3D surface plot in R, given x,y,z coordinates

Question

I have the following data-set, and need to plot a surface based on this set of data (of 60 3D points). Here, X, Y is the horizontal plane coordinates, and Z is the vertical / height coordinate.

p = read.csv("points.csv")
   PTS        X       Y        Z
1  101 481897.9 5456408 94.18695
2  102 481888.8 5456417 94.30702
3  103 481877.0 5456410 94.29034
4  104 481879.9 5456425 94.25546
5  105 481872.7 5456424 94.09370

After looking through several posts and trying to use functions in several libraries, I still cannot figure out a way to properly plot the surface. I've tried the following:

library(plotly)
plot_ly( y= Y, x = X, z = Z, data=p, type = "surface") #returns empty graphic frame

PX = data.matrix(p$X)
PY = data.matrix(p$Y)
PZ = data.matrix(p$Z)

library(plot3D)
surf3D(PX, PY, PZ)
#returns: Error in if (is.na(var)) ispresent <- FALSE else if (length(var) == 1) if (is.logical(var)) if (!var) ispresent <- FALSE : 
  argument is of length zero
library(lattice)
wireframe(p$Z ~ p$X*p$Y, data = p) #returns just a cube
library(rgl) 
surface3d(p$X,p$Y,p$Z)
#returns: Error in rgl.surface(x = c(481897.916, 481888.8482, 481876.9524, 481879.9393,  : y' length != 'x' rows * 'z' cols; 
#although there are 60 data points in the form (X,Y,Z) in the data set, with no points missing any coordinate

I must have been doing something horribly wrong here. Would anyone mind to point out what the mistake is?

Answer 1

You cannot make a 3D surface plot with this data because to do it you have to have Z value for each (X,Y) couple, like this :

   X1  X2  X3  ... Xn
Y1 Z11 Z12 Z13 ... Z1n
Y2 Z21 Z22 Z23 ... Z2n
Y3 Z31 Z32 Z33 ... Z3n
.                   .
.                   .
.                   .
Ym Zm1 Zm2 Zm3 ... Zmn

For example you don't have Z value for (481897.9,5456417) couple.

So, all you can do is a scatter3d plot :

plot_ly(data = p,x = X,y = Y, z = Z,type = "scatter3d",showlegend = FALSE)