使用R中的栅格包为特定国家的省份自定义颜色 [英] custom color for provinces in specific country using raster package in R

问题描述

我正在尝试为特定国家的每个省(绿、红、蓝)贴标签，如下所示：

library(raster)
library(rgeos)
library(ggplot2)
library(dplyr)

iran <- getData("GADM", country = "Iran", level = 1)
map <- fortify(iran)
map$id <- as.integer(map$id)
dat <- data.frame(id = 1:(length(iran@data$NAME_1)),
                  state = iran@data$NAME_1,
                  pr = c(530, -42, 1673, 75, 206, 544, 1490, 118, 75,
                         40, 105, 191, 111, 810, 609, 425, 418, 550, 40, 425, -54, -50,
                         16, 18, 133,425, -30, 241,63, 191,100))
dat <- dat %>% mutate(color_province = case_when(pr <= 50 ~ 'green',
                                                 pr > 150 ~ 'red',
                                                 TRUE ~ 'yellow'))

我想设置每个省的自定义颜色(使用‘dat’数据框)，如下所示：

> dat
   id                       state   pr color_province
1   1                      Alborz  530            red
2   2                     Ardebil  -42          green
3   3                     Bushehr 1673            red
4   4 Chahar Mahall and Bakhtiari   75         yellow
5   5             East Azarbaijan  206            red
6   6                     Esfahan  544            red
7   7                        Fars 1490            red
8   8                       Gilan  118         yellow
9   9                    Golestan   75         yellow
10 10                     Hamadan   40          green
11 11                   Hormozgan  105         yellow
12 12                        Ilam  191            red
13 13                      Kerman  111         yellow
14 14                  Kermanshah  810            red
15 15                   Khuzestan  609            red
16 16  Kohgiluyeh and Buyer Ahmad  425            red
17 17                   Kordestan  418            red
18 18                    Lorestan  550            red
19 19                     Markazi   40          green
20 20                  Mazandaran  425            red
21 21              North Khorasan  -54          green
22 22                      Qazvin  -50          green
23 23                         Qom   16          green
24 24             Razavi Khorasan   18          green
25 25                      Semnan  133         yellow
26 26      Sistan and Baluchestan  425            red
27 27              South Khorasan  -30          green
28 28                      Tehran  241            red
29 29             West Azarbaijan   63         yellow
30 30                        Yazd  191            red
31 31                      Zanjan  100         yellow

举个例子，‘阿德比勒’省是绿色的，‘赞江’省是黄色的。我使用‘Fill’参数设置自定义颜色，下面是我的尝试：

map_df <- inner_join(map, dat, by = "id")
centers <- data.frame(gCentroid(iran, byid = TRUE))
centers$state <- dat$state
ggplot() +  
geom_map(data = map_df, map = map_df,
         aes(map_id = id, group = group, 
             x = long, y = lat,
             fill = as.factor(color_province))) +
geom_text(data = centers, aes(label = state, x = x, y = y), size = 3) +
coord_map() +
labs(x = "", y = "", title = "Iran Province") +
scale_fill_manual(values = list(yellow = 'yellow', red = 'red', green = 'green'))

但它不起作用。根据"日期"数据框，"阿德比勒"是"绿色"的，但在地图上是红色的。有没有更好的方法来做这件事？

推荐答案

既然您问我，我就快速浏览了一下您的代码。您对id的假设基本上是错误的。当您使用fortify()时，您可能认为id是按正常顺序分配的(例如，1到n)。但事实并非如此。只需运行unique(mymap$id)即可。你会明白我的意思的。那么解决方案是什么呢？创建dat时，需要rownames(iran@data)。一旦这样做了，你应该会没事的。请参阅最终图形。

library(raster)
library(rgeos)
library(ggplot2)
library(dplyr)

iran <- getData("GADM", country = "Iran", level = 1)
mymap <- fortify(iran) # This is the hidden cause
mymap$id <- as.integer(mymap$id)

dat <- data.frame(id = rownames(iran@data), # This is what you needed.
                  state = iran@data$NAME_1,
                  pr = c(530,-42,1673,75,206,544,1490,118,75,
                         40,105,191,111,810, 609,425,418,550, 40, 425, -54,-50,
                         16, 18, 133,425, -30, 241,63, 191,100)) %>% 
       mutate(color_province = case_when(pr <= 50 ~ 'green',
                                         pr > 150 ~ 'red',
                                         TRUE ~ 'yellow'))

mydf <- inner_join(mymap, dat, by = "id")
centers <- data.frame(gCentroid(iran, byid = TRUE))
centers$state <- dat$state

ggplot() +
geom_map(data = mydf,
         map = mydf,
         aes(map_id = id, group = group,
             x = long, y = lat,
             fill = as.factor(color_province))) +
geom_text(data = centers,
          aes(label = state, x = x, y = y), size = 3) +
coord_map() +
labs(x = "", y = "", title = "Iran Province") +
scale_fill_manual(values = c("green", "red", "yellow"),
                  name = "Province") 

这篇关于使用R中的栅格包为特定国家的省份自定义颜色的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！