Projects

Novel view on the study of glacier kinematics in the context of global climate change

Role: principal investigator
Duration: 02/2023 - 02/2026
Source of funding: National Scientific Centre of Poland

Main goals: The project aims to develop a new method for monitoring glacier movements using SAR data and machine learning. By combining the Offset-Tracking (OT) and DInSAR technique, with high-resolution SAR images, it will analyze changes in glacier kinematics, detect anomalies, and assess speed trends in Greenland and Svalbard. The research involves testing machine learning algorithms to optimize displacement assessment, followed by comparative analyses using GIS software. Ultimately, the goal is to provide a more accurate and efficient way of monitoring glacier behavior and environmental changes.

Keywords: InSAR, Offset-Tracking, machine learnig, marine-terminating glaciers, long-term analysis of kinematics, surface collapses

More info: link to the project website here

Novel algorithm of sinkhole precursors detection

Role: team member
Principal Investigator: Dr Eng. Wojciech Witkowski (AGH University of Krakow)
Duration: 07/2022 - 07/2025
Source of funding: National Scientific Centre of Poland

Main goals: The natural environment is experiencing new phenomena as a result of global warming. Particularly, massive sinkholes are caused by the rapid thawing of the Arctic permafrost. Sinkholes not only cause severe environmental changes, but they are also primarily linked to risinsg CO2 emissions. Climate change, on the other hand, indicates an intensification of the drought, which could be also accompanied by the sinkhole hazard. Sinkholes caused by humans have been observed in many countries, including the United Kingdom, the United States, China, and RPA. Nevertheless, they are poorly monitored in comparison to other deformation processes such as landslides or compaction-induced subsidence. Traditional observation techniques such as levelling, GNSS, and tachymetry face difficulties in monitoring the movement of the terrain surface prior to the occurrence of sinkholes. However, remote techniques such as Satellite Radar Interferometry (InSAR) can be useful in resolving this problem. Simultaneously, the intensive development of satellite technologies promotes an effective ability to identify precursors of the sinkhole. Furthermore, Machine Learning (ML) tools, which has grown in popularity in recent years, are increasingly being used to identify patterns in big data. They enable evaluating phenomena for which a strict algorithm cannot be developed due to the multiplicity of factors, allowing for an unambiguous mathematical description of the phenomenon under study. As a result, advanced InSAR tools combined with ML algorithms will allow for a better understanding of the physics of sinkhole formation as well as the effective detection of developing sinkholes. The investigation of the displacement field characteristics in the sinkhole area will raise awareness of the nature of such accelerated deformations caused by climate change.

Keywords: InSAR, machine learning, sinkholes, mutlispectral data, geomechanical modelling, precursors, climate change

More info: Sonata website

AlignSAR

Role: team member
Principal Investigators: Dr Eng. Ling Chang (TU Twente), Dr Eng. Wojciech Witkowski (AGH University of Krakow)
Duration: 02/2023 - 02/2024
Source of funding: European Space Agency

Main goals: The AlignSAR project aims to provide FAIR-guided open datasets and tools designed for SAR applications, ensuring interoperability and consistency with existing and upcoming initiatives and technologies. The project facilitates a wider exploitation of SAR data and its integration and combination with other datasets.

Keywords: SAR signatures, machine-learning, benchmark dataset

More info: AlignSAR website

Assessing the feasibility of using high-resolution radar images for rapid landslide monitoring

Role: principal investigator
Duration: 04/2022 - 06/2023
Source of funding: Ministry of Science and Higher Education of Poland

Main goals: This project aimed to check wheter high-resolution SAr datsets and Offset-Tracking method can deliver infromation about the displacement field for a rapid landslide. The case study is located in central Italy, Citivella del Tronto. The calculation of movements fot the main displacement phase as well as for the post-landslide period were perofrmed with both Sentinel-1 and TerraSAR-X data to compare the utility of those dataset and OT method for such application.

Keywords: SAR data, Offset-Tracking, rapid landslide, high-resolution dataset, TerraSAR-X