2nd ACM SIGSPATIAL International Workshop on Polar Data Science (PolDS 2026)

November 3, 2026 in Riverside, CA, USA
Half Day workshop at the 34th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems PolDS is announcing a Call for Papers for 2nd ACM SIGSPATIAL International Workshop on Polar Data Science (PolDS 2026).

PolDS 2026 aims to connect the polar science community with the spatial computing community to foster convergent approaches that will address significant questions in the Arctic and Antarctic regions. Addressing phenomena such as dynamic modeling of the ice bed, tracking the internal layers of the ice sheet, and investigating causal relationships between ice sheets, sea ice, and atmosphere requires developing new techniques for spatial and spatio-temporal data analysis, spatial machine learning, and spatial datainfrastructure.
Papers are being accepted on a broad range of topics exploring geospatial AI/ML techniques to detect novel patterns within data and promote scientific discovery in the polar regions.

Themes may include (but are not limited to):

• Improving our ability to project the future ice sheet contribution to sea-level rise across varying spatial scales
• New insights into ice-dynamics
• Improved estimate and visualization of the ice sheet and subglacial topography at a global scale
• Lining satellite-based observations of the near-surface with both atmospheric drivers and effects on the ice sheet
• Ensuring FAIR reproducibility of these scientific discoveries
• Accelerating discoveries in the polar regions with geospatial AI
• Improving our understanding of global and local sea level rise.

Submission Deadline: August 15, 2026
Call for Papers Link: https://easychair.org/cfp/polds2026
Workshop Link: https://iharp.umbc.edu/polds26/

Additional Information: 
ACM SIGSPATIAL 2026 Conference website

Francis successfully defended on Thursday, July 16, 2026

Francis successfully defended his dissertation on Thursday, July 16, 2026. Francis's dedicated contributions have made him an important member of the iHARP and UMBC communities. This achievement is a true testament to his scholarly rigor and hard work.

Congratulations, Francis, on this incredible milestone! We have been honored to witness your growth and look forward to your future successes. The iHARP community wishes you the very best as you embark on the next chapter of your journey!

_____________________________

Dissertation Title: Accurate Clustering of High-Dimensional Multivariate Spatiotemporal Data

Committee

• Dr. Jianwu Wang, Chair/Advisor, UMBC
• Dr. Vandana P. Janeja, Co-Advisor, UMBC
• Dr. James Foulds, UMBC
• Dr. Yiqun Xie,  University of Maryland, College Park.
• Dr. Aneesh Subramanian, University of Colorado Boulder

Abstract

The rapid growth of multidimensional multivariate spatiotemporal data from Earth observation systems, sensor networks, climate reanalysis products, and large- scale monitoring platforms has created unprecedented opportunities for understanding complex natural and human-driven processes. These datasets simultaneously vary across space, time, and multiple variables, enabling the discovery of hidden patterns, climate regimes, anomalies, and evolving system behaviors. However, clustering such data remains challenging due to high dimensionality, nonlinear interactions, spatial autocorrelation, long-range temporal dependencies, noise, missing values, nonstationarity, and multiscale structures. Traditional clustering approaches
primarily rely on distance-based or correlation-based similarities and often fail to capture the complex spatial, temporal, and causal mechanisms underlying these systems.

To address these challenges, this dissertation develops three novel deep clustering frameworks for multivariate spatiotemporal data. First, Hybrid Ensemble Deep Graph Temporal Clustering (HEDGTC) integrates homogeneous and heterogeneous ensemble clustering with a dual-consensus strategy and a Deep Graph Attention Autoencoder to improve clustering robustness, stability, and accuracy. By combining object co-occurrence consensus and non-negative matrix factorization consensus, HEDGTC effectively reduces noise and misclassification while preserving temporal and relational structures. 

Building on the foundation established by HEDGTC, we introduce the Bi-directional Temporal Graph Attention Transformer (B-TGAT), an end-to-end deep clustering framework for multivariate spatiotemporal data. The proposed architecture combines a ConvLSTM-based U-Net autoencoder for learning rich spatiotemporal representations, graph attention mechanisms for modeling spatial relationships, and a bi-directional transformer for capturing long-range temporal dependencies in both forward and backward directions. This integrated design enables B-TGAT to effectively learn complex spatial and temporal patterns from climate data. By jointly modeling local spatial interactions and long-term temporal dynamics, B-TGAT un- covers meaningful climate regimes, large-scale teleconnections, abrupt climate transitions, and extreme weather events. The learned latent representations improve cluster compactness, stability, and separation, leading to more accurate, robust, and interpretable clustering of multivariate spatiotemporal climate datasets. This unified framework provides a powerful solution for discovering hidden climate patterns that are difficult to identify using conventional clustering approaches. 

Recognizing that correlation-based representations alone may not reveal the true drivers of system evolution, this dissertation further proposes Causal Adversarial Subspace Clustering (CASC), a causality-aided deep clustering framework that transforms spatiotemporal regime discovery from a correlation-driven process into a causal-temporal learning paradigm. CASC integrates a U-Net-inspired adversarial autoencoder, stacked FAConvLSTM layers, graph attention-based self-expressive learning, and two novel objectives: Causal Subspace Preservation (CSP) Loss and Dynamic Temporal Subspace Evolution (DTSE) Loss. These mechanisms enable the model to discover clusters that are not only geometrically compact but also causally meaningful and temporally coherent. A Subspace-Aware Energy-Based TemporalDiscriminator further enhances clustering stability, interpretability, and robustness by evaluating latent representations according to cluster-specific subspace structures.
Extensive experiments on multiple real-world climate and environmental datasets demonstrate  that the proposed frameworks consistently outperform state-of-the-art traditional and deep clustering approaches across clustering quality, robustness, stability, and interpretability metrics. Collectively, the proposed methods establish a unified framework for learning hierarchical spatial, temporal, and causal representations from complex spatiotemporal data. The resulting models provide powerful tools for climate regime discovery, environmental monitoring, anomaly detection, and scientific understanding of dynamic Earth systems, while advancing the state of the art in deep unsupervised spatiotemporal clustering.

Call for UMBC SCIPE (UM-CIP Fellows) Graduate Fellows - Fall 2026
Submission Deadline: July 10, 2026

Funded by the National Science Foundation (NSF) SCIPE program (Enhancing the Transdisciplinary Research Ecosystem for Earth and Environmental Science with Dedicated Cyber Infrastructure Professionals), CGC-SCIPE is a collaborative effort between the University of Maryland, Baltimore County (UMBC) - iHARP and the University of Maryland Center for Environmental Sciences (UMCES). The project aims to strengthen the research capacity in Earth and Environmental Sciences by leveraging High Performance Computing (HPC), Artificial Intelligence, and data-driven modeling, while developing the next generation of Cyber Infrastructure Professionals (CIPs). This project, as part of the iHARP center and Mdata lab, invites applications for UMBC CIP Graduate Fellows (UM-CIP Fellows).

About the Fellowship
The UM-CIP Graduate Fellowship supports a senior PhD student whose dissertation research aligns with Earth or Environmental Science and who seeks to incorporate HPC or advanced AI into their work. The Fellows will receive guidance from an HPC mentor to expand their technical and computational expertise.

Fellowship Details
* Duration: One Year (Starting Fall 2026)
* Funding: Support for Fall 2026 and Spring 2027, including tuition, health insurance, and stipend. Note that summer will be supported at 20 hours per week.
* Focus Areas: Earth and Environmental Science (Students may come from any department but must have these focus areas for their research). 

Fellowship Eligibility
* Must be a PhD candidate at UMBC who has completed their dissertation proposal defense by the time of the appointment.
* Must have a research focus in Earth or Environmental Science.
* Must demonstrate interest in using advanced computing High Performance Computing (HPC) in the research. They should describe how they plan to incorporate HPC into their research.
* Students who are already using advanced computing will not be considered for this call. Fellowship Expectations
* A fellow is expected to continue with their research topics identified in their dissertation proposal defense and continue to be mainly advised by their PhD advisor(s)
* A fellow is expected to engage with the SCIPE team and Cyber Infrastructure Professionals (CIPs)
* A fellow will be expected to present the progress of their work during the fellowship year.
* A fellow's work will have the UMBC SCIPE grant in acknowledgment of all work developed through the fellowship. A final publication is an expected outcome of this fellowship.
* A fellow is expected to participate in CGC-SCIPE related events, such as workshops and seminars.
* Fellowship stipend will be based upon the current Information Systems Department stipend rate.
* Previously awarded fellows are not eligible.
* As this is a full time fellowship, fellows may not be supported by another grant funding source or full time support.

Applicants should submit a one-page research statement outlining the research question and approach, the type of data used or planned, and how HPC will be incorporated or enhanced through the fellowship. Please attach a copy of your CV or Resume and provide at least one Letter of Recommendation (preferably from your PhD advisor).

Please submit via this Google Submission Form by July 10, 2026.
If you have any questions, please contact Dr. Vandana Janeja (vjaneja@umbc.edu).

Undergraduate Data Science and AI Scholars

Fall 2026 Call for Applications 

Deadline: Friday, July 10

We are pleased to announce the Undergraduate Data Science and AI Scholars program for supporting students in Fall 2026. The Data Science and AI Scholars program welcomes undergraduate scholars from multiple disciplines, including non-STEM majors, who will examine all aspects of data and its impact on society.

This year, scholars will have the opportunity to be teaching fellows or research fellows. Scholars will be able to work on research projects in Climate Science, Deepfake detection, and other data science-related projects. Students selected through this effort will be part of the Data Science and AI Scholars program, which will be run in partnership with the Center for Women in Technology (CWIT), the Information Systems Department (IS), and the Center for Social Science Scholarship (CS3). Selected scholars will be assigned to respective affiliations (CWIT/IS or CS3) depending on the student's major, and the scholars' expectations per affiliation are as follows:

Center for Women in Technology (CWIT) Responsibilities:

Research Fellows

• Scholars who will work as research fellows will work on a supervised research project and be mentored by faculty. Scholars will be paid $20.00/hour for specific tasks, for 5 - 10 hours per week during Fall 2026.
• Scholars will attend at least two events in Fall 2026 and/or participate in research initiatives and professional development for their leadership growth and mentoring. Mentors will coordinate this at CWIT.
• Scholars will need to register as CWIT Affiliates to participate in the 2 required events.
  Scholars will be required to enroll in a Zero-credit Practicum course on research experience (PRAC 98C) through the UMBC Career Center.
  

Teaching Fellows

• Scholars will work as teaching fellows and peer mentors for undergraduate students in IS 296 - Foundations of Data Science, and support students throughout the course.. The scholars will devote time to mentoring and leadership development efforts for their advancement. The cohort of scholars will meet as a group under the supervision of the IS 296 instructor. Scholars will be paid $20.00/hour for specific tasks, for 5 - 10 hours per week during Fall 2026.

• Scholars who will work as teaching fellows will be expected to be available to assist with classes and also present during lectures. Fall 2026 class schedule is as follows: Mondays 4:30 - 7:00 pm (Instructor: Dr. Karen Chen), and Wednesdays 4:30 - 7:00 pm (Instructor: Zehra Zaidi)

• Scholars will be required to attend one mentoring meeting per month with the coordinator
  Scholars will be required to enroll in a Zero-credit Practicum course on research experience (PRAC 98C) through the UMBC Career Center. 

Center for Social Science Scholarship (CS3) Responsibilities:

Research Fellows 

• Scholars who will work as research fellows will work on a supervised research project and be mentored by faculty. Scholars will be paid $20.00/hour for specific tasks, for 5 - 10 hours per week during Fall 2026.
• Scholars will attend at least two events in Fall 2026 and/or participate in research initiatives and professional development for their leadership growth and mentoring. Mentors will coordinate this at CS3.
• Scholars will also be expected to attend one mentoring meeting with the faculty.
  Scholars will be required to enroll in a Zero-credit Practicum course on research experience (PRAC 98C) through the UMBC Career Center.

Background and Skills:

ALL applicants should have knowledge in AT LEAST ONE of the following:

• Analysis of social, behavioral, economic, or geographic data
• Python or one of the data science tools and/or languages (such as R, Rapid Miner, Weka, Orange, Knime, ML on cloud computing)
• Jupyter Notebooks with Python or taken IS 296 in a prior semester.

Application Form:

Interested students should complete the Google Submission Form, OR access the form using the following link: https://forms.gle/FFUAa5RpDNMGYvTY9 by Friday, July 10, 2026.

Questions:
For questions, please email our team at DataScienceAIScholars@umbc.edu
To learn more about potential projects, please check out the following websites:
➔ https://www.umces.edu/chesapeake-global-collaboratory
➔ https://iharp.umbc.edu
➔ https://mdata.umbc.edu/deep-fake-detection/
➔ https://socialscience.umbc.edu/

The Polar Ice Museum :: From Greenland to South Baltimore

• Opening Reception Thursday, April 30, 4:00 - 6:00 pm
• Imaging Research Center, ITE 108 (first floor) UMBC
• Additional hours: Friday, May 1, 10–4pm, and Saturday, May 2, 11 to 4pm

An Imaging Research Center (IRC) collaboration with iHARP presents an arc of monitors displaying an ice cave within a game-like environment. As the viewer's own CO2 (breath) enters the space, a series of events animates the monitors. These global events, including the emergence of lakes in Greenland, temperature rising, and global CO2 levels emitting particles, lakes, and letters, include maps containing windheads, icebergs, ice skaters, and more. The unfolding glacial movements turn the earth’s data into something visceral and immediate.   

A sculpture shows the impact of a South Baltimore shop whose flooding is twice that of the rest of the Chesapeake Bay due to events in the polar regions.

Viewers will also see the IMDA MFA Graduate student collaboration “Climate Matters”, and a VR game to test your commitment to change things!

________________________________________________________

A preview of the project, please visit the Polar Ice Museum Site.

We hope you’ll take a moment to see how research can be made accessible and engaging for everyone, not just experts.

Rhoda Nankabirwa and and Alabi Jamiu Ahmed, UMBC PhD students and Graduate Researchers, recently attended and served on the organizing committee for the Imageomics Institute NextGen Day workshop. During the event, Rhoda and Jamiu engaged in meaningful dialogue with fellow emerging scientists and researchers. The day-long event featured insightful sessions on interdisciplinary career pathways, science communication, and strategies for thriving in collaborative research environments.

Thank you to Imageomics for hosting an invaluable workshop for the NextGen of Researchers and Scientists and allowing us to be apart of it!

iHARP is honored to have participated as both a sponsor and presenter at the FAIR in ML, AI Readiness & Reproducibility (FARR) Workshop, held in Washington, D.C., on April 8–9. We want to thank the leadership behind FARR for organizing an amazing and impactful workshop: Christine Kirkpatrick, Julie Christopher, Kevin Coakley, Daniel S. Katz, Douglas Rao, Lynne Schreiber, and Karen Stocks. The workshop was a resounding success, highlighting the critical importance of implementing and promoting data practices that ensure reproducible research across the AI, Data, and ML domains.

Key Highlights from the Workshop:

• Poster Flash Talks: Mostafa Cham, Achala Denagamage, Emam Hossain, Ellie Davidson, Rhoda Nankabirwa Dr. Chhaya Kulkani presented flash talks on a diverse range of topics, including implementing Open Science workflows, AI-ready and reproducible data, and evaluating the reproducibility of benchmark algorithms.

• The 2nd FAIR HDR ML Challenge: This session, co-led by three HDR centers—iHARP, A3D3, and Imageomics—offered attendees a deep dive into the complexities of managing a multi-faceted ML competition guided by FAIR principles. This year’s theme, Scientific Modeling out of Distribution (Scientific-Mood), featured distinct datasets curated by each institution to reflect their specific research areas.During the session, awards were presented to the winners of each individual sub-challenge as well as the overall grand prize winner.  

Winner: iHARP would like to extend a huge congratulations to Dony Darmawan Putra for taking first place sub-challenge on its challenge: Predicting Coastal Flooding Events. 

We would like to acknowledge and thank the leadership team who designed and ran iHARP’s challenge: Dr. Josephine Namayanja, Dr. Ratnaksha Lele, Dr. Aneesh Subramanian, Dr. Bayu Adhi Tama, and Dr. Vandana Janeja. 

Our gratitude also goes to our dedicated support team who worked tirelessly behind the scenes: Sai Vikas Amaraneni, Emam Hossain, Dr. Maloy Kumar Devnath, and Subhankar Ghosh.

Being part of a mission focused on Findable, Accessible, Interoperable, and Reusable (FAIR) practices is vital to our work. These standards ensure that iHARP’s research breakthroughs remain accessible and maintain a lasting impact on the scientific community.

Akila successfully defended on Monday, March 9, 2026

Throughout her time with us, Akila's dedicated contributions have made her an invaluable member of both the iHARP and UMBC communities. This significant achievement is a true testament to her scholarly rigor, hard work, and immense talent.

Congratulations, Akila, on this incredible milestone! We have been honored to witness your growth and look forward to your future successes. The entire iHARP community wishes you the very best as you embark on your next great adventure!

• Dr. Jianwu Wang, Chair/Advisor (UMBC)
• Dr. Vandana Janeja, Co-Chair (UMBC)
• Dr. Houbing Song (UMBC)
• Dr. James Foulds  (UMBC)
• Dr. Donald K Perovich (Dartmouth College)
• Dr. Nicole Schlegel (NOAA)

Rapid and accelerating Arctic climate change poses significant challenges for artificial intelligence (AI) systems, primarily due to severe data scarcity, inherent nonlinearities, and complex spatiotemporal interactions within the ocean–ice–atmosphere system. Conventional deep learning approaches rely heavily on large data volumes and often lack physical consistency. Consequently, purely data-driven models may produce physically implausible predictions and offer limited interpretability, thereby reducing their utility for scientific discovery and decision-making. To enable reliable and trustworthy sea ice prediction, physics-embedded learning architectures are required to leverage domain-specific priors that encode known physical laws, constraints, and causal mechanisms.

This dissertation presents a physics-integrated deep learning framework for modeling and analyzing the time-series evolution of Arctic sea ice. The proposed framework systematically combines physical knowledge with data-driven learning to enhance predictive performance, interpretability, and scientific validity. Specifically, this work introduces three complementary strategies for embedding physical constraints and governing principles directly into the learning process, each addressing a distinct scientific challenge in Arctic climate research.

First, a physics-informed deep learning model is developed for sea ice thickness prediction by explicitly incorporating thermodynamic constraints and governing energy-balance laws into the loss function. This approach ensures that model predictions respect known physical relationships while remaining flexible enough to learn from sparse observational data. Second, the dissertation introduces physics-encoded neural network architectures that embed established physical relationships directly into the model structure. These architectures enable the inference of latent physical parameters from noisy, incomplete proxy data, facilitating physically meaningful representation learning. Third, knowledge-guided temporal causal models are formulated to quantify the causal impact of sea ice variability on coupled oceanic and atmospheric processes. By incorporating time-varying treatments, causal structural constraints, and physics-based priors, these models provide interpretable estimates of causal effects that are consistent with established physical mechanisms rather than spurious correlations.

Across all evaluation settings, the results demonstrate that physics-integrated models consistently outperform conventional deep learning baselines. Overall, this work highlights the critical role of physics-integrated machine learning in advancing predictive capability, causal understanding, and trustworthiness in climate and Earth system modeling.

Tolulope (Tolu) successfully defended on Tuesday, November 18, 2025. 

Tolu's dedicated contributions have made him an invaluable member of the iHARP and UMBC community.  Following 3 successful internships with NASA GESTAR II, Amazon, and Microsoft, Tolu has secured a full-time Microsoft Data Scientist position commencing in 2026., This significant achievement is a testament to his hard work and talent.

Congratulations, Tolu, on this incredible accomplishment! We look forward to your continued growth and success. The entire iHARP community wishes you great success in your next adventure!

• Dr Vandana Janeja, UMBC, Advisor/Chair  
• Dr. Jianwu Wang,  UMBC
• Dr. Patricia (Patti)  Ordóñez, UMBC
• Dr. Nicole Schlegel,  NOAA
• Dr. Sudip Chakraborty, iHARP/ UMBC,
• Dr. Ratnaksha Lele, iHARP/ UMBC

Maloy successfully defended on Tuesday, November 11, 2025. Maloy has been an invaluable and dedicated member of the iHARP and UMBC community. We look forward to seeing what he does next!

Congratulations to Maloy on all of his hard work!!! iHARP wishes him great success in his next adventure!

• Dr. Vandana P. Janeja, Chair/Advisor, Department of Information Systems, University of Maryland, Baltimore County
• Dr. Sudip Chakraborty, Co-Chair/Co-Advisor, iHARP, University of Maryland, Baltimore County
• Dr. James Foulds, Department of Information Systems, University of Maryland, Baltimore County
• Dr. Jianwu Wang, Department of Information Systems, University of Maryland, Baltimore County
• Dr. Md Osman Gani, Department of Information Systems, University of Maryland, Baltimore County
• Dr. Aneesh Subramanian, Department of Atmospheric and Oceanic Sciences, University of Colorado Boulder