如何向使用具有双 y 轴的 ggplot2 完成的绘图添加更多线条

Question

数据框df1 总结了一条鱼在一段时间内的平均每日深度（meanDepth），以及不同深度的每日平均水温（T5m、T15m、T25m 和 T35m）以及整个水柱的日平均温度（meanT）（不考虑不同深度）。举个例子：

df1<- data.frame(Date=c("2016-08-05","2016-08-06","2016-08-07","2016-08-08","2016-08-09","2016-08-10"),
                 meanDepth=c(15,22,18,25,27,21),
                 T5m=c(17,18,21,23,21,18),
                 T15m=c(16,17,18,19,18,17),
                 T25m=c(16,17,17,18,18,17),
                 T35m=c(15,16,17,17,17,16),
                 meanT=c(16,17.2,17.8,18.3,17.8,17.4))
df1$Date<-as.Date(df1$Date)

df1

        Date meanDepth T5m T15m T25m T35m meanT
1 2016-08-05        15  17   16   16   15  16.0
2 2016-08-06        22  18   17   17   16  17.2
3 2016-08-07        18  21   18   17   17  17.8
4 2016-08-08        25  23   19   18   17  18.3
5 2016-08-09        27  21   18   18   17  17.8
6 2016-08-10        21  18   17   17   16  17.4

我想在一张图中绘制鱼的深度剖面和不同深度的日平均温度。

到目前为止，我在一个 Y 轴上绘制 meanDepth，在另一个 Y 轴上绘制 meanT。但我不知道如何添加更多与右 y 轴（=温度）相关的线，这些线代表不同深度的日平均温度。这里有我迄今为止能够构建的代码。

p <- ggplot(df1, aes(x = Date))
p <- p + geom_line(aes(y = meanDepth, colour = "Overall daily mean depth"))
p <- p + geom_line(aes(y = meanT/(max(range(df1$meanT,na.rm=TRUE)/max(range(df1$meanDepth,na.rm=TRUE)))), colour = "Mean Water T"))
p <- p + scale_y_continuous(sec.axis = sec_axis(~.*(max(range(df1$meanT,na.rm=TRUE)/max(range(df1$meanDepth,na.rm=TRUE)))), name = "Mean daily Temp"))
p <- p + scale_colour_manual(values = c("blue", "red"))
p <- p + labs(title="Mean daily depth and water temperature through time",
              y = "Mean daily depth",
              x = "Date",
              colour = "Parameter")
p <- p + theme(legend.position = c(0.8, 0.9), plot.title = element_text(hjust=0.5, face="bold",margin = margin(0,0,12,0) ),axis.title.y =element_text(margin = margin(t = 0, r = 12, b = 0, l = 0)),axis.title.x =element_text(margin = margin(t = 12, r = 0, b = 0, l = 0)),axis.text.x=element_text(angle=60, hjust=1))
p <- p + scale_x_date(date_breaks = "1 days", labels = date_format("%Y-%m-%d"))
p

这就是我得到的情节：

有没有人如何添加与 5、15、25 和 35 米的温度有关的线？

Answer 1

辅助轴几乎从来都不是一个好主意，但你可以这样做：

library(ggplot2)
library(tidyr)
library(dplyr)
df1 %>% 
  mutate(meanDepthNormalized = meanDepth * max(meanT) / max(meanDepth)) %>% #1
  select(-meanDepth) %>% 
  # could have changed meanDepth before directly, but wanted to be more verbose
  gather(type, value, -Date) %>% #2
  ggplot(aes(x = Date, y = value, color = type, linetype = type == "meanT")) + #3
    geom_line(size = 1.5) +
    scale_y_continuous(sec.axis = sec_axis(~ . * max(df1$meanDepth) / max(df1$meanT))) +
    scale_color_manual("", values = c("#E41A1C", "#984EA3", "#BDD7E7",
                                      "#6BAED6", "#3182BD", "#08519C")) +
    theme_minimal() + 
    guides(linetype = FALSE) +
    theme(legend.position = "top")

说明

1. 您首先将 Depth 转换为与温度测量值相同的比例
2. 然后通过gather 将数据从宽格式转换为长格式
3. 然后，您可以将颜色映射到类型变量，该变量基本上包含以前的列名。

Answer 2

如果我正确理解了您的问题，则为每个温度/列添加 geom_line 就足够了。这是T5m 和T35m 的示例。

df1<- data.frame(Date=c("2016-08-05","2016-08-06","2016-08-07",
                        "2016-08-08","2016-08-09","2016-08-10"),
                 meanDepth=c(15,22,18,25,27,21),
                 T5m=c(17,18,21,23,21,18),
                 T15m=c(16,17,18,19,18,17),
                 T25m=c(16,17,17,18,18,17),
                 T35m=c(15,16,17,17,17,16),
                 meanT=c(16,17.2,17.8,18.3,17.8,17.4))
df1$Date<-as.Date(df1$Date)
norm <- max(df1$meanT,na.rm=TRUE)/max(df1$meanDepth,na.rm=TRUE)
p <- ggplot(df1, aes(x = Date)) +
  geom_line(aes(y = meanDepth, colour = "Overall daily mean depth")) +
  geom_line(aes(y = meanT/norm, colour = "Mean Water T")) +
  geom_line(aes(y = T5m/norm, colour = "T5m")) +
  geom_line(aes(y = T35m/norm, colour = "T35m")) +
  scale_x_date(date_breaks = "1 days", labels = date_format("%Y-%m-%d")) +
  scale_y_continuous(sec.axis = sec_axis(~.*norm, name = "Mean daily Temp")) +
  scale_colour_manual(values = c("blue", "red", "orange", "black")) +
  labs(title="Mean daily depth and water temperature through time",
       y = "Mean daily depth",
       x = "Date",
       colour = "Parameter") +
  theme(legend.position = c(0.8, 0.9),
        plot.title = element_text(hjust=0.5, face="bold", margin = margin(0,0,12,0)),
        axis.title.y = element_text(margin = margin(t = 0, r = 12, b = 0, l = 0)),
        axis.title.x = element_text(margin = margin(t = 12, r = 0, b = 0, l = 0)),
        axis.text.x=element_text(angle=60, hjust=1)) 
p

Answer 3

也许与 facet_grid(y ~., scales = "free")

这里是一个例子

ggplot(mtcars, aes(factor(cyl), mpg)) + 
   geom_boxplot() + 
   facet_grid(cyl ~., scales = "free")