gganimate::transition_time results in flying polygons

Question

Trying to create an animation with simple features, essentially a time series of a spatial variable in the form of choropleth maps.

The issue is that the polygons fly all over the plot at the time of the animation.

Here's a reproducible example, lifted in part from https://www.blog.cultureofinsight.com/2017/09/animated-choropleth-maps-in-r/ (which used an older version of gganimate.

Load packages:

library(tidyverse) # dev ggplot version required: devtools::install_github("hadley/ggplot2")
library(sf)
library(readxl)
library(httr)
library(gganimate)

Get and organize data for reproducible example.

# download the natural earth shapefile we need into your working directory
URL <- "http://www.naturalearthdata.com/http//www.naturalearthdata.com/download/110m/cultural/ne_110m_admin_0_map_units.zip"
temp <- tempfile()
download.file(URL, temp)
unzip(temp)
unlink(temp)

# read in shapefile as an sf object and set the projection
# this will be our base world map for plot sans Antarctica
world <- st_read("ne_110m_admin_0_map_units.shp") %>% 
  st_transform(crs = "+proj=longlat +datum=WGS84") %>% 
  filter(!NAME %in% c("Fr. S. Antarctic Lands", "Antarctica"))

# download dataset into your working directory
url <- "https://www.blog.cultureofinsight.com/data/wc.xlsx"
GET(url, write_disk("wc.xlsx", overwrite=TRUE))

# read in our the massive 20 rows of data and get the winner/runner-up variable in 1 column #tidyafdata
# setting factor for winner to show first in the legend
winners <- read_excel("wc.xlsx") %>% 
  gather(w_l, country, winner:runner_up) %>% 
  mutate(w_l = factor(w_l, levels = c("winner", "runner_up")))

# merge our world shape file with our main dataset
# this will add the polygon for the appropriate country to each row of our winners dataset
# and remove any countries that haven't won or come 2nd in the WC
wc_geo <- left_join(world, winners, by = c("NAME" = "country")) %>%
  st_as_sf() %>%
  drop_na(Year)

Create the gganimate object:

wc_map <- ggplot() +
  geom_sf(data = world, colour = "#ffffff20", fill = "#2d2d2d60", size = .5) +
  geom_sf(data = wc_geo, aes(fill = w_l)) +
  coord_sf(crs = st_crs(world), datum = NA) +
  scale_fill_manual(values = c("#D9A441", "#A8A8A8"), name = NULL, labels = c("Winner", "Runner-Up")) +
  theme(legend.position = c(0.9, 1.01), legend.direction = "horizontal", axis.text = element_blank(), 
        panel.grid.minor = element_blank(), panel.grid.major = element_blank()) +
  transition_time(Year)

Animate:

animate(wc_map)

This is the result:

Flying countries

score 3 · Accepted Answer · answered Apr 24 '20 at 08:39

This is a grouping issue. You need to be explicit with your groups. In this case, create an interaction of Year and w_l

You can use the package rnaturalearth to get world / country maps more conveniently.

library(tidyverse) 
library(gganimate)

world <- rnaturalearth::ne_countries( returnclass = 'sf') %>% filter(admin!= 'Antarctica')

url <- "https://www.blog.cultureofinsight.com/data/wc.xlsx"
httr::GET(url, httr::write_disk("wc.xlsx", overwrite=TRUE))

winners <- readxl::read_excel("wc.xlsx") %>% 
  gather(w_l, country, winner:runner_up) %>% 
  mutate(w_l = factor(w_l, levels = c("winner", "runner_up")))

wc_geo <- left_join(world, winners, by = c("admin" = "country")) %>%
  drop_na(Year)

p <- 
  ggplot() +
  geom_sf(data = world, colour = "#ffffff20", fill = "#2d2d2d60", size = .5) +
  geom_sf(data = wc_geo, aes(fill = w_l, group = interaction(w_l, Year))) +
  scale_fill_manual(values = c("#D9A441", "#A8A8A8"), name = NULL, labels = c("Winner", "Runner-Up")) +
  theme(legend.position = c(0.9, 1.01), legend.direction = "horizontal", axis.text = element_blank(), 
        panel.grid.minor = element_blank(), panel.grid.major = element_blank())

anim <- p + transition_time(Year)

animate(anim)

^{Created on 2020-04-24 by the reprex package (v0.3.0)}

P.S. It's always the same countries, isn't it...

gganimate::transition_time results in flying polygons

1 Answers1

Linked