Problem with non-Zero Matrix Appearing Zero in Command Window and in Imagesc

I am generating a matrix "allpowpf" which shows the power generated as a function of wave slowness and frequency. When I double click on the matrix under the variable listing, it shows a matrix filled with numbers on the magnitude of e10 (screenshot attached). However, when I type "powpf" into the command window, it shows a matrix filled with zeros (screenshot attached). Most importantly, when I plot allpowpf using imagesc, I get an image of an array of zeros instead of the proper numbers.

I generate the zero matrix allpowpf using:

allpowpf = zeros(np,nf/2);

Then, I fill allpowpf using the following:

Y = fft(taup(k,:));

for j = 1:nf/2 % samples in frequency domain, stop at Nyquist

allpowpf(k,j) = allpowpf(k,j) + real(Y(j)^2);

end

And finally, I plot allpowpf using:

imagesc(allpowpf) % plot of energy slowness and frequency

xlabel('frequency (Hz)')

ylabel('slowness (ms/m)')

I believe that my loops are working correctly because the variable allpowpf is correct, but I do not understand why it plots as a zero matrix or why it appears in the command window as a zero matrix when it is not a zero matrix as a stored variable.

Answer 1

[John D'Errico]

They are still (non-zero) numbers. Just too small to display in the chosen format. If you close your eyes, the world does not disappear. ;-)

X = [123456789012345 0.00000123456789]

X = 1×2

1.0e+14 * 1.2346 0.0000

But that second element is still there. Not zero.

X(2)

ans = 1.2346e-06

You can generally force MATLAB to show the true values using a scientific display format.

format short g
X

X = 1×2

1.0e+00 * 1.2346e+14 1.2346e-06

[Rose]

Thank you! Do you have advice for the imagesc plot? Right now, it only plots one solid color at 0, which doesn't help me much. I need to see which frequencies and slownesses have the most power across my seismic array.

[Steven Lord]

What's the range of data stored in your matrix? What do you see when you run these commands?

format longg
[minvalue, maxvalue] = bounds(allpowpf, "all")
therange = maxvalue - minvalue