add new task       

Task ID:  998  Barley max_grain_size too large    
Created by hol353 on 2009-08-13 3:31 PM, 3444 days ago
Project:  Module Code
Category:  Bug
Priority:  high
Assigned to:  hol353  
Status:  closed
Apsim Version:7.0
Your Name:
Contact Details:

show inline images      show change history

 comment 30759 posted by hol353 on 2009-08-13 3:32 PM, 3444 days ago 
Done. Will be released in Apsim 7.1
 comment 30757 posted by hol353 on 2009-08-13 3:31 PM, 3444 days ago 
Hi Tim,
I think that provided the phenology is tracking correctly there is no point in changing varieties.
I am more or less relaxed about the possibility of achieving more than 8 t/ha in a "perfect finish" to the season.
Dean, I think we could, as an interim measure reduce the maximum grain size to 0.065 for all barley varieties.

From: Tim McClelland []
Sent: Monday, 3 August 2009 9:28 AM
To: Huth, Neil (CSE, Toowoomba); Hochman, Zvi (CSE, St. Lucia)
Cc: Holzworth, Dean (CSE, Toowoomba)
Subject: RE: Barley problem in YP

Hi All,

Thanks for your further analysis and comments relating to the potential yield of this paddock.

At this point I am still unclear what I tell the client. Has the grain size been altered so that the yield potential is reduced? Am I telling him that we are confident that under these conditions the current yield potential is achievable? Should I try to change the cultivar so that the yield potential may be reduced?

Can you please offer me some advice?

From: Neil Huth []
Sent: Friday, 31 July 2009 9:52 AM
To: Hochman, Zvi (CSE, St. Lucia)
Cc: Holzworth, Dean (CSE, Toowoomba); Tim McClelland
Subject: Re: Barley problem in YP

We have wheat varieties in the current configuration that do have bigger grains.  One of the bigger NZ varieties I recently worked on (>10 t) does get up to 53mg, but it also has a higher grain number I think.  We've never managed to get hartog bigger than about 41mg with +/- irrigation and applied N up to 360 kg N.  So basically, I'm seeing grain size as a cultivar specific thing with size determined via similar mechanisms as for leaves (i.e. cell division during early formation).

More experiments are always good.


Hochman, Zvi (CSE, St. Lucia) wrote:

Hi All,

Clearly barley does not have a 50% higher yield potential than wheat. I'd argue that wheat has a greater yield potential than implied by max grain weight of 0.41. I checked my MScAgr thesis and found that the wheat I grew under irrigation in the Negev desert in 1979 had a 1000 grain weight of 56.3g (se=0.6g) for the unstressed treatment. This was achieved in a cool environment (500m above sea level) with no clouds and produced an 8 t/ha crop. We won't resolve this debate until we conduct such a trial in the Australian wheat belt with a couple of recent high yielding wheat and barley varieties. This might be a good trial for Allan Peake's project.


From: Huth, Neil (CSE, Toowoomba)
Sent: Thursday, 30 July 2009 7:15 PM
To: Hochman, Zvi (CSE, St. Lucia)
Cc: Holzworth, Dean (CSE, Toowoomba); Tim McClelland
Subject: Re: Barley problem in YP

Hi all,

In the wheat model, grains grow from flowering to maturity.  Normally, grain growth from flowering to the start of linear grain fill is about half the rate during linear grain filling.  The relative durations, and filling rates are what determine the final protein content of the grain as grain n filling rate is constant from flowering to maturity.
The values used in the wheat model have been fitted to grain and head growth rate data and seem to catch the trends between flowering-GF and GF-Mat.
The maximum grain size parameter comes also from the experimental data.  95% of final yield at Gatton, for instance, can be determined from grain number.  Final grain size is relatively constant.  Some wheat cultivars do have larger grains and where this is known we put it into the cultivar description.  Some have questioned if this is a pragmatic approach, or a fix that would preclude some responses in very different environments.  I'm yet to find anything better than the concept of a maximum grain size.  In theory, we could use growth rates around flowering to determine amounts of cell division and get a more dynamic maximum grain size, but this is likely to be difficult.
From what I see in the paper, maximum grain sizes are likely to be about 60mg (average for whole ear).  TT for grain filling is comparable to wheat and so grain filling rates should therefore be about 50 % higher (=60/40).  If barley uses similar grain number formulations, then yields should be 50% higher than wheat.  You won't get the 10t but maybe it's closer to the mark.  I'm not an expert on barley yield potentials though.

That's my 2 cents' worth,

Hochman, Zvi (CSE, St. Lucia) wrote:

Hi All,

I agree that max grain size should be reduced from 0.1 perhaps to 0.07 (see attached paper) but I'd need to look closely at the literature. I didn't put that number in there so it must have come from Ahmed. By the way, I'm not sure what the "potential_grain_growth_rate    =     0.0010 (g/grain/day)" does.

The wheat numbers (.002 and .041) are too low and I bet they were put in as a fudge at some time and have never been challenged. Barley does have bigger grains than wheat so I guess this highlights that we need to do the work to determine these numbers in a non limiting environment.

Anyway, I'm not entirely convinced that a yield of 9.5 t/ha is not possible for barley. Looks like growth so far must be ideal and if ( a big if!) the season continues with a 1 in 20 best finish, without disease pest and other problems like frost, heat stress, lodging etc.

So if you are really keen to do something here I'd change the max grain size to 0.07 till we get a better estimate.



From: Dean Holzworth []
Sent: Thursday, 30 July 2009 12:31 PM
To: Hochman, Zvi (CSE, St. Lucia)
Cc: Tim McClelland; Huth, Neil (CSE, Toowoomba)
Subject: Barley problem in YP

Hi Zvi

Tim has alerted me to a barley problem. Here's the graph that isn't believable:

It's the N unlimited line that is too high. I've had Neil Huth have a look at this and we found this in the summary file for an 8.3 t/ha crop (also attached).
     flowering day          =  265                           stover (kg/ha)         =     8262.1
     maturity day           =  313                           grain yield (kg/ha)    =     8311.9
     grain % water content  =   12.5                         grain yield wet (kg/ha)=     9499.3
     grain wt (g)           =    0.073                       grains/m^2             =    11335.3
     grains/plant           =   75.6                         maximum lai            =      5.906
     total above ground biomass (kg/ha)    =    16574.0
     live above ground biomass (kg/ha)     =    16574.0
     green above ground biomass (kg/ha)    =    14090.3
     senesced above ground biomass (kg/ha) =     2483.7
     number of leaves       =     68.5
     DM Root:Shoot ratio    =     0.11                       Harvest Index          =       0.50
     Stover C:N ratio       =    39.31                       Root C:N ratio         =      17.01
     grain N percent        =     1.89                       total N content (kg/ha)=     240.93
     grain N uptake (kg/ha) =   156.87                       senesced N content (kg/ha)=   12.42
     green N content (kg/ha)=    71.64

The grain wt number of 0.073g is way too high. Looking at the .ini file for barley gairdner we see this:

     <tt_start_grain_fill units="oCd">545</tt_start_grain_fill>
     <tt_end_grain_fill units="oCd">35</tt_end_grain_fill>

Neil is wondering why the potential_grain_filling rate is 0.0033 when in the wheat cultivars it is 0.0020. The max_grain_size in barley is also too high at 0.1 (in wheat it is 0.041).

As the barley module owner can you suggest a way forward.