Chhaya Kulkarni

 

iHARP Research Assistant | UMBC Ph.D.

Email: ckulkar1@umbc.edu
See me at Google Scholar Citation Page, LinkedIn

Research Interests

  • Anomaly Detection
  • Deep Learning
  • Spatial Data Science
  • Environmental Informatics

Successfully defended the Ph.D. dissertation in April 2025

Short Biography

Chhaya Kulkarni completed her PhD in Information Systems at the University of Maryland, Baltimore County (UMBC) in May 2025. Her research interests include advanced spatiotemporal data mining, machine learning, and geographic information systems (GIS), with a particular focus on polar regions impacted by climate change. Chhaya’s interdisciplinary approach integrates satellite data, climate reanalysis, and computational methods to analyze complex environmental patterns. She will join Towson University as an Assistant Professor in the Department of Computer & Information Sciences in August 2025.

Research Summary

Chhaya’s research investigates ice mass changes in Southeast Greenland using a physics-guided analytical framework. Her approach combines satellite observations with atmospheric and oceanographic datasets to examine glacier dynamics. Through advanced spatial clustering techniques, she identifies submarine melting as the dominant process affecting northern marine-terminating glaciers, while surface processes dominate in southern regions. Her findings provide robust evidence of ocean-driven glacier dynamics, offering insights valuable for improving predictive modeling of sea-level rise.

Publications
  • Sudip Chakraborty, Maloy Kumar Devnath, Atefeh Jabeli, Chhaya Kulkarni, Gehan Boteju, Jianwu Wang, and Vandana P. Janeja. 2025. Impact of increased anthropogenic Amazon wildfires on Antarctic Sea ice melt via albedo reduction. Environmental Data Science 4, (2025), e18. https://doi.org/10.1017/eds.2025.1
  • Chhaya Kulkarni, Bayu Adhi Tama, Nicole-Jeanne Schlegel, and Vandana P. Janeja. 2025. Anomaly Detection Using Graph Deviation Networks Within Spatiotemporal Neighborhoods: A Case Study in Greenland. IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing 18, (2025), 1362–1375. https://doi.org/10.1109/JSTARS.2024.3501092
  • Chhaya Kulkarni, Nikki Privé, and Vandana P. Janeja. 2024. Interactive Assessment of Variances of High-Resolution Model Features in Digital Twin Simulations. In Proceedings of the 32nd ACM International Conference on Advances in Geographic Information Systems, October 29, 2024. ACM, Atlanta GA USA, 685–688. https://doi.org/10.1145/3678717.3691302
  • Sudip Chakraborty, Chhaya Kulkarni, Atefeh Jebeli, Jianwu Wang, and Vandana Janeja. “Extreme Slash and Burn Practices over the Amazon Rainforest in 2019 Wreaked Havoc on Sea Ice Extent Over the Antarctic,” December 2023. Poster at the AGU23 meeting, C51D-0975. [Poster]
  • Sudip Chakraborty, Chhaya Kulkarni, Atefeh Jabeli, Akila Sampath, Gehan Boteju, Jianwu Wang, and Vandana Janeja. 2023. Understanding the Role of 2019 Amazon Wildfires on Antarctic Ice Sheet Melting Using Data Science Approaches. Accepted 2023 ACM SIGKDD International Conference on Knowledge Discovery and Data Mining Workshop: Fragile Earth: AI for Climate Sustainability – from Wildfire Disaster Management to Public Health and Beyond, 2023 [Conference/Workshop Paper]
  • Chhaya Kulkarni, Vandana Janeja, and Nicole-Jeanne Schlegel. 2023. Multi-Contextual Learning: Analyzing Melt Over the Greenland Ice Sheet. In IGARSS 2023 – 2023 IEEE International Geoscience and Remote Sensing Symposium, July 16, 2023, Pasadena, CA, USA. IEEE, Pasadena, CA, USA, 6736–6739. https://doi.org/10.1109/IGARSS52108.2023.10281954 [Symposium Paper]
Poster Presentations
  • Kulkarni, C., Privé, N., & Janeja, V. (2024). Analyzing the Variance of Simulated Brightness Temperatures Within Footprints Using Machine Learning. Presented at the American Geophysical Union (AGU) Fall Meeting, Washington DC, December 12, 2024
  • Kulkarni, C., Tama, B. A., & Janeja, V. (2024). Integrating Spatiotemporal Anomaly Detection in Polar Ice Melt Analysis: A Graph Deviation Network Approach. Presented at the inaugural College of Engineering and Information Technology (COEIT) Research Day, University of Maryland, Baltimore County, April 19, 2024.
Dissertation Defense

Successfully Defended On: April 24, 2025

Abstract

This dissertation presents a multi-contextual learning framework within spatio-temporal neighborhoods to tackle analytical challenges in Earth observation data. The rapid rise in the amount of environmental data – as demonstrated by the European Space Agency’s Copernicus program expanding from 2 to 20 petabytes per year and NASA’s Earth Observing System producing 16 terabytes per day – poses difficulties to conventional analytical methods for managing multi-source data fusion, cross-scale pattern recognition, and spatial autocorrelation.

The work develops a context-aware spatio-temporal data analysis approach with a neighborhood-based spatio-temporal framework at its foundation. The framework employs Voronoi tessellation for micro-neighborhood generation and attribute-based grouping for macro-neighborhood generation. By incorporating contextual information from both spatial proximity and attribute similarity, the approach captures nuanced patterns that traditional methods tend to ignore.

This multi-contextual learning framework is validated through two complementary application domains that serve as case studies. The Greenland Ice Sheet case demonstrates how the application of neighborhood analysis successfully encapsulates intricate melt behavior, accounting for local variability, seasonality, and couplings between temperature, albedo, and other variables. Importantly, the framework aids in comparing surface and subsurface processes influencing change in the ice mass in marine-terminating glaciers in Southeast Greenland. Digital twin simulations form a second test case by demonstrating that the same neighborhood-based approach is capable of defining areas with analogous variance structures within high-resolution atmospheric data.

The methodological contributions of the dissertation are (1) multi-contextual learning for spatio-temporal neighborhood formation, (2) Graph Deviation Networks for multivariate anomaly detection in such neighborhoods, (3) a technique for differentiating surface versus subsurface process dominance in analyzing ice mass change through hotspot analysis, and (4) spatial clustering for variance analysis. Each of these components tackles intrinsic challenges in spatio-temporal data analysis while offering practical solutions for environmental monitoring application scenarios. Results indicate that multi-contextual learning in spatio-temporal neighborhoods substantially enhances detection and interpretation capability for complex Earth observation data, with immediate implications for environmental monitoring, modeling, and satellite-based observational systems.

Committee

  • Dr. Vandana Janaja – Advisor and Committee Chair (UMBC)
  • Dr. Jianwu Wang – Committee Member (UMBC)
  • Dr. Karuna Joshi – Committee Member (UMBC)
  • Dr. Bayu Adhi Tama – Committee Member (UMBC/iHARP)
  • Dr. Nicole Schlegel – Committee Member (NOAA)

Last updated 16 June 2025