ggplot双纵轴图表：geom_tile与geom_line组合的图例异常问题

你的推测完全正确：shape是geom_point()的专属美学属性，当前图例仅基于geom_line()的color映射，所以设置shape参数会被忽略。要让图例显示柱状图形，需要添加一个虚拟的geom层来承载shape属性，同时不影响实际图表显示。

解决方案代码 #

library(ggplot2)
library(dplyr)
library(plotly)

# 保留原数据准备代码（此处省略，使用你提供的dfAll、brks2等变量）

# 准备geom层：保持原geom_tile和geom_line设置，geom_tile仍需show.legend=FALSE
ggtiles <- geom_tile(aes(x = Date, y = Val1/2, height = Val1, fill = ID, group = ID),
                     na.rm = TRUE, stat = "identity", position = position_dodge(preserve = "single"),
                     show.legend = FALSE)

gglines <- geom_line(aes(x = Date, y = Val2_scl, color = ID, group = ID), na.rm = TRUE)

# 添加虚拟geom_point层：仅用于图例显示，实际图表中不可见
# 只针对左侧的ID（A、B），设置shape=15（方块），alpha=0让点不可见
gg_dummy <- geom_point(
  aes(x = Date, y = Val1, color = ID),
  data = dfAll %>% filter(ID %in% c("A", "B")),
  shape = 15, size = 5, alpha = 0
)

# 调整图例设置：指定不同ID的显示样式
legguides <- guides(
  color = guide_legend(
    title.position = "top",
    override.aes = list(
      shape = c(15, 15, NA, NA),  # A、B显示方块，J、K隐藏形状
      linetype = c(0, 0, 1, 1),   # A、B隐藏折线，J、K显示折线
      fill = c("A" = "#F8766D", "B" = "#00BFC4", "J" = NA, "K" = NA) # 匹配geom_tile的填充色
    )
  )
)

# 生成图表：添加虚拟层，手动指定颜色确保一致性
ggplot(dfAll) + ggtiles + gglines + gg_dummy +
  scale_y_continuous(sec.axis = dup_axis(breaks = rev(brks2), labels = rev(labs2), name = "Val2")) +
  coord_cartesian(ylim = ylim1) + ylab("Val1") +
  scale_color_manual(values = c("A" = "#F8766D", "B" = "#00BFC4", "J" = "#F8766D", "K" = "#00BFC4")) +
  legguides

关键说明 #

1. 虚拟geom层的作用：通过alpha=0让点在实际图表中不可见，但它会让ggplot识别shape属性，从而在图例中生效。
2. 颜色一致性：手动指定scale_color_manual和override.aes中的fill值，确保geom_tile的填充色和图例方块颜色一致。
3. 图例样式控制：通过override.aes分别设置不同ID的shape和linetype，实现左侧ID显示柱状方块、右侧ID显示折线的效果。

另一种思路：统一美学映射 #

如果不想用虚拟层，可以将geom_tile的fill和geom_line的color都映射到ID，然后合并图例：

# 修改geom_tile去掉show.legend=FALSE
ggtiles <- geom_tile(aes(x = Date, y = Val1/2, height = Val1, fill = ID, group = ID),
                     na.rm = TRUE, stat = "identity", position = position_dodge(preserve = "single"))

# 合并fill和color图例
legguides <- guides(
  fill = guide_legend(title = "ID", title.position = "top",
                      override.aes = list(shape = 15, linetype = 0)),
  color = guide_legend(title = "ID", title.position = "top",
                       override.aes = list(shape = NA, linetype = 1))
)

# 统一颜色映射
ggplot(dfAll) + ggtiles + gglines +
  scale_y_continuous(sec.axis = dup_axis(breaks = rev(brks2), labels = rev(labs2), name = "Val2")) +
  coord_cartesian(ylim = ylim1) + ylab("Val1") +
  scale_color_manual(values = c("A" = "#F8766D", "B" = "#00BFC4", "J" = "#F8766D", "K" = "#00BFC4")) +
  scale_fill_manual(values = c("A" = "#F8766D", "B" = "#00BFC4", "J" = NA, "K" = NA)) +
  legguides +
  theme(legend.position = "top")

这种方式会生成两个图例，你可以通过theme(legend.box = "horizontal")让它们并排显示，或者用guide_merge合并为一个（需要ggplot2 3.5.0及以上版本）。

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