2020
Ce document est lié à :
info:eu-repo/semantics/altIdentifier/doi/10.5937/zrgfub2068099l
http://creativecommons.org/licenses/by/ , info:eu-repo/semantics/OpenAccess
Polina Lemenkova, « Using R packages 'tmap', 'raster' and 'ggmap' for cartographic visualization: An example of dem-based terrain modelling of Italy, Apennine Peninsula », HAL-SHS : géographie, ID : 10.5937/zrgfub2068099l
The main purpose of this article is to present the use of R programming language in cartographic visualization demonstrating using machine learning methods in geographic education. Current trends in education technologies are largely influenced by the possibilities of distance-learning, e-learning and selflearning. In view of this, the main tendencies in modern geographic education include active use of open source GIS and publicly available free geospatial datasets that can be used by students for cartographic exercises, data visualization and mapping, both at intermediate and advanced levels. This paper contributes to the development of these methods and is fully based on the datasets and tools available for every student: the R programming language and the free open source datasets. The case study demonstrated in this paper show the examples of both physical geographic mapping (geomorphology) and socio-economic geography (regional mapping) which can be used in the classes and in self-learning. The objective of this research includes geomorphological modelling of the terrain relief in Italy and regional mapping. The data include dem SRTM90 and datasets on regional borders of Italy embedded in R packages 'maps' and 'mapdata'. Modelling references to the characteristics of slope, aspect, hillshade and elevation, their visualization using R packages: 'raster' and 'tmap'. Regional mapping of Italy was made using main package 'ggmap' with the 'ggplot2' as a wrapper. The results present five thematic maps (slope, aspect, hillshade, elevation and regions of Italy) created in R language. Traditionally used in statistical analysis, R is less known as a perfect tool in geographic education. This paper contributes to the development of methods in geographic education by presenting new technologies of the machine learning methods of mapping.