Put arrow and its value in a plot

Hello there,

I want to show a row and its value in a plot. Please find attached the data. My intended plot should be like this:

load('data_ask_MLD')
figure;
plot(PTn,z);
set(gca,'ydir','reverse')

Thank you!

Answer 1

[Mathieu NOE]

hello

this is a simple example , based on the fex submission :

arrow - File Exchange - MATLAB Central

load('data_ask_MLD')
figure;
plot(PTn,z);
set(gca,'ydir','reverse')

% define which row for display
r = 100;
x = PTn(r);
y = z(r);
al = 1; % arrow length (in x direction)

arrow([x-al,y],[x,y]); % Fex : https://fr.mathworks.com/matlabcentral/fileexchange/278-arrow
text(x-4*al, y, ['MLD = ' sprintf('%0.5g', y)])

[Adi Purwandana]

Great to know this function. Thank you @Mathieu NOE! but since I have lots of profiles, I need an automatically adjusted the MLD value instead of defining manually the r = 100.

[Mathieu NOE]

hello again

try this :

load('data_ask_MLD')
figure;
plot(PTn,z);
set(gca,'ydir','reverse')
% define which row for display
value = 24.7; % MLD value to display 

[x,~] = find_zc(PTn,z,value);
y = value;
al = 1; % arrow length (in x direction)
arrow([x-al,y],[x,y]); % Fex : https://fr.mathworks.com/matlabcentral/fileexchange/278-arrow
text(x-3*al, y, ['MLD = ' sprintf('%0.5g', y)])


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function [ZxP,ZxN] = find_zc(x,y,threshold)
% find x values corresponding to y =  threshold (like a zero crossing detector) : 
% ZxP is x when signal slope is positive at the crossing point
% ZxN is x when signal slope is negative at the crossing point

% put data in rows
x = x(:);
y = y(:);

% positive slope "zero" crossing detection, using linear interpolation
y = y - threshold;
zci = @(data) find(diff(sign(data))>0);  %define function: returns indices of +ZCs
ix=zci(y);                      %find indices of + zero crossings of x
ZeroX = @(x0,y0,x1,y1) x0 - (y0.*(x0 - x1))./(y0 - y1); % Interpolated x value for Zero-Crossing 
ZxP = ZeroX(x(ix),y(ix),x(ix+1),y(ix+1));

% negative slope "zero" crossing detection, using linear interpolation
zci = @(data) find(diff(sign(data))<0);  %define function: returns indices of +ZCs
ix=zci(y);                      %find indices of + zero crossings of x
ZeroX = @(x0,y0,x1,y1) x0 - (y0.*(x0 - x1))./(y0 - y1); % Interpolated x value for Zero-Crossing 
ZxN = ZeroX(x(ix),y(ix),x(ix+1),y(ix+1));

end

[Mathieu NOE]

and if you need to plot more than one MLD value, you can use this modified code :

NB that I am not using interp1 because your profile is not monotonic everywhere , so interp1 may fail sometimes.

load('data_ask_MLD')
figure;
plot(PTn,z);
set(gca,'ydir','reverse')
% define which row for display
value = [24.7 33.1  54.7 78.9]; % MLD values to display 
al = 1; % arrow length (in x direction)

for k = 1:numel(value)
    [x,~] = find_zc(PTn,z,value(k));
    x = x(1); % take the first value
    arrow([x-al,value(k)],[x,value(k)]); % Fex : https://fr.mathworks.com/matlabcentral/fileexchange/278-arrow
    text(x-3*al, value(k), ['MLD = ' sprintf('%0.5g', value(k))])
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function [ZxP,ZxN] = find_zc(x,y,threshold)
% find x values corresponding to y =  threshold (like a zero crossing detector) : 
% ZxP is x when signal slope is positive at the crossing point
% ZxN is x when signal slope is negative at the crossing point

% put data in rows
x = x(:);
y = y(:);

% positive slope "zero" crossing detection, using linear interpolation
y = y - threshold;
zci = @(data) find(diff(sign(data))>0);  %define function: returns indices of +ZCs
ix=zci(y);                      %find indices of + zero crossings of x
ZeroX = @(x0,y0,x1,y1) x0 - (y0.*(x0 - x1))./(y0 - y1); % Interpolated x value for Zero-Crossing 
ZxP = ZeroX(x(ix),y(ix),x(ix+1),y(ix+1));

% negative slope "zero" crossing detection, using linear interpolation
zci = @(data) find(diff(sign(data))<0);  %define function: returns indices of +ZCs
ix=zci(y);                      %find indices of + zero crossings of x
ZeroX = @(x0,y0,x1,y1) x0 - (y0.*(x0 - x1))./(y0 - y1); % Interpolated x value for Zero-Crossing 
ZxN = ZeroX(x(ix),y(ix),x(ix+1),y(ix+1));

end

Answer 2

[Pramil]

Hi Adi,

You can use the "text" function available in MATLAB to create plot shown in your image. Here is the documentation link for the same for your reference:

https://www.mathworks.com/help/releases/R2022a/matlab/ref/text.html

And here is the updated code:

load('data_ask_MLD')
figure;
plot(PTn,z);
set(gca,'ydir','reverse')

[~, idx] = min(abs(z - MLD));
x_value = PTn(idx);
text(x_value, MLD, 'MLD = 24.7\rightarrow ','HorizontalAlignment','right');

[Adi Purwandana]

Thank you @Pramil; but since I have lots of profiles, I need an automatically adjusted the MLD value.

Answer 3

[Star Strider]

The 'textarrow' coordinates have to adapt to the data and then correct for the reversed y-axis direction.

Try this —

load('data_ask_MLD')
figure;
plot(PTn,z);
set(gca,'ydir','reverse')
xlabel('PTn')
ylabel('z')
xapf = @(x,pos,xl) pos(3)*(x-min(xl))/diff(xl)+pos(1);                % 'x' Annotation Position Function
yapf = @(y,pos,yl) pos(4)*(y-min(yl))/diff(yl)+pos(2);                % 'y' Annotation Position Function
xl = xlim;
yl = ylim;
pos = gca().Position;
MLDval = 24.7;
PTnval = interp1(z, PTn, MLDval)                                      % X-Coordinate Value (Derived From Data) Right End Of The Arrow

PTnval = 27.2905

zval = interp1(PTn, z, PTnval)                                        % Y-Coordinate Value (Derived From Data)

zval = 24.7000

annotation('textarrow', xapf(PTnval+[-2.25 -0.25],pos,xl), 1-yapf([1 1]*zval,pos,yl), String=("MLD = "+MLDval)+" m ", HeadStyle='vback3')

EDIT — (23 Nov 2024 at 12:45)

Changed 'HeadStyle'.

.

[Adi Purwandana]

Thank you @Star Strider. Are you sure the arrow points the value 24.7? It should be point of MLD = 24.7 or z = y = 24.7.

[Star Strider]

It was not originally obvious to me what axis ‘MLD’ refers to, since they are both quite similar. It plots correctly if the y-axis is not reversed. The absolute references do not change when the axis direction is reversed. (I consider that to be a bug.) Fudging an ‘ofst’ (offset) value is the best I can do in this instance.

load('data_ask_MLD')
figure;
plot(PTn,z);
set(gca,'ydir','reverse')
xlabel('PTn')
ylabel('z')

MLDval = [24.7; 100; 150];
for k = 1:numel(MLDval)
    txt_arrow(z,PTn,MLDval(k),gca)
end

function txt_arrow(z,PTn,MLDval,gca)
PTnval = interp1(z, PTn, MLDval);                                     % X-Coordinate Value (Derived From Data) Right End Of The Arrow
zval = interp1(PTn, z, PTnval);                                       % Y-Coordinate Value (Derived From Data)
xapf = @(x,pos,xl) pos(3)*(x-min(xl))/diff(xl)+pos(1);                % 'x' Annotation Position Function
yapf = @(y,pos,yl) pos(4)*(y-min(yl))/diff(yl)+pos(2);                % 'y' Annotation Position Function
xl = xlim;
yl = ylim;
pos = gca().Position;
minz = yapf(min(z),pos,yl);
ofst = 1+minz/4;
annotation('textarrow', xapf(PTnval+[-2.25 -0.25],pos,xl), ofst-yapf([1 1]*zval,pos,yl), String=("MLD = "+MLDval)+" m ", HeadStyle='vback3')
end

Plotting more arrows would simply require running the annotation call in a loop, with appropriate values for ‘PTNval’ and ‘zval’. I wrote one here that seems to work.

EDIT — (25 Nov 2024 at 12:17)

I submitted a bug report on this problem this morning (‘Report a bug Case 07439706 created successfully’).

EDIT — (26 Nov 2024 at 11:57)

MathWorks corroborated my fiindings and was able to reproduce this behaviour. There may be a fix in a future release.

.

[Adi Purwandana]

Great! Thanks for let us know the update.

[Star Strider]

My pleasure!

It appears that my ‘kludge’ fix is the best option just now. I have not checked to see if it might work in other instances, however I believe it is likely a ‘one-off’ that will not generalise to other situations.

My only consolation is that my code works correctly, and would continue to do so if the absolute references changed with the axis direction (and possibly axis scaling, however I have not experimented with it with a diifferent axiis scale).