问题
I am trying to generate the map of U.S. in which each state can have one of the following colors:
EScolors <- c("#7aad42","#4a77bb","#f7931e","#d3dfbd","#787878")
I have created a data frame, states_info, to match each state with it's color.
head(states_info)
State.Code region St_Abbr Num_Estab colors
1 1 alabama AL 13123 #f7931e
3 4 arizona AZ 18053 #f7931e
4 5 arkansas AR 9154 #4a77bb
5 6 california CA 143937 #787878
6 8 colorado CO 21033 #d3dfbd
7 9 connecticut CT 17176 #f7931e
I have tried various ways to get the colors for each state correct, but my code is not working. (btw, "colors" is a factor variable and contain the hex value of specific colors)
Approach 1:
map('state',fill=TRUE,col=states_info$colors)
I get a map, but the colors for the states are not correct. This approach probably requires matching, but I cannot figure it out.
Approach 2: I create a data frame by merging the latitudes and longitudes for each state with my state_info dataframe to draw the map
states_location <- map_data("state")
map.df <- merge(states_location,states_info, by=intersect(states_location$region, states_info$region), all=TRUE)
map.df <- map.df[order(map.df$order),]
ggplot(map.df, aes(x=long,y=lat,group=group))+
geom_polygon(aes(fill=region.x))+
geom_path()+
scale_color_hue(states_info$colors)
This approach generates a map using its own color gradient and not the colors I specified. What am I doing wrong? Thank you.
回答1:
Let ggplot2 do the hard work for you:
library(ggplot2)
read.table(text="State.Code region St_Abbr Num_Estab colors
1 1 alabama AL 13123 #f7931e
3 4 arizona AZ 18053 #f7931e
4 5 arkansas AR 9154 #4a77bb
5 6 california CA 143937 #787878
6 8 colorado CO 21033 #d3dfbd
7 9 connecticut CT 17176 #f7931e",
stringsAsFactors=FALSE, header=TRUE, comment.char="") -> df
usa_map <- map_data("state")
gg <- ggplot()
gg <- gg + geom_map(data=usa_map, map=usa_map,
aes(long, lat, map_id=region),
color="#2b2b2b", size=0.15, fill=NA)
gg <- gg + geom_map(data=df, map=usa_map,
aes(fill=colors, map_id=region),
color="#2b2b2b", size=0.15)
gg <- gg + scale_color_identity()
gg <- gg + coord_map("polyconic")
gg <- gg + ggthemes::theme_map()
gg
回答2:
Using the base maps package can be a little tricky. The order and the naming of the states is not standard, several states have more than one region, i.e. Manhattan island in New York. A little manipulation is required to properly label/color the map.
In this solution, I created a dataframe statelist to hold the state names, islands and an index. Then merge this with your dataframe, state_info, and then plotted.
#function to split strings an return a dataframe
strtodf<-function (list){
slist<-strsplit(list, ":")
x<-sapply(slist, FUN= function(x) {x[1]})
y<-sapply(slist, FUN= function(x) {x[2]})
df<-data.frame(state=x, island=y, stringsAsFactors = FALSE)
return(df)
}
#user defined coloring scheme
# Example data for to test solution
colors<- c("#7aad42","#4a77bb","#f7931e","#d3dfbd","#787878")
region<-c("washington", "new york", "virginia", "pennsylvania", "ohio")
states_info<-data.frame(region, colors)
# End of example data
#dataframe to hold state names for mapping purposes
library(maps)
maplist<-map("state", namesonly = TRUE, plot=FALSE)
statelist<-strtodf(maplist) #convert to dataframe
statelist$row<-as.numeric(rownames(statelist)) #index column
#merge the data from and resort into proper order
statelist<-merge(statelist, states_info, by.x = "state", by.y="region", sort=FALSE, all.x=TRUE)
statelist<-statelist[order(statelist$row),]
#plot the map
maplist<-map("state", fill=TRUE, col=statelist$colors)
来源:https://stackoverflow.com/questions/38777732/r-code-to-generating-map-of-us-states-with-specific-colors