Sun 19 Jul 2020 9:30PM

Implementation of Plantain in OverseerFM

CR Caroline Read Public Seen by 28

This discussion provides information on, and an opportunity to discuss, the way Plantain is to be implemented in OverseerFM


Poll Created Mon 20 Jul 2020 3:48AM

Group discussion on the implementation of Plantain in OverseerFM Closed Fri 31 Jul 2020 5:02AM

We have invited you to this Overseer Science Discussion thread because you have either been involved with or have shown an interest in the implementation in plantain in OverseerFM. The information posted here aims to provide clarity on how we propose to implement plantain in OverseerFM.

The aim of this online discussion forum is to give people an opportunity to read about and then comment on the implementation approach before it goes live. We will attempt to answer questions that relate to the topic, but the forum is open to discussion and so please feel free to post responses.

This group is public and all posts by members are visible. Others can also request to join the group.

About the Topic:

The Foragers for Reduced Nitrate Leaching programme identified how to reflect their research findings in the Overseer model. Overseer Limited have implemented these recommendations and they have been tested.

Three documents are attached with information on this:

  1. The Implementation Document - which sets out the FRNL findings and how they are to be implemented in OverseerFM.

  2. The Test Report that was produced to assess that the implementation is delivering the results that are expected.

  3. A Question and Answer document with some potential questions on the implementation.

The Implementation Document includes links to the FRNL research reports and publications that set out the scientific research behind the new modelling.


Results Option % of points Voters
Undecided 0% 10 GR AW CR AM AR IP IP MS MD

0 of 10 people have voted (0%)


Mike Dodd Tue 21 Jul 2020 1:07AM

The 60% impact threshold has been noted by our Tararua dairy farmers and required some clarification for them. Possibly just need to check over the way it is described in the Implementation text, so it is clear when referring to content in a paddock vs content in the animal intake. it will have implications for how blocks are set up to reflect management. For example a farmer who has a pure plantain paddock but break feeds it to give the stock a 6-7kg allowance each day (c. 40% of intake rather than 100% of intake).


Mark Shepherd Tue 21 Jul 2020 2:54AM

Yes, the model coefficients are modified by proportion of plantain expressed as intake. When considering the model, we were thinking about dilution by supplement but, clearly, the grazing strategy you describe also ends up with a dilution too. will need to check how this works in the model.


Gerald Rys Tue 21 Jul 2020 7:58PM

Has there been any work looking at leaching response to different levels of intake eg intakes during winter or intake at peak lactation at he same proportion of plaintain?

there any impacts on rates of leaching at different times of the year?


Mark Shepherd Wed 22 Jul 2020 1:08AM

There has been no direct measurement of N leaching, the model is based on first principles of effects on urination characteristics. We received no information of differences in effect at different times of the production cycle. When constructing the Overseer model, we did indicate the need for further information on seasonality of plantain and how this might effect leaching reductions. We indicated in the slide set that this might be a future refinement to the model as more information becomes available.


Gerald Rys Wed 22 Jul 2020 1:09AM

Ta Mark

Dr Gerald Rys

Principal Science Advisor

Science Policy

Policy and Trade Group

Ministry for Primary Industries

Charles Ferguson Building

34-38 Bowen Street

PO Box 2526


Ph 64 4894 0711 wk

Ph 0211398348 pvt cell

gerald.rys@mpi.govt.nz ( gerald.rys@mpi.govt.nz )

www.mpi.govt.nz ( http://www.mpi.govt.nz/ )

https://www.researchgate.net/profile/Gerald_Rys ( https://www.researchgate.net/profile/Gerald_Rys )

https://orcid.org/0000-0002-0049-0543 ( https://orcid.org/0000-0002-0049-0543 )

“In times of crisis, facts matter most”


Greg Sneath Wed 22 Jul 2020 4:55AM

If using first principles on urination characteristics, then I assume the modelling will also reflect a reduction in GHG emissions from reduced nitrogen being partitioned to urine. However, as yet, there is insufficient science to attribute any changes to soil nitrous oxide emissions resulting from bioactive compounds of plantain, or effects of dietary intake on enteric methane emissions. Is that correct? Presumably as the science develops, any GHG emissions effects will be introduced? Do you have an estimate for how long this will take?


Mark Shepherd Wed 22 Jul 2020 8:58PM

There is a small effect on nitrous oxide emissions as a result of reduced partitioning of excreta to urine. As more research becomes available on other aspects, these can be captured.


Greg Sneath Wed 22 Jul 2020 9:33PM

Cheers Mark. The proposed implementation looks good and sensible . I support Mark Dodd's comment above about providing clarity for the OverseerFM users on what constitutes up to 60% of dietary intake.


Caroline Read Wed 22 Jul 2020 10:04PM

Thanks Greg and Mike - we will make sure we have some clarification on the 60% for users in FM and the knowledge base.


Alister Metherell Mon 27 Jul 2020 3:36AM

I am wondering what the origin of the equation for partitioning of N between urine and dung is?

ie. (11.9 NutrientDietConcentration 100 + 29.9)/100.0

Was this from studies using just grass clover pastures or was it from a wide range of diets? Given that the plantain adjustment is a maximum factor of 1-0.6/3 = 0.8 is there evidence that the N partitioning factor for a plantain diet is actually significantly different to the original relationship? Alternatively do the recent studies give good support for the original relationship for grass clover pastures and other supplementary feeds used in pasture based systems?


Mark Shepherd Thu 30 Jul 2020 8:58PM

Sorry for the delay. As part of the FRNL programme, we did compare the estimates of measured urine N production and Overseer estimates. There was reasonable agreement when collated against all of the diet types, (plantain, fodder beet and standard pasture), especially given the challenge of measurement. This is reported in the 'integration' report on the FRNL website. I haven't checked that the scalar was significantly different. Of the two factors, this has the smallest effect on N leaching estimates.

As for a more general comment about the Ledgard et al. relationship based on %N in diet, as described above it gave a reasonable fit . However, by applying that scalar, we have introduced indirectly another aspect of feed composition - water soluble carbohydrate - but only for plantain. Longer-term, it may be that another look at that equation is needed and other dietary components included.


Alister Metherell Mon 27 Jul 2020 3:38AM

The Urine patch load equation of 750-(PP*500) is presumably for used for cattle urine patches.

Are there equivalent equations for sheep and deer?


Mark Shepherd Thu 30 Jul 2020 8:58PM

Scalars are applied to N leaching for other species to reflect smaller UPN loads.


Gerald Rys Mon 27 Jul 2020 8:13PM

We recenttly updated the GHG inventory partitioning model.


Gerald Rys Mon 27 Jul 2020 11:13PM

Might be interested in this publication on plantain 'the role of lucerne and plantain on the East Coast of the north island

cheers Gerald



Penny MacCormick Thu 30 Jul 2020 9:06PM

Hi everyone,

Hoping someone can help out with my question?

We have a few landowners in our catchment that have plantain in their pasture and we are interested in gauging the % plantain value to enter into Overseer. Can you please indicate whom is coming up with the recommended sampling process for plantain data collection to enter into Overseer ie present/absence or % plantain in pasture by reference photos or any other method? Is the sampling method defined somewhere or is it still a work in progress?

Nga mihi



Mike Dodd Thu 30 Jul 2020 10:26PM

HI Penny,

I am involved in a bit of the work here. We have a couple of methods that we (AgR, DairyNZ and Massey) have evaluated for estimating plantain content in mixed pasture i.e. grass - based. At the moment we are working on a visual estimation approach which seems to do a good job at the block scale. That is, while there may be some variance in estimate for an individual paddock, at the block scale it's pretty solid. Which will be the number Overseer will likely need. Pip Hedley at DairyNZ will work on a booklet with guidance for the method and reference photos, a bit like Body Condition Scoring.


Penny MacCormick Fri 31 Jul 2020 12:04AM

Thanks Mike - do you know when the above mentioned guidance booklet is likely to become available and is it planned to be available through a hyperlink in OverseerFM? Also is there any planned date for the plantain functionality in Overseer to be activated?

Nga mihi


Caroline Read Fri 31 Jul 2020 1:58AM

Thanks Penny and Mike - once any guidance on this has been made we will look at how to link it to FM. With this discussion ending today, we will discuss on Monday the timeframe for releasing the functionality - we may have a couple of other changes to roll into a release and so will let you know timing as soon as we can. Just in case you had missed it - users can already enter their plantain pastures into FM and once the modelling is released their analysis results will be automatically updated.


Alister Metherell Fri 31 Jul 2020 2:28AM

Have pasture quality differences between plantain and grass/clover pasture been considered? Did the FRNL project provide data which could be used to update the Overseer pasture quality database?


Mark Shepherd Fri 31 Jul 2020 2:45AM

The FRNL team might be able to add further information. They have collated a large database. From this, we concluded that there was no consistent difference in N concentration between standard and plantain-rich pasture, even though the starting hypothesis in FRNL was that there would be a difference. Analysis of the database did pick up the difference in WSC, hence we have captured that in our partitioning scalar. There is potential to use the FRNL database to update the OvS database with other quality parameters (P, K, etc for example).


Alister Metherell Fri 31 Jul 2020 2:51AM

Thanks Mark. I was thinking more in terms of ME differences which are an important driver of intake in Overseer, which in turn has a big impact on N excretion.


Mike Dodd Fri 31 Jul 2020 2:56AM

From the Tararua plantain rollout project sampling over the last 18 months there is no evidence of significant ME differences for grass-based pastures with plantain in them. Plantain-clover mixes are a different story, as one might expect.


Mark Shepherd Fri 31 Jul 2020 3:01AM

Yeah, sorry. I missed the point. The database was unable to show consistent differences in ME between standard pasture and plantain-rich pasture


Barenbrug Fri 31 Jul 2020 4:12AM

Barenbrug's comments are in the attached.