Extract data of a irregular shaped region from global data set

Hi,

I would like to extract the field values corresponding to an irregular shaped region (E.g. a city, a country or a continent) from a global data set.

If I have the geographical boundary (longitudes and latitudes) of such region, is there a way to extract data within that region (including the boundary) from the available global data?

I tried using 'find' command but it is taking hours together just to search one day of the global data.

There must be a smart way of doing it.

Please help.

Thankyou.

Answer 1

[Keegan Carvalho]

You will have to try indexing. Assuming you have a 3-d gridded data file (netcdf, etc.):

t       % t is a variable (like temperature, precipitation, etc.) 
lat  
lon     % lat and lon are latitude and longitude respectvely

longindex = find(lon==70.5);
latindex  = find(lat==15.5);

newt = t(longindex,latindex,:); % newt is t values for a region of coordinates 70.5, 15.5

If your looking to find t values for an entire region, try using loops:

for i=1:length(latrange)
    latindex(i)= find(lat==latrange(i));
end
for i=1:length(lonrange)
    longindex(i)= find(lon==lonrange(i));
end 

Hope this helps

Answer 2

[Image Analyst]

Yes but what are you going to do? The easiest is if you convert the coordinates to a digital image, and then make a make a mask from the coordinates of the one region with poly2mask(). With that you can do things on "the field values" inside the mask such as get their mean or distribution or whatever.