using ms_calc_flux_rsfme with monthly and annual periods, I find that (depending in the method) there are concerning differences between the sum of monthly estimates and annual estimate for a given year. I will summarize my initial findings, and then provide screenshots of the data which shows the problem below.
summary
troubling:
- pw: sum of period weighted is roughly 100,000 times smaller than annual flux estimate (but, if multiplied by 100,000, is pretty close)
- average: the mean of the monthly estimates is approximately the annual estimate, sum of monthly is well above annual value
good so far:
- composite: monthly sum is very close to annual
- beale: monthly sum is very close to annual
- rating: monthly sum is very close to annual
screenshots from bad ones:
pw
average
example of good one:
composite
using ms_calc_flux_rsfme with monthly and annual periods, I find that (depending in the method) there are concerning differences between the sum of monthly estimates and annual estimate for a given year. I will summarize my initial findings, and then provide screenshots of the data which shows the problem below.
summary
troubling:
good so far:
screenshots from bad ones:
pw
average
example of good one:
composite