Plotting variables of continuously array with different colors

Hello everyone,

Is it possible to :

(1) Change the value of Station column of this array (excel attached) into ordered rank Station? Anyone knows the code to do this? FYI, the number of stations is in hundreds (I put here for example only 3 stations) so, doing this in loop is preferable solution.

(2) Plot those three Station's profile of variable "s" vs depth from those three stations with different colors of line plots? Anyone knows the code to do this also?

Thank you!

Answer 1

[Star Strider]

You can use either unique or findgroups for this to get serial indices —

T1 = readtable('datax.xlsx')

T1 = 17×5 table
    Station     lon       lat     depth    salinity
    _______    ______    _____    _____    ________

       2       114.64    -8.16      0       33.995 
       2       114.64    -8.16      2       33.994 
       2       114.64    -8.16      4       33.994 
       2       114.64    -8.16      6       33.993 
       2       114.64    -8.16      8       33.993 
       2       114.64    -8.16     10       33.992 
       7       114.72    -8.16      0       33.976 
       7       114.72    -8.16      2       33.978 
       7       114.72    -8.16      4       33.977 
       7       114.72    -8.16      6       33.977 
      31        114.8    -8.16      0       33.983 
      31        114.8    -8.16      2       33.981 
      31        114.8    -8.16      4       33.981 
      31        114.8    -8.16      6       33.981 
      31        114.8    -8.16      8       33.981 
      31        114.8    -8.16     10        33.98 

[Us,ix1,ix2] = unique(T1.Station, 'stable');
T1.Station = ix2

T1 = 17×5 table
    Station     lon       lat     depth    salinity
    _______    ______    _____    _____    ________

       1       114.64    -8.16      0       33.995 
       1       114.64    -8.16      2       33.994 
       1       114.64    -8.16      4       33.994 
       1       114.64    -8.16      6       33.993 
       1       114.64    -8.16      8       33.993 
       1       114.64    -8.16     10       33.992 
       2       114.72    -8.16      0       33.976 
       2       114.72    -8.16      2       33.978 
       2       114.72    -8.16      4       33.977 
       2       114.72    -8.16      6       33.977 
       3        114.8    -8.16      0       33.983 
       3        114.8    -8.16      2       33.981 
       3        114.8    -8.16      4       33.981 
       3        114.8    -8.16      6       33.981 
       3        114.8    -8.16      8       33.981 
       3        114.8    -8.16     10        33.98 

figure
stem3(T1.Station, T1.depth, T1.salinity, ':k', 'Marker','none')
hold on
scatter3(T1.Station, T1.depth, T1.salinity, 150, T1.salinity, 'p', 'filled')
hold off
xlabel('Station')
ylabel('Depth')
zlabel('Salinity')
xticks(1:3)

This uses the third output of unique to get serial indices (note that I use the 'stable' argument to avoid sorting the results). The findgroups function works similarly to produce a grouping variable. The grouping variable vector is used here as the new value for ‘Station’.

EDIT — Added plots.

.

[Dyuman Joshi]

+1, unique/findgroups is the way to go here.

[Adi Purwandana]

Thank you @Star Strider. You've answered my 1st problem.

Anyway, for the 2nd question, is it possible to get a simple line plot of all stations (not in 3D plot as your suggestion) with color difference, for example, in jet color (e.g., Station 1-red,....., station 3-blue)?

[Dyuman Joshi]

Yes, use a for loop to go through each station and plot the data and setting the plot color corresponding to each station.

[Adi Purwandana]

yeah, that's my problem... could you please @Dyuman Joshi write a piece of codes to do that? I don't get it if the stations are located continously in one column.

[Star Strider]

I am not certain what you want to plot.

Plotting ‘salinity’ here —

T1 = readtable('datax.xlsx')

T1 = 17×5 table
    Station     lon       lat     depth    salinity
    _______    ______    _____    _____    ________

       2       114.64    -8.16      0       33.995 
       2       114.64    -8.16      2       33.994 
       2       114.64    -8.16      4       33.994 
       2       114.64    -8.16      6       33.993 
       2       114.64    -8.16      8       33.993 
       2       114.64    -8.16     10       33.992 
       7       114.72    -8.16      0       33.976 
       7       114.72    -8.16      2       33.978 
       7       114.72    -8.16      4       33.977 
       7       114.72    -8.16      6       33.977 
      31        114.8    -8.16      0       33.983 
      31        114.8    -8.16      2       33.981 
      31        114.8    -8.16      4       33.981 
      31        114.8    -8.16      6       33.981 
      31        114.8    -8.16      8       33.981 
      31        114.8    -8.16     10        33.98 

[Us,ix1,ix2] = unique(T1.Station, 'stable');
T1.Station = ix2

T1 = 17×5 table
    Station     lon       lat     depth    salinity
    _______    ______    _____    _____    ________

       1       114.64    -8.16      0       33.995 
       1       114.64    -8.16      2       33.994 
       1       114.64    -8.16      4       33.994 
       1       114.64    -8.16      6       33.993 
       1       114.64    -8.16      8       33.993 
       1       114.64    -8.16     10       33.992 
       2       114.72    -8.16      0       33.976 
       2       114.72    -8.16      2       33.978 
       2       114.72    -8.16      4       33.977 
       2       114.72    -8.16      6       33.977 
       3        114.8    -8.16      0       33.983 
       3        114.8    -8.16      2       33.981 
       3        114.8    -8.16      4       33.981 
       3        114.8    -8.16      6       33.981 
       3        114.8    -8.16      8       33.981 
       3        114.8    -8.16     10        33.98 

StaData = accumarray(T1.Station,  (1:size(T1,1)).', [], @(x){T1(x,:)})

StaData = 3×1 cell array
    {6×5 table}
    {4×5 table}
    {7×5 table}

n = numel(StaData);
colororder(jet(n))
cm = colormap(jet(n));

figure
hold on
for k = 1:n
    % StaData{k}
    scatter(StaData{k}.Station, StaData{k}.salinity, 50, cm(k,:), 's', 'filled')
end
% scatter3(T1.Station, T1.depth, T1.salinity, 150, T1.salinity, 'p', 'filled')
hold off
xlabel('Station')
ylabel('Salinity')
% ylabel('Depth')
% zlabel('Salinity')
xticks(1:3)
xlim([min(xticks)-1 max(xticks)+1])
colormap(jet(n))
hcb = colorbar;
hcb.Ticks = (1:n)/(n);
hcb.TickLabels = 1:n;
hcb.Label.String = 'Stations';

This approach creates individual table arrays for each station in the file, and plots them in a scatter plot.

.

[Adi Purwandana]

Sorry for confusing you @Star Strider, actually I want to plot it as the comment from @Sulaymon Eshkabilov below. But, since the stations can be many, I need any other way. So, is it possible to modify your code so that I can get the plot as the one plotted by @Sulaymon Eshkabilov below?

[Star Strider]

I do not know what you want to do.

Using my code, it is relatively straightforward to plot anything against anything —

T1 = readtable('datax.xlsx')

T1 = 17×5 table
    Station     lon       lat     depth    salinity
    _______    ______    _____    _____    ________

       2       114.64    -8.16      0       33.995 
       2       114.64    -8.16      2       33.994 
       2       114.64    -8.16      4       33.994 
       2       114.64    -8.16      6       33.993 
       2       114.64    -8.16      8       33.993 
       2       114.64    -8.16     10       33.992 
       7       114.72    -8.16      0       33.976 
       7       114.72    -8.16      2       33.978 
       7       114.72    -8.16      4       33.977 
       7       114.72    -8.16      6       33.977 
      31        114.8    -8.16      0       33.983 
      31        114.8    -8.16      2       33.981 
      31        114.8    -8.16      4       33.981 
      31        114.8    -8.16      6       33.981 
      31        114.8    -8.16      8       33.981 
      31        114.8    -8.16     10        33.98 

[Us,ix1,ix2] = unique(T1.Station, 'stable');
T1.Station = ix2

T1 = 17×5 table
    Station     lon       lat     depth    salinity
    _______    ______    _____    _____    ________

       1       114.64    -8.16      0       33.995 
       1       114.64    -8.16      2       33.994 
       1       114.64    -8.16      4       33.994 
       1       114.64    -8.16      6       33.993 
       1       114.64    -8.16      8       33.993 
       1       114.64    -8.16     10       33.992 
       2       114.72    -8.16      0       33.976 
       2       114.72    -8.16      2       33.978 
       2       114.72    -8.16      4       33.977 
       2       114.72    -8.16      6       33.977 
       3        114.8    -8.16      0       33.983 
       3        114.8    -8.16      2       33.981 
       3        114.8    -8.16      4       33.981 
       3        114.8    -8.16      6       33.981 
       3        114.8    -8.16      8       33.981 
       3        114.8    -8.16     10        33.98 

StaData = accumarray(T1.Station,  (1:size(T1,1)).', [], @(x){T1(x,:)})

StaData = 3×1 cell array
    {6×5 table}
    {4×5 table}
    {7×5 table}

n = numel(StaData);
colororder(jet(n))
cm = colormap(jet(n));

figure
hold on
for k = 1:n
    % StaData{k}
    plot(StaData{k}.depth, StaData{k}.salinity, '-s', 'Color',cm(k,:), 'MarkerSize',10, 'MarkerEdgeColor',cm(k,:), 'MarkerFaceColor',cm(k,:))
end
% scatter3(T1.Station, T1.depth, T1.salinity, 150, T1.salinity, 'p', 'filled')
hold off
xlabel('Depth')
ylabel('Salinity')
% ylabel('Depth')
% zlabel('Salinity')
% xticks(1:3)
xlim([min(xticks)-1 max(xticks)+1])
colormap(jet(n))
xlim([min(xlim)-1 max(xlim)+1])
hcb = colorbar;
hcb.Ticks = (1:n)/(n)-(1/(2*n));
hcb.TickLabels = 1:n;
hcb.Label.String = 'Stations';

.

[Adi Purwandana]

Voila, that's what I want. Thank you very much @Star Strider!

[Star Strider]

As always, my pleasure!

[Adi Purwandana]

@Star Strider Anyway...my last question to this thread... could you please help me to define LON, that is the value of lon from each StaData?

for k = 1:n

% StaData{k}

LON = StaData{k}.lon; % I want to pick lon value from each table of StaData and store it as LON

end

FYI, I can solve if it is not a continous array. I do this code when running multi excel files:

for k = 1:n

filename = A(k).name;

data = readmatrix(filename);

LON(k,1)= data(1,2); %pick lon value from each table from each file/data and store it as LON

end

But I dont know to adapt this in your code.

[Adi Purwandana]

I think I solved it...

LON(k,:) = StaData{k}.long(1);

[Star Strider]

That is correct.

Actually:

LON = StaData{k}.lon(1);

or:

LON(k) = StaData{k}.lon(1);

depending on the result you want.

(I was away for a few minutes running errands.)

Answer 2

[Sulaymon Eshkabilov]

Logical indexing would be an easy way of sorting the data and assigning new values to column 'station'. Then use the sorted data indecies to plot the data: s vs. depth

Here is one of the ways of doing it:

F = readmatrix('datax.xlsx');
Idx1=F(:,1)==2;
Idx2=F(:,1)==7;
Idx3=F(:,1)==31;
F(Idx1,1) = 1;
F(Idx2,1) = 2;
F(Idx3,1) = 3;
% Plotting the data
plot(F(Idx1, 5),F(Idx1, 4), '-o','linewidth', 2), hold all
plot(F(Idx2, 5),F(Idx2, 4), '-s','linewidth', 2)
plot(F(Idx3, 5),F(Idx3, 4), '-d','linewidth', 2)
legend('Station: 1', 'Station: 2', 'Station: 3', 'location', 'best')
grid on
xlabel('s')
ylabel('depth')

[Adi Purwandana]

Thank you @Sulaymon Eshkabilov. It's solved if the number of station is not much. In my case, the number of stations is in hundreds and even thousands. Do you know to do that in something like looping lines to do this?