使用ggplot2绘制农田月度任务堆叠条形图：问题修正与优化建议

需求 #

需要绘制可视化图表展示3块农田全年各月份的任务安排，目标效果类似分块堆叠的条形图：

• X轴为月份（按全年顺序排列）
• Y轴为农田ID
• 不同任务用4种颜色区分

现有代码 #

# 加载所需包
library(ggplot2)
library(tidyr)

# 数据
df34278 <- structure(list(fieldID = c("fieldA", "fieldB", "fieldC"), 
                          January = c("task2", "task3", "task2"), 
                          February = c("task2", "task3", "task2"), 
                          March = c("task2", "task3", "task2"), 
                          April = c("task2", "task3", "task1"), 
                          May = c("task2", "task3", "task4"), 
                          June = c("task1", "task3", "task4"), 
                          July = c("task2", "task3", "task4"), 
                          August1 = c("task2", "task2", "task4"), 
                          August15 = c("task2", "task2", "task2"), 
                          September = c("task1", "task2", "task2"), 
                          October = c("task3", "task2", "task2"), 
                          November = c("task3", "task2", "task2"), 
                          December = c("task3", "task2", "task2")), 
                     class = "data.frame", row.names = c(NA, -3L))

# 转换为长格式并将Task转为因子
df34278_longer <- pivot_longer(df34278,
                               cols = 2:14,
                               names_to = "Month",
                               values_to = "Task")

df34278_longer$Task <- as.factor(df34278_longer$Task)

# 创建绘图
plot34278 <- ggplot(df34278_longer, aes(x = fieldID, y = Month, fill = Task)) +
  geom_bar(stat = "identity", position = "stack") +
  coord_flip()
  
plot34278

遇到的问题 #

运行代码后结果不符合预期：

• 月份未按全年顺序排列
• 月份标签显示异常（如August1、August15格式不统一）
• 常规堆叠条形图的用法无法匹配需求，需要调整绘图逻辑

运行环境 #

R version 4.3.0 (2023-04-21)
Platform: x86_64-apple-darwin20 (64-bit)
Running under: macOS Monterey 12.6.9

解决方案建议 #

你的需求更适合用**瓷砖图（tile plot）**而非堆叠条形图，瓷砖图能清晰展示每个农田在每个时间节点的任务类型，调整步骤如下：

1. 数据预处理：规范月份并设置有序因子 #

先处理August1和August15的格式，再将Month设置为有序因子，确保按时间顺序排列：

# 统一八月子节点的命名格式
df34278_longer$Month <- gsub("August(\\d+)", "August_\\1", df34278_longer$Month)

# 定义正确的月份顺序
month_order <- c("January", "February", "March", "April", "May", "June", 
                 "July", "August_1", "August_15", "September", "October", 
                 "November", "December")

# 将Month转为有序因子
df34278_longer$Month <- factor(df34278_longer$Month, levels = month_order, ordered = TRUE)

2. 使用geom_tile绘制瓷砖图 #

瓷砖图直接映射每个(农田, 月份)组合的任务颜色，完全匹配需求：

ggplot(df34278_longer, aes(x = Month, y = fieldID, fill = Task)) +
  geom_tile(color = "white", size = 1) + # 白色边框区分每个任务块
  scale_x_discrete(guide = guide_axis(n.dodge = 2)) + # 避免月份标签重叠
  scale_fill_brewer(palette = "Set2") + # 用预设配色区分任务
  labs(x = "月份", y = "农田ID", fill = "任务类型") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) # 旋转X轴标签提升可读性

3. 若坚持堆叠条形图形式 #

如果需要类似堆叠条形的视觉效果，可以给每个月份分配连续X轴位置，结合分面实现：

# 给每个月份分配连续的X轴位置
df34278_longer$x_pos <- match(df34278_longer$Month, month_order)

ggplot(df34278_longer, aes(x = x_pos, y = 1, fill = Task)) +
  geom_col(position = "stack", width = 1) +
  facet_grid(fieldID ~ ., switch = "y") + # 按农田分面
  scale_x_continuous(breaks = seq_along(month_order), labels = month_order) +
  scale_fill_brewer(palette = "Set2") +
  labs(x = "月份", y = "", fill = "任务类型") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1),
        strip.text.y.left = element_text(angle = 0)) # 调整分面标签方向

内容的提问来源于stack exchange，提问作者Moritz Twente