18  数据可视化基础

19 数据可视化基础

好的图表胜过千言万语。本章介绍 ggplot2 的语法体系和科研论文级图表的制作方法。

library(tidyverse)
library(patchwork)

19.1 ggplot2 语法体系

ggplot2 基于”图形语法”（Grammar of Graphics），将图表分解为几个核心组件：

ggplot(data, aes(x, y)) +   # 数据和映射
  geom_xxx() +                # 几何对象（画什么）
  scale_xxx() +               # 标度（怎么映射）
  labs() +                    # 标签
  theme()                     # 主题（外观）

Noteaes() 是什么？

aes() 是 aesthetic mapping（美学映射）的缩写，它告诉 ggplot2 如何将数据变量映射到图形属性上：

• aes(x = soil_ph) → 将 soil_ph 映射到 x 轴
• aes(color = habitat) → 将 habitat 映射到颜色
• aes(size = elevation) → 将 elevation 映射到点的大小

关键区别：写在 aes() 里面的属性由数据驱动（每个数据点可以不同），写在 aes() 外面的属性是固定值（所有点相同）。例如 geom_point(aes(color = habitat)) 按生境着色，而 geom_point(color = "red") 所有点都是红色。

Tip+ 运算符

ggplot2 使用 + 将图层逐层叠加。每个 + 后面添加一个新的组件（几何对象、标度、主题等）。注意 + 必须放在行末，不能放在下一行的开头：

# 正确
ggplot(data, aes(x, y)) +
  geom_point()

# 错误（+ 不能在行首）
ggplot(data, aes(x, y))
  + geom_point()

19.2 示例数据

set.seed(2027)
n <- 100

ecology <- tibble(
  species_richness = rpois(n, 15),
  soil_ph = rnorm(n, 6.2, 0.8),
  organic_carbon = rlnorm(n, log(25), 0.4),
  elevation = runif(n, 200, 1200),
  habitat = sample(c("森林", "草地", "湿地", "灌丛"), n, replace = TRUE,
                   prob = c(0.4, 0.25, 0.15, 0.2)),
  season = sample(c("春", "夏", "秋", "冬"), n, replace = TRUE)
)

19.3 常用图表类型

19.3.1 散点图

ggplot(ecology, aes(x = soil_ph, y = species_richness)) +
  geom_point(alpha = 0.6, size = 2) +
  geom_smooth(method = "lm", se = TRUE, color = "steelblue") +
  labs(x = "土壤 pH", y = "物种丰富度", title = "土壤 pH 与物种丰富度的关系") +
  theme_minimal()

19.3.2 分组散点图

ggplot(ecology, aes(x = soil_ph, y = species_richness, color = habitat)) +
  geom_point(alpha = 0.7, size = 2) +
  geom_smooth(method = "lm", se = FALSE) +
  labs(x = "土壤 pH", y = "物种丰富度", color = "生境类型") +
  theme_minimal()

19.3.3 箱线图

ggplot(ecology, aes(x = habitat, y = species_richness, fill = habitat)) +
  geom_boxplot(alpha = 0.7, outlier.shape = 21) +
  geom_jitter(width = 0.2, alpha = 0.3, size = 1) +
  labs(x = "", y = "物种丰富度", title = "不同生境的物种丰富度") +
  scale_fill_brewer(palette = "Set2") +
  theme_minimal() +
  theme(legend.position = "none")

19.3.4 直方图与密度图

p1 <- ggplot(ecology, aes(x = organic_carbon)) +
  geom_histogram(bins = 20, fill = "steelblue", alpha = 0.7, color = "white") +
  labs(x = "有机碳 (g/kg)", y = "频次", title = "直方图") +
  theme_minimal()

p2 <- ggplot(ecology, aes(x = organic_carbon, fill = habitat)) +
  geom_density(alpha = 0.5) +
  labs(x = "有机碳 (g/kg)", y = "密度", title = "密度图", fill = "生境") +
  theme_minimal()

p1 + p2

19.3.5 柱状图

ecology |>
  group_by(habitat) |>
  summarise(
    mean_richness = mean(species_richness),
    se = sd(species_richness) / sqrt(n())
  ) |>
  ggplot(aes(x = reorder(habitat, -mean_richness), y = mean_richness, fill = habitat)) +
  geom_col(alpha = 0.8) +
  geom_errorbar(aes(ymin = mean_richness - se, ymax = mean_richness + se),
                width = 0.2) +
  labs(x = "", y = "平均物种丰富度 (± SE)") +
  scale_fill_brewer(palette = "Set2") +
  theme_minimal() +
  theme(legend.position = "none")

19.3.6 热力图（相关性矩阵）

# 计算相关系数
cor_matrix <- ecology |>
  select(species_richness, soil_ph, organic_carbon, elevation) |>
  cor()

# 转为长格式绘图
cor_matrix |>
  as.data.frame() |>
  rownames_to_column("var1") |>
  pivot_longer(-var1, names_to = "var2", values_to = "cor") |>
  ggplot(aes(x = var1, y = var2, fill = cor)) +
  geom_tile() +
  geom_text(aes(label = round(cor, 2)), size = 4) +
  scale_fill_gradient2(low = "steelblue", mid = "white", high = "coral",
                       midpoint = 0, limits = c(-1, 1)) +
  labs(title = "变量相关性矩阵", x = "", y = "", fill = "相关系数") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

19.4 科研论文级图表

19.4.1 自定义主题

theme_paper <- function(base_size = 12) {
  theme_bw(base_size = base_size) +
  theme(
    panel.grid.minor = element_blank(),
    panel.grid.major = element_line(color = "grey92"),
    strip.background = element_rect(fill = "grey95", color = NA),
    legend.background = element_rect(fill = "white", color = NA),
    plot.title = element_text(face = "bold", size = base_size + 2),
    axis.title = element_text(size = base_size),
    axis.text = element_text(size = base_size - 1)
  )
}

19.4.2 多面板组合图

p_scatter <- ggplot(ecology, aes(x = elevation, y = species_richness)) +
  geom_point(aes(color = habitat), alpha = 0.6) +
  geom_smooth(method = "lm", color = "black", linewidth = 0.8) +
  labs(x = "海拔 (m)", y = "物种丰富度", color = "生境") +
  theme_paper()

p_box <- ggplot(ecology, aes(x = habitat, y = organic_carbon, fill = habitat)) +
  geom_boxplot(alpha = 0.7) +
  labs(x = "", y = "有机碳 (g/kg)") +
  scale_fill_brewer(palette = "Set2") +
  theme_paper() +
  theme(legend.position = "none")

# 组合并添加标签
p_scatter + p_box +
  plot_annotation(tag_levels = "a", tag_prefix = "(", tag_suffix = ")") +
  plot_layout(widths = c(1.5, 1))

19.4.3 导出高质量图片

# 将图表保存到变量中
p_final <- p_scatter + p_box +
  plot_annotation(tag_levels = "a", tag_prefix = "(", tag_suffix = ")")

# 保存为 PDF（矢量图，论文首选）
ggsave("output/figures/fig1_diversity.pdf", plot = p_final, width = 10, height = 5, dpi = 300)

# 保存为 PNG（位图，PPT/网页用）
ggsave("output/figures/fig1_diversity.png", plot = p_final, width = 10, height = 5, dpi = 300)

# 保存为 TIFF（某些期刊要求）
ggsave("output/figures/fig1_diversity.tiff", plot = p_final, width = 10, height = 5, dpi = 600,
       compression = "lzw")

Tip期刊图表要求

• 分辨率：通常 ≥ 300 dpi（线图 600 dpi）
• 格式：PDF/EPS（矢量）或 TIFF（位图）
• 字体大小：图中文字 ≥ 8pt
• 配色：考虑色盲友好（避免红绿搭配，推荐 viridis 色板）
• 图注：在图下方，不在图内

19.5 色盲友好配色：viridis

viridis 色板在亮度上均匀变化，对色盲用户友好，打印为灰度图时也能区分。ggplot2 内置了 viridis 系列函数：

# 离散变量：scale_color_viridis_d()
p1 <- ggplot(ecology, aes(x = soil_ph, y = species_richness, color = habitat)) +
  geom_point(alpha = 0.7, size = 2) +
  scale_color_viridis_d(option = "D") +
  labs(x = "土壤 pH", y = "物种丰富度", color = "生境", title = "viridis 离散配色") +
  theme_minimal()

# 连续变量：scale_color_viridis_c()
p2 <- ggplot(ecology, aes(x = soil_ph, y = species_richness, color = elevation)) +
  geom_point(alpha = 0.7, size = 2) +
  scale_color_viridis_c(option = "C") +
  labs(x = "土壤 pH", y = "物种丰富度", color = "海拔 (m)", title = "viridis 连续配色") +
  theme_minimal()

p1 + p2

Tipviridis 色板选项

option 参数可选 "A"（magma）、"B"（inferno）、"C"（plasma）、"D"（viridis，默认）、"E"（cividis）。科研论文推荐使用 "D" 或 "E"。

19.6 课后练习

1. 用 iris 数据集绘制花瓣长度 vs 花瓣宽度的散点图，按物种着色
2. 绘制三个物种花萼长度的箱线图，添加抖动点
3. 创建一个 2×2 的组合图（散点图 + 箱线图 + 直方图 + 密度图）
4. 应用自定义主题，导出为 PDF
5. 尝试使用 scale_color_viridis_d() 替换默认配色