IN FOCUS: GRSS Earth Science Informatics Technical Committee

The ESI TC operates under the leadership of a chair and supporting committee members drawn from academia, government agencies, and industry. “We encourage participation from all our members,” explains Chair Dr. Manil Maskey, Manager and Senior Research Scientist at NASA’s Marshall Space Flight Center. His vision for the ESI TC is to accelerate the responsible adoption of advanced data-driven technologies as a transformative force for Earth science discovery and decision-making across geoscience and remote sensing research and application.

“I took on the role of chair of the TC because of a rare moment in the field’s history,” explains Dr. Maskey. “With such significant Earth observation data, large-scale AI models, and infrastructure available to Earth science, it is clear that this is a moment to take part and to help define the future of Earth science.”

“My goal with the TC is twofold,” he adds. “First, to create a connection between the technology community and Earth science to ensure people understand the significance of the data. Second, to grow the community and specifically include new researchers and practitioners from different regions of the world who are doing significant work in the field but do not have a home within IEEE GRSS yet.”

Dr. Maskey observes that the field of AI for Earth science is rapidly developing, and that large-scale models trained on Earth observation data are performing tasks beyond what was imagined just three years ago. “To remain at the forefront of the field, I will bring to the committee the perspective of someone who has worked with AI solutions being deployed at scale and for significant outcomes to ensure the TC remains connected to the challenges of Earth science research and applications,” he explains.

Trends and Challenges in Earth Science Informatics

Earth science informatics has become a critical research area as Earth observation data grow in volume, variety, and velocity, moving beyond traditional pixel‑based analysis toward multi-dimensional, scalable approaches. Advances in artificial intelligence, cloud computing, and interoperable data infrastructures now enable foundation models that integrate optical, radar, and other sensor modalities for spatiotemporal analysis, supporting applications such as monitoring vegetation dynamics, surface deformation, and long-term climate signals. At the same time, analysis‑ready datacubes, spatiotemporal databases, and knowledge-representation frameworks provide standardized, multidimensional structures and query capabilities that make large geoscience datasets more accessible and actionable. By connecting these innovations, the ESI Technical Committee fosters the development and exchange of informatics methods that support reproducible, interdisciplinary research at scale.

“The timing couldn’t be more critical,” says Dr. Maskey. “We are in the midst of a shift in how AI is interacting with Earth observation data. Geospatial foundation models trained on petabytes of Earth observation data are generalizing and performing various tasks ranging from identifying flooded areas and crop conditions to detecting wildfires and lunar surfaces. The ESI TC is in the midst of this shift.”

“The opportunity to take advantage of these advancements comes with challenges,” cautions Dr. Maskey. “Most published research in the area outpaces the community’s ability to validate the findings of those publications. Most AI models are being deployed into critical contexts before there is general agreement on how to evaluate or fail those models. The technical committee has the opportunity to close this gap. Furthermore, there’s a data moment happening simultaneously. Data from commercial satellite constellations, Earth observation archives, and AI-ready datasets have become more available to the research community than ever before. The challenge isn’t access to these data but how to make sense of it all at scale. The Earth observation AI community and the ESI TC will have a role to play in these discussions, and I’m glad they’re having them now instead of after the fact,” he says.

Technical Scope and Dedicated Working Groups

A defining feature of the ESI TC is its emphasis on well-defined working groups that address core challenges in Earth science informatics. Together, these working groups define the technical core of the ESI TC and provide a structured framework for ongoing efforts, collaboration and outreach.

The High-Performance and Distributed Computing for Remote Sensing (HDCRS) working group focuses on advancing the processing and analysis of large-scale Earth observation data through cutting-edge computational paradigms. Its technical scope spans high-performance and distributed computing (including cloud and supercomputing platforms), specialized hardware such as GPUs, FPGAs, and ASICs, and emerging quantum computing approaches and technologies, all applied to scalable machine learning and geospatial analytics. Alongside this technical focus, HDCRS actively fosters community engagement and knowledge transfer through educational initiatives, conference sessions, and popular training events, including its annual HDCRS Summer School. 

The Databases in Remote Sensing (DBRS) working group focuses on advancing database technologies for the efficient storage, management, and analysis of large-scale Earth observation data. Its technical scope includes spatio-temporal databases, array-based datacubes, and knowledge graph representations of geospatial information. The group is particularly active in bridging the gap between database research and remote sensing applications, promoting approaches that support efficient querying, indexing, and visualization of complex datasets, while fostering collaboration and knowledge exchange through workshops, conference sessions, and related community activities.

The ESI Technical Committee has expanded into artificial intelligence–driven Earth science through the Earth Science Foundation Models (Earth² or E²) initiative. This effort centers on the development and optimization of geoscience-focused foundation models (GeoFMs) and data-centric Geo-AI, highlighting the growing role of large-scale machine learning in Earth observation. The Earth² group emphasizes cross-disciplinary collaboration between AI and geoscience experts, promotes open science through the sharing of data, models, and knowledge, and supports community training and education on the effective development and application of these models, with the goal of maximizing the utility of Earth science data across a wide range of applications.

Broad Range of Activities and Initiatives

The ESI Technical Committee plays a visible role in major GRSS activities, particularly through the society’s flagship global and regional conferences such as IEEE International Geoscience and Remote Sensing Symposium (IGARSS) and IEEE InGARSS (India Geoscience and Remote Sensing Symposium. Additionally, the ESI TC is leading multiple projects under IEEE GRSS Big Ideas Grant involving the TC working groups and partners from the industry. The committee regularly organizes and contributes to special sessions, workshops, webinars, and tutorials covering topics in Earth science informatics, including cloud computing, big data analytics, and AI applications for geoscience. In addition to conference engagement, the committee promotes community building and knowledge exchange through its working groups, which focus on emerging areas such as foundation models and large-scale multidimensional data management. 

For example, at IGARSS 2025, the ESI TC presented a full-day tutorial, “Lifecycle of Large-Scale AI Models in the Cloud: A Focus on Deploying and Fine-Tuning Geospatial Foundation Models.” Participants explored key aspects of geospatial data analysis and tackle challenges unique to Earth Observation (EO), such as processing multi-source and multitemporal satellite remote sensing datasets, and acquired skills to use FMs effectively across various stages of geoscience research and practical applications.

In addition, the ESI TC have developed a webinar series  that showcases topics directly aligned with its Earth science informatics mission, featuring presentations on frameworks and technologies that support scalable data management, interoperable analysis, and advanced analytics for Earth observation. Recent webinars have included “GeoCroissant – A Metadata Framework for Geospatial ML‑ready Datasets,” which discusses metadata standards to support machine learning workflows; “Towards Digital Twin: Introduction to Foundation Models for Geoscience,” which explores how foundation models can advance integrated Earth system representations toward digital twin concepts; and “VEDA: An Open, Interoperable Platform for Open Science,” which highlights an open platform for visualizing and analyzing NASA Earth science data. These events illustrate the committee’s engagement with metadata frameworks, foundation models, open science infrastructure, and other emerging informatics topics of interest to the community.

“Outreach is central to achieving our goals,” says Dr. Maskey. “We catalyze advances in geoscience and remote sensing through innovation, collaboration, and discovery. We inspire and empower the next generation of scientists and engineers,” he adds.

EO DataCubes

A major activity of the ESI TC is the IEEE GRSS EO Datacube effort. The EO-Cube is a cloud-native, high-performance Earth Observation (EO) data cube designed to modernize how researchers interact with massive, multi-dimensional satellite datasets. Functioning as a centralized archival and curation resource, it organizes data across space, time, and spectral bands, allowing users to perform complex server-side analytics using the rasdaman datacube engine. This infrastructure is unique because it eliminates the need for data downloading and local preprocessing by providing Analysis-Ready Data (ARD) through a “location-transparent” federation. This means it can automatically fuse and query data from various global sources, such as Copernicus hubs, through a single access point, making it an essential tool for scalable global environmental monitoring.

The initiative addresses the community’s critical need for advanced data stewardship and standardized informatics tools. “It directly supports the ESI TC’s mission to bridge the gap between database experts and remote sensing scientists,” says Dr. Maskey. “Ultimately, it lowers the barrier to entry for complex Earth science research, ensuring that both seasoned experts and students have the infrastructure support required to derive new insights from voluminous remote sensing data,” he adds.

Engage with the ESI TC

“We encourage interested students, researchers, or practitioners to engage with the ESI TC,” says Dr. Maskey. “Emerging areas such as foundation models, real-time data processing, and integrated Earth system analytics are expected to shape the committee’s future activities, and our members contribute to our field in so many meaningful ways. Involvement in ESI TC provides a direct way to stay connected, share ideas, and advance the field,” he adds. 

Find out how to support the initiatives of the ESI TC.