This is a guide about how to check MOSART water budget between two components in E3SM
Introduction
In current E3SM versions (V1 and V2), the river component, MOSART, only does the internal water balance check to make sureinput - output = delta storage at every coupling time step (3 hourly). It does not average the water budget terms on monthly or annual basis in the log files. This makes it difficult to do water balance check between the river component and the coupler, in which the water budgets from each model component are summarized at monthly time scale in the log file like this:
(seq_diag_print_mct) NET WATER BUDGET (kg/m2s*1e6): period = monthly: date = 19800201 0
atm lnd rof ocn ice nh ice sh glc *SUM*
wfreeze 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000
wmelt 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000
wrain -29.57235627 5.41307635 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 -24.15927992
wsnow -2.05972140 0.98147580 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 -1.07824559
wevap 0.00000000 -2.82794138 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 -2.82794138
wrunoff 0.00000000 -0.75014053 0.36133891 0.00000000 0.00000000 0.00000000 0.00000000 -0.38880162
wfrzrof 0.00000000 -0.18369858 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 -0.18369858
*SUM* -31.63207767 2.63277166 0.36133891 0.00000000 0.00000000 0.00000000 0.00000000 -28.63796710
The rof column indicates the river component (MOSART) which has only one positive value 0.36133891 at row wrunoff. This means that river is taking water from the coupled system at a monthly averaged rate of 0.36133891 (kg/m2s*1e6) . This rate is calculated based on the total surface area of the earth, including both ocean and land.
If we have the same value calculated from the river component, we can say that the coupler handles the water budget terms correctly, in another word, there’s no water conservation issue when passing water from river to the coupler. Such check is getting more and more necessary as more features are added to MOSART.
Before we introduce the monthly budget table into MOSART, there’s a quick and dirtyway to do the check. Of course, it involves some efforts.
Steps to do the water budget checks
1. turn on a few flags in MOSART source code so that it could output the information needed for the check
- In
\mosart\src\riverroute\RtmMod.F90, findoutput_all_budget_termsand change it to.true.. - In the same file, find
budget_writeand change it to.true.. - compile the model.
2. make sure the BUDGETS is .true. in env_run.xml so that the coupler will output the budget table above.
3. run the model for at least a few months
4. find the budget terms from the MOSART log file
-
The water balance equation for any given time period in MOSART is
volume change - input + output - other = 0where
volume changeis the water volume change in river channels, hillslopes, floodplains, and reservoirs during the period;otheris a lag volume term of the channel outflow caused by the MOSART subcycling scheme which should also be accounted for the total volume change during the period. To calculate the monthly water budget and to compare it with the budget in the coupler, we need to find out the initial volume and the final volume of the month, as well as the total “other volume” over the month from the MOSART log file. - The water budget for the very first time step in the log file is like this (note that the water budget of the first time step is at 10800 second (3 hr) of the first day of the first month):
(Rtmrun) MOSART BUDGET diagnostics (million m3) for 19800101 10800 (Rtmrun)----------------------------------------------------------------- (Rtmrun) tracer = 1 (Rtmrun) dvolume wh = 1 0.000000 (Rtmrun) dvolume wt = 1 0.000000 (Rtmrun) dvolume wr = 1 0.000000 (Rtmrun) dvolume dstor = 1 0.000000 (Rtmrun) * dvolume total = 1 0.000000 (Rtmrun) x dvolume check = 1 0.000000 (should be zero) (Rtmrun) x volume init = 1 0.000000 (Rtmrun) x volume final = 1 0.000000 (Rtmrun) x volumeh init = 1 0.000000 (Rtmrun) x volumeh final = 1 0.000000 (Rtmrun) x volumet init = 1 0.000000 (Rtmrun) x volumet final = 1 0.000000 (Rtmrun) x volumer init = 1 0.000000 (Rtmrun) x volumer final = 1 0.000000 (Rtmrun) x storage init = 1 0.000000 (Rtmrun) x storage final = 1 0.000000 ... - The water budget for the first time step of the second month (or first step after the final step of the first month):
(Rtmrun) MOSART BUDGET diagnostics (million m3) for 19800201 10800 (Rtmrun)----------------------------------------------------------------- (Rtmrun) tracer = 1 (Rtmrun) dvolume wh = 1 172.580688 (Rtmrun) dvolume wt = 1 281.340362 (Rtmrun) dvolume wr = 1 -44.947927 (Rtmrun) dvolume dstor = 1 0.000000 (Rtmrun) * dvolume total = 1 408.973123 (Rtmrun) x dvolume check = 1 0.000000 (should be zero) (Rtmrun) x volume init = 1 491655.966962 (Rtmrun) x volume final = 1 492064.940085 (Rtmrun) x volumeh init = 1 46749.098846 (Rtmrun) x volumeh final = 1 46921.679533 (Rtmrun) x volumet init = 1 70426.281867 (Rtmrun) x volumet final = 1 70707.622229 (Rtmrun) x volumer init = 1 374480.586249 (Rtmrun) x volumer final = 1 374435.638322 (Rtmrun) x storage init = 1 0.000000 (Rtmrun) x storage final = 1 0.000000 (Rtmrun)---------------- ... -
From the two tables we can see the initial volume is 0 and the final volume is 491655.966962 (million m3) for the first month.
-
The “other volume” term is written at each time step, we can extract it from the log file using
grep 'sum other = 1' rof.log.XXXXXX > other.txt - We will get a text file that lists the “other volume” at each time step:
(Rtmrun) sum other = 1 0.000000 (Rtmrun) sum other = 1 -0.000000 (Rtmrun) sum other = 1 -0.000001 (Rtmrun) sum other = 1 -0.000021 (Rtmrun) sum other = 1 -0.000207 (Rtmrun) sum other = 1 -0.001263 (Rtmrun) sum other = 1 -0.003941 (Rtmrun) sum other = 1 -0.007846 (Rtmrun) sum other = 1 -0.012761 (Rtmrun) sum other = 1 -0.019006 (Rtmrun) sum other = 1 -0.027986 ... ... - Then there’s a tricky step. We need to sum the “other volume” over the time period we are examining. For example, if we are examining the first month, which is January (31 days), then we need to sum the first 8*31 = 248 values to get the total “other volume” for this month. In this case, the total “other volume” over the first month is -2023.689434 (million m3).
5. Calculate the monthly water budget from MOSART component
- Now we can convert the values to monthly water budget using the values from the steps above. To be consistent with the budget table, the unit of water budget will be converted to kg/m2s*1e6:
R = 6.37122e6; % earth radius (m)
area = 4*pi()*R^2; % earth surface area (m2)
rof_beginS = 0 % initial volume (million m3)
rof_endS = 491655.966962 % final volume (million m3)
sum_other = -2023.689434 % total other volume (million m3)
days = 31
budget = (rof_endS - rof_beginS - sum_other)/area/(days*24*3600)*1e15; % kg/m2s*1e6
Using the values we extract from the first month above, we can calculate the budget is 0.361338907794895
6. Compare it with the value from the coupler budget table
From the coupler budget table of the same month (the one showed in the introduction section), the value is 0.36133891, equals to the value calculated from the MOSART component. The check is passed!