MATLAB：2D 数据到 3D ！ 如何？

Question

I have a 34200 x 4 table. This table shows the 30 years monthly amount of pr (precipitation) in some latitudes and longitudes. So the columns are lat , lon , date , and pr . I want to convert it to a 3D matrix that longitude x latitude x month . I attach my table. Please tell me how to do it I'm a beginner. If I don't want to convert it based on the month I could but this issue is so complicated for me. Please look at my table it's only 235 KB I upload it to my DropBox so please click on Open in the top right side and click download.

Here is my image

Answer 1

Inspecting your data, your latitude and logitude values actually represent 95 unique locations which are scattered seemingly randomly. You can see that in the figure below.

length(unique(C.lat)) % 95
length(unique(C.lon)) % 95
scatter(C.lat, C.lon)

If the locations were spaced in a grid, it would make sense to use lat and lon as the axes of a data matrix. But instead, it is better to use only one axis representing the unique locations. This then leaves you with a second axis representing the date.

length(unique(C.date)) % 360
360 * 95 % 34200 - the number of values we have

Reformatting the data

Therefore, I would store the data in a 2D matrix as follows.

locations_lat = unique(C.lat, 'stable');
locations_lon = unique(C.lon, 'stable');
dates = unique(C.date, 'stable');
data = reshape(C.pr, length(dates), length(locations_lat)); % size 360 x 95

Then, to check that this has worked, choose a random example.

location_num = 27;
date_num = 242;
lat = locations_lat(location_num) % 14.68055556
lon = locations_lon(location_num) % 65.23111111
date = dates(date_num) % 2/1/2009
precipitation = data(date_num, location_num) % 16.7179

Searching for that position and date in the original tale, we have:

9602|   14.6805555600000    65.2311111100000    '2/1/2009'  16.7179000000000

Answer 2

If A is your data with the 4 columns [34200x4] then you can create a 3-dimensional matrix like this:

B = zeros(len(A),3)
B(:, 1) = A(:, 1)
B(:, 2) = A(:, 2)
B(:, 3) = A(:, 3)

Possibly even:

B(:,1:3) = A(:, 1:3)

Depending how your data is set, you may need to transpose which is:

B = 'B

You can map data over as long as the dimensions match.

You can implement a for loop if you have even more columns or for dynamic data entries.