Plot bathymetry and coastline using ggplot2 and ma

问题:

I looked around and I have not found a nice solution for my goal. I want to plot some data on a longitude/ latitude plot using ggplot2 and the coastline plus bathymetry with marmap, everything in one single plot.

This script is to plot mydata

ggplot(data = ctd, aes(x = Longitude, y = Latitude)) +
  geom_raster(aes(fill = Temp)) +
  scale_fill_gradientn(colours = rev(my_colours)) +
  geom_contour(aes(z = Temp), binwidth = 2, colour = "black", alpha = 0.2) +

  #plot stations locations
  geom_point(data = ctd, aes(x = Longitude, y = Latitude),
             colour = 'black', size = 3, alpha = 1, shape = 15) +

  #plot legends
      labs(y = "Latitude", x = "Longitude", fill = "Temp (°C)") +
      coord_cartesian(expand = 0)+
      ggtitle("Temperature distribution") 

Using marmap I download the bathymetry

library(marmap)
Bathy <- getNOAA.bathy(lon1 = 37, lon2 = 38.7,
                       lat1 = -45.5, lat2 = -47.3, resolution = 1)

The result I would like to obtain is the distribution of mydata on Lon/Lat with the land colored in black plus grey lines for the bathymetry.

回答1:

Here is an approach:

get bathy data:

library(marmap)
Bathy <- getNOAA.bathy(lon1 = 37, lon2 = 38.7,
                       lat1 = -45.5, lat2 = -47.3, resolution = 1)

convert it to matrix:

Bathy <- as.matrix(Bathy)
class(Bathy) <- "matrix"

now reshape it to long format and plot

library(tidyverse)

Bathy %>%
  as.data.frame() %>%
  rownames_to_column(var = "lon") %>%
  gather(lat, value, -1) %>%
  mutate_all(funs(as.numeric)) %>%
  ggplot()+
  geom_contour(aes(x = lon, y = lat, z = value), bins = 10, colour = "black") +
  coord_map()

回答2:

Well, there's a marmap function for that. It's called autoplot.bathy(). Have you checked it's help file?

library(marmap) ; library(ggplot2)

library(marmap)
Bathy <- getNOAA.bathy(lon1 = 37, lon2 = 38.7,
                       lat1 = -45.5, lat2 = -47.3, resolution = 1)

ctd <- data.frame(Longitude = c(37.5, 38, 38.5), Latitude = c(-47, -46.5, -46))

autoplot.bathy(Bathy, geom=c("tile","contour")) +
    scale_fill_gradient2(low="dodgerblue4", mid="gainsboro", high="darkgreen") +
    geom_point(data = ctd, aes(x = Longitude, y = Latitude),
               colour = 'black', size = 3, alpha = 1, shape = 15) +
    labs(y = "Latitude", x = "Longitude", fill = "Elevation") +
    coord_cartesian(expand = 0)+
    ggtitle("A marmap map with ggplot2") 

Or, with base graphics (and the correct aspect ratio):

# Creating color palettes
blues <- c("lightsteelblue4", "lightsteelblue3", "lightsteelblue2", "lightsteelblue1")
greys <- c(grey(0.6), grey(0.93), grey(0.99))

# Plot
plot(Bathy, image = TRUE, land = TRUE, n=30, lwd = 0.1, bpal = list(c(0, max(Bathy), greys), c(min(Bathy), 0, blues)), drawlabels = TRUE)

# Add coastline
plot(Bathy, deep = 0, shallow = 0, step = 0, lwd=2, add = TRUE)

# Add stations
points(ctd, pch=15, cex=1.5)