search

APSIMInitiative / ApsimX

ApsimX is the next generation of APSIM
http://www.apsim.info
Other
132 stars 161 forks source link

AgPasture science review. #5356

Closed hol353 closed 3 years ago

hol353 commented 4 years ago

It is time for a review of the science in AgPasture. Two things to kick off the list:

peter-devoil commented 4 years ago

Can we also have establishment / killing routines so it works in a rotation?

hol353 commented 4 years ago

@peter-devoil Can you elaborate a bit on this? What do you mean by establishment routines.

The model does respond to a Kill management call. It also implements the plant and organ damage routines so that crop consumers (pest and diseases) can damage a pasture species.

peter-devoil commented 4 years ago

Sowing routines. Perhaps I am missing something, or thinking of classic. One day it would be good to run a crop/pasture rotation.

sarahcleary commented 4 years ago

@jwhish - you may wish to comment?

hol353 commented 4 years ago

I've found a problem in AgPasture that needs fixing.

Soil Arbitrator sends 4 soil N states to a pasture species and expects an estimate back for each on how much N would be taken up. AgPasture does this but always returns the same estimate back to soil arbitrator regardless of the soil state passed to it. The result is that each species will be given full access to all the available N in the soil which means that there is no competition between the species.

When I fix the bug I get a dramatic over prediction in the validation harvested wt: image The blue dots are with the bug fixed. The orange dots are the original validation values. RMSE changes from 2045 kg/ha to 3529 kg/ha NSE changes from 0.57 to -0.27 R2 changes from 0.57 to 0.56

This needs further investigation.

rcichota commented 4 years ago

Sowing routines. Perhaps I am missing something, or thinking of classic. One day it would be good to run a crop/pasture rotation.

@peter-devoil , The model does have a sowing routine now (I had implemented in late versions of Classic). I have run simulations with rotations (and multi-cropping) involving PMF crops and AgPasture (they may need a better look/validation, but they are possible at least). Phenology is very simplified in AgPasture, so germination and early development would benefit of having some development... For pastures being able to run simulations without needing to sow (with the crop already established) can be quite handy, maybe there is room for thought and development in both AgPasture and PMF in this regard...

rcichota commented 4 years ago

Regarding the 4 options to do the water/N uptake (from first comment), I have implemented them, trying to test different ways to do these processes in Apsim. Thus, I probably should have either selected one deleting the rest or make clearer what was going on (I probably have some documentation if needed). AgPasture had a very simple (too simple) way of doing uptake, one of the implementation was to mimic what Plant was doing. The others, I think, were attempts to get rid of KL. I had no time (nor good data) to properly test them, it would be good to check whether it is worth pursuing them (maybe not only for AgPasture) or simply delete and make the code cleaner... Happy to give some help if needed

demlie257 commented 3 years ago

I am using APSIM Next Generation for modelling the pasture (Ryegrass and white clover) and eucalyptus tree growth. I want to extract information on the water and nitrogen stress in both the pasture and tree production systems, and I would like to kindly ask you to give me a hint on the methods/steps to extract such outputs from APSIM Next Generation.

sarahcleary commented 3 years ago

@HamishBrownPFR; @her123, @sno036 -are you able to help @demlie257?

MarkLieffering commented 3 years ago

from the documentation (https://apsimnextgeneration.netlify.app/modeldocumentation/ and for example the pdf for ryegrass https://apsimdev.apsim.info/ApsimX/Releases/2021.04.15.6139/AGPRyegrass.pdf) choose the appropriate output e.g. GlfNSupply Gets the growth limiting factor due to soil N availability or GlfWaterSupply Gets the growth limiting factor due to water deficit. Then just output these in report e.g. [AGPRyegrass].GlfWaterSupply or [AGPRyegrass].GlfNSupply

MarkLieffering commented 3 years ago

Also check out the AgPasture example simulation for more detailed examples of outputs

sno036 commented 3 years ago

All the outputs (and inputs, parameters, ...) can be found in https://apsimnextgeneration.netlify.app/modeldocumentation/ then go to the "Params/Inputs/Outputs" column and open the PDF. In AgPasture species the stresses all start with "Glf" so you will find them together in the middle of Page 11.

@hut104 can you point the stress outputs for Eucs?

@demlie257 - to help us, can you say where you initially looked for this information but couldn't find it? We do attempt to provide options for new users but one thing that we don't know is where new users look for information.

demlie257 commented 3 years ago

Thank you all a lot for providing me your relevant feedback for my enquiries.

Kind regards,

Demlie

On Wed, May 5, 2021 at 6:16 AM Val Snow @.***> wrote:

All the outputs (and inputs, parameters, ...) can be found in https://apsimnextgeneration.netlify.app/modeldocumentation/ then go to the "Params/Inputs/Outputs" column and open the PDF. In AgPasture species the stresses all start with "Glf" so you will find them together in the middle of Page 11.

@hut104 https://github.com/hut104 can you point the stress outputs for Eucs?

@demlie257 https://github.com/demlie257 - to help us, can you say where you initially looked for this information but couldn't find it? We do attempt to provide options for new users but one thing that we don't know is where new users look for information.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/APSIMInitiative/ApsimX/issues/5356#issuecomment-832214523, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR4FSF5OUH57C75SEFXP7ULTMBIZRANCNFSM4OPIHAIA .

demlie257 commented 3 years ago

Dear all,

I was simulataing N and Water stress in APSIM, and I obtained different values in the range between 0 and 1. What does ‘0’, ‘0.45’ and ‘1’ values in water and nitrogen stress simulation result in APSIM?

Kind regards,

Demlie

On Wed, May 5, 2021 at 5:26 PM Demlie Gebrsellassie Zelelew < @.***> wrote:

Thank you all a lot for providing me your relevant feedback for my enquiries.

Kind regards,

Demlie

On Wed, May 5, 2021 at 6:16 AM Val Snow @.***> wrote:

All the outputs (and inputs, parameters, ...) can be found in https://apsimnextgeneration.netlify.app/modeldocumentation/ then go to the "Params/Inputs/Outputs" column and open the PDF. In AgPasture species the stresses all start with "Glf" so you will find them together in the middle of Page 11.

@hut104 https://github.com/hut104 can you point the stress outputs for Eucs?

@demlie257 https://github.com/demlie257 - to help us, can you say where you initially looked for this information but couldn't find it? We do attempt to provide options for new users but one thing that we don't know is where new users look for information.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/APSIMInitiative/ApsimX/issues/5356#issuecomment-832214523, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR4FSF5OUH57C75SEFXP7ULTMBIZRANCNFSM4OPIHAIA .

demlie257 commented 3 years ago

Dear all,

When simulating cow weight gain in APSIM, the cow weight doesn't not sometimes increase as proportional to increased supplementary feeding instead the cow weight decreases over time. Is there any mechanism related to climatic or any other conditions for that cow weight is not increasing regardless of the amount of supplementary feeding?

Kind regards,

Demlie

On Wed, May 12, 2021 at 2:28 PM Demlie Gebrsellassie Zelelew < @.***> wrote:

Dear all,

I was simulataing N and Water stress in APSIM, and I obtained different values in the range between 0 and 1. What does ‘0’, ‘0.45’ and ‘1’ values in water and nitrogen stress simulation result in APSIM?

Kind regards,

Demlie

On Wed, May 5, 2021 at 5:26 PM Demlie Gebrsellassie Zelelew < @.***> wrote:

Thank you all a lot for providing me your relevant feedback for my enquiries.

Kind regards,

Demlie

On Wed, May 5, 2021 at 6:16 AM Val Snow @.***> wrote:

All the outputs (and inputs, parameters, ...) can be found in https://apsimnextgeneration.netlify.app/modeldocumentation/ then go to the "Params/Inputs/Outputs" column and open the PDF. In AgPasture species the stresses all start with "Glf" so you will find them together in the middle of Page 11.

@hut104 https://github.com/hut104 can you point the stress outputs for Eucs?

@demlie257 https://github.com/demlie257 - to help us, can you say where you initially looked for this information but couldn't find it? We do attempt to provide options for new users but one thing that we don't know is where new users look for information.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/APSIMInitiative/ApsimX/issues/5356#issuecomment-832214523, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR4FSF5OUH57C75SEFXP7ULTMBIZRANCNFSM4OPIHAIA .

MarkLieffering commented 3 years ago

with regard to the stress (GLF) out outputs, as described in the AgPasture example ReportExample:

"// outputs that indicate what conditions are limiting growth - each growth-limiting factor is 0 (full stress) to 1 (no stress) but there are two components and the values are added so 2 indicates no stress"

So 0 is full stress and 1 is no stress - these are for each of the stresses (water, temperature etc.) and multiplied with the growth without stress - so a value of 0 for water stress and 1 for the other stresses for example will give no growth (0 1 1 * x = 0)

demlie257 commented 3 years ago

Thank you very much for the feedback, and the clarification made.

Kinde regards,

Demlie

On Wed, 12 May 2021, 2:38 pm Makushuzan, @.***> wrote:

with regard to the stress (GLF) out outputs, as described in the AgPasture example ReportExample:

"// outputs that indicate what conditions are limiting growth - each growth-limiting factor is 0 (full stress) to 1 (no stress) but there are two components and the values are added so 2 indicates no stress"

So 0 is full stress and 1 is no stress - these are for each of the stresses (water, temperature etc.) and multiplied with the growth without stress - so a value of 0 for water stress and 1 for the other stresses for example will give no growth (0 1 1 * x = 0)

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/APSIMInitiative/ApsimX/issues/5356#issuecomment-839427071, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR4FSF3X4GHJU22PEJDRA23TNIA2XANCNFSM4OPIHAIA .