Abstract

This master thesis aims to develop a novel methodology that can leverage old maps without precise geodetic measures to reconstruct historical land use/cover change through spatial extrapolation and applies this methodology to the Czech Republic from the 18th century onward. The input data in this study includes three old maps from the 18th century and other land use/cover data from 1840, 1870 and 1950 for selected areas, moreover, Corine Land Cover data from 1990, 2000, 2006, 2012 and 2018 for the entire Czech Republic. This thesis analyses the grid-based Pearson correlation between land use/cover distribution and slope in the selected areas and uses this correlation as a basis for assuming that areas with similar slopes exhibit similar land use/cover patterns. Consequently, the land use/cover distribution within specific slope intervals in selected areas can be extrapolated to represent land use/cover distribution in the same slope intervals across other regions of the Czech Republic. The study successfully reconstructed land use/cover change from the 18th century to 2018 for almost 70% of the Czech Republic within valid slope intervals, ranging from flat lowlands to steep highlands. Most importantly, the findings reveal a relatively balanced land use/cover distribution in the 18th century, with 38.04% arable land, 29.09% forest, 31.42% grassland, and only 1.30% in other land use/cover categories. These results can serve as a valuable reference for future land use/cover studies. Moreover, the methodology helps to mitigate the limitations caused by the scarcity of input data and has the potential for application in other historical land use/cover studies. …víceméně