I need help importing this complicated data set into MATLAB.

I am trying to import a data set that contains the latitude, longitude, and the thickness of the Earth's crust at that coordinate. The difficult part is the formatting of the data within the text file. It is fomatted as follows:

_______________________________________________________

column 1 ; column 2

latitude_1 ; crustal thickness w/ sediment_1

longitude_1 ; crustal thickness w/o sediment_1

latitude_2 ; crustal thickness w/ sediment_2

longitude_2 ; crustal thickness w/o sediment_2

latitude_3 ; crustal thickness w/ sediment_3

longitude_3 ; crustal thickness w/o sediment_3

_______________________________________________________

And it continues like this for a few thousand data points.

I would like to be able to have the latitudes in one column, the longitudes in once column, and the crustal thicknesses in their own columns.

I have tried all sorts of stuff, and have searched through a ton of forum posts, but I have not been able to figure this out.

Here is what I have so far, but this method has not produced usable results. More specifically, the for loops don't always pick out the correct values. Latitude should never be greater than 90, yet the for loop is producing values greater than 100. I dont know how. I wouldnt mind abandoning the for loop method if I can find a way to textscan the data into columns.

fid = fopen('nam-data-xl_2.txt');
N = 3902;
formatSpec = '%s%f%f';
Mat = textscan(fid,formatSpec,N);
fclose(fid);

coord=Mat{3};
thick=Mat{2};

lati=[];
k=1;
for i = 1:2:3902
    lati(k)= coord(i);
    k=k+1;

end

longi=[];
j = 1;
for ii=2:2:3902
    longi(j)=coord(ii);
    j=j+1;
end
coordinates = [lati; longi];

crsti=[];
l = 1;
for iii = 2:2:3902
    crsti(l) = thick(iii);
    l = l+1;
end

Answer 1

[Kirby Fears]

Drew,

I made a text file called DrewL.txt which literally has the faux-data you posted in your question. I am using the format string '%s%s' since the faux-data is semicolon separated strings.

The code below reads the data, separates everything out into 4 columns, and puts them all into 1 table. You should be able to adapt this to your actual data. If you post a more realistic portrayal of your data, I could possibly help make the required adaptations.

formatSpec = '%s%s';
fid = fopen('DrewL.txt');
read = textscan(fid,formatSpec,'Delimiter',';','HeaderLines',1);
fclose(fid);
read = [read{:}];
latitude = read(1:2:end,1);  
longitude = read(2:2:end,1);
crustSed = read(1:2:end,2);  
crustNoSed = read(2:2:end,2);
data = table(latitude,longitude,crustSed,crustNoSed);

Hope this helps.

[Kirby Fears]

I tested this out with formatSpec = '%f%f' on a text file that looks like this:

column 1 ; column 2
10 ; 1234
11 ; 2234
20 ; 3234
21 ; 4234
30 ; 5234
31 ; 6234

It worked for me.

Let me know if this works for you.

[Drew L]

This should work, however the latitude and longitude are on top of each other in the same column. So the format is

column 1

lat1

long1

lat2

long2

I need it to be:

column 1

lat 1

lat 2

column 2

long 1

long 2

Thank you

[Kirby Fears]

I don't understand your last comment. Did you try the code I posted?