Unable to plot 3D graph

I am trying to plot velocity data contained within a 2D plane (x-y). I would like to plot a 3D graph to present this. However, when trying to plot, I get an error stating

'Error using surf

Z must be a matrix, not a scalar or vector.'

The data is imported from an excel file and the appropriate columns are turned into arrays of x, y and U as shown below:

tbl = readtable('2D1mat.xlsx');

x=tbl.Var1;

y=tbl.Var2;

U=tbl.Var3;

surf(x,y,U)

I have also ensured that the input arguments for surf are all matrices with the help of ismatrix.

Answer 1

[Torsten]

Doesn't look very impressive, does it ?

A = readmatrix("https://de.mathworks.com/matlabcentral/answers/uploaded_files/1184273/2D1mat.xlsx");
x = A(:,1);
y = A(:,2);
u = A(:,3);
F = scatteredInterpolant(x,y,u);
xq = 33.5:1:53.5;
yq = 45:0.25:48.25;
[XQ,YQ] = meshgrid(xq,yq);
UQ = F(XQ,YQ);
surf(XQ,YQ,UQ)

[vauntedmango]

Not impressive at all, but atleast I know what to work on now :D

Thank you very much for the solution!

[Torsten]

Or the adapted solution from Star Strider:

A = readmatrix("https://de.mathworks.com/matlabcentral/answers/uploaded_files/1184273/2D1mat.xlsx");
x = A(:,1);
y = A(:,2);
z = A(:,3);
N = 100;
xv = linspace(min(x), max(x), N);
yv = linspace(min(y), max(y), N);
[X,Y] = ndgrid(xv, yv);
Z = griddata(x, y, z, X, Y, 'linear');
figure
surfc(X, Y, Z)
grid on
colormap(turbo)
colorbar
xlabel('X')
ylabel('Y')
zlabel('Z')

Answer 2

[Star Strider]

The ismatrix function wil return true for a vector and for a 2D matrix. It returns false for arrays with more than 2 dimensions.

The vectors may actually represent gridded data and so would form matrices when the vectors are appropriately reshaped.

An approach that may work (since we do not have the vectors to work with) —

x = rand(1,10);
y = rand(1,10);
z = rand(1,10);

N = 25;
xv = linspace(min(x), max(x), N);
yv = linspace(min(y), max(y), N);

[X,Y] = ndgrid(xv, yv);
Z = griddata(x, y, z, X, Y, 'linear');

figure
surfc(X, Y, Z)
grid on
colormap(turbo)
colorbar
xlabel('X')
ylabel('Y')
zlabel('Z')

.

[vauntedmango]

Thank you for your response. However, I am still struggling to present this data. Just to clarify, the dataset being imported consists of experimental measurements of mean velocity in a 2D plane. This was done by following a grid pattern (this means there are repeating values in x and in y when looking at the columns individually). I have attached the Excel file containing the data. Column 1, 2 and 3 correspond to X, Y and mean velocity respectively. I would much appreciate it if you can go through that too

[Star Strider]

There is actually no need to interpolate this, since the data are gridded, as the stem3 plot demonstrates.

For some reason, ‘Var2’ does not work with reshape as it should, despite a fair amount of effort to get it to. I finally gave up on that and went with repmat. Anyway, if all goes well, all the matrices are (14x21) and the surface plots correctly.

I checkked the surf plot against the stem3 plot, and they essentially agree. (The log transformation is necessary to get the stem3 plot to show any variation in Z, since the magnitudes of the third column don’t vary much. Without it, the stem3 plot hides what little variation there is.)

T1 = readtable('https://www.mathworks.com/matlabcentral/answers/uploaded_files/1184273/2D1mat.xlsx')

T1 = 294×3 table
    Var1    Var2      Var3 
    ____    _____    ______

    33.5       45    2.2377
    33.5    45.25    2.2382
    33.5     45.5    2.2408
    33.5    45.75     2.239
    33.5       46    2.2403
    33.5    46.25    2.2397
    33.5     46.5    2.2407
    33.5    46.75     2.239
    33.5       47    2.2378
    33.5    47.25    2.2401
    33.5     47.5    2.2343
    33.5    47.75    2.2339
    33.5       48    2.2378
    33.5    48.25    2.2332
    34.5    48.25    2.2385
    34.5       48    2.2376

figure
stem3(T1{:,1}, T1{:,2}, T1{:,3}, '.')
hold on
scatter3(T1{:,1}, T1{:,2}, T1{:,3}, 25, T1{:,3}, 'filled')
hold off
grid on
colormap(turbo)
xlabel('X')
ylabel('Y')
zlabel('Z')
set(gca, 'ZScale','log')

% view(0,90)

[U1,ix1] = unique(T1{:,1});
U1

U1 = 21×1

33.5000 34.5000 35.5000 36.5000 37.5000 38.5000 39.5000 40.5000 41.5000 42.5000

rowlen = numel(U1)

rowlen = 21

[U2,ix2] = unique(T1{:,2});
U2

U2 = 14×1

45.0000 45.2500 45.5000 45.7500 46.0000 46.2500 46.5000 46.7500 47.0000 47.2500

collen = numel(U2)

collen = 14

% X = reshape(T1{:,1}, rowlen, collen)
% Y = reshape(T1{:,2}, rowlen, collen)

X = repmat(U1,1,collen).';
Y = repmat(U2,1,rowlen);

Z = reshape(T1{:,3}, collen, rowlen);

figure
surfc(X, Y, Z)
grid on
colormap(turbo)
xlabel('X')
ylabel('Y')
zlabel('Z')
set(gca, 'ZScale','log')

% shading('interp')

.