Making a diagram map in r?

Question

This is my dataframe:

df:

Country          Total       lon        lat
United Kingdom   5000       -3.43597    55.37805
China            4000       104.1954    35.86166
France           4000       2.213749    46.22764
Australia        4500       133.7751    -25.2744
Mexico           0          -102.553    23.6345

I am new to R. I am trying to make a map showing travel from many different countries to Mexico. The destination is Mexico, so all the data flow will be in one direction. I don't know how to make the connections between countries. I have used the following codes to make a layout:



world <- ne_countries(scale = "medium", returnclass = "sf")
class(world)
new_df <- read.csv('df.csv')
ggrepel::geom_label_repel

ggplot(data = world) +
  geom_sf()+geom_point(data=new_df, aes(x=lon, y=lat), colour='red')+
  ggrepel::geom_label_repel(data = new_df, 
                            aes(x = lon, y = lat), 
                            label = new_df$Country,
                            size = 2,
                            col = "blue")

A good starting point https://github.com/rafapereirabr/flow-map-in-r-ggplot — M.Viking, Sep 03 '22 at 13:33

Maël · Accepted Answer · 2022-09-03T19:15:06.650

You can use geom_curve and use the size argument. You can also use color to augment contrast:

new_df <- new_df %>% 
  mutate(Total = as.numeric(gsub(",", "", Total)),
         lon2 = lon[Country == "Mexico"],
         lat2 = lat[Country == "Mexico"])
#          Country   Total         lon       lat     lon2    lat2
# 1 United Kingdom 7200000   -3.435970  55.37805 -102.553 23.6345
# 2          China 5359000  104.195400  35.86166 -102.553 23.6345
# 3         Canada 6016000 -106.347000  56.13037 -102.553 23.6345
# 4         Brazil 5269000  -51.925300 -14.23500 -102.553 23.6345
# 5         France 4484000    2.213749  46.22764 -102.553 23.6345
# 6      Australia 4014000  133.775100 -25.27440 -102.553 23.6345
# 7         Mexico       0 -102.553000  23.63450 -102.553 23.6345   


ggplot(data = world) +
  geom_sf() + 
  geom_curve(data = new_df[-7, ], 
             aes(x = lon, y = lat, xend = lon2, yend = lat2, size = Total, color = Total),
             curvature = -0.2, 
             lineend = "round",
             arrow = arrow(length = unit(0.1, "cm"))) +
  scale_size(range = c(.5, 3), label = scales::comma) +
  scale_color_distiller(label = scales::comma, 
                        palette = "Reds", 
                        direction = 1) +
  guides(color = guide_legend(), 
         size = guide_legend()) +
  ggrepel::geom_label_repel(data = new_df, aes(x = lon, y = lat), 
                            label = new_df$Country, size = 2, col = "blue") +
  ggthemes::theme_map()

Data

new_df <- read.table(header = T, text = "Country          Total       lon        lat
'United Kingdom'   7,200,000   -3.43597   55.37805
China            5,359,000   104.1954   35.86166
Canada           6,016,000   -106.347   56.13037
Brazil           5,269,000   -51.9253   -14.235
France           4,484,000   2.213749   46.22764
Australia        4,014,000   133.7751   -25.2744
Mexico           0           -102.553   23.6345")

This is beautiful. I have a bit of an irrelevant question. How did you manage to copy the data frame from the clipboard? Did you use datapasta? — LDT, Sep 03 '22 at 16:28
datapasta is a good option, but I guess I'm doing it the old fashioned way, with `read.table` and a simple copy paste. When it does not work, it's usually because this method does not handle string with spaces, so one has to manually add quotation mark (e.g. here with United Kingdom). I edited the question to include the data.frame (and you can see how I retrieve it. — Maël, Sep 03 '22 at 19:14

Making a diagram map in r?

1 Answers1