The southern African savanna landscapes are among the most striking vegetation types where contrasting plant life forms co-dominate and are shaped by interactions of various top-down and bottom-up processes such as rainfall, fires, herbivory and human activities. These ecosystems are critical to a vast variety of wildlife forms and a considerable portion of the global human population. Today, southern African savanna landscapes are threatened by various challenges related to population growth, climate change, and land degradation. In addition, trends in woody encroachment have caused speculations about the greening of savannahs and the underlying causes. In this 4-hour seminar we will analyse and discuss how savanna ecosystems in Southern Africa are impacted by humans, climate change and, specifically, woody encroachment. The seminar will focus on state-of-the-art remote sensing methods to analyze woody plant encroachment in the country of Namibia. Namibia is the driest country in sub-Saharan Africa and therefore, ecological and socio-economic consequences might be pronounced in that region. In the first part of the course, students will learn the theory and practice of implementing relevant remote sensing methods using R/Python and cloud computing. Students will work with a variety of data sources ranging from drone data to satellite time series. This way, students will learn how to scale from field measurements to larger regions and to derive vegetation trends from satellite time series. A special emphasis is given to unmixing approaches for mapping fractional woody cover as well as time series methods for analyzing seasonal and inter-annual vegetation changes. In the second part, students will apply the learned methods and knowledge in a class project focusing on woody encroachment. The goal is to learn how to design, develop, and implement a small research project that combines remote sensing and other geodata. The course builds on Module 5.1 Earth Observation.
Literature is provided during the seminars and in moodle.
Students will write a report of about 10 pages about their own model experiment and the analysis of the model results.
Students with prior knowledge in remote sensing and satellite image analysis.
Die Veranstaltung wurde 2 mal im Vorlesungsverzeichnis SoSe 2024 gefunden:
