how to calculate 3D netcdf in matlab?

I have 300-by-240-by-365 3D netcdf file for 16 years rainfall data. I need to have mean monthly data which contains 300-by-240-by-12. how can I do this?

here is my trial script

clear all;

MP2(1:300,1:240,1:12)=0; % in case of IMERG lat is arranged in row where as lon is arranged in column((lat:lon,year)=0) is Monthly Precipitation (MP) with zero intial matrix to start for loop

for yy=2002:2017 % the year from 2002 to 2017

for mm=1:12

ap2(1:300,1:240)=0;

E = eomday(yy,mm);

dy=YYYYMMDD2doy(yy,mm,1);

hits1 = ['D:\Rainfall_Data_from_Satelite\CHIRPS/','chirps-v2.0.',num2str(yy,'%02d'),'.days_p05.nc'];

fe = dir(hits1);

ncid = netcdf.open(hits1);

lat=netcdf.getVar(ncid,0,'single'); % latitude

lon=netcdf.getVar(ncid,1,'single'); % longitude

p1=ncread(hits1,'precip',[4269,1061,dy],[300,240,E]);% days in a month

p1_sum=nansum(p1,3); % a monthly rainfall using sum of daily data

% scale=netcdf.getAtt(ncid,3,'scale_factor');

% offset=netcdf.getAtt(ncid,3,'add_offset');

netcdf.close(ncid);

end

MP2(:,:,mm)=p1_sum;

end

Answer 1

[Ronit]

Hello Habtamu,

To calculate the mean monthly rainfall data from your "3D NetCDF file", you need to ensure that you are correctly iterating through the years and months, and accumulating the monthly precipitation data.

clear all;
meanMonthlyPrecipitation = zeros(300, 240, 12);

for mm = 1:12
    % Initialize a temporary matrix to accumulate monthly precipitation
    tempMonthlySum = zeros(300, 240);
    yearCount = 0;

    for yy = 2002:2017
        E = eomday(yy, mm);
        dy = YYYYMMDD2doy(yy, mm, 1);
        
        % Construct the file path and add here
        
        if exist(hits1, 'file')
            ncid = netcdf.open(hits1);
            % Read the precipitation data for the current month
            p1 = ncread(hits1, 'precip', [4269, 1061, dy], [300, 240, E]);
            
            % Replace NaN values with zeros for summing
            p1(isnan(p1)) = 0;
            % Sum the daily precipitation data to get monthly totals
            p1_sum = sum(p1, 3);
            % Accumulate the monthly totals
            tempMonthlySum = tempMonthlySum + p1_sum;
            
            yearCount = yearCount + 1;
            netcdf.close(ncid);
        end
    end
    
    % Calculate the mean monthly precipitation by dividing by the number of years
    if yearCount > 0
        meanMonthlyPrecipitation(:, :, mm) = tempMonthlySum / yearCount;
    end
end
% Now, meanMonthlyPrecipitation contains the mean monthly data over 16 years

Make sure that the "YYYYMMDD2doy" function is correctly defined and available in your environment.

Hope it helps!