Before attaching this data frame to the other one I am using I nullify month, year, and day as I added them for assessing the monthly mean so it is no problem if the day column is lost. As for sample date, the date itself does not really matter. It could be one of the two dates, or even the first or last date of the month over which the results are being averaged. I don't think the nested version will work for me as the point of this code is to count exceedances in water quality objectives after assessing them over their averaging periods. Having both rows with repeated data would result in two counts if the result exceeded my objective value.