CDO

I have monthly data. I want to interpolate it to daily timesteps. For instance, using the code below I interpolated daily data for time period 2016-2030. After checking the output I found that the daily values in a month are approximately equal to the monthly values (i.e. daily values are kind of an average of monthly values and not so much as distributed), which isn't an issue if we were working on average daily/monthly values. But for processes (like SWAT which takes the summation of daily values to run a monthly simulation), this method severely overestimates the data.

I wanted to understand the way inttime or intntime interpolation worked. Furthermore, is it possible to "distribute" the monthly values into daily values? (so that the sums of all values in a month are equal to the total monthly value) I would like to stress that the quantum of daily values isn't that important. As long as their sum is equal to the total monthly value, it is good to go.

cdo -O -mergetime -selmon,12 -selyear,2015 pr_ensmean-2011-2030.nc -selyear,2016/2030 pr_ensmean-2011-2030.nc -selmon,1 -selyear,2030 pr_ensmean-2011-2030.nc ifile1.nc
cdo -outputtab,date,lat,lon,value -selyear,2016/2030 -inttime,2016-01-01,12:00:00,1day ifile1.nc > test.txt

Please help me out.

Thanks,
Akhilesh