Trial Lecture
Title: "Applications of artificial intelligence in nature management"
Time and place: October 2, 2025, 10:15 AM, Tøyen hovedgård
Public Defence
Adjudication Committee
- Professor Niina Käyhkö, Department of Geography and Geology, University of Turku, Finland
- Professor James D.M. Speed, Department of Natural History, NTNU University Museum, Trondheim, Norway
- Professor Øyvind Hammer, Natural History Museum, University of Oslo
Chair of the Defence
- Professor Brita Stedje, Natural History Museum, University of Oslo
Supervisors
- Professor Rune Halvorsen, Natural History Museum, University of Oslo
- Professor Anders Bryn, Natural History Museum, University of Oslo
- Senior Engineer, Dr. Peter Horvath, Natural History Museum, University of Oslo
- Professor Olav Skarpaas, Natural History Museum, University of Oslo
- Research Scientist, Dr. Liv Guri Velle, Norwegian Institute of Bioeconomy Research, Ås, Norway
Summary
Ecosystem maps are used to support environmental impact assessments, maintain ecosystem accounts, and calibrate climate models – which are in turn used to make environmental policy decisions. It is therefore concerning that the quality of detailed ecosystem maps is often low and variable, leading to the risk of decisions being based on incorrect information. The aim of this dissertation was to explore causes of variation in the quality of ecosystem maps.
Paper I shows that the quality of field-based ecosystem maps depends on the experience of the mapper and the properties of the area being mapped. Paper II demonstrates that the quality of model-based ecosystem maps significantly improves with the use of remote sensing data for a selection of ecosystem types. Paper III illustrates how model-based ecosystem maps can be used in spatial planning and emphasises that decisions can be improved considerably if a quality assessment of the map product is available.
Together, the papers highlight the challenges associated with creating detailed ecosystem maps of high quality. The dissertation provides a foundation for further method development and improved practical applications of ecosystem maps.
Contact