Computation of root zone water for irrigation demand

[sim.moulds]

Hi,

I have a question about the calculation of root zone water contents used to compute irrigation demand. Sorry if it’s difficult to follow – bear with me!

Variables ws1/2/3, wres1/2/3 are calculated in source_py3/hydrological_modules/landcoverType.py lines 218-224 by multiplying the corresponding volumetric water content, denoted by thetas and thetar respectively, by the depth of each soil layer (rootDepth[0/1/2]), so that the resulting quantity is a depth (e.g. if rootDepth[0] is 0.05m and thetas1 is 0.5, ws1 will equal 0.025m). Variables wrange1/2/3, wfc1/2/3, wwp1/2/3 are computed from ws1/2/3 and wres1/2/3 and hence are also depths which must be some quantity less than the corresponding soil layer depth.

When irrigation demand is computed, in source_py3/hydrological_modules/waterdemand.py lines 300-365, availWaterPlant1/2 is computed as max(0, w1 – wwp1) * rootDepth[0] and max(0, w2 – wwp2) * rootDepth[1]. Variables critWaterPlant1/2 and totAvlWater1/2 are also scaled by multipling by rootDepth[1/2]. So far as I can see (and of course I could be missing something), this means that these variables have dimensions of L^2, when they should have dimensions of L. If I’m right, the values for readAvlWater, totAvlWater and critAvlWater will be incorrect because they have been scaled by rootDepth twice.

As I say, I may well have missed something here, and the units of w/wwp,wfc etc. may not be depths - apologies if this is the case.

Thanks,

Simon

[PeBu]

Hi Simon,

thanks for diving in so deep.

I really appreciate, because that is what we need.

I have to check and dive into the code as well.

Please give me some time!

Cheers,

Peter

[sim.moulds]

No worries Peter - take your time! 

Thanks a lot,

Simon

[Mikhail Smilovic]

Hello Simon,

It seems that you’re probably quite right, and that there is a double scaling of rootDepth!

A first thought is that it should be instead scaled by rootFrac, since it developed just beforehand but not applied.

I’ll work on this early next week and come back to you.

Cheers,

Mikhail

[sim.m...@gmail.com]

Thanks Mikhail.

Yes, maybe that is right. On the other hand, if the goal of irrigation is to bring the water content in the root zone above the critical water content then perhaps the root distribution is irrelevant. This is because readily available water and the critical water content are properties of the soil, not the plant, and regardless of the root distribution the water content in all layers must exceed this point in order for the plant not to suffer water stress. The root fraction then becomes relevant when you model transpiration (all else being equal, more transpiration occurs where the root system is denser).

Let me know what you decide next week :)

Cheers,

Simon

[Mikhail Smilovic]

Simon,

Yes, you're right. Root distribution is relevant when modeling transpiration. 

Peter has removed the scaling with rootDepth, and we'll update the version on GitHub shortly. Thank you!

Cheers,

Mikhail