r - 使用极坐标（epsg：3995）投影和ggspatial时如何在ggplot中的第66个平行北方画一条线

Question

我正在尝试在北纬 66 度线上画一个蓝色圆圈。在我正在制作的地图上。

我通常是fortify()一个SpatialPolygonsDataFrame，然后使用组合geom_polygon()，coord_map("ortho" , ylim = c(50, 90))并geom_hline()实现这一点。

但是，对于一个新项目，我将不得不ggspatial::layer_spatial()在一个已经有 CRS 日期的对象上使用，所以这种直线绘制和自动重新投影不起作用。

这是我常用方法的演示，该方法有效：

library(tidyverse)

map_data <- rnaturalearth::ne_countries()
map_data_df <- fortify(map_data)

ggplot() + 
  geom_polygon(data = map_data_df, 
               aes(x = long,
                   y = lat,
                   group = group),
               col = "black",
               fill = "lightgrey") + 
  coord_map("ortho" , ylim = c(50, 90))  +
  geom_hline(yintercept = 66, col = "blue")

我试图在这段代码中复制它，但这显然不起作用，因为geom_hline()没有正确的投影：

map_data_transformed <- sp::spTransform(map_data, sp::CRS("+init=epsg:3995"))
# needs some fixing to close all the polygons
map_data_transformed <- rgeos::gBuffer(map_data_transformed, byid=TRUE, width=0)


ggplot() + 
  ggspatial::layer_spatial(data = map_data_transformed, 
                           fill = "lightgrey", col = "black", size = .2)+
  coord_sf( xlim = c(-4500000 , 4500000),
            ylim = c(-4500000 , 4500000)) +
  geom_hline(yintercept = 66, col = "blue")

Answer 1

看看这对你有用吗？

# define parallel north line as a spatial polygon with latitude / longitude projection
new_line <- sp::Polygon(coords = matrix(c(seq(-180, 180),
                                          rep(66, times = 361)), # change this for other
                                                                 # intercept values
                                        ncol = 2))
new_line <- sp::SpatialPolygons(Srl = list(sp::Polygons(srl = list(new_line), 
                                                        ID = "new.line")),
                                proj4string = sp::CRS("+proj=longlat +datum=WGS84"))

# change projection to match that of data
new_line <- sp::spTransform(new_line, sp::CRS("+init=epsg:3995"))

# plot
ggplot() + 
  ggspatial::layer_spatial(data = map_data_transformed, 
                           fill = "lightgrey", col = "black", size = .2) +
  ggspatial::layer_spatial(data = new_line,
                           fill = NA, col = "blue") +
  coord_sf( xlim = c(-4500000 , 4500000),
            ylim = c(-4500000 , 4500000))