Make a surface plot

I need to make a surface plot where I also need to colour the surface formed from the x,y and z coordinates based on a fourth variable, c.

Is there any way to do this.

I am attaching the excel file with the variables in them.

Answer 1

[Star Strider]

Creating surface plots from your data is straightforward using the scatteredInterpolant function. Once the ‘Cm’ matrix is created (and it has to be created using the same interpolation method that created ‘Zm’), the colour mapping to the surface works, and matches reasonably well with the initial scatter3+stem3 plot that I created to see how the final result should look (and to see if there was any pattern to the data that does not appear to have any patterns). The key to getting the colours to work was to use the interpolated colour matrix ‘Cm’ and clim.

Try this —

T1 = readtable('Surface plot data.xlsx')

T1 = 128x4 table
     x       y        z         c   
    ____    ____    ______    ______

    3500    1300       0.3    370.66
    6000    2200      0.45    316.32
    7250    1750     0.375    327.03
    4750    2650     0.525    305.79
    5375    1975    0.4875    323.89
    7875    2875    0.3375    295.35
    6625    1525    0.5625    329.26
    4125    2425    0.4125    321.77
    4437    1862    0.5812    324.85
    6937    2762    0.4312     299.4
    8187    1412    0.5062    330.26
    5687    2312    0.3562    318.33
    5062    1637    0.3937    342.51
    7562    2537    0.5437    296.44
    6312    2087    0.3187    321.77
    3812    2987    0.4687     306.8

figure
stem3(T1.x, T1.y, T1.z)
hold on
scatter3(T1.x, T1.y, T1.z, 75, T1.c, 'filled')
hold  off
cm = colormap(turbo(height(T1)));
CL1 = clim

CL1 = 1x2

285.5780 370.6595

<mw-icon class=""></mw-icon>

<mw-icon class=""></mw-icon>

colorbar
xlabel('X')
ylabel('Y')
zlabel('Z')
title('Scatter + Stem Plot')

xv = linspace(min(T1.x), max(T1.x), fix(size(T1,1)));
yv = linspace(min(T1.y), max(T1.y), fix(size(T1,1)));
[Xm,Ym] = ndgrid(xv, yv);                                               % Create Interpolation Matrices

Ci = scatteredInterpolant(T1.x, T1.y, T1.c, 'natural','none');          % Colour  Matrix
F = scatteredInterpolant(T1.x, T1.y, T1.z, 'natural','none');           % 'Z' Matrix
Cm = Ci(Xm,Ym);
Zm = F(Xm,Ym);

[c1,c2] = bounds(T1.c)

c1 = 285.5780

c2 = 370.6595

[Cm1,CM2] = bounds(Cm(:))

Cm1 = 285.5890

CM2 = 370.6595

figure
surfc(Xm, Ym, Zm, Cm)
clim(CL1)
colormap(cm)
colorbar
xlabel('X')
ylabel('Y')
zlabel('Z')
title('Surface — ‘natural’ Method, No Extrapolation')

Ci = scatteredInterpolant(T1.x, T1.y, T1.c, 'linear','none');
F = scatteredInterpolant(T1.x, T1.y, T1.z, 'linear','none');
Cm = Ci(Xm,Ym);
Zm = F(Xm,Ym);

figure
surfc(Xm, Ym, Zm, Cm)
clim(CL1)
colormap(cm)
colorbar
xlabel('X')
ylabel('Y')
zlabel('Z')
title('Surface — ‘linear’ Method, No Extrapolation')

Ci = scatteredInterpolant(T1.x, T1.y, T1.c, 'nearest','none');
F = scatteredInterpolant(T1.x, T1.y, T1.z, 'nearest','none');
Cm = Ci(Xm,Ym);
Zm = F(Xm,Ym);

figure
surfc(Xm, Ym, Zm, Cm)
clim(CL1)
colormap(cm)
colorbar
xlabel('X')
ylabel('Y')
zlabel('Z')
title('Surface — ‘nearest’ Method, No Extrapolation')

The same general approach is repeated with each interpolation method.

My code does not extrapolate. There is no harm in setting an extrapolation method — I am simply averse to extrapolating.

.

[Edin Michael]

Thanks a lot for the help.

[Star Strider]

As always, my pleasure!