MPTABLE.TBL CBIOM values

[atmosmattbrewer]

Hello,

I am using the Noah-MP for WRF reanalysis simulations, in previous simulations we had noticed that the CBIOM values of 0.02 gave extreme cold biases and "zombie fog" in the Pacific Northwest. We changed the values to 0.002 similar to: https://www2.mmm.ucar.edu/wrf/users/workshops/WS2025/presentations/6_Lin.pdf , and found this gave more accurate surface temperature.

I wanted to know if there are updated parameter values for the different vegetation classes? We are open to testing new values.

Thank you!

[cenlinhe]

Thank you for reporting this. We are planning to tune down the default 0.02 value for CBIOM to a smaller value (e.g., 0.002) based on some other tests as well. Currently, we do not have vegetation type-dependent values for CBIOM to test but you may want to optimize these values independently by analyzing your results across different vegetation type categories to derive a set of vegetation type-dependent parameter values.

[cenlinhe]

@tslin2 We may need to update the default CBIOM 0.02 to 0.002. Could you please provide the obs-contrained/derived justification from your earlier Atlanta presentation slide to show that 0.02 is actually too large? Thanks!

[tslin2]

yes, the simple rule of thumb (slide 9, https://www2.mmm.ucar.edu/wrf/users/workshops/WS2025/presentations/6_Lin.pdf) of calculation of areametric heat capacity of biomass [J/m2/K] can be calculated as follows:
forest biomass per area in the Southeast US is about 10 [kg/m2] (Gu et al., 2019) and multiply by the specific heat capacity of biomass either dry wood: around 1700 [J/kg/K] (White 2003) or moist biomass: 2000 ~ 3500 [J/kg/K] depending on water content, this would lead to around 2e04~4e04.

Noah-MP’s formula of areametric heat capacity of biomass in evergreen needleleaf forest with plant area index, 4.4 from the NoahMP parameter table, NoahmpTable.TBL is
0.02 (CBIOM) * 4.4 (plant area index) * 4.188e06 (volumetric heat capacity of water) ~ 3.69e05

So 2e04~4e04 is around 10 times lower than 3.69e05, so changing the default CBIOM from 0.02 to 0.002

I think the 0.02 maybe is okay for tropical forests such as the Amazon with dense and high biomass broadleaf forests (although I haven't tested it) but not for temperate forests in the midlatitudes. Grass, crops, deserts, tundra, savannas, and shrubs also have lower biomass. You may refer to this aboveground biomass carbon density figure a in https://www.nature.com/articles/s41597-020-0444-4/figures/4

Future work will be needed to enhance this.

[scappsphd]

So, for other vegetation types, the CBIOM should be approximated. For less dense/lower SAI/LAI types including grasslands, shrublands, and savannas I expect 0.002 to be overestimated:

VEGTYP = 7 and 9 biomass per area [kg/m2]: ~2 kg/m2 -> 2 * ~2000 = ~4e03
CBIOM (0.02) * pant area index (1.2) * volumetric heat capacity of water (4.188e06) = 1.00512e05
4e03/1.00512e05 = 0.039796
CBIOM (0.0008) * pant area index (1.2) * volumetric heat capacity of water (4.188e06) = 3.350e03 to 4.020e03

VEGTYP = 10 biomass per area [kg/m2]: ~0.8 kg/m2 -> 0.8 * ~2000 = ~1.6e03
CBIOM (0.02) * pant area index (0.6) * volumetric heat capacity of water (4.188e06) = 5.0256e04
1.6e03/5.0256e04 = 0.032
CBIOM (0.0007) * pant area index (0.6) * volumetric heat capacity of water (4.188e06) = 1.8e03

I am testing NoahMP in the western United States and will try these values as well as others. I am seeing a dramatic warm nighttime minimum temperature bias in the summer and early fall (especially pronounced when temperatures dip below 35 degrees F). This is mostly sensitive to soil moisture but, hopefully, these CBIOM values could explain a small fraction of that as well. So, if I understand correctly, if the biomass heat capacity is overestimated, the vegetation temperature will have more "inertia" to external temperature swings.

[cenlinhe]

this has been fixed in this PR: #219