Blockmean for multiple parameter


I used GMT blockmean for regridding datasets. As we know blockmean is used [x,y,z] as standard input. I am curious to know that can we use the blockmean for multiple parameters such as [x,y,z1,z2,z3]? or Is there any other potential solution for this?

I’d do something like
for i in 1 2 3;do
gawk ‘{print($1,$2,$(2+i))}’ file.dat | blockmedian …

Yep. Or eliminate one process (gawk) by using -i0,1,$i in a loop for i = 2,3,…

This showed solution works one by one, am I right?
It means that on the first cycle x,y,z1 go then x,y,z2 and so on…
I am looking for something which works together such as x,y,z1,z2… in a one cycle.
So my output file will be regrdded output with x,y,z1,z2.
Hoope this make sense

I understand. But the block* modules only operate on one data column at the time, hence you will need to run one block* command per data column you want processed.

Thank you @pwessel and @thwbecker for a solution, but I think I am looking for something else.
As @pwessel replied, I think we are unable to do it at the time.