High-resolution land use/cover forecasts for Switzerland in the 21st century

Luca Bütikofer; Antoine Adde; Davnah Urbach; Silvia Tobias; Matthias Huss; Antoine Guisan; Christophe Randin

doi:10.1038/s41597-024-03055-z

High-resolution land use/cover forecasts for Switzerland in the 21st century

Sci Data. 2024 Feb 23;11(1):231. doi: 10.1038/s41597-024-03055-z.

Authors

Luca Bütikofer^{1

2}, Antoine Adde³, Davnah Urbach^{4

5}, Silvia Tobias⁶, Matthias Huss^{6

7}, Antoine Guisan³, Christophe Randin^{8

9

5}

Affiliations

¹ Centre alpien de phytogéographie CAP, Fondation Aubert, Route de l'Adray 27, CH-1938, Champex-Lac, Switzerland. luca.butikofer@unil.ch.
² Department of Ecology and Evolution, University of Lausanne, Biophore, CH-1015, Ecublens, Switzerland. luca.butikofer@unil.ch.
³ Institute of Earth Surface Dynamics, University of Lausanne, Geopolis, Quartier Mouline, CH-1015, Lausanne, Switzerland.
⁴ Global Mountain Biodiversity Assessment, Institute of Plant Sciences, University of Bern, Altenbergrain 21, CH-3013, Bern, Switzerland.
⁵ Interdisciplinary Centre for Mountain Research (CIRM), University of Lausanne, Chemin de l'Institut 18, CH-1967, Bramois/Sion, Switzerland.
⁶ Swiss Federal Research Institute WSL, Zürcherstrasse 111, CH-8903, Birmensdorf, Switzerland.
⁷ Laboratory of Hydraulics, Hydrology and Glaciology (VAW), Swiss Federal Institute of Technology ETH, Hönggerbergring 26, CH-8093, Zürich, Switzerland.
⁸ Centre alpien de phytogéographie CAP, Fondation Aubert, Route de l'Adray 27, CH-1938, Champex-Lac, Switzerland.
⁹ Department of Ecology and Evolution, University of Lausanne, Biophore, CH-1015, Ecublens, Switzerland.

Abstract

We present forecasts of land-use/land-cover (LULC) change for Switzerland for three time-steps in the 21^st century under the representative concentration pathways 4.5 and 8.5, and at 100-m spatial and 14-class thematic resolution. We modelled the spatial suitability for each LULC class with a neural network (NN) using > 200 predictors and accounting for climate and policy changes. We improved model performance by using a data augmentation algorithm that synthetically increased the number of cells of underrepresented classes, resulting in an overall quantity disagreement of 0.053 and allocation disagreement of 0.15, which indicate good prediction accuracy. These class-specific spatial suitability maps outputted by the NN were then merged in a single LULC map per time-step using the CLUE-S algorithm, accounting for LULC demand for the future and a set of LULC transition rules. As the first LULC forecast for Switzerland at a thematic resolution comparable to available LULC maps for the past, this product lends itself to applications in land-use planning, resource management, ecological and hydraulic modelling, habitat restoration and conservation.

Publication types

Dataset