Addition of dynamic root scheme (Fan et al 2017; Bieri et al 2025)

[bieri2]

These changes incorporate a dynamic root scheme (DynaRoot). A summary of the changes are below.

Four new output variables:
EASYXY ! ease function (unitless)
ROOTACTIVITYXY ! root activity function (unitless, ranges from 0-1)
INACTIVEXY ! number of timesteps without active roots (-)
MAXUPTAKEXY ! bottom of deepest root water uptake layer (m)

Two new submodules:
DynaRootGrowDieMod: Determine "growth" or "death" of roots based on root water uptake profile
SoilMatricPotentialMod: Compute soil matric potential that will be used for soil moisture transpiration factor and dynamic root calculations

One new namelist option:
Opt_Root: root scheme option (0->off; 1->Fan2017; Bieri2025 dynamic root scheme)

The publication describing the root scheme implementation is linked here (see section 2.2.3): Bieri et al. (2025)

Note that the calculation of the maximum depth of root water uptake (D_RWU) has been updated to be slightly different from what is described in the publication linked above. It is now simply the bottom of the deepest layer with active root water uptake at a given time step.

[bieri2]

Adding plots from tests using a small 4 km grid over the state of Rondonia, Brazil (southern Amazon). Output extracted at the model grid point closest to -10.1866 N, -61.8725 W (location of an Ameriflux site). This point is classified as evergreen broadleaf vegetation.

Three cases are shown in the plots below:

• v5.2-dr-gw: Noah-MP version 5.2 with the root scheme activated and the MMF groundwater scheme activated
• v5.2-nodr-gw: Noah-MP version 5.2 with the root scheme turned off and the MMF groundwater scheme activated
• v5.2-dr-nogw: Noah-MP version 5.2 with the root scheme activated and RUN_OPT = 3 (free drainage)

Accumulated water balance fluxes (UGDRNOFF, SFCRNOFF, QRFS, QSLAT, ACC_PRCP, ACC_ETRAN, ACC_EDIR, ACC_ECAN). Note that ACC_PRCP, ACC_ETRAN, ACC_EDIR, and ACC_ECAN may be underestimated since they are output as accumulations over a time step (and then accumulated again for the plot), but I only have the model set to write output once a day.

10-day running means of surface/canopy fluxes (ETRAN, ECAN, EDIR)

Water balance storage terms (only for v5.2-dr-gw) (SOIL_M, CANLIQ, CANICE)

10-day running means of sensible and latent heat fluxes (HFX, LH)

[cenlinhe]

Thanks for the test results. Looks like dr is doing the right thing and gw has stronger impacts than dr. Am I interpreting your results correctly?