Uncertainty Quantification

Satellite-based products and information are prone to uncertainties from different sources, namely the aleatoric (or statistical) uncertainty, referring to the notion of randomness, and the epistemic (or systematic) uncertainty, which is caused by insufficient amount of information. Improvements to uncertainty quantification can provide additional information for informed decision-making, especially in response to non-trivial problems where wrong decisions can result in adverse consequences.

Sources

Hertel, Victor, Candace Chow, Omar Wani, Marc Wieland, and Sandro Martinis. 2023. “Probabilistic SAR-Based Water Segmentation with Adapted Bayesian Convolutional Neural Network.” Remote Sensing of Environment 285 (February): 113388. https://doi.org/10.1016/j.rse.2022.113388.

Related Content

Article

Interview with Victor Hertel, PhD candidate at the German Aerospace Centre

Interview with Victor Hertel, PhD candidate at the German Aerospace Centre

Photo of VIctor Hertel showing on EO projecttions
Victor Hertel is a doctoral researcher specializing in the field of environmental risks and human security. He currently works at the German Aerospace Center (DLR) on the development of (physics-informed) deep learning methods in the context of emergency response and disaster preparedness. With an academic background in aerospace engineering, he previously worked with organizations like Human Rights Watch and the United Nations Office for Outer Space Affairs’ UN-SPIDER program, using geospatial analyses to address environmental and social challenges. His primary area of interest is data-informed decision-making and policy, with a focus on practical and implementation-oriented solutions for humanitarian emergencies caused by climate shocks and conflict.
Read More

Interview with Victor Hertel, PhD candidate at the German Aerospace Centre

Interview with Victor Hertel, PhD candidate at the German Aerospace Centre

Photo of VIctor Hertel showing on EO projecttions
Victor Hertel is a doctoral researcher specializing in the field of environmental risks and human security. He currently works at the German Aerospace Center (DLR) on the development of (physics-informed) deep learning methods in the context of emergency response and disaster preparedness. With an academic background in aerospace engineering, he previously worked with organizations like Human Rights Watch and the United Nations Office for Outer Space Affairs’ UN-SPIDER program, using geospatial analyses to address environmental and social challenges. His primary area of interest is data-informed decision-making and policy, with a focus on practical and implementation-oriented solutions for humanitarian emergencies caused by climate shocks and conflict.
Read More