF)16
DO - DOI:10.1109/TGRS.2025.3642610
UR - https://juser.fz-juelich.de/record/1050065
ER -