Avg over a 3d array?

Hi there!

So I have monthly gridded datasets in lat/lon/time(monthly): Example:

>> whos KoppenD_airT
  Name                Size                     Bytes  Class     Attributes
    KoppenD_airT      360x720x361            374284800  single              
>> whos KoppenDsa_airT
  Name                  Size                     Bytes  Class     Attributes
    KoppenDsa_airT      360x720x361            374284800  single              
>> whos KoppenDsb_airT
  Name                  Size                     Bytes  Class     Attributes
    KoppenDsb_airT      360x720x361            374284800  single

I need to create climatologies of these regions so that I can plot monthly changes in temperature over my 30 year climate period. What is the best way to do this?

I need to aggregate my monthly data over the entire region to a single value so that I can plot all of my months.

Here's one attempt I've made that isn't working:

idx=find(Dsa);
for ii=idx
    y=ind2sub(size(KoppenDsa_airT),ii);
    x=squeeze(KoppenDsa_airT(y,:));
end

I get:

Index exceeds matrix dimensions.

Thanks!

Answer 1

[Matt J]

aggregate my monthly data over the entire region to a single value

Does "aggregate" mean average? If so, this might be what you want,

 mean(reshape(KoppenDsa_airT,[],361),1);

[Kate]

Wait, question: Now my values are ranging between -1 and 1. Is the above code referencing the 1 dimension of the array instead of the 3rd?

[Matt J]

It should be averaging over each KoppenDsa_airT(:,:,i)

[Kate]

I realized my problem, I have a mask on my data so that non Koppen Dsa's are zero, I need to turn those into NaN's, thanks again for your help!

[Matt J]

If you turn them into NaNs, you would probably have to use nanmean()