如何让ggplot2使用指定颜色绘制浮游生物叶绿素a箱线图

我有墨西哥湾5个区域两类浮游生物（'WSW'和'<20 µm'）的叶绿素a生物量数据，每个区域都包含两类数据。想要绘制箱线图实现以下效果：

• 为每个区域指定特定颜色（需与其他图表统一）
• 为其中一类尺寸添加图案区分两类数据

目前图案已成功添加，但ggplot2始终使用默认颜色，无法应用我指定的region_colors颜色。以下是我原代码及数据：

region_colors <- c('WFS' = 'darkgreen', 'TLS' = 'cyan3', 'EMS' = 'orangered2', 'CB' = 'yellow3', 'YC/FS' = 'plum2')

surf_boxplot= ggplot(surf_box, aes(x = region, y = Chla, fill = region, pattern = Size)) + 
  geom_boxplot_pattern(pattern_fill = "white", pattern_color = "white",
                       pattern_angle = 45, pattern_size = 0, 
                       pattern_spacing = 0.025) + 
  scale_pattern_fill_manual(values = region_colors) + 
  scale_pattern_manual(values = patterns) +
  theme(panel.background = element_blank(), 
        legend.position = 'top',
        axis.line.x = element_line(color="black", size = 1), 
        axis.line.y = element_line(color="black", size = 1),
        plot.title = element_text(margin = margin(t = 0, r = 0, b = 0, l = 0),
                                  hjust = 0.5, size = 15, face = 'bold'),
        legend.key = element_blank(),
        legend.title = element_text(margin = margin(t = 0, r = 0, b = 0, l = 0), 
                                    size = 10, face = "bold")) + 
  labs(x=expression(bold(paste("Regions of the Gulf of Mexico"))), 
       y=expression(bold(paste("Chl-",bolditalic ("a"), 
                               " biomass (µg L"^'-1',')'))), 
       title = expression(bold(paste("Surface Chl-", 
                                     bolditalic ("a"), ' Biomass'))))

数据：

structure(list(Size = c("WSW", "WSW", "WSW", "WSW", "WSW", "WSW", 
"WSW", "WSW", "WSW", "WSW", "WSW", "WSW", "WSW", "WSW", "WSW", 
"WSW", "WSW", "WSW", "WSW", "WSW", "WSW", "WSW", "WSW", "WSW", 
"WSW", "<20 µm", "<20 µm", "<20 µm", "<20 µm", "<20 µm", 
"<20 µm", "<20 µm", "<20 µm", "<20 µm", "<20 µm", "<20 µm", 
"<20 µm", "<20 µm", "<20 µm", "<20 µm", "<20 µm", "<20 µm", 
"<20 µm", "<20 µm", "<20 µm", "<20 µm", "<20 µm", "<20 µm", 
"<20 µm", "<20 µm"), Chla = c(0.0158553374250702, 0.0681835535205316, 
1.69758559710116, 0.38623655064116, 0.18999889055035, 0.0475280806809533, 
0.0518441820414833, 1.38653044994664, 0.811092911431239, 0.0627435113039509, 
0.0600206049988672, 0.76576153727216, 0.0131834301368552, 0.0139065718743198, 
0.0801018539960821, 0.0476673877342761, 0.0170883088104008, 0.0353610062847066, 
0.0862740442259294, 0.54829425475163, 0.354272113611421, 0.0419050024352668, 
0.14044768258058, 0.0693766746908316, 0.0539223064788303, 0.0159113633523673, 
0.033895686014726, 0.592754310802978, 0.251252264220314, 0.177521351469432, 
0.0574533958589567, 0.0447197172789116, 1.75919476060424, 0.789171481392557, 
0.104392221393643, 0.0613148498505948, 0.646043888487357, 0.0200254634990205, 
0.0165766336741892, 0.0619120579844718, 0.0324340598144779, 0.0264756363476934, 
0.0276985089356931, 0.105004593301296, 0.288904529677619, 0.295226761342851, 
0.0419623280199936, 0.10089302911911, 0.0580098748827292, 0.0647851594974109
), region = c("WFS", "WFS", "WFS", "WFS", "WFS", "WFS", "TLS", 
"TLS", "TLS", "TLS", "TLS", "TLS", "EMS", "EMS", "EMS", "EMS", 
"EMS", "CB", "CB", "CB", "CB", "CB", "CB", "YC/FS", "YC/FS", 
"WFS", "WFS", "WFS", "WFS", "WFS", "WFS", "TLS", "TLS", "TLS", 
"TLS", "TLS", "TLS", "EMS", "EMS", "EMS", "EMS", "EMS", "CB", 
"CB", "CB", "CB", "CB", "CB", "YC/FS", "YC/FS"), depth_name = c("Surface", 
"Surface", "Surface", "Surface", "Surface", "Surface", "Surface", 
"Surface", "Surface", "Surface", "Surface", "Surface", "Surface", 
"Surface", "Surface", "Surface", "Surface", "Surface", "Surface", 
"Surface", "Surface", "Surface", "Surface", "Surface", "Surface", 
"Surface", "Surface", "Surface", "Surface", "Surface", "Surface", 
"Surface", "Surface", "Surface", "Surface", "Surface", "Surface", 
"Surface", "Surface", "Surface", "Surface", "Surface", "Surface", 
"Surface", "Surface", "Surface", "Surface", "Surface", "Surface", 
"Surface"), distance = c("Offshore", "Offshore", "Nearshore", 
"Nearshore", "Intermediate", "Offshore", "Offshore", "Nearshore", 
"Nearshore", "Intermediate", "Intermediate", "Nearshore", "Nearshore", 
"Intermediate", "Offshore", "Intermediate", "Offshore", "Offshore", 
"Intermediate", "Nearshore", "Nearshore", "Offshore", "Intermediate", 
"Offshore", "Intermediate", "Offshore", "Offshore", "Nearshore", 
"Nearshore", "Intermediate", "Offshore", "Offshore", "Nearshore", 
"Nearshore", "Intermediate", "Intermediate", "Nearshore", "Nearshore", 
"Intermediate", "Offshore", "Intermediate", "Offshore", "Offshore", 
"Intermediate", "Nearshore", "Nearshore", "Offshore", "Intermediate", 
"Offshore", "Intermediate")), row.names = c(NA, -50L), class = c("tbl_df", 
"tbl", "data.frame"))

问题核心在于你错误地使用了scale_pattern_fill_manual来设置箱线图的主体填充色，实际上fill美学映射的是region，应该用scale_fill_manual来指定区域颜色。同时补充patterns变量的定义，确保图案映射生效。

修正后的代码：

# 定义区域颜色
region_colors <- c('WFS' = 'darkgreen', 'TLS' = 'cyan3', 'EMS' = 'orangered2', 'CB' = 'yellow3', 'YC/FS' = 'plum2')
# 定义图案：给<20 µm添加条纹，WSW无图案
patterns <- c('WSW' = '', '<20 µm' = 'stripe')

surf_boxplot <- ggplot(surf_box, aes(x = region, y = Chla, fill = region, pattern = Size)) + 
  geom_boxplot_pattern(
    pattern_fill = "white",    # 图案内部填充色
    pattern_color = "white",   # 图案线条颜色
    pattern_angle = 45,        # 图案角度
    pattern_size = 0.2,        # 图案线条粗细（原0可能导致图案不显示）
    pattern_spacing = 0.025    # 图案间距
  ) + 
  scale_fill_manual(values = region_colors) +  # 关键：用这个设置fill=region的颜色
  scale_pattern_manual(values = patterns) +
  theme(
    panel.background = element_blank(), 
    legend.position = 'top',
    axis.line.x = element_line(color="black", size = 1), 
    axis.line.y = element_line(color="black", size = 1),
    plot.title = element_text(margin = margin(t = 0, r = 0, b = 0, l = 0),
                              hjust = 0.5, size = 15, face = 'bold'),
    legend.key = element_blank(),
    legend.title = element_text(margin = margin(t = 0, r = 0, b = 0, l = 0), 
                                size = 10, face = "bold")
  ) + 
  labs(
    x=expression(bold(paste("Regions of the Gulf of Mexico"))), 
    y=expression(bold(paste("Chl-",bolditalic ("a"), 
                            " biomass (µg L"^'-1',')'))), 
    title = expression(bold(paste("Surface Chl-", 
                                  bolditalic ("a"), ' Biomass')))
  )

# 查看图形
print(surf_boxplot)

关键修正说明： #

1. 替换scale_pattern_fill_manual为scale_fill_manual(values = region_colors)：这是控制fill = region映射颜色的正确scale，直接作用于箱线图的主体填充。
2. 补充patterns变量定义：明确两类尺寸对应的图案类型，确保图案映射生效。
3. 调整pattern_size从0到0.2：原0值会导致图案线条不可见，适当调整后条纹图案能正常显示。

内容的提问来源于stack exchange，提问作者garfunkl_30