Proposed Abstract Title

Using the geospatial capabilities of R for an integrated data analysis workflow

Type of Presentation

Snapshot

Session Title

Mapping and Data

Location

2016SSEC

Description

When most scientists and analysts have geospatial tasks to perform, they instinctively turn to traditional GIS software such as ESRI ArcGIS or open source tools such as QGIS or GRASS. These tools are powerful and well suited to their specialties. However, when working with spatial data it is rarely the case that the entire analytical workflow can be completed in a GIS environment alone: data must often be cleaned or restructured before being imported into the GIS, or statistical analysis or visualization is performed after spatial analysis is completed. Many analysts also use the statistical programming language R to perform this related data analysis work. Switching back and forth between GIS software and R breaks up the workflow both cognitively and practically, and reduces reproducibility of analyses. R is very well known for its advanced statistical capabilities, but it also has a suite of powerful and extensive spatial packages. These include functionality for basic GIS operations such as clipping, buffering, and spatial overlays, but also include advanced techniques such as point pattern analysis, kriging, and spatial regression, as well as excellent spatial visualization capabilities. I will give an overview of the spatial capabilities of R, how using R as a comprehensive data analysis environment is beneficial, as well as some challenges.

This document is currently not available here.

Share

COinS
 

Using the geospatial capabilities of R for an integrated data analysis workflow

2016SSEC

When most scientists and analysts have geospatial tasks to perform, they instinctively turn to traditional GIS software such as ESRI ArcGIS or open source tools such as QGIS or GRASS. These tools are powerful and well suited to their specialties. However, when working with spatial data it is rarely the case that the entire analytical workflow can be completed in a GIS environment alone: data must often be cleaned or restructured before being imported into the GIS, or statistical analysis or visualization is performed after spatial analysis is completed. Many analysts also use the statistical programming language R to perform this related data analysis work. Switching back and forth between GIS software and R breaks up the workflow both cognitively and practically, and reduces reproducibility of analyses. R is very well known for its advanced statistical capabilities, but it also has a suite of powerful and extensive spatial packages. These include functionality for basic GIS operations such as clipping, buffering, and spatial overlays, but also include advanced techniques such as point pattern analysis, kriging, and spatial regression, as well as excellent spatial visualization capabilities. I will give an overview of the spatial capabilities of R, how using R as a comprehensive data analysis environment is beneficial, as well as some challenges.