for reserves to ROOT if conc too great (overflow!)

[gelianxing]

IF (((LFWT+GROLF-SENLFG-SENLFGRS) ! Prevent divide by zero
& +(STWT+GROST)+(RSWT+GRORS)) .GT. 0.0)THEN
TVR1 = ! Conc
& (RSWT+GRORS-SENRS)/
& ((LFWT+GROLF-SENLFG-SENLFGRS)
& +(STWT+GROST)+(RSWT+GRORS))
ELSE
TVR1 = 0.0
END IF

SENLFG、SENLFGRS why these variables use yesterday’s variables，maybe can move to integr

[chporter]

This is in CER_Growth.for, line 737. TVR1 uses values SENLFG and SENLFGRS which are calculated later in the routine. Maybe this TVR1 calculation could be moved later in the routine to use today's senescence values. I don't think it should go to integrate because it is still a rate calculation.
@palderman @GerritHoogenboom @fabiooliveira72 - if you have time, please have a look.

[palderman]

Linking directly for convenience:

dssat-csm-os/Plant/CERES-Wheat_Barley/CER_Growth.for

Lines 737 to 745 in 185324f

	IF (((LFWT+GROLF-SENLFG-SENLFGRS) ! Prevent divide by zero
	& +(STWT+GROST)+(RSWT+GRORS)) .GT. 0.0)THEN
	TVR1 = ! Conc
	& (RSWT+GRORS-SENRS)/
	& ((LFWT+GROLF-SENLFG-SENLFGRS)
	& +(STWT+GROST)+(RSWT+GRORS))
	ELSE
	TVR1 = 0.0
	END IF

Just did a quick search and we'd need to be careful about where we move it, because TVR1 is used to calculate RTWTGRS just below:

dssat-csm-os/Plant/CERES-Wheat_Barley/CER_Growth.for

Lines 748 to 763 in 185324f

	IF(TVR1.LT.0.0.AND.TVR1.GT.-1.0E-07) TVR1 = 0.0
	IF (TVR1.GT.RSPCX/100.0) THEN ! If potential>max
	TVR2 = RSWT+GRORS-SENRS ! What rswt could be
	TVR3 = ! What rswt should be
	& ((RSPCX/100.0)
	& *(LFWT+GROLF-SENLFG-SENLFGRS
	& +STWT+GROST))
	& /(1.0-(RSPCX/100.0))
	RTWTGRS = (TVR2 - TVR3)
	! Determine FINAL new concentration
	IF (LFWT+GROLF+STWT+GROST.GT.0.0) TVR5 =
	& (RSWT+GRORS-SENRS-RTWTGRS)/
	& ((LFWT+GROLF-SENLFG-SENLFGRS)
	& +(STWT+GROST)
	& +(RSWT+GRORS-SENRS-RTWTGRS))
	ENDIF

and RTWTGRS is then used subsequently:

dssat-csm-os/Plant/CERES-Wheat_Barley/CER_Growth.for

Line 779 in 185324f

RTWTG = AMAX1(0.0,(CARBOR+RTWTGRS)*(1.0-RRESP)) ! LAH 280211

dssat-csm-os/Plant/CERES-Wheat_Barley/CER_Growth.for

Line 884 in 185324f

RTRESP = (CARBOR+RTWTGRS) * RRESP

The dependency graph seems to balloon rather quickly after that. Would it be easier to move senescence calculations earlier?

[chporter]

@palderman Thanks for looking. Yes, maybe. But it might be circular logic and we just need to live with the fact that this is a daily model.

[gelianxing]

I agree move senescence calculations earlier