Wrangling spatio-temporal data with R

Welcome to the WOMBAT 2022 Tutorial!

About the workshop:

This tutorial is designed to get you up and running with multivariate spatio-temporal data analysis and visualisation in R quickly. It will cover importing and combining spatial and temporal data, filtering on space and time, aggregating on space and time, making basic maps and creating interactive plots.

About the presenters:

H. Sherry Zhang is a third year PhD student in Econometrics and Business Statistics at Monash University. Her research focuses on multivariate spatiotemporal data wrangling and visualisation tools. She will be assisted by Professor Di Cook, who is well-known for her work in data visualisation and teaching of tutorials.

Target audience:

This workshop is suitable for those who know how to use R and need to develop more skills and experience with wrangling and visualising spatio-temporal data in R.

Sessions

• Session 0: Introduction slides
• Session 1: Importing various spatio-temporal data formats slides, exercises
  • csv: spatial coordinates, time index and measured variables
  • netcdf: array of large gridded data
  • shape files: to provide the map background
  • R data structures: sf, tsibble, cubble
• Session 2: Spatial and temporal operations, and basic mapping slides, exercises, covid data
  • casting temporal and spatial polygon data into a cubble to make a Covid-19 map
  • filtering spatio-temporal data, based on spatial coordinates (using weather station data)
  • aggregating measured variables over time (using weather station data)
• Session 3: Creating interactive graphics, with linking between plots slides, exercises, bonus leaflet popup
  • mouse-over tooltips with plotly
  • using crosstalk to link spatial plots with time series plots
  • sharing your interactive graphic

Requirements:

Participants need a basic knowledge of R, ideally with familiarity with the topics covered in R for Data Science.

Desirable:

• Know about tidy data (i.e. importing data and putting data into the right format for plotting)
• Some familiarity with tidyverse

Software to install:

The latest version of R (>= 4.2.1) and RStudio (>= 2022.07.1) should be installed, and these packages with their dependencies installed prior to joining the tutorial:

pkgs <- c("colorspace", "crosstalk", "cubble", "dplyr", "ggplot2", "htmltools", "ozmaps", "plotly", "rmapshaper", "rnaturalearth", "sf", "tsibble", "webshot")
              
install.packages(pkgs)

The strayr package needs to be installed from GitHub:

remotes::install_github("runapp-aus/strayr")