Joshua Dimasaka
NH Natural Hazards
The 2026 Outstanding Student and PhD candidate Presentation (OSPP) Award is awarded to Joshua Dimasaka for the poster/PICO entitled:
METEOR 2.5D: An Open Geospatial Dataset of the Spatiotemporal Evolution of Physical Vulnerability in UN-recognized Least Developed Countries (as of 2020) at Five-year Intervals, 1975-2030 (Dimasaka, J.; Geiß, C.; So, E.)
Click here to download the poster/PICO file.
Joshua Dimasaka began his career as a civil engineer specialising in structural engineering in the Philippines before transitioning to disaster risk research, data science, and artificial intelligence (AI). His work focuses on the intersection of geospatial data science, AI, Earth Observation (EO), and Climate & Disaster Risk Assessment (CDRA). He has worked at the University of Cambridge as a PhD researcher, Stanford University as a Knight-Hennessy Scholar, the German Aerospace Center (DLR) as a Helmholtz and Digital GreenTalents Visiting Scientist, and the Earthquakes and Megacities Initiative as a Global Urban Resilience Fellow.
At the conference, Joshua presented a novel deep learning approach to updating global building exposure datasets across Least Developed Countries (LDCs) using EO data and probabilistic spatiotemporal modelling. Moving beyond conventional analytical Bayesian approaches, his research explores how deep neural networks can use noisy and uncertain signals from satellite imagery and other multimodal data sources to approximate the changing or evolving probability distributions (e.g., multinomial) of exposure indicators, such as building presence and typologies, across space and time.