One key task in environmental science is to map environmental variables continuously in space or even in space and time as a baseline to inform decision-making in various applications, such as agriculture, land-use planning, or natural resource management; or to study ecological research questions based on spatial patterns. However, most ecological variables are only available as point data, e.g. from field surveys. Modelling approaches are hence required to move from local field observations to continuous maps of ecological variables by estimating the value of the variable of interest in places where it has not been measured.
In recent years, machine learning methods have become a popular tool to learn patterns in nonlinear and complex systems. They have been applied to map various ecological variables, even ambitiously on a global scale, such as land cover, soil properties, plant traits, occurrence and abundance of plant or animal species.
In this session, you will learn the basic concepts and techniques of how to apply remote sensing and machine learning for spatial mapping. However, we will also discuss current challenges of using machine learning in the context of environmental monitoring.
Basic R skills required, knowledge in GIS or remote sensing is advantageous
The course is fully booked. Registration is no longer possible (also not for the waiting list).
University of Göttingen, Ecosystem Modelling