- Extract the size of data using “[lon, lat, time] = size(climateData);”
- You can reshape the data as per your requirement with the following command: “reshapedData = reshape(climateData, [[lon,lat], time]);”. As the data I created was already in the format 1x2x12, the size of the “reshapedData” will remain same.
- Convert the time vector data to datetime format using: “timeVector = datetime(timeVector);”
- Now you can use this time data to extract unique years or any other information. If you wish to extract unique years from the data you can use: “years = unique(year(timeVector));”
- To plot the created time series data use: “scatter3(x_lon,y_lat, timeVector)”. Where “x_lon” and “y_lat” contains the longitudes and latitudes respectively.
Create timeseries with 3d climate data
7 views (last 30 days)
Show older comments
Hi,
I have 3d climate data in the form lonxlatxtime. I would like to create a timeseries for these data for every year, but have no idea how to do this since I am new to Matlab. Can someone please assist?
Thank you
0 Comments
Answers (1)
Ayush
on 5 Oct 2023
Hi Aarde,
I understand that you want to create time series of the 3D data points which is in the form longitude x latitude x time.
Since the data was not provided, I generated my own data points, which is of the dimension “1x2x12”. Meaning that there will be 1 row and 2 columns where first column will contain the data of the longitude and the second column will contain the data of the latitude and there are total of 12 years.
I randomly generated data of longitude and latitude using the “randi” command, keeping the range from 0 to 180 (representing degrees).
Considering that the data is ready with us, following steps can be used to display the data in timeseries:
Final code snippet is given below:
%% Data Generation
lon = 1; % Number of longitudes
lat = 2; % Number of latitudes
time = 12; % Number of time steps
% Generate random climate data
climateData = randi([0 180],lon, lat, time);
% Display the size of the climate data
disp(size(climateData));
% Generate a random time vector
numTimeSteps = 12; % Number of time steps
% Generate random values for year, month, day, hour, minute, and second
yearValues = randi([2000, 2021], numTimeSteps, 1); % Random years between 2000 and 2021
monthValues = randi([1, 12], numTimeSteps, 1); % Random months
dayValues = randi([1, 28], numTimeSteps, 1); % Random days
hourValues = randi([0, 23], numTimeSteps, 1); % Random hours
minuteValues = randi([0, 59], numTimeSteps, 1); % Random minutes
secondValues = randi([0, 59], numTimeSteps, 1); % Random seconds
% Create the random time vector
timeVector = datetime(yearValues, monthValues, dayValues, hourValues, minuteValues, secondValues);
% Display the time vector
disp(timeVector);
%% Time Series Conversion
[lon, lat, time] = size(climateData);
reshapedData = reshape(climateData, [[lon,lat], time]);
timeVector = datetime(timeVector);
years = unique(year(timeVector));
disp("Unique year: " + years)
%% Data Plotting
x_lon = squeeze(reshapedData(:,1,:));
y_lat = squeeze(reshapedData(:,2,:));
scatter3(x_lon,y_lat, timeVector)
1 Comment
Ronald
on 8 Apr 2024
Hi Ayush, could you modify this code so it can be used to reshape a 3D timeseries matrix(lon,lat,time) from daily to monthly? or from monthly to annual?
See Also
Categories
Find more on Climate Science and Analysis in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!