R语言绘制按GaPa_NaPa分组的Forest.Area.ha.饼图方法

现有数据结构 #

你提供的数据框结构如下：

structure(list(GaPa_NaPa = c("Gaidahawa", "Kanchan", "Kotahimai", 
"Marchawari", "Mayadevi", "Omsatiya", "Rohini", "Sammarimai", 
"Siyari", "Sudhdhodhan", "Devdaha", "Lumbini Sanskritik", "Sainamaina", 
"Siddharthanagar", "Tillotama", "Butwal"), Total.Area..ha. = c(9657L, 
5835L, 5812L, 4844L, 7228L, 4844L, 6449L, 5066L, 6620L, 5743L, 
13667L, 11194L, 16082L, 3595L, 12592L, 10139L), Forest.Area.ha. = c(114.91, 
178.19, 31.37, 43.43, 152.87, 29.12, 63.16, 59.81, 36.4, 16.42, 
113.13, 422.87, 186.13, 167.2, 60.27, 45.3), Forest.Percent = c(6.67, 
10.35, 1.83, 2.52, 8.88, 1.69, 3.67, 3.47, 2.11, 0.95, 6.57, 
24.57, 10.81, 9.71, 3.5, 2.63), Forest.Area..Fraction. = c(0.07, 
0.1, 0.02, 0.03, 0.09, 0.02, 0.04, 0.03, 0.02, 0.01, 0.07, 0.25, 
0.11, 0.1, 0.04, 0.03), Household.No = c(8612L, 9828L, 5939L, 
5305L, 8003L, 6683L, 6349L, 5164L, 7889L, 7619L, 15624L, 10736L, 
17572L, 12329L, 30452L, 36989L), Family.Size = c(10020L, 10483L, 
7921L, 6972L, 10040L, 8218L, 8096L, 7303L, 9060L, 8717L, 17582L, 
13854L, 19657L, 16011L, 36399L, 51099L), Total = c(56529L, 42528L, 
46417L, 41058L, 57341L, 41080L, 43277L, 43300L, 45274L, 41472L, 
71806L, 88090L, 78477L, 76307L, 149657L, 195054L)), row.names = c(NA, 
16L), class = "data.frame")

原有代码问题 #

你写的代码有两个核心错误，导致无法出图：

• pie()函数第一个参数必须传入数值型向量定义每个扇区的大小，你传入了字符型的行政区名称列GaPa_NaPa，无法计算扇区占比
• 读入数据时赋值的对象名是ForestArea2010，调用时却写了不存在的RupandehiLULC19，对象名不匹配会直接抛出错误
  原有错误代码：

setwd("C:/Users/lenovo/Desktop/AllAboutR/AssignmentDocs")
ForestArea2010<-read.csv("Forest2010.csv")
View(ForestArea2010)
pie(RupandehiLULC19$GaPa_NaPa, main="Piechart of Forest Area", las=3, col=hsv(12))

正确实现代码 #

基础R原生函数实现 #

# 设置工作路径
setwd("C:/Users/lenovo/Desktop/AllAboutR/AssignmentDocs")
# 读入数据
ForestArea2010 <- read.csv("Forest2010.csv")

# 生成扇区标签：行政区名+森林面积占比
pie_labels <- paste0(
  ForestArea2010$GaPa_NaPa, "\n",
  round(ForestArea2010$Forest.Area.ha. / sum(ForestArea2010$Forest.Area.ha.) * 100, 1),
  "%"
)

# 绘制饼图
pie(
  x = ForestArea2010$Forest.Area.ha.,
  labels = pie_labels,
  main = "各行政区森林面积占比（单位：公顷）",
  col = rainbow(nrow(ForestArea2010)),
  border = "white"
)

行政区数量较多时，原生pie()函数的标签容易重叠，更推荐用ggplot2绘制可读性更高的版本。

ggplot2版本实现（推荐） #

如果未安装ggplot2，先运行install.packages("ggplot2")完成安装，再执行以下代码：

library(ggplot2)

ggplot(ForestArea2010, aes(x = "", y = Forest.Area.ha., fill = GaPa_NaPa)) +
  geom_col(color = "white", linewidth = 0.5) +
  # 转换为极坐标生成饼图
  coord_polar(theta = "y", start = 0) +
  # 添加扇区占比标签
  geom_text(
    aes(label = paste0(round(Forest.Area.ha./sum(Forest.Area.ha.)*100, 1), "%")),
    position = position_stack(vjust = 0.5),
    size = 3
  ) +
  # 设置图表标注
  labs(
    title = "各行政区森林面积占比（单位：公顷）",
    fill = "行政区",
    x = NULL,
    y = NULL
  ) +
  theme_bw() +
  # 清理冗余坐标轴元素
  theme(
    axis.text = element_blank(),
    axis.ticks = element_blank(),
    panel.grid = element_blank()
  )

内容的提问来源于stack exchange，提问作者Prabhakar