Plant Water Uptake threshold in very dry climates

[Baya]

Good morning,

we have an issue regarding modeling of dry environments. In the field, we have very low water contents measured (< 1 vol.-%). When we try to fit the model, we are not able to achieve these low water contents with the model (which we observed in the calibration profiles). We were able to improve it a bit by increasing EpFac, LAI, and some other plant parameters. Ultimately (from my understanding, h_wp (matric potential at wilting point) should fix this issue, but still we don't get lower water contents than 1 vol. - % in the model. Does anybody have an idea how to fix this or how to approach this issue?

Cheers,

Matthias

[Per Abrahamsen]

Hi Baya.

You need to look at both the retention curve and the hydraulic conductivity curve for your soil. As you note, The water content at wilting point (pF 4.2) should be below 1%, but the hydraulic conductivity at wilting point must also be sufficiently high for the roots to be able to extract water from the soil.

You will find both the water content (Theta) and hydraulic conductivity (K) at wilting point for each horizon listed in the daisy.log file.

Regards,

Per A.

[Baya]

Hey Per,

thanks a lot. For implementation, however, it seems difficult, because when using such soils, conductivity will be fairly low at such water contents. Is there a way to constrain the lower end of the conductivity curve? For example, lower than -10000 cm of matric potential use always specified conductivity value? 

The Daisy log says:

h h Theta K

cm pF cm/h

0 0.3567 125

-1 0 0.354755 105.044

-10 1 0.213609 5.04702

-100 2 0.0247976 0.000121013

-1000 3 0.00906207 1.21859e-009

-15848.9 4.2 0.00803856 1.21864e-015

That means below 100 cm there is a very low conductivity. In arid areas, however, certain plants are still able to extract water, and this is what we cannot reproduce so far.

We tried the following:

(hydraulic M_vG (Theta_res .0080 [cm^3 H2O/cm^3])     ;Mualem-van Genuchten model parameters calculated from field sample - BGR (best 0.005)

                    (Theta_sat .3567 [cm^3 H2O/cm^3]) ;(best 0.3567)

                    (alpha .1250 [cm^-1]) ;(best 0.125)

                    (n 2.200 []) ;(best 2.20)

                    (K_at_h -100 0.00012)

                    (K_sat 3000.0 [cm/d])) ;(best 3000)

but K_at_h does not work when using KSAT.

Do you have any idea? Thanks,

Baya

[Per Abrahamsen]

K_at_h and K_sat just scale the hydraulic conductivity curve, i.e. multiply all points with a number to get the specified point right. That is why you can only set one of them.

To change the shape of the curve, you need to specify the l parameter.

Other suggestions to things to try: 

1) Set Theta_res to 0. 

2) Try to use hypres (hydraulic hypres) instead of your field sample parameters.

Regards,

Per