The new crop types in CLM5

[Xueli Huo]

Hi,

I am reading 2.26.2.2. Crop plant functional types 2.26. Crops and Irrigation — ctsm release-clm5.0 documentation

There is a paragraph describing the representation of the sugarcane, rice, cotton, tropical corn, and tropical soy that are new in CLM5.

"Sugarcane and tropical corn are both C4 plants and are therefore represented using the temperate corn functional form. Tropical soybean uses the temperate soybean functional form, while rice and cotton use the wheat functional form. In tropical regions, parameter values were developed for the Amazon Basin, and planting date window is shifted by six months relative to the Northern Hemisphere."

I have no idea what "represented using the temperate corn functional form" or "use the wheat function form" means. If the new crop functional types use the existing function form, what's the sense of representing these five new crop function types in CLM5 ? I am confused and hope anyone could help.

Sincerely,

Xueli

 

[dll@ucar_edu]

Hi Xueli,

New crop types allow for wider spatial coverage of managed crops. Each of the new crop types has their own set of parameters related to phenology and allocation, which can differ from the original functional form that they were derived from. These differences allow the model to more accurately simulate yields of each specific crop type.

There is not a great definition of "functional form" other than classifying some broad similarities across crop types. For example, sugarcane and corn are both C4 crops and therefore have similar characteristics (they grow tall and have similar leaf and stem structures). Similarities like this that are broadly classified by "crop functional forms" are used in the code, such as calculating temperature thresholds for maturity.

Danica

 

[Xueli Huo]

Hi Danica,

Thanks for your answer. It's clear to me now.

Sincerely,

Xueli

 

[ShirleyX]

Hi Xueli,

New crop types allow for wider spatial coverage of managed crops. Each of the new crop types has their own set of parameters related to phenology and allocation, which can differ from the original functional form that they were derived from. These differences allow the model to more accurately simulate yields of each specific crop type.

There is not a great definition of "functional form" other than classifying some broad similarities across crop types. For example, sugarcane and corn are both C4 crops and therefore have similar characteristics (they grow tall and have similar leaf and stem structures). Similarities like this that are broadly classified by "crop functional forms" are used in the code, such as calculating temperature thresholds for maturity.

Danica

Hi Danica,
If I use PFT=21( rainfed winter wheat ),then the planting window will from Sep. to Dec.( northern hemisphere ) and it "Crop Parameters" will use rainfed spring wheat (eg. temperature thresholds , heat accumulated and so on).
Am I correct in my understanding of the crop plant function types in CLM5.0?
Thank you for your help !

 

[dll@ucar_edu]

Hi Shirley,

This is not exactly how the crop model works. All area that is 'winter wheat' on the land use time series file (or the surface dataset if running a time slice) will get merged into the 'spring wheat' crop functional type and therefore simulated as if it is spring wheat. If you'd like to run a simulation with winter wheat, Yaqiong Lu developed code specifically for this crop type that is available in CTSM (see CNPhenologyMod.F90 around line 1880: ESCOMP/CTSM). You will have to update the parameter file so that the winter wheat crop type is no longer merged into spring wheat crop type (I believe this is the 'mergetoclmpft' parameter), and then make sure that the region(s) that you are interested in contain winter wheat crop area on the land surface dataset.

I hope this helps!
Danica

 

[Xueli Huo]

Hi Danica,

As what you said, all area that is 'winter wheat' on the land use time series file (or the surface dataset if running a time slice) will get merged into the 'spring wheat' crop functional type and therefore simulated as if it is spring wheat. However, the fact is that the winter wheat performs differently comparing with the spring wheat. Could you explain the benefit in or reason for merging the inactive management class crop function type into the active management class crop function type, for example merging winter wheat into spring wheat ?

Sincerely,

Xueli

 

[dll@ucar_edu]

Hi Xueli,

Previous to CLM5, all crop area that was not one of the active crop types was simulated as a generic C3 crop, which functioned the same as a C3 grass. The C3 generic crop does not use fertilization or irrigation management, and does not use temperature-triggered phenology or allocation like the active crops. Even though there are inaccuracies in simulating the inactive crop types using one of CLM's active crop types, we think it is more accurate to represent all crop area as managed crop area (fertilization and irrigation) than to represent it similar to unmanaged grassland. We acknowledge that this is an imperfect solution, and are working toward developing functionality to explicitly represent the inactive crop types with help from the community.

Danica

 

[ShirleyX]

Hi Shirley,

This is not exactly how the crop model works. All area that is 'winter wheat' on the land use time series file (or the surface dataset if running a time slice) will get merged into the 'spring wheat' crop functional type and therefore simulated as if it is spring wheat. If you'd like to run a simulation with winter wheat, Yaqiong Lu developed code specifically for this crop type that is available in CTSM (see CNPhenologyMod.F90 around line 1880: ESCOMP/CTSM). You will have to update the parameter file so that the winter wheat crop type is no longer merged into spring wheat crop type (I believe this is the 'mergetoclmpft' parameter), and then make sure that the region(s) that you are interested in contain winter wheat crop area on the land surface dataset.

I hope this helps!
Danica

Hi Danica,

My research area grows winter wheat. I want to simulate winter wheat rather than combine winter wheat into spring wheat. You mentioned the related content in "CNPhenologyMod.F90"(line 1880).I would like to know how to modify the parameter file in this part.Thank you for your help!

Sincerely
Shirley

 

[dll@ucar_edu]

Hi Shirley,

If you would like to simulate winter wheat, you will need to change the 'mergetoclmpft' variable in the parameter file in your simulation so that winter wheat is not merged to spring wheat. You will also need to make sure you have winter wheat on your land use or surface dataset.

Please keep in mind that this code has not been thoroughly tested or evaluated within CLM, so you may find you have to make some changes for simulations to perform well.

Danica

 

[knreddy]

Hi Danica,

My research area focuses on spring wheat and rice, especially in the Indian region. The issue I found while evaluating the pft specific LAI for spring wheat is that the crop phenology variables like LAI and yield are summations of many other crops such as barley, citrus, rye, rapseed, sugarbeet, sunflower, potato, etc. I am attaching a document that shows, in the first two slides, the difference in spring wheat planted area. The crop areas of merged crops are shown in CLM in the following slides.
This has made it difficult to compare the spring wheat yield with FAO observations, and crop phenology. How do I resolve this issue?
As suggested in an earlier comment in this thread, providing land-use data for all the above-mentioned crops would be difficult. How should I proceed with making comparisons with the observations?

Thank you

 

[knreddy]

Hi Danica,

My research area focuses on spring wheat and rice, especially in the Indian region. The issue I found while evaluating the pft specific LAI for spring wheat is that the crop phenology variables like LAI and yield are summations of many other crops such as barley, citrus, rye, rapseed, sugarbeet, sunflower, potato, etc. I am attaching a document that shows, in the first two slides, the difference in spring wheat planted area. The crop areas of merged crops are shown in CLM in the following slides.
This has made it difficult to compare the spring wheat yield with FAO observations, and crop phenology. How do I resolve this issue?
As suggested in an earlier comment in this thread, providing land-use data for all the above-mentioned crops would be difficult. How should I proceed with making comparisons with the observations?

Thank you

I am attempting to replicate the yield plots of spring wheat and rice figure 2 from Lombardozzi et al. (2020) [https://doi.org/10.1029/2019JG005529]

 

[dll@ucar_edu]

I'm sorry that I missed this thread -- it looks like you posted awhile ago! From a separate thread, it looks like perhaps you have already figured out a method to do this. I use the land use time series to apply a mask to the area for the individual crop types. One other thing to note is that I masked all areas that had <0.5% area occupied by a crop in the publication you point to.