convert orderd vector into a matrix

Hello everyone,

I am trying to understand how to convert ordered vector into a matrix. Here is my problem:

lat=[4,7,4,8,3,5]
lon=[32,31,32,33,35,38]
wind_speed=[12,24,45,56,78,24]

The correspoding values are identified by their position in the vector. This is to say that at the location lat(1)=4 e lon(1)=32 there was a wind speed of wind_speed(1)=12.

What I would like to have is two vector of lat and lon with unique and ordered values and a 2D matrix of wind values. My aim is to plot it with the function contour as

contour(X,Y,win_spd)

where X and Y are the new vector of ordered and unique lon and lat values and win_spd is a matrix with the the dimesion of X by Y.

Any help is greatly appricieted!

Giacomo

[Stephen23]

Do you have a wind_speed value for every lat and lon combination? Or are some of them missing?

[giacomo labbri]

Some values are NaN

[Stephen23]

Then you will need to interpolate scattered data:

https://www.mathworks.com/help/matlab/math/interpolating-scattered-data.html

[giacomo labbri]

thanks a lot! just out of curiosity what if I had no NaN?

Btw I would like to accept your answer but it seems I cannot because you posted it as a comment. If you put it in the aswer section I will accept it!

[Stephen23]

"just out of curiosity what if I had no NaN?"

The issue is not the NaN per se, but what kind of interpolation depends on if the input data are gridded or scattered:

https://www.mathworks.com/help/matlab/math/overview-of-interpolation-techniques.html

There are many ways that scattered data could be stored: in vectors, in a matrix (with NaN values), etc.

Gridded data cannot contain NaN values.