How Thermal conductivity is calculated in the code

[MostafaGomaa93]

Hi all
@yijianzeng @bobzsu @EntingTang @Crystal-szj @Yunfei-Wang1993

I have a problem. I have high peaks for the simulated soil temperature as compared to the observations (see below). I am trying to lower those simulated peaks, and I am thinking to do so by lowering the thermal hydraulic conductivity. I checked the code and the thermal hydraulic conductivity is calculated here. The two if conditions are valid, meaning Lambda_eff(i, j) = ETCON(i, j) and ETCON(i, j) is calculated in calculateSoilThermalProperites. I think ETCON(i, j) is calculated in Line 65, however, it is a bit hard to follow the equations and know what is what. Could you please @yijianzeng explain a bit about those calculations and where their reference is? Others are also very welcome to share their thoughts

Note: The soil moisture in this site is low, and I can imagine it influences the increase of the soil temperature. So, I am looking for a way (like: a parameter) that can help to reduce the thermal conductivity (within the same low soil moisture). I noticed that the Chung and Horton 1987 equation in Line 44, which is based on the soil texture and soil moisture, is not used.

[EntingTang]

Hi Mostafa. I learnt that your simulation overestimated the soil temperature at all the soil layers. As shown on your attached figure that compares the soil temperature at 20 cm depth, I can see that your simulations are not only overestimated at the noon but also underestimated at night. Personally, I never touched thermal hydraulic conductivity so I am sorry that I cannot help you on this part.

From my experience: before you tune anything (or say you already have the correct soil parameters), I think first you can plot the simulated soil surface temperature from SCOPE model, i.e., Tsave calculated from ebal.m and compare Tsave with your observed soil temperature at the surface or at the nearst surface. Then you can see if this deviation is rooted from the simulated surface temperature (i.e., the enery balance module).

[MostafaGomaa93]

I think first you can plot the simulated soil surface temperature from SCOPE model, i.e., Tsave calculated from ebal.m and compare Tsave with your observed soil temperature at the surface or at the nearst surface. Then you can see if this deviation is rooted from the simulated surface temperature (i.e., the enery balance module).

Hi @EntingTang. Thanks for the suggestion. Indeed, it looks like the simulated soil temperature from SCOPE is the issue, in particular, as you can see from the figure below, the simulated sunlight soil temperature is too high and doesn't make sense I think. However, I still can't figure out what could be the reasons for such issue

[Crystal-szj]

Hi @MostafaGomaa93, Could you please try t to change the value of rbs in the input_data.xlsx, which will change the simulated soil boundary resistance see here and further impact the calculation of soil temperature.

Additionally, do you have the longwave upward and downward radiation data? If yes, please try to retrieve the land surface temperature (LST) based on the Stefan-Boltzmann law:
$$LST = (\cfrac{LWup - (1-\varepsilon)*LWdown}{\varepsilon * \sigma})^{\frac{1}{4}}$$, where LWup is longwave upward radiation, LWdown is longwave downward radiation, $\varepsilon$ is emissivity of land surface (normally, 0.99 for vegetation and 0.96 for bare soil). $\sigma$ is the Stefan-Boltzmann constant.

Please plot LST, observed soil temperature (3 cm), simulated soil temperature (3 cm), air temperature, and simulated canopy temperature (Ts1, Ts2, and Tsave) in the same figure for further discussion.

[MostafaGomaa93]

Could you please try t to change the value of rbs in the input_data.xlsx, which will change the simulated soil boundary resistance see here and further impact the calculation of soil temperature.

Yeah I try to change the rbs but doesn't affect the soil temperature

Additionally, Please plot LST, observed soil temperature (3 cm), simulated soil temperature (3 cm), air temperature, and simulated canopy temperature (Ts1, Ts2, and Tsave) in the same figure for further discussion.

Okay, here it is

[MostafaGomaa93]

One note I want to highlight
The soil moisture measurement at the Loobos site is very low. In one of my trials, there was a wrong rise of the groundwater head (see 1st figure below), then the head went close to the land surface, and the soil was saturated from shallow soil thickness. The soil moisture increased to the saturated value (0.18, see 2nd figure below), and the soil temperature dropped (see yellow circle in the 3rd figure) to be closer to the observed values.

However, when I correct the groundwater head, soil moisture became low (close to observed values), but temperature got higher than observed values

[bobzsu]

Can you replot the component temperatures for the couple of days as well as the heat fluxes? Prajwal and I have discussed the parameterizations for the different resistances. There seems some inconsistences in many places.

[MostafaGomaa93]

Can you replot the component temperatures for the couple of days as well as the heat fluxes? Prajwal and I have discussed the parameterizations for the different resistances. There seems some inconsistences in many places.

Hi Bob, Is this figure clear or needs any clarification?

[bobzsu]

HI Mustafa, these look not bad. If you have the observed fluxes, please plot them. Also choose one week in the summer when it is hot and plot the temperatures. This gives more insights.

[MostafaGomaa93]

[MostafaGomaa93]

One week plot

[MostafaGomaa93]

We checked the images together in the group meeting and agreed that I would rerun the same model after the implementation of the new resistance scheme by Prajwal. The new resistance scheme is expected to change the sensible and latent heat fluxes and consequently will affect the soil temperature profile