如何在R的tmap分面图中为每个面板添加对应数据注释？

问题描述 #

我在使用R语言tmap包创建分面栅格图时，尝试为每个面板添加对应图层的均值（Mean）和标准差（SD）注释，但所有面板都重叠显示了所有注释，如何让每个面板仅显示自身对应的注释？

可复现代码：

library(tmap)
library(stars)
library(terra)
library(tidyverse) # 提供pivot_longer等数据处理函数

# 读取测试数据
tif = system.file("tif/L7_ETMs.tif", package = "stars")
x = terra::rast(tif)

# 计算每个栅格图层的均值和标准差
summ <- x |> 
  as.data.frame() |> 
  pivot_longer(everything()) |> 
  group_by(name) |> 
  summarise(Mean = mean(value, na.rm = TRUE), 
            SD = sd(value, na.rm = TRUE)) %>% 
  mutate(across(c(Mean, SD), ~round(., 3))) %>% 
  mutate(lab = paste0("Mean = ", Mean, "\nSD = ", SD)) %>% 
  dplyr::select(name, lab)

# 设置注释的显示位置（左下角偏移）
bbox <- st_bbox(x)
annotations <- summ %>%
  mutate(
    x = bbox["xmin"] + 0.05 * (bbox["xmax"] - bbox["xmin"]),
    y = bbox["ymin"] + 0.1 * (bbox["ymax"] - bbox["ymin"])
  )

# 转换为sf空间对象
annotations_sf <- st_as_sf(annotations, coords = c("x", "y"), crs = st_crs(x))

# 绘制基础分面栅格图
map <- tm_shape(x) + 
  tm_raster(style="quantile") + 
  tm_facets(nrow = 3) +
  tm_layout(panel.labels = names(x), attr.outside = T, 
            attr.outside.position = "bottom", attr.just = "right",
            legend.outside = T, legend.outside.position = "right",
            legend.position = c("left", "top"),
            legend.text.size = 0.8, legend.title.size = 0.9,
            legend.format = list(digits = 2, text.separator = "-"))+
  tm_compass(position = c("left", "top"))+
  tm_scale_bar(text.size = 1, position = c("RIGHT", "top"))

解决方法 #

核心是让注释图层遵循和栅格图层一致的分面规则，通过为注释图层添加tm_facets(by = "name")，将每个注释分配到对应名称的面板中：

修改后的注释添加代码：

map +
  tm_shape(annotations_sf) +
  tm_text("lab", xmod = 0.05, ymod = -0.05, col = "black", just = "left", size = 1) +
  tm_facets(by = "name", free.coords = FALSE)

关键参数说明 #

• by = "name"：根据注释数据中的name字段，将注释匹配到同名的栅格面板
• free.coords = FALSE：强制所有面板使用相同的坐标范围，避免注释位置偏移

内容的提问来源于stack exchange，提问作者UseR10085