Research

Spatiotemporal Modeling & Earth Observation

Spatiotemporal data – originating from sources such as remote sensing, mobile and ground-based sensors, numerical models, social media, and citizen science – are rapidly expanding in both diversity and volume. These data offer unprecedented opportunities for innovative applications, but they also require robust approaches to represent abstract spatiotemporal phenomena digitally, and to manage, integrate, and analyze them effectively.

Our group specializes in statistical and machine learning methods for the integration, analysis, and interpretation of spatial and spatiotemporal data. We also develop agent-based models and leverage remote sensing techniques to address diverse geoscientific challenges.

Currently, our research focuses on three main themes:

• Air pollution mapping and exposure assessment, with particular attention to social inequality and public health outcomes.
• Urban and natural feature extraction and classification from remote sensing imagery.
• Water quality assessment using remote sensing data, through the integration of radiative transfer models and machine learning approaches.

Geoinformatics lies at the intersection of cutting-edge technology and geoscientific applications. Breakthroughs often emerge when addressing domain-specific challenges. Our team has contributed to a range of real-world applications, including forest dynamics monitoring, air quality prediction, environmental modeling, coastal geomorphology, and geo-health.

Please explore our ongoing projects and long-term research themes below. We warmly welcome new collaborators!

We collaborate closely with

ifgi	UU	ITC	UMC	Swiss THP	KAUST