[R] minimum distances

[Ben Bolker]

Dave Depew <ddepew <at> sciborg.uwaterloo.ca> writes:

> 
>  Hi,
> I have a question about computing shortest Euclidean distances between 
> two data frames of spatial points...
> 
> I have 2 dataframes (not yet converted to spatial class)
> 
>  >Sewers<-data.frame(x=c(10,20,35,50),y=c(100,150,200,300))
>  >transect <- data.frame(x=seq(from=0, to=50, by=1),y=seq(from=100, 
> to=150, by=1))
> 
> I would like to be able to compute the shortest distance from the 
> transect points the nearest sewer (Euclidean distance)
> 
> I've tried a number of different loops, but so far ave had no luck.
> 
> Any help would be appreciated
> 


Sewers<-data.frame(x=c(10,20,35,50),y=c(100,150,200,300))
transect <- data.frame(x=0:50,y=100:150)


dist <- sqrt((outer(Sewers$x,transect$x,"-"))^2+
             (outer(Sewers$y,transect$y,"-"))^2)
mindist = apply(dist,2,min)
closest = apply(dist,2,which.min)

library(MASS)
eqscplot(Sewers,pch=16,col=1:4)
points(transect,col=closest)