How can I use pcolor for rings ?

Hello,

I am trying to highlight a correlation between the position of earthquakes and some parameters such as depth or magnitude... I have already divided the map according to a grid. Then I get this correlation for each small square of the grid. </matlabcentral/answers/uploaded_files/29722/Capture%20d%E2%80%99%C3%A9cran%202015-04-24%20%C3%A0%2023.45.01.png>

Up to now, I succeeded to correlate the positions to the parameters, be it in cartesian or in polar coordinates.

Yet, I would like to correlate the mean depth of earthquakes to their positions according to the RADIUS (and only the radius, NOT the angle).

In other words, I would like to work with rings. (like the following figure)

Thank you very much for your help.

Answer 1

[Star Strider]

This looks more like an archery target than an earth section, but it at least seems to give the appearance you want. Experiment with the colours to get the result you want.

The code uses the cylinder object to create a cone, then uses the colormap to colour it by height. There might be better ways to code it, but no one has come up with one yet, so I offer mine:

c = colormap(jet(4));
[X,Y,Z] = cylinder([1:-0.25:0], 256);

figure(1)
hs = surf(X, Y, Z)
axis square
set(hs, 'EdgeColor','none')
view([0  90])

The plot:

[Star Strider]

JLM’s‘Answer’ moved here...

Thank you very much for your answer. It must be a good alternative to solve the problem.

Though, I do not understand how I can introduce the depth (or an other parameter) in such a code since :

cylinder(r,n) returns the x-, y-, and z-coordinates of a cylinder based on the profile curve defined by vector r (n equally spaced points around its circumference).

I mean, I have to calculate the average depth of all the earthquakes located in every section of the cylinder. For all the previous plots, I did it by loops. And for every average I get, I had a special color which enabled me to say if there was a correlation or not thanks to the colorbar.

Thank you in advance.

JLM

[JLM]

x0=-110.30;y0=44.30;      %New origin

%B2 is for Longitudes
%B1 is for Latitudes
%B3 is for Depth

% Polar coordinates 

for i=1:size(B2)
    R(i,1)=sqrt((B1(i)-y0)^2+(B2(i)-x0)^2); %rho
    T(i,1)=atan2((B1(i)-y0),(B2(i)-x0));    %theta

figure(1)

% Grid

n = 8;
r = (0:n)'/n;
theta = pi*(-n:n)/n;

% Average Depth

Pas=1/n;      %Pas=Step

for i = 1:length(r)
    for j = 1:length(theta)
        hypocentres = R <= r(i)+Pas./2 & R >= r(i)-Pas./2 & T <= theta(j)+Pas./2 & T >= theta(j)-Pas./2;
        Positions = find(hypocentres);
        ProfMoy(i,j) = mean(ProfSeismes(Positions));
    end
end

X = r*cos(theta)+x0;
Y = r*sin(theta)+y0;
C = ProfMoy;
pcolor(X,Y,C)
axis equal tight 
xlabel('Longitude')
ylabel('Latitude')
colorbar
colormap(jet)
hold on

[Star Strider]

‘I would like to correlate the mean depth of earthquakes to their positions according to the RADIUS (and only the radius, NOT the angle).’

You only need a 2D plot for that, not 3D. Consider something like this:

plot(radius, depth)

[JLM]

Yes, sure, there are many ways to highlight this correlation but I was asked to do it on the map, in a visual way. Thank you, Star Strider, for your answer.

[Star Strider]

I’ve been giving some thought to this. What sort of profile do you want to plot? If you have that profile function, you can probably plot it using (X,Y) matrices created by meshgrid, then depending on how you want to plot it, use cart2pol or pol2cart. Since I still don’t understand how you want to map a 2D relation to a 3D relation, I can’t be more specific.

Answer 2

[Ahmet Cecen]

I am not sure what the question here is. If all you need to check is whether there is a correlation between mean depth and radius, just convert your Cartesian coordinates to polar using:

 [THETA,RHO] = cart2pol(X,Y)

then just do not use theta ... which is effectively

 [~,RHO] = cart2pol(X,Y)

So then you would just plot RHO vs Mean Depth for a scatter plot, and do any sort of correlation analysis you want with RHO as your x and Mean Depth as your y.

If then you want to plot it, you will plot using:

 polar(theta,rho)

but instead use:

 polar(1:2*pi,rho)

so you will plot rings instead of points.

[JLM]

Ahmet, Thank you very much for your help. I am afraid it does not work properly. I may be mistaken but it writes "Error using polar THETA and RHO must be the same size."

I have already calculated the radius and the angle corresponding to the longitudes and latitudes of my earthquakes , by myself because I had an issue with the origin of my system. I finally get the figure above (kind of wheel).

Yet, what I would like to have is the average of the depth for example, of all the earthquakes located between radius1 and radius2 and to make it appear with colors depending on the average obtained.

If you need more info about my code or what I am trying to do, please, let me know. I appreciate what you do.

Thank you everybody for helping me.