How to add up multiple values at every co-ordinate.

Hi,

I have a data (LIG6) set which has values for iron (Fer) at multiple depths, for multiple co-ordinates across the globe. I can plot

contourf(LIG6.Fer(:,:,1)');

However I do not want just the Fer at the surface (1), I want the sum of Fer at every depth at each lat/long, and then be able to plot those values over the globe.

Any ideas?

Alex

Answer 1

[John BG]

Alexandra

understanding that

contourf(LIG6.Fer(:,:,1)');

is a matrix 3x3xN where N is constant or varies from Lat Long point

if N is constant, you have all depths over all points, even if only filled with 0 or NaN then

[sz1 sz2 sz3]=size(LIG6.Fer) % measure size array
X=LIG6.Fer(:,1)              % extract latitudes
Y=LIG6.Fer(:,2)              % extract longitudes

contourf(X,Y,sum(LIG6.Fer(:,:,[1:end])'));

if N is not a constant, meaning the amount of Fe measurements vary between [Lat Long] then you have to extract first each depth and apply it to where I wrote 'end'

I would you to have my answer as accepted so please let me know whether the constant N assumption is correct or you want to proceed extracting each individual and varying depths.

Appreciating time and attention, awaiting answer

John BG

<mailto:[e-mail] [e-mail]>

[Alexandra Ridgway]

Hi John, Thank you for your help! The values are varying so I need to extract and then sum them...

Alex

[Alexandra Ridgway]

Ive tried

for k=1:30
    newFe(:,:,k) = anomglobalFe68(:,:,k)*dz(k);
end

bob = nansum(newFe,3) %to integrate over the third dimension

but it doesn't seem to like this as newFe is still 3 dimensions.

[Jan]

@Alexandra: After this operation bob should be 2D. Please explain exactly, what differs from your expectations.

[John BG]

Alexandra

ok, implementing the extraction of varying amounts of depths is easy with cells,

but to avoid guess work, would it be possible for you to provide a sample,

just a few points of countourFe68, points that have different amounts of depths, ok?

please attach it to your question

awaiting answer

John BG <mailto:[e-mail] [e-mail]>

[Alexandra Ridgway]

Hi, I have rerun it and bob is 2D (360x180) my plot just looks very odd (see picture), and I've no idea why! Thank you for trying to help me:)

For context it should look like the second image (note the axis)..

Answer 2

[John BG]

thanks for attaching the images showing what you get so far and what you expect.

Are you reading with MATLAB from XML format data, from for instance a GOCAD export?

http://www.mirageoscience.com/data/media/all/original/tips13.pdf

or do you already have a MATLAB structure in the workspace?

anyway, it would greatly help to know the specifics of the structure LIG6.

can you attach a sample? if you already have a MATLAB variable for the LIG6 structure, save LIG6 to a .mat file

save sample1.mat LIG6

or

save('sample.mat','LIG6')

if the name of the structure or sample structure is not exactly LIG6 please use the name of the variable accordingly

and attach the .mat file to your question or comment

John BG

<mailto:[e-mail] [e-mail]>