3

我想在 ggplot 中绘制一些地图数据，其中根据连续变量的值填充多边形。我遇到的一个问题是，数据中只有少数极端异常值（约 3000 个中的 10 个）阻止了大多数其他观察结果以足够有意义的颜色变化显示。

基本上我希望能够为我的大部分数据指定配色方案，然后将所有正异常值都涂成亮蓝色，将所有负异常值涂成亮红色。

示例代码 - 举例来说，假设大多数数据介于 -0.01 和 0.01 之间，但有一些值低于/高于这些水平......

library(ggplot2)

DAT <- structure(list(long = c(848025.138769486, 827715.400155344, 819783.06692123, 
    811273.46685221, 807112.24435793, 825218.389316822, 824653.263676394, 
    851268.022958262, 854876.20206792, 848025.138769486, 824710.033713034, 
    808060.98832434, 791388.306466439, 791011.02400712, 790256.857765555, 
    790077.133379578, 792915.8857683, 803799.11479841, 812042.67207171, 
    811086.452951819, 818801.811870496, 824623.483334553, 829923.458416495, 
    844732.373710393, 843509.962977133, 840793.468791028, 844661.128104187, 
    841054.158477243, 824318.536510595, 824710.033713034, 957744.153625377, 
    964378.49410997, 963723.372709918, 963536.949007208, 950656.36657876, 
    923479.701493794, 918733.432704199, 950416.579446804, 951806.914646236, 
    952020.610154412, 957744.153625377, 838474.166485858, 839637.34900613, 
    846161.553188089, 854105.657535852, 850749.837928768, 851268.022958262, 
    824653.263676394, 825218.389316822, 807112.24435793), lat = c(-755063.209671518, 
    -751075.227428769, -758486.024742793, -742209.818502709, -725584.096412241, 
    -723927.958082718, -719335.285579824, -716264.904081879, -747370.371610989, 
    -755063.209671518, -607299.311749675, -601135.889250199, -605563.568360578, 
    -602482.006717006, -602560.163831817, -601169.088819494, -595490.118653201, 
    -589823.673980013, -577642.515107293, -574207.831414412, -570499.93978401, 
    -555676.769699399, -554574.027849141, -569849.400696679, -581914.546950335, 
    -587110.213020993, -586640.550459734, -595326.416960721, -603469.770252399, 
    -607299.311749675, -686468.891810573, -696059.793425604, -696994.182842627, 
    -698894.537935268, -702169.738050773, -705710.496557967, -666147.504830981, 
    -662203.856856129, -666852.009638632, -667571.668902733, -686468.891810573, 
    -684524.444323129, -694047.01193517, -705551.617269712, -711060.296586236, 
    -711747.897688833, -716264.904081879, -719335.285579824, -723927.958082718, 
    -725584.096412241), group = c("0.1", "0.1", "0.1", "0.1", "0.1", 
    "0.1", "0.1", "0.1", "0.1", "0.1", "1.1", "1.1", "1.1", "1.1", 
    "1.1", "1.1", "1.1", "1.1", "1.1", "1.1", "1.1", "1.1", "1.1", 
    "1.1", "1.1", "1.1", "1.1", "1.1", "1.1", "1.1", "2.1", "2.1", 
    "2.1", "2.1", "2.1", "2.1", "2.1", "2.1", "2.1", "2.1", "2.1", 
    "3.1", "3.1", "3.1", "3.1", "3.1", "3.1", "3.1", "3.1", "3.1"
    ), VALUE = c(0.0197178619295337, 0.0197178619295337, 0.0197178619295337, 
    0.0197178619295337, 0.0197178619295337, 0.0197178619295337, 0.0197178619295337, 
    0.0197178619295337, 0.0197178619295337, 0.0197178619295337, 0.0144402919365254, 
    0.0144402919365254, 0.0144402919365254, 0.0144402919365254, 0.0144402919365254, 
    0.0144402919365254, 0.0144402919365254, 0.0144402919365254, 0.0144402919365254, 
    0.0144402919365254, 0.0144402919365254, 0.0144402919365254, 0.0144402919365254, 
    0.0144402919365254, 0.0144402919365254, 0.0144402919365254, 0.0144402919365254, 
    0.0144402919365254, 0.0144402919365254, 0.0144402919365254, -0.00812118892018265, 
    -0.00812118892018265, -0.00812118892018265, -0.00812118892018265, 
    -0.00812118892018265, -0.00812118892018265, -0.00812118892018265, 
    -0.00812118892018265, -0.00812118892018265, -0.00812118892018265, 
    -0.00812118892018265, 0.00751936235807205, 0.00751936235807205, 
    0.00751936235807205, 0.00751936235807205, 0.00751936235807205, 
    0.00751936235807205, 0.00751936235807205, 0.00751936235807205, 
    0.00751936235807205)), .Names = c("long", "lat", "group", "VALUE"
    ), row.names = c(6L, 5L, 4L, 10L, 7L, 8L, 9L, 2L, 3L, 1L, 151L, 
    150L, 156L, 162L, 168L, 163L, 159L, 154L, 160L, 149L, 161L, 158L, 
    164L, 152L, 153L, 155L, 165L, 167L, 166L, 157L, 226L, 223L, 233L, 
    232L, 225L, 227L, 224L, 230L, 228L, 229L, 231L, 258L, 253L, 257L, 
    262L, 254L, 256L, 261L, 260L, 269L), class = "data.frame")

#Construct plot, placing limits on color scale. How to make outliers bright red or blue?
gmp <- ggplot(data=DAT, aes(x=long,y=lat,group=group)) +
scale_fill_gradient2(low=rgb(0.8,0.2,0.2),high=rgb(0.2,0.3,0.8),mid=rgb(0.9,0.9,0.9),limits=c(-0.01,0.01)) + 
geom_polygon(aes(fill=VALUE,group=group),colour=NA, size = .3) + 
coord_fixed()
plot(gmp)
4

1 回答 1

4

这是一个选项：

为每个标准（最小极端、最大极端和中间范围）创建子集。为了说明，我在您的数据中添加了一个最低限度的极端值。

DAT$VALUE[DAT$group == 0.1] <- -0.019
DAT.mid <- DAT[abs(DAT$VALUE) < 0.01,]
DAT.max <- DAT[DAT$VALUE >= 0.01,]
DAT.min <- DAT[DAT$VALUE <= -0.01,]

然后用中间数据构建一个基础图：

base <- ggplot(DAT.mid, aes(x=long,y=lat,group=group)) +
  geom_polygon(aes(fill=VALUE,group=group),colour=NA, size = .3) + 
  scale_fill_gradient2(low=rgb(0.8,0.2,0.2),high=rgb(0.2,0.3,0.8),
  mid=rgb(0.9,0.9,0.9),limits=c(-0.01,0.01)) + 
  coord_fixed()
base

在此处输入图像描述

然后你可以用你想要的任何颜色添加极端的图层：

ext <- base +
  geom_polygon(data = DAT.max, fill = "blue") +
  geom_polygon(data = DAT.min, fill = "red") 
ext

在此处输入图像描述

棘手的一点是要获得一个代表您的极端情况的图例。从这里开始，您可以添加映射到 non-existent 的不可见 geom aes，然后 ab/使用图例来获得您想要的结果。在这里，我已经习惯了geom_point。第一个，我将大小映射到“> = 0.01”，这将为我们提供一个带有该文本的图例，第二个类似，除了我使用 alpha 来获取另一个图例条目。然后，您可以使用guides()覆盖图例属性，并获得一个具有正确填充的正方形。它并不完美，但在大多数情况下都有效。

gmp <- ext +
  geom_point(data = DAT.max, aes(size = ">= 0.01", shape = NA), colour = "blue") +
  geom_point(data = DAT.min, aes(alpha = "<= -0.01", shape = NA), colour = "red") +
  guides(size = guide_legend(title = "", override.aes = list(shape = 15, size = 10)),
         alpha = guide_legend(title = "", override.aes = list(shape = 15, size = 10, alpha = 1)))
gmp

在此处输入图像描述

于 2014-03-30T23:11:28.287 回答