Mapping, monitoring and modelling the dynamics of land surface morphology

Traditional methods for the monitoring of land surface morphology such as field investigations are cost-intensive, time consuming and limited regarding spatial and temporal coverage. Nowadays, the increasing availability of remote sensing data allows for the comprehensive mapping of geomorphological features and the continuous monitoring of surface morphology changes at high spatial and temporal resolutions. Just as the spatial and temporal resolution of remote sensing data steadily increases, so do the demands for extracting timely and relevant geospatial information in an automated manner. Within this project novel methods for mapping, monitoring and modelling spatial- temporal dynamics of surface morphology will be developed by an integrated analysis of various remote sensing data (e.g. optical satellite images, satellite radar data, digital elevation data). This project will focus on the investigation of landslides and volcanic deposits in two study areas in Iceland, which are highly dynamic in their geomorphic evolution and are characterized by progressive mass displacements and surface deformation. One major objective will be the development of an efficient image analysis method for the multi-scale mapping of slope instabilities and volcanic deposits that is transferable across various remote sensing data. Automated methods for time series analysis will be set up for monitoring spatio-temporal changes of land surface morphology. Surface displacements and deformation will be measured by analyzing satellite radar data. To gain added value from the analysis of different remote sensing data a flexible digital data model, which allows for the integration of data/analysis results at multiple scales for spatio-temporal modelling of land surface trends and dynamics, will be implemented. Novel approaches for the analysis and visualization of spatial change patterns and for hotspot mapping will be introduced to produce aggregated and meaningful added-value results. Outcomes will be validated against field reference data, visual image interpretation and existing landslide inventories or geomorphological maps. From mapping to monitoring to modelling this concept, based on an integrated use of various remote sensing data, allows for the provision of results with high information content. Existing mapping approaches can be complemented, and new insights on spatio-temporal changes of the environment can be achieved. Monitoring surface changes can contribute to a better understanding of mass-transport systems, to detect related environmental variability and to assess natural hazards. Hence, project results could be supportive for hazard and risk analysis by identifying hotspots of erosion/deposition or landslide- affected areas. It is expected that knowledge on landscape evolution can be improved.

The project MORPH (Mapping, Monitoring and Modelling the Spatio-Temporal Dynamics of Land Surface Morphology) aimed at developing novel methods for mapping, monitoring and modelling spatial-temporal dynamics of surface morphology by an integrated analysis of various optical and radar remote sensing data from different sensors as well as digital elevation models (DEMs). MORPH focused primarily on study areas in Iceland, which are highly dynamic in their geomorphic evolution and are characterized by progressive mass displacements and surface deformation. The increasing availability of remote sensing data from a wide range of sensors allows for the comprehensive mapping of geomorphological features and the continuous monitoring of surface morphology changes with advanced methods. We developed an object-based image analysis (OBIA) approach to semi-automatically map geomorphological features and related changes in the Htardalur valley, western Iceland. By combining optical and radar imagery, we were able to assess the landslide and the landslide-induced geomorphological changes. This includes the semi-automated differentiation of landslide source and deposition area, as well as changes in the watercourses and the landslide-dammed lake. We also explored the applicability of Sentinel-1 radar image pairs to generate DEMs for landslide volume estimation. Moreover, we developed an object-linking approach to automatically track rock avalanches on glaciers over long periods based on Landsat time series in south-eastern Iceland. The approach allows to track objects despite changing shape and movement rates, which can assist automated change analysis of rock avalanche deposits transportation on glaciers and glacier velocity estimation. Furthermore, we investigated the applicability of Persistent Scatterer Interferometry (PSI) to infer local surface deformation, as well as image correlation techniques based on optical imagery to measure ground displacements and to detect slow-moving mass movements in the Öræfajökull study area. In this region, we also explored multi-temporal changes in glacial lake area from 2015 to 2020 in the southern ice-marginal lakes of the Vatnajökull glacier using radar data within an OBIA approach. Moreover, we developed new approaches for combining machine learning and OBIA for lava flow morphology classification in the Krafla lava field, for rock glacier mapping and landslide detection. MORPH also benefited from the successful application for an add-on FWF Top Citizen Science (TCS) project (citizenMorph), where a technological solution for citizens to contribute (spatial) information on geomorphological features and landscape dynamics was developed. The outcomes of citizenMorph increased in-situ knowledge with geomorphological features locations and 3D models from field photographs that supported the validation of the remote sensing-based results. By the integrated use of various remote sensing data, MORPH allowed us to gain new insights on spatio-temporal surface changes. Consequently, the project results contributed to a better understanding of mass-transport systems, to detect related environmental variability and to assess natural hazards.