How to regrid into different resolution without interpolation?

I have a 2d data of resolution 0.5 deg. lat=-89.75:0.5:89.75 and lon=-179.75:0.5:179.75. data has dimension (720*360), first dim=lon, 2nd dim=lat.

I want to make data to a 2 deg resolution by taking 2 deg * 2 deg grid box. Averaging all the values coming inside this box. In this way it would be a 2 deg resolution without interpolation.

I have done it using griddata interpolation. I want to crossverify it with the above method.

Thanking you in advance.

Answer 1

[Matt J]

Using,

https://www.mathworks.com/matlabcentral/fileexchange/48089-separable-block-wise-operations

sepblockfun(data,[2,2],'mean')

[Deepshikha Dadhwal]

Hello Matt, I want to do the same to my variable but it is not in uniform grid as in Subodh's statement. My variable is irregularly spaced and changes size per file. I want to collect all the datas falling into a 1*1 grid and count & average them. I used Griddata but it averages out sparse data as well and hence the data represntation is affected.

[Matt J]

@Deepshikha Dadhwal

It sounds like you need histcounts2() rather than any kind of interpolation.