This past December, Clark Center for Geospatial Analytics (Clark CGA) joined the global scientific community in New Orleans for the American Geophysical Union (AGU) Fall Meeting. As the world’s largest gathering of Earth and space scientists, the conference brings together over 25,000 attendees from more than 100 countries to tackle urgent environmental challenges. For the Clark University team, attending this year was about more than just presence. It was an opportunity to engage directly with the global community regarding the pressing climate issues that define their work.

The delegation included Clark CGA Director Hamed Alemohammad, Postdoctoral Researchers Sam Khallaghi, Ph.D.’24 and Varun Tiwari, as well as doctoral students Denys Godwin and Rufai Omowunmi Balogun. Together, they observed a distinct change in the atmosphere of the conference. While AI has been growing in recent years, the team noted that 2025 marked the moment it became an inevitable part of the discipline. From using Large Language Models to enhance user interfaces to building specialized GeoAI models for analyzing complex data, AI was almost everywhere.

A major highlight of this year was receiving the AGU Open Science Recognition Prize as part of the Prithvi Geospatial AI Foundation Model Team. This award recognizes a person or team for outstanding work in advancing Open Science related to Earth and space science and its impact globally. The Prithvi Team was recognized for the development of the open-source foundation models  that allow researchers to develop task specific models like flood detection or burn scar mapping by fine-tuning them with limited data, rather than building new models from scratch.

The team also demonstrated real-world applications across climate and ecosystems through several technical presentations. These included research on developing novel self-supervised AI model for P-Band polarimetric SAR sensors, and tools for embedding exploration of GeoAI Foundation Models. The latter project builds on the research carried out at Clark CGA in the last three years to advance geospatial foundation models to learn spatial and temporal properties of satellite imagery. The team also discussed strategies for operationalizing these foundation models to solve near real-time Earth observation challenges.

Beyond presenting research, the meeting aligned strongly with the Clark CGA’s commitment to open science and interdisciplinary collaboration. Alemohammad and Balogun helped facilitate this exchange by chairing sessions on AI Foundation Models and GeoAI Best Practices. These sessions created a space to learn from complementary efforts across academia, industry, and government, helping to translate cutting-edge research into actionable impact.

By empowering students and early-career researchers to present on this global stage, the trip exemplified the core mission of Clark CGA. It provided a space for Clark students to connect with the broader AGU community and grow as researchers. As the team returns to Clark, they bring back not only new insights but a reinforced commitment to driving the progress of geospatial analytics for sustainable environmental stewardship.

The post Shaping the Future of Earth Science: Clark CGA at AGU 2025 appeared first on Center for Geospatial Analytics.

Clark CGA is one of the organizations contributing to JupyterGIS. JupyterGIS is a JupyterLab extension which allows collaborative Geographical Information System (GIS) development. It supports QGIS project files and also provides Python API for Python developers to integrate with Python geospatial libraries. As much of this open-source collaboration happens digitally, the conference allowed Yao-Ting to connect in person with the developers she usually only sees online. It was a chance to strengthen those relationships.

The most significant takeaway from the conference was the rapid evolution of AI Agents. Yao-Ting noted that the AI technology is blooming and has moved beyond just the simple tools. Developers are now learning how to work efficiently alongside AI agents to optimize their workflow and significantly reduce their development time. Additionally, the industry is seeing new capabilities where different AI agents can collaborate with one another. The challenge for developers now is to learn how to partner with these intelligent systems efficiently. Additionally, in one of the AI lectures, Zach Sailer from Apple said that while humans and machines work together, humans remain heavily in the loop, and creating more advanced AI systems is still a deeply human-focused process.

Back at the Center, Yao-Ting is focused on working to integrate these insights into our daily operations. She is further introducing the team to these new AI tools and outlining the pros and cons of incorporating them into our workflows. This experience is helping Clark CGA stay ahead of the curve as we learn to incorporate these powerful new workflows into our research.

The post Beyond the Code: How AI Agents are Reshaping the Jupyter Ecosystem appeared first on Center for Geospatial Analytics.

Clark Center for Geospatial Analytics (Clark CGA) operates on the philosophy of open science, prioritizing the creation of tools and data that are accessible to the wider public. This commitment to sharing resources aligns closely with the global community surrounding QGIS, the world’s leading open source Geographic Information System software. To stay up-to-date on this technology, Research Scientist Rishi Singh and Research Associate Eli Simonson recently traveled to Sweden to attend the annual QGIS Conference.

Unlike proprietary software which is owned and managed by a single company, QGIS relies on a community-driven model. This unique structure allows contributors from around the world to develop features, fix bugs, and share their innovations freely. Furthermore, the conference in Sweden served as a meeting ground for these contributors, bringing together the core developers who build the software and the diverse user base that relies on it for everything from municipal asset tracking to experimental research.

For Simonson and Singh, the trip was an opportunity to engage directly with this international network. Their goal was to immerse themselves in the community and generate new ideas for Clark CGA. A major highlight of the event was seeing how developers are integrating cutting-edge technologies into established platforms. Moreover, Simonson noted that many sessions focused on combining novel methods, such as GeoAI, with the user-friendly interface of QGIS. This trend demonstrates how new innovations can be made accessible to a broader audience without requiring them to learn entirely new software environments.

This focus on accessibility connects directly to the ongoing development work at Clark CGA. As an analyst and developer, Simonson has a particular interest in QGIS plugins, which are custom tools that extend the functionality of the software. Last year, he successfully created a plugin for the UDef-ARP tool, a project originally developed by the Clark CGA software team. By converting this standalone tool into a QGIS plugin, he made it easier for users to visualize data and analyze results within a familiar map interface.

By participating in global events like the QGIS User Conference, the Clark CGA ensures its team remains connected to the latest industry standards. The insights gathered by Simonson and Singh in Sweden strengthen the Center’s ability to build better tools.

The post Connecting with the Global QGIS Community: Clark CGA in Sweden appeared first on Center for Geospatial Analytics.

In an era where Artificial Intelligence (AI) is rapidly reshaping the digital landscape, maintaining the technology behind a research center requires constant adaptation. Staying ahead of these trends is critical for ensuring that Clark Center for Geospatial Analytics (Clark CGA) remains a secure and efficient environment for students and faculty. To navigate these shifting standards, Clark CGA looks to industry leaders and peer communities for the latest insights in IT management and cybersecurity.

This focus on community and shared knowledge was the driving force behind Clark CGA IT Specialist Pete Wason’s recent attendance at the SpiceWorld Conference 2025. Organized for users of the Spiceworks software ecosystem, the event distinguished itself from massive industry trade shows by maintaining an intimate, community-first atmosphere. Moreover, the conference prioritized peer-to-peer problem solving over sales pitches, allowing IT professionals to discuss real-world solutions.

A central theme of this year’s gathering was the complex role of AI in modern networks. While AI was integrated into every session and keynote, the reception among attendees was mixed. Wason noted a distinct divide between industry voices positioning AI as an inevitable tool for efficiency and the professionals who remain skeptical about its reliability. He further highlighted a compelling statistic shared at the event which suggested that nearly two-thirds of IT professionals are hesitant to integrate AI into their systems.

This hesitation often stems from the nature of Large Language Models, which can be difficult to secure and prone to errors. Despite this caution, the conference emphasized the necessity of adapting to AI-driven threats. Additionally, these sessions illustrated how automation is accelerating the speed of external cyber attacks, leaving internal teams with less time to react to security breaches. The consensus was that even if IT managers are wary of using AI, they must be prepared to defend against it.

For Clark CGA, participation in these events provides tangible benefits beyond high-level theory. The conference served as a vital networking hub where Wason connected directly with key vendors, including Dell, whose hardware is integral to the center’s operations. By maintaining these relationships and understanding the problems other technical teams have already solved, Clark CGA ensures its infrastructure is ready to support the next generation of geospatial research at Clark University.

The post Learning from the IT Community: Clark CGA at SpiceWorld Conference 2025 appeared first on Center for Geospatial Analytics.

Since 2014, Center for Geospatial Analytics (Clark CGA) has partnered with the Gordon and Betty Moore Foundation to tackle critical questions in environmental conservation particularly as it relates to sustainable seafood production. Established by Intel co-founder Gordon Moore and his wife Betty Moore, the foundation uses science and technology to solve pressing global problems. One of these challenges is finding the balance between growing food demands and the preservation of natural ecosystems.

To address this delicate balance, and with support from the foundation’s Oceans and Seafood Markets Initiative, this project brings transparency to the global seafood supply chain by monitoring the rapid expansion of aquaculture. Specifically, the team utilizes satellite imagery to identify exactly where shrimp farming is encroaching on vital coastal mangrove forests.

The scope of this monitoring has grown significantly since the project first launched with a focus on Thailand, Cambodia, and Vietnam. Over the last decade, the initiative expanded to track land cover changes across 17 different countries. By analyzing satellite data over specific time intervals, the team can visualize exactly where mangrove forests are being cleared to make way for aquaculture ponds or where reforestation of mangroves is happening.

Now in its sixth phase, the project has reached a new level of technical maturity. According to Rishi Singh, Clark CGA Research Scientist who started as a student researcher on the project nearly a decade ago, this final year has been defined by a major increase in efficiency. After ten years of development, the team now utilizes robust baseline maps and specialized software. Singh notes that these tools allow the current team of student Research Assistants to process data and generate maps faster and more accurately than ever before.

This speed and accuracy are crucial because, unlike traditional academic research that often stays within the university ecosystem, this project delivers tangible products used by major industry players. The maps and data produced by Clark CGA are distributed through a web dashboard and downloaded by organizations responsible for auditing sustainability of food production.

One of the most prominent partners relying on this data is the Monterey Bay Aquarium’s Seafood Watch program. When Seafood Watch assesses a region for its consumer sustainability ratings, they use maps from Clark CGA to audit the environmental impact of farming in that area. Their findings directly influence the purchasing decisions of major retailers like Whole Foods and Costco, as well as certification bodies like the Aquaculture Stewardship Council (ASC). By empowering major market leaders, the project exemplifies the core mission of Clark CGA: to drive the progress of geospatial analytics for sustainable environmental stewardship. This initiative grants Clark University students the opportunity to address pressing global conflicts while producing data that guides sustainable purchasing decisions. As the current phase concludes, it leaves behind a legacy of high-quality data and a proven methodology for safeguarding vital coastal ecosystems.

The post Mapping the Blue Economy: A Decade of Monitoring Aquaculture with the Moore Foundation appeared first on Center for Geospatial Analytics.

The ESA-NASA International Workshop on AI Foundation Model for Earth Observations (EO) was held at ESA-ESRIN in Frascati, Italy from May 5–7. This was a dynamic gathering focused on recent advances in foundation models for EO (EO FMs). Clark CGA’s team participated in the workshop, presented results of our research on explainable AI methods for EO FMs, and moderated multiple sessions during the week.

The workshop brought together experts working on the development and application of these models. NASA’s team presented its Prithvi family of models for EO and weather applications, while ESA and IBM introduced TerraMind, a new model featuring an innovative pretext task that showed promising results on Sentinel satellite products.

A major theme across the sessions was the direct use, fusion, and interpretation of EO FM embedding outputs. Presenters explored how to design models that effectively address the multi-modal nature of remote sensing by leveraging data from diverse imaging sensors, particularly optical, thermal, and SAR. The workshop also highlighted growing efforts to create massive datasets to support self-supervised training, semantic search, and inference directly within learned embedding spaces.

“It was a pleasure to collaborate with our NASA and ESA colleagues to organize this flagship workshop, and join other experts to discuss the latest advancements in this field,” says Dr. Hamed Alemohammad, Director of Clark CGA, who presented at the workshop. “The workshop provided a forum to both engage in technical discussions with other experts, and explore joint opportunities to bring the impact of advancements in EO FMs to solve societal problems”.

“Beyond networking opportunities, it was exciting to see how the field of EO FMs is evolving: there is a clear shift toward more sophisticated handling of multi-modal remote sensing data and a growing interest in how to effectively leverage the embedding space of foundation models,” says Dr. Sam Khallaghi, Postdoctoral Researcher at Clark CGA who attended the workshop. “These developments offered new perspectives that will directly inform the self-supervised learning section of a book chapter I’m currently working on.”

The post Signals from Space, Insights from Embeddings: A Recap of the ESA-NASA EO FM Workshop appeared first on Center for Geospatial Analytics.

For Tammy, this was a conference experience unlike any before. “I don’t think I’ve ever attended a conference as a learner,” she shared. “It was a refreshing change to just be there to listen, connect, and grow.” The sessions—some familiar, others eye-opening offered a deeper understanding of the cloud-native ecosystem and highlighted the power of collaboration over siloed development.

A recurring theme was the importance of building capacity in this space. Conversations touched on the steep learning curves, the need for better tools and training, and the challenge of navigating a growing landscape of formats and standards. Still, the atmosphere was one of encouragement and shared momentum. One quote that resonated with attendees came from Julie Stewart Lowndes of OpenScapes: “Transformation occurs from consistently doing key activities that add up over time.”

Yao-Ting dove into technical sessions that explored formats like Zarr and Icechunk, which are transforming how we manage large geospatial datasets. With Landsat alone expected to reach 240 petabytes of data, these formats promise smarter, more efficient storage and analysis. She also participated in a hands-on Earthmover workshop converting GeoTIFFs into Zarr, gaining insights into workflows that could benefit CGA’s own data pipeline.

A highlight for Yao-Ting was the lightning talk session, where approximately twenty people each presented their work in five minutes.”  She also enjoyed talking with other conference participants, who showed a lot of interest in CGA’s work on Geospatial Foundation Models (GFMs). The response was overwhelmingly positive, with several organizations expressing interest in future collaborations. “CNG felt like a warm retreat for the global geospatial community,” she noted. “It offered chances to connect deeply and exchange ideas with people from all corners of the field.”

Attending CNG was more than just an opportunity to absorb knowledge. It was a moment of reflection and realignment for CGA’s place in a rapidly evolving space. Tammy and Yao-Ting returned with renewed clarity, energized by new tools, new connections, and a stronger sense of how Clark CGA can contribute to and thrive within the growing cloud-native geospatial movement.

The post Inspiring Insights on the Cloud-Native Geospatial (CNG) Conference by CGA Staff appeared first on Center for Geospatial Analytics.