Density plots/ kernel densities

[CKres]

0

I am trying to create a map from latitude and longitudes of catch locations. I was able to plot this on the map, however I would like to make densities of these locations. Therefore I have a few questions... (Image is in link below)
[1]: https://i.stack.imgur.com/SvqNH.jpg

1. Would that require me to zoom in on only the Atlantic ocean?
2. Should the points be smaller?
3. How the heck can I map kernel densities or densities of the most used locations of these coordinates?

library(maps)
library(ggplot2)
library(stringr)
world_map <- map_data("world")
#Creat a base plot with gpplot2
p <- ggplot() + coord_fixed() +
  xlab("") + ylab("")

#Add map to base plot
base_world_messy <- p + geom_polygon(data=world_map, aes(x=long, y=lat, group=group),
                                     colour="black", fill="light green") + 
                                      ggtitle("striper reports")

base_world_messy

#Strip the map down so it looks clean 
cleanup <- 
  theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), 
        panel.background = element_rect(fill = 'blue', colour = 'blue'), 
        axis.line = element_line(colour = "white"), legend.position="none",
        axis.ticks=element_blank(), axis.text.x=element_blank(),
        axis.text.y=element_blank())

base_world <- base_world_messy + cleanup

base_world


map_data <- 
  base_world +
  geom_point(data=striper, 
             aes(x=Long, y=Lat), colour="Red", 
             fill="Red",pch=1, size=.2, alpha=I(0.5)) + ggtitle("striper report")
map_data

[nirgrahamuk]

try

map_data <- 
  base_world +
  geom_density2d(data=striper, 
             aes(x=Long, y=Lat), colour="Red", 
             ) + ggtitle("striper report")

instead of your geom_points

[CKres]

Hello,
it yielded a blank map without any data points. Is this because I replace the geom_points with the geom_density2d?

[nirgrahamuk]

I took your code. made random striper data. and changed geom_points to geom_density2d
the result on my machine is:

library(maps)
library(ggplot2)
library(stringr)
world_map <- map_data("world")
#Creat a base plot with gpplot2
p <- ggplot() + coord_fixed() +
  xlab("") + ylab("")

#Add map to base plot
base_world_messy <- p + geom_polygon(data=world_map, aes(x=long, y=lat, group=group),
                                     colour="black", fill="light green") + 
  ggtitle("striper reports")

base_world_messy

#Strip the map down so it looks clean 
cleanup <- 
  theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), 
        panel.background = element_rect(fill = 'blue', colour = 'blue'), 
        axis.line = element_line(colour = "white"), legend.position="none",
        axis.ticks=element_blank(), axis.text.x=element_blank(),
        axis.text.y=element_blank())

base_world <- base_world_messy + cleanup

base_world

set.seed(42)

striper <- data.frame(Long=rnorm(1000,-60,10),
                      Lat = runif(100,19,71.4))

map_data <- 
  base_world +
  geom_density2d(data=striper, 
                 aes(x=Long, y=Lat), colour="Red", 
  ) + ggtitle("striper report")
map_data

[CKres]

Thank you for your time, I believe it is from my initial csv file.
I have two columns :
Lat Long
78.945 45.656
.... .......
...... .....

This extends for a few thousand, is there a more compatible format I should use?

[nirgrahamuk]

@CKres

I have two columns :
Lat Long
78.945 45.656

that doesnt seem different to what I had assumed when I made an artificial striper....
what output do you get when you ignore the map part and simply plot points and density of your striper ?

m <- ggplot(striper, aes(x = Long, y = Lat)) +
    geom_point()

# contour lines
m + geom_density_2d()

[CKres]

I receive an output of a plot with lat and long on each axis. It appears fine in a plot format.

[nirgrahamuk]

and can you share the visual output of ?

 base_world +
  geom_density2d(data=striper, 
                 aes(x=Long, y=Lat), colour="Red"  )

[CKres]

@nirgrahamuk

base_world +
  geom_density2d(data=striper, 
                 aes(x=Long, y=Lat), colour="Red"  )

I think it worked! The density is just very contained...

[nirgrahamuk]

great, perhaps play around with bins param

base_world +
  geom_density2d(data=striper, 
                 aes(x=Long, y=Lat), colour="Red" ,
                 bins=5 )

base_world +
  geom_density2d(data=striper, 
                 aes(x=Long, y=Lat), colour="Red" ,
                 bins=50 )

And as you say, you might consider zooming the map

[CKres]

@nirgrahamuk

base_world +
  geom_density2d(data=striper, 
                 aes(x=Long, y=Lat), colour="Red" ,
                 bins=5 )

Thank you for all your help!!! I will adjust the map.