Barplots on a Map

Question

I am trying to plot vertical bar plots over a map. I went through examples online but somehow not being able to.

My data is currently in this format:

University| Count | Category | lat | long

Here is the code that I am trying execute:

library(ggplot2)
library(ggmap)
library(ggsubplot)

df1 <- data.frame(
  University = c(rep("University1", 4), rep("University2", 4), rep("University3", 4), 
    rep("University4", 4)),
  Count = sample(1:10, 16, replace = T),
  Category = rep(c("A", "B", "C", "D")),
  lat = c(rep(10.902469, 4), rep(17.921959, 4), rep(18.606910, 4), rep(13.202366, 4)),
  long = c(rep(76.90020, 4), rep(83.42510, 4), rep(73.87501, 4), rep(77.62340, 4))
)

india <- get_map("India", zoom = 5)
p <- ggmap(india)
p + geom_subplot(data = df1, mapping=aes(x = long, y = lat, group = University,
subplot= geom_bar(aes(x = Category, y = Count, color = Category, stat = "identity"))))

When I run the above code, I get the following error:

Error in get(x, envir = this, inherits = inh)(this, ...) : 
   could not find function "%:::%"

Pedi

I didn't downvote your question but I assume that it is because your question isn't reproducible - we don't have your data frame so we can't reproduce your code. See this for reproducible example and question – Waldowaldon

Made the necessary change. Now the problem can be reproduced. – Pedi

I think the problem has to do with the version of R. My answer below was produced with R 2.15.3. On my other computer, which is running R 3.0.2, I got the same error message. – Expansionism

I updated my answer. It now also works in R 3.0.2. If it also works for you, it would be appreciated if you accept the answer. This will give future readers a clue about the value of the solution. – Expansionism

Jaap, I haven't been able to install R 3.0.2 and test out. Anyway I managed to make some plots using CartoDB and hence this requirement is no longer there. Maybe somebody else will get chance to verify. – Pedi

Possible duplicate of Plots on a map using ggplot2 – Variorum

Answer 1

You should also use the mapproj package. With the following code:

ggmap(india) +
  geom_subplot(data = df1, aes(x = long, y = lat, group = University,
                           subplot = geom_bar(aes(x = Category, y = Count,
                                                  fill = Category, stat = "identity"))))

I got the following result:

As noted in the comments of the question: this solution works in R 2.15.3 but for some reason not in R 3.0.2

---

UPDATE 16 januari 2014: when you update the ggsubplot package to the latest version, this solution now also works in R 3.0.2

---

UPDATE 2 oktober 2014: Below the answer of the package author (Garret Grolemund) about the issue mentioned by @jazzuro (text formatting mine):

Unfortunately, ggsubplot is not very stable. ggplot2 was not designed to be extensible or recursive, so the api between ggsubplot and ggplot2 is very jury rigged. I think entropy will assert itself as R continues to update.

The future plan for development is to implement ggsubplot as a built in part of Hadley's new package ggvis. This will be much more maintainable than the ggsubplot + ggplot2 pairing.

I won't be available to debug ggsubplot for several months, but I would be happy to accept pull requests on github.

---

UPDATE 23 december 2016: The ggsubplot-package is no longer actively maintained and is archived on CRAN:

Package ‘ggsubplot’ was removed from the CRAN repository.

Formerly available versions can be obtained from the archive.

Archived on 2016-01-11 as requested by the maintainer .

Expansionism

Thanks I will try and get 2.15.3. BTW, I tried with R 2.15.1 and it did not work. – Pedi

I am testing this code with R3.1.1., and I am receiving errors. Any updates? – Ike

@Expansionism I used your answer and got Error : Mapping a variable to y and also using stat="bin". I, then, compared your code to one in this link. I thought ) was missing after fill = Category. Once I corrected this bracket issue, I got the following error message, Error in layout_base(data, vars, drop = drop) : At least one layer must contain all variables used for facetting. Do you have any ideas of what is happening? – Ike

@Ike I added an update from the package author: unfortunately, no solution on the short term – Expansionism

@Expansionism Thank you for that. So it seems that we will have to wait for a little bit to see this kind of graphics then. – Ike

Answer 2

In fact we can use geom_errorbar() to add a bar in map actually. As long as we set the x and y of geom_errorbar() is longitude and latitude; and the ymin and ymax is stat of your data .

library(tidyverse)

plant_data <- read.csv("DV.csv")
colnames(plant_data) <- c("name", "lat", "lon", "p", "t", "hmi")
head(plant_data)
# name      lat      lon    p    t         hmi
# 1 Dodonaea viscosa subsp. viscosa -17.3000 145.9667 4084 24.1 8.905719509
# 2 Dodonaea viscosa subsp. viscosa -18.0833 146.0000 2562 24.0 13.18851823
# 3 Dodonaea viscosa subsp. viscosa -18.0833 146.0000 2562 24.0 13.18851823
# 4 Dodonaea viscosa subsp. viscosa -18.0833 146.0000 2562 24.0 13.18851823
# 5 Dodonaea viscosa subsp. viscosa -18.0833 146.0000 2562 24.0 13.18851823
# 6 Dodonaea viscosa subsp. viscosa -18.0833 146.0000 2562 24.0 13.18851823

plot part

aus <- map_data('world', region = "(Australia)|(Indonesia)|(Papua New Guinea)")
ggplot() +
  geom_polygon(data = aus, aes(x = long, y = lat, group = group), 
               fill = "gray80") + 
  geom_point(data = plant_data, aes(x = lon, y = lat), color = '#00CC00') +
  coord_map(xlim = c(110, 160), ylim = c(-45, -5)) +
  ggtitle("植物分布") + 
  theme(plot.title = element_text(hjust = 0.5)) + 
  geom_errorbar(data = plant_data[c(50, 99), ], 
                aes(x = lon,ymin=lat-3,ymax=lat+1), 
                color='blue', size = 3, width=0)

this part is very smart:

geom_errorbar(data = plant_data[c(50, 99), ], 
                aes(x = lon,ymin=lat-3,ymax=lat+1), 
                color='blue', size = 3, width=0)

it is very amazing!!

Gannie