Deploying Geospatial Foundation Models in the Real World: Lessons from WorldCereal

Christina Butsko, Gabriel Tseng, Kristof Van Tricht, Giorgia Milli, David Rolnick, Ruben Cartuyvels, Inbal Becker-Reshef, Zoltan Szantoi, Hannah Kerner
Proceedings of The TerraBytes {ICML} Workshop: Towards global datasets and models for Earth Observation, PMLR 292:13-31, 2025.

Abstract

The increasing availability of geospatial foundation models has the potential to transform remote sensing applications such as land cover classification, environmental monitoring, and change detection. Despite promising benchmark results, the deployment of these models in operational settings is challenging and rare. Standardized evaluation tasks often fail to capture real-world complexities relevant for end-user adoption such as data heterogeneity, resource constraints, and application-specific requirements. This paper presents a structured approach to integrate geospatial foundation models into operational mapping systems. Our protocol has three key steps: defining application requirements, adapting the model to domain-specific data and conducting rigorous empirical testing. Using the Presto model in a case study for crop mapping, we demonstrate that fine-tuning a pre-trained model significantly improves performance over conventional supervised methods. Our results highlight the model’s strong spatial and temporal generalization capabilities. Our protocol provides a replicable blueprint for practitioners and lays the groundwork for future research to operationalize foundation models in diverse remote sensing applications. Application of the protocol to the WorldCereal global crop-mapping system showcases the framework’s scalability.

Cite this Paper

BibTeX
@InProceedings{pmlr-v292-butsko25a, title = {Deploying Geospatial Foundation Models in the Real World: Lessons from WorldCereal}, author = {Butsko, Christina and Tseng, Gabriel and Van Tricht, Kristof and Milli, Giorgia and Rolnick, David and Cartuyvels, Ruben and Becker-Reshef, Inbal and Szantoi, Zoltan and Kerner, Hannah}, booktitle = {Proceedings of The TerraBytes {ICML} Workshop: Towards global datasets and models for Earth Observation}, pages = {13--31}, year = {2025}, editor = {Audebert, Nicolas and Azizpour, Hossein and Barrière, Valentin and Castillo Navarro, Javiera and Czerkawski, Mikolaj and Fang, Heng and Francis, Alistair and Marsocci, Valerio and Nascetti, Andrea and Yadav, Ritu}, volume = {292}, series = {Proceedings of Machine Learning Research}, month = {19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v292/main/assets/butsko25a/butsko25a.pdf}, url = {https://proceedings.mlr.press/v292/butsko25a.html}, abstract = {The increasing availability of geospatial foundation models has the potential to transform remote sensing applications such as land cover classification, environmental monitoring, and change detection. Despite promising benchmark results, the deployment of these models in operational settings is challenging and rare. Standardized evaluation tasks often fail to capture real-world complexities relevant for end-user adoption such as data heterogeneity, resource constraints, and application-specific requirements. This paper presents a structured approach to integrate geospatial foundation models into operational mapping systems. Our protocol has three key steps: defining application requirements, adapting the model to domain-specific data and conducting rigorous empirical testing. Using the Presto model in a case study for crop mapping, we demonstrate that fine-tuning a pre-trained model significantly improves performance over conventional supervised methods. Our results highlight the model’s strong spatial and temporal generalization capabilities. Our protocol provides a replicable blueprint for practitioners and lays the groundwork for future research to operationalize foundation models in diverse remote sensing applications. Application of the protocol to the WorldCereal global crop-mapping system showcases the framework’s scalability.} }
Endnote
%0 Conference Paper %T Deploying Geospatial Foundation Models in the Real World: Lessons from WorldCereal %A Christina Butsko %A Gabriel Tseng %A Kristof Van Tricht %A Giorgia Milli %A David Rolnick %A Ruben Cartuyvels %A Inbal Becker-Reshef %A Zoltan Szantoi %A Hannah Kerner %B Proceedings of The TerraBytes {ICML} Workshop: Towards global datasets and models for Earth Observation %C Proceedings of Machine Learning Research %D 2025 %E Nicolas Audebert %E Hossein Azizpour %E Valentin Barrière %E Javiera Castillo Navarro %E Mikolaj Czerkawski %E Heng Fang %E Alistair Francis %E Valerio Marsocci %E Andrea Nascetti %E Ritu Yadav %F pmlr-v292-butsko25a %I PMLR %P 13--31 %U https://proceedings.mlr.press/v292/butsko25a.html %V 292 %X The increasing availability of geospatial foundation models has the potential to transform remote sensing applications such as land cover classification, environmental monitoring, and change detection. Despite promising benchmark results, the deployment of these models in operational settings is challenging and rare. Standardized evaluation tasks often fail to capture real-world complexities relevant for end-user adoption such as data heterogeneity, resource constraints, and application-specific requirements. This paper presents a structured approach to integrate geospatial foundation models into operational mapping systems. Our protocol has three key steps: defining application requirements, adapting the model to domain-specific data and conducting rigorous empirical testing. Using the Presto model in a case study for crop mapping, we demonstrate that fine-tuning a pre-trained model significantly improves performance over conventional supervised methods. Our results highlight the model’s strong spatial and temporal generalization capabilities. Our protocol provides a replicable blueprint for practitioners and lays the groundwork for future research to operationalize foundation models in diverse remote sensing applications. Application of the protocol to the WorldCereal global crop-mapping system showcases the framework’s scalability.
APA
Butsko, C., Tseng, G., Van Tricht, K., Milli, G., Rolnick, D., Cartuyvels, R., Becker-Reshef, I., Szantoi, Z. & Kerner, H.. (2025). Deploying Geospatial Foundation Models in the Real World: Lessons from WorldCereal. Proceedings of The TerraBytes {ICML} Workshop: Towards global datasets and models for Earth Observation, in Proceedings of Machine Learning Research 292:13-31 Available from https://proceedings.mlr.press/v292/butsko25a.html.

Related Material