The PhD defence and trial lecture are fully digital and streamed using Zoom. The host of the session will moderate the technicalities while the chair of the defence will moderate the disputation.
Trial lecture
As a recording:
The potential and limitations of using social media for landscape research
Conferral summary (in Norwegian)
Landskapsovervåking gir et viktig grunnlag for å kunne følge landskapsendringer over lang tid. Det er også en forutsetning for å evaluere effekt av politiske virkemidler og ulike tiltak. Avhandlingen viser hvordan etablerte overvåkingsprogrammer kan dra nytte av å inkludere nye metoder og teknologier som har blitt tilgjengelige de siste årene. Samtidig setter avhandlingen lys på utfordringene med å iverksette endringer i metodikken uten å true dataintegritet og kontinuiteten i datagrunnlaget.
Main research findings
Popular scientific article about Bayr’s dissertation:
Survey techniques in landscape monitoring: testing new quantitative methods to assess landscape change
Long-term landscape monitoring plays an important role in assessing the state of landscapes, identifying trends and drivers of landscape change and in evaluating the effectiveness of political instruments. Despite the technological advances of the past years, many well-established monitoring programs still rely on rather conventional approaches as they ensure data integrity and consistency.
The aim of this dissertation was to test and examine the use of modern quantitative methods in landscape monitoring, with focus on the Norwegian monitoring program for agricultural landscapes (3Q). The program utilizes a broad range of conventional data sources such as aerial photographs, terrestrial repeat photographs and field inventories of birds and plants.
Monoplotting is one of the methods that were tested on repeat landscape photographs to extract geospatial data. In addition, machine learning approaches were applied on repeat landscape photographs (Convolutional Neural Network) and field registrations of breeding birds (Random Forest). Findings showed that the tested methods provided good results on the respective data sources, but they also showed some limitations in the practical usability for landscape monitoring on the national scale. As the used methods focused on improving data analysis of already existing data sources, data integrity and consistency can be maintained.
Photo and other information:
Press photo: Ulrike Bayr, portrait; 500px. Photo: Erling Fløistad/NIBIO
Other photo material: Figure with description and credit as specified in the article above, size 1000px.