Original scientific paper
https://doi.org/10.21861/HGG.2019.81.01.02
Simulation model of land cover changes in a post-socialist peripheral rural area: Požega-Slavonia County, Croatia
Tomislav Jogun
orcid.org/0000-0002-8685-8328
; Promet i prostor d.o.o., Zagreb, Croatia
Aleksandar Lukić
orcid.org/0000-0002-7884-5818
; University of Zagreb, Faculty of Science, Department of Geography, Zagreb, Croatia
Mateo Gašparović
orcid.org/0000-0003-2345-7882
; University of Zagreb, Faculty of Geodesy, Institute for Cartography and Photogrammetry, Zagreb, Croatia
Abstract
Simulation modelling is a useful method for the explanation and prediction of land use and land cover changes, which are fundamentally interconnected with broader social and natural elements. Significant land cover changes, predominantly land abandonment, occurred in peripheral rural areas across Central and Eastern Europe after the collapse of socialism. One such area is Požega-Slavonia County in Croatia, for which we modelled land cover changes from 1985 to 2027. The prevailing change from 1985 to 2013 was secondary succession. The simulation model of future land cover changes was based on cellular automata and artificial neural networks, and was implemented in the MOLUSCE plugin for QGIS. Validation of the test models showed that they more successfully predicted the quantity (rather than the location) of the changes, and the post-simulation combination of individual results slightly improved their agreement with reference data. The final model predicted that forests would cover over 60% of the whole area by 2027 if the current negative demographic and economic trends continue into the future.
Keywords
artificial neural networks; cellular automata; land use and land cover change; peripheral area; simulation model; MOLUSCE
Hrčak ID:
221020
URI
Publication date:
13.6.2019.
Visits: 3.871 *