Organizer(s): Monika Mętrak Ph.D. - assistant professor at the Faculty of Biology, University of Warsaw)
Date: 1 July 2025
Time: TBC
Room: TBC
Max number of participants: 30
Fee: free of charge
Description
Future design is an approach used in strategy and policy development. It provides a set of tools to identify and understand drivers of change (events and processes shaping the future) and explore their implications for making current decisions. It is a creative, exploratory process based on divergent thinking, acknowledging the complexity of studied problems and potential uncertainties. Future design does not aim to provide a correct prediction of the future. Instead, it gives us the possibility to actively shape the future through the informed decisions we make today. A specific example of such an approach is the IPCC’s projections of climate change and its diversified consequences (climate change scenarios), which are used for the formulation of both state and international climate policies.
During the workshop, we will collect signals of change for the topic of wetland conservation based on the diversified experiences of all participants. Working in groups, we will prepare a matrix for future scenarios for wetland conservation and arrange the collected signals into more general trends. Finally, using the prepared matrix, we will try to define a preferable scenario for wetland conservation for 2035. The workshop will be performed using non-digital participatory techniques. A week after the workshop, all participants will be provided with a leaflet summarizing the main findings of this activity.
The idea behind this workshop is to provide space and tools for structured discussion and reflection on potential developments in wetlands conservation in a broad, systemic context. A process of creating a preferable scenario for wetlands conservation for 2035 stimulates a search for potential solutions to current problems and opens the way to designing system interventions that these ecosystems may benefit from.
Organizer(s): Daniel Gann, Florida International University, United States of America
Date: 1 July 2025
Time: TBC
Room: TBC
Max number of participants: 12
Fee: free of charge
Description
Landcover class definitions are scale-dependent. Upscaling categorical data must account for that scale dependence, but most scaling algorithms for categorical data (e.g., majority rule, nearest neighbor) assume validity of the classification system at the lower resolution regardless of the scale factor which leads to high and uncontrolled information loss in the scaled dataset. This workshop will demonstrate why it is important to consider information loss and how to quantify it. Participants will then be introduced to multi-dimensional grid-point (MDGP) scaling (Gann 2019, https://doi.org/10.1111/2041-210X.13301), a new scaling algorithm that accounts for the scale dependence of classification systems as data are spatially aggregated. This algorithm aggregates categorical data while simultaneously controlling information loss and generating a non‐hierarchical, representative, classification system for the aggregated scale. The hands-on component of this course introduces an R package for scaling of categorical spatial data. The algorithm evaluates information retention and class representativeness of scaled classification systems while aggregating spatial data. Applications of the scaling algorithm will be presented with examples in wetland vegetation ecology.
Course Objectives: The participants of the workshop will understand the complexity of spatial aggregation and scaling of categorical data; Learn how to evaluate the loss of information when scaling categorical data; Evaluate tradeoffs of class detail, class representativeness, and detection accuracy of classes from multi-spectral medium-resolution satellite data to optimize parameter selection for multi-dimensional grid point scaling.
Course Outline:
