ggplot2中同一特征的颜色与形状图例合并技术问询

要合并颜色（colour）和形状（shape）图例，核心是让这两个美学映射绑定到同一个变量，并且统一图例的配置（名称、排列等），避免生成两个独立的引导项。下面针对你的代码给出具体解决方案：

问题分析 #

你当前的代码中，虽然colour和shape都映射到了EQ变量，但因为以下两点导致图例分开：

• 在labs()中设置了shape = element_blank()，使得形状图例的标题为空，与颜色图例的默认标题（EQ）不一致
• 在guides()中分别配置了shape和colour的引导规则，强制生成两个独立图例

修改后的完整代码 #

library(ggplot2)
library(dplyr) # 确保加载dplyr以支持sample_n函数

# 你的示例数据
ALL_Spans <- structure(list(UG = c("100", "124", "108", "92", "124", "84", "100", "100", "124", "84"), S = c(12, 12, 12, 12, 12, 12, 12, 12, 12, 12), EQ = c("Cai", "AASHTO LRFD", "Suksawang & Nassif", "AASHTO LRFD", "Modified Henry", "Modified Henry", "Analytical Data", "Rigid Rotation", "Rigid Rotation", "Rigid Rotation"), DF = c(0.818313323177077, 1.01028571424963, 0.741281190719393, 1.08244897959184, 0.858545454545455, 0.858545454545455, 0.789710362673591, 0.71825, 0.926666666666667, 0.926666666666667), Girder = c("Exterior Girder", "Exterior Girder", "Exterior Girder", "Interior Girder", "Interior Girder", "Exterior Girder", "Exterior Girder", "Interior Girder", "Exterior Girder", "Exterior Girder" ), Type = c("Moment LLDF", "Shear LLDF", "Moment LLDF", "Shear LLDF", "Moment LLDF", "Moment LLDF", "Moment LLDF", "Moment LLDF", "Moment LLDF", "Moment LLDF"), L = c(240L, 300L, 260L, 220L, 300L, 200L, 240L, 240L, 300L, 200L), group = c("Cai 12", "AASHTO LRFD 12", "Suksawang & Nassif 12", "AASHTO LRFD 12", "Modified Henry 12", "Modified Henry 12", "Analytical Data 12", "Rigid Rotation 12", "Rigid Rotation 12", "Rigid Rotation 12" )), row.names = c(NA, -10L), class = "data.frame")

b <- ggplot(data = ALL_Spans, aes(x = L, y = DF, colour = EQ, shape = EQ)) + 
  geom_point(stroke = 0.2, size = 2) + 
  geom_line(size = 0.2, linetype = "dotted") + 
  # 保持两个scale的名称一致（默认均为EQ，也可手动指定name参数）
  scale_shape_manual(values = c(0,19,3,2,4,5,6,8)) + 
  scale_colour_manual(values = c("black", "red", rep("black",6))) + 
  scale_x_continuous(breaks = seq(200,300,20), limits = c(200,300)) + 
  labs(x = "x", y = element_blank()) + # 移除shape的空标题设置，统一图例标题
  theme_classic() + 
  theme(legend.title.align = 0.5, legend.position = "bottom", 
        axis.text = element_text(size=8, colour="black"), 
        axis.title.x = element_text(size=8, colour="black"), 
        axis.line = element_blank(), 
        axis.ticks = element_line(colour = "black", size = 0.2), 
        legend.text = element_text(size=8, margin = margin(-10,5,-10,-4.5)), 
        legend.title = element_text(size=8, margin = margin(-10,1.5,-10,0)), 
        legend.spacing.x = unit(0.1, 'cm'), 
        legend.spacing.y = unit(0, 'cm'), 
        legend.background = element_blank(), 
        panel.margin.y = unit(0.3, "lines"), 
        legend.box.background = element_rect(color = "black", size = 0.2)) + 
  # 仅配置colour的引导规则，同时控制颜色和形状，关闭独立的shape引导
  guides(colour = guide_legend(nrow=2, byrow=TRUE), 
         linetype = FALSE, 
         shape = FALSE) + 
  facet_grid(Type~Girder)

print(b)

关键修改点说明 #

1. 统一图例标题：移除了labs()中的shape = element_blank()，让colour和shape的图例标题保持一致（默认都是EQ），这是合并图例的核心前提。如果需要自定义标题，只需在labs()中添加colour = "你的自定义标题"，shape图例会自动继承该标题。
2. 合并引导配置：在guides()中仅保留colour的引导设置，同时将shape设为FALSE，这样ggplot会自动把形状和颜色合并到同一个图例中，共享nrow=2, byrow=TRUE的排列规则。
3. 优化映射写法：将shape = EQ移到了顶层的aes()中，让代码更整洁（多个图层共享的映射推荐放在顶层）。

这样修改后，颜色和形状就会合并成一个图例，每个条目同时显示对应的颜色和形状，完全符合你的需求。

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