combine Latitude and longitude country data from multiple arrays into one array

I have an array (MYD_countries) of size 249*64800 in which the first dimension represents different countries in the world and the second dimension represents global data grid cells of 1-degree data (180*360 = 64800). Each country data has data values within the country grid points and NaN elsewhere. I want to combine the data of all countries together and reduce its dimension to 1*64800 so that I can plot it at one call using geoshow. I have tried it using a loop as below but it is only storing the last country data, not all country data. Could anybody help please?

for i = 1:249;
    for j = 1:64800; 
                if(MYD_countries(i, j) ~= NaN);
                MYD_all_countries_plot (1, j) = MYD_countries (i, j);
                else;
                end;
    end;
end;

Answer 1

[Sagar Parajuli]

I solved this problem with a simple technique using 'max' function.

data_2d = max(MYD_countries, [], 1); %taking maximum along the first dimension retains the data for each country since values are NaN elsewhere.

Answer 2

[KSSV]

Let A be your data of size 249*64800

[m,n] = size(A) ; 
iwant = zeros(180,360,m) ; 
for i = 1:m
    iwant(:,:,i) = reshape(A(i,:),180,360) ; 
end

Now iwant variable is a 3D matrix of size 180*360*249; each 2D matrix represents a country. You can plot it now.

[Sagar Parajuli]

Thank you for your prompt answer. But that is not what I want. I could plot the 3-d data of all the countries together using pcolor in a loop but I need to plot it using geoshow which doesn't allow me to use the 3-d data as far as I know. That is why I want to reduce it to 2-d data (64800=180*360) so that I can plot all the data at one call.

[KSSV]

USe the variable iwant(:,:,i) which is a 2D matrix using geoshow.

[Sagar Parajuli]

I need to plot all the data together at once, not only one country. I can't use a loop in geoshow, that is my problem. The real problem I want to solve creatively is to reduce the data dimension from 249*64800 to 1*64800 which should be possible because the countries are unique in each country layer. I basically want to extract only those grid cells with data (all other grid cells have NaN) in each country and put it in a 2-d array so that data from all the countries is there in a single 2-d array.