A newly published PollinERA paper in the Agricultural and Environmental Modelling journal introduces a methodology to describe the land-use/land-cover of agricultural landscapes and generate detailed landscape representations for use in the Animal Landscape and Man Simulation System (ALMaSS).

ALMaSS is a landscape-scale simulation system for investigating the effects of changes in landscape structure and management on animal populations and ecosystem services in European agricultural landscapes. As an open-source, open-science platform hosted on GitLab since 2010, ALMaSS enables researchers, regulators, and policymakers worldwide to assess the impacts of agricultural policies, pesticide use, and land management decisions on biodiversity. By combining agent-based animal models with a dynamic landscape simulation, the system aims to improve the realism and predictive capacity of ecological assessments.

The new project publication focuses specifically on the landscape component of the system, providing a detailed methodology for ALMaSS users on how to generate the spatial and temporal landscape representations for simulations. To do this, the authors, Elżbieta Ziółkowska, Barbara Jaśkowiec, Geoffrey Brian Groom, and Christopher John Topping, provide a mapping algorithm for generating a landscape (land-use/land-cover) raster map and a method for classification and definition of farm types and crop rotations.

The updated method improves how different spatial data layers are combined by using an improved algorithm to fix overlaps and inconsistencies, producing cleaner and more internally consistent landscape maps with less need for manual correction. This is especially useful when working with national datasets that may not align well or may classify areas differently, making the process more reliable and easier to transfer to other regions. 

Interested in learning more? Read the full publication here.