Opened 4 years ago

Closed 4 years ago

# Bug in soil heat capacity

Reported by: Owned by: fwang somebody major Physical processes trunc heat capacity

### Description

Problem: The term (1-mcs(jst)) is missing when computing the soil heat capacity (pcapa, pcapa_tmp) in thermosoil.f90.

pcapa_tmp(ji, jg) = so_capa_dry_ns(jst) + water_capa * tmc_layt(ji,jg)/mille/dlt(jg)

The dry soil heat capacity values in ORCHIDEE is from Pielke [2002, 2013].
The dry soil density defination in current code and in Pielke [2002, 2013] is not consistent.
The description of the equations in ORCHDIEE:
http://forge.ipsl.jussieu.fr/orchidee/attachment/wiki/TracTickets/problems_in_soil_heat_capacity_equation.pdf

A reference by MICHAEL C. MCCUMBER AND ROGER A. PIELKE, JGR, 1981:
http://forge.ipsl.jussieu.fr/orchidee/attachment/wiki/TracTickets/Mccumber%26Pileke1981.pdf

The values and equations given by Pielke:
http://forge.ipsl.jussieu.fr/orchidee/attachment/wiki/TracTickets/Pileke3.tiff
http://forge.ipsl.jussieu.fr/orchidee/attachment/wiki/TracTickets/pielke4.tiff

Solution:

so_capa_dry_ns(jst) should become so_capa_dry_ns(jst) * (1-mcs(jst)) in many places. For example,
pcapa_tmp(ji, jg) = so_capa_dry_ns(jst) * (1-mcs(jst)) + water_capa * tmc_layt(ji,jg)/mille/dlt(jg)

### Changed 4 years ago by fwang

problems_in_soil_heat_capacity_equation

### comment:1 Changed 4 years ago by fwang

(1) In the equation (3') in above domucment (The description of the equations in ORCHDIEE):
Cv,t = (1-theta) * Cv,d + theta * Cv,w
should be
Cv,t = (1-theta_s) * Cv,d + theta * Cv,w
It has been corrected in the document:
http://forge.ipsl.jussieu.fr/orchidee/attachment/ticket/410/problems_in_soil_heat_capacity_equation.pdf

(2) Sincerely thanks Anne Verhoef (U. of Reading) for spending time on this problem.

Fuxing

### Changed 4 years ago by fwang

test of corrected capacity in offline model

### comment:2 Changed 4 years ago by fwang

The tests were done for:
(1) Corrected capacity
(2) Corrected capacity and corrected conductivity (http://forge.ipsl.jussieu.fr/orchidee/ticket/406)

All results can be found: https://forge.ipsl.jussieu.fr/orchidee/attachment/ticket/410/Capacity.pdf

16/11/2017 by Fuxing W.
-- A lower capacity (after correction) leads to a lower surface temperature (0.8 maximum).
-- The impacts of correcting capacity (decrease Ts) and correcting conductivity (increase Ts) on Ts are in opposite direction.
-- The impacts of capacity correction on sensible heat flux (around 6W/m2 maximum) are higher than latent heat flux.

17/11/2017 by Catherine O.:
The plots demonstrate indeed that the corrections of heat capacity and thermal conductivity compensate mostly each other. The residual differences concern the dryer areas and winter seasons and are always
lower than 1K for the surface temperature and a few W/m2 for H and LE on the seasonal means. Since Frederique has shown that in coupled mode, the differences are even smaller (in fact, on the atlas presented in https://vesg.ipsl.upmc.fr/thredds/fileServer/IPSLFS/fabric/lmdz/MultiSimu/THERMPROP/BIASGLOBJJA.html, I can not manage to see any differences between the simulations ! ), I suggest that we committ the two corrections.

Fuxing

### comment:3 Changed 4 years ago by jgipsl

Done in the trunk rev [4768]

### comment:4 Changed 4 years ago by jgipsl

• Resolution set to fixed
• Status changed from new to closed
Note: See TracTickets for help on using tickets.