Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

CLM5 crop yield

xgao304

Member
Dear Scientist,

I would like to compare the crop yield from CLM5 with some observations (i.e. from USDA). I am wondering what variable is the best to use.
Is "GRAINC_TO_FOOD" the right one if I generate the pft-based output? I am running compset "I2000Clm50BgcCrop" with GSWP3 data.

Thanks,

Xiang
 

oleson

Keith Oleson
CSEG and Liaisons
Staff member
I believe that is correct. See section 2.3 of this paper:

Lombardozzi, D.L., Y. Lu, P. Lawrence, D.M. Lawrence, S. Swenson, K.W. Oleson, W.R. Wieder, and E.A. Ainsworth, 2020: Simulating transient crop management in the Community Land Model version 5, J. Geophys. Res., doi: 10.1029/2019JG005529.
 

Han Lubin

Han Lubin
Member
Dear scientists,

I am running the same compset "I2000Clm50BgcCrop" with GSWP3 data as xgao304. According to grain yield calculation in the CLM50_Tech_Note, I have got the result of GRAINC_TO_FOOD with 0.5° resolution. I am confused with the global spatial patterns of individual crop yields, such as maize or soybean. How can I get the spatial patterns of individual crop yields from GRAINC_TO_FOOD or CLM5.0? Do I need to extract it from the global crop distribution map by myself?

Thanks,
Han
 

dll@ucar_edu

Danica Lombardozzi
New Member
By default, CLM simulations only output grid-cell averaged data (also only monthly averages). To get yields for individual crop types, you need to specify that you would like PFT-level data in the user_nl_clm file prior to running your simulation. Here is an example of how I would add 'GRAINC_TO_FOOD' as a monthly-averaged PFT-level variable in the user_nl_clm:

hist_mfilt = 1,1
hist_nhtfrq = 0,0
hist_fincl2 = 'GRAINC_TO_FOOD'
hist_dov2xy = .true.,.false.

Note that the hist_mfilt and hist_nhtfrq variable specify the temporal frequency and averaging (in this example, for both the default output, or h0 files, and the PFT-level output). hist_fincl2 is the list of variables that you would like written out a different time or spatial resolution (these variables will be on a different file stream, e.g., 'h1') and hist_dov2xy specifies whether the variables are averaged by the gridcell (.true.) or provide sub-grid (e.g., PFT-level) information (.false.).
 

Han Lubin

Han Lubin
Member
Hi Danica,

Thank you very much for your reply! I ran into some problems after setting user_nl_clm as you said:
(1) How should I set PFT-level data in the user_nl_clm file prior to running my simulation? e.g. rainfed temperate corn (PFT=17).
(2) I tried to analyze the effect of soil-water stress on corn yield, so that I turned off the irrigation process of crop in the user_nl_clm file (irrigate = .false.). Is there any difference in the simulation effect between use_hydrstress=.true. (Medlyn2011) and use_hydrstress=.false. (Ball-Berry1987)? What conditions do the two models of plant hydraulics stress apply to?
(3) I can't find anything about crop water stress in CLM50_Tech_Note. Is it that crops considered as general vegetation and share the same water stress process with natural vegetation?
(4) I thought of one question temporarily. I want to simulate the global corn yield from 1991 to 2010, so I create a new case with the compset I2000Clm50BgcCrop and the res hcru_hcru. Will the yield be any different between one task with continuously simulation 20 years from 1991 to 2010 or 20 tasks independently simulation each year, eg. 1991, 1992, ..., 2010. If there are some differences, is it because the successive simulations change the initial surface field, e.g., vegetation distribution, or is it due to the change in soil carbon after crop harvest?

Thanks in advanced!
Han
 

dll@ucar_edu

Danica Lombardozzi
New Member
Hi Han,

I will try to answer your questions below:
  1. You cannot specify history output for only a single PFT. The information I included above (Jan. 14, 2022 post) will write the 'GRAINC_TO_FOOD' as a PFT-level vector (including all PFTs) in the *h1* history files. You can add additional variables to this stream if you would like additional PFT-level variables. You can also find additional information about writing PFT-level variable here in the CLM Users' Guide
  2. Turning off irrigation is fine. Note that this flag only affects irrigated crop areas, so if you are looking at a single point without any irrigated crops, you are unlikely to see any differences. The use_hydrstress option turns on or off the plant hydraulic stress option (if false, CLM will use the older soil moisture stress function to calculate water stress) and does not change the stomatal conductance calculation. You can find out more about CLM's representation of plant hydraulic stress in the technical documentation, and read Daniel Kennedy's paper comparing the two methods of calculating plant water stress. If you want to toggle stomatal conductance methods, change this variable in the namelist file: 'stomatalcond_method'.
  3. Crop water stress is calculated the same way that natural vegetation water stress is calculated. In CLM5, the default is to use plant hydraulic stress.
  4. In theory, there should not be a difference between running one continuous simulation or 20 one-year simulations. However, we recently identified that this may not be the case for tropical crops, including tropical corn, due to the dependence of phenology triggers that tend to trigger planting in December or January (around the start of the year) that may or may not be reset properly. I've not investigated this, but keep in mind that it may influence your simulations.
Hope these answers help!

Danica
 
Top