Include N2O emission predictions into APSIM-SoilN

[ApsimBot]

Legacy Bug ID: 1166 Author: Guest Val Snow Date: 2010-06-28 15:43:26 +0000 UTC

We propose and provide methodology so that the capability of APSIM to predict denitrification be extended to include predictions  of nitrous oxide (N2O). An important greenhouse gas.

This work has been published in a referred journal paper.

Thorburn, PJ, Biggs, JS, Collins, K, and Probert, ME. (2010) Using the APSIM model to estimate nitrous oxide emissions from diverse Australian sugarcane production systems. Agricultural, Ecosystems and Environment, 136, 343-350.

Details of code (written in Manager) are available on request.

[ApsimBot]

Author: jody.biggs@ccsiro.au Date: 2010-07-09 11:01:00 +0000 UTC

Subject: Include N2O emission predictions into APSIM-SoilN  (DO NOT EDIT THIS:1166)Hi Jody, Thanks for submitting you APSIM Task. In order to better assess these major tasks the Reference Panel are asking for some additional information from Developers. To this end could you please address the following:

Minimum information required about a Proposal from Developers:

1. Short description of the concept or proposed improvements

2. Rationale and Justification

3. Details of the proposed improvements

4. Implications for users

5. Supporting documentation (optional)

Any questions please don't hestiate to contact me.

Regards Chris

Chris Murphy Principal Project Officer (APSIM) Agri-science Queensland Department of Employment, Economic Development & Innovation 203 Tor Street, Toowoomba, Queensland 4350, Australia PO Box 102, Toowoomba, Queensland 4350, Australia t: +61 7 4688 1596 f: +61 7 4688 1193 e: apsim@dpi.qld.gov.au w: www.dpi.qld.gov.au

The APSIM (Agricultural Production systems siMulator) Initiative is an unincorporated joint venture between the Commonwealth Scientific and Industrial Research Organisation, the State of Queensland and The University of Queensland. Website www.apsim.info

    We propose and provide methodology so that the capability of APSIM to predict denitrification be extended to include predictions  of nitrous oxide (N2O). An important greenhouse gas.

    This work has been published in a referred journal paper.

    Thorburn, PJ, Biggs, JS, Collins, K, and Probert, ME. (2010) Using the APSIM model to estimate nitrous oxide emissions from diverse Australian sugarcane production systems. Agricultural, Ecosystems and Environment, 136, 343-350.

    Details of code (written in Manager) are available on request.

[ApsimBot]

Author: murphy Date: 2010-09-09 11:32:00 +0000 UTC

Thorburn etal 2010 Using the APSIM to model N2O emissions from sugarcane. Agric.Ecosys.Environ. 136, 343-350.pdf Size: 359718

[ApsimBot]

Author: murphy Date: 2010-09-09 11:33:00 +0000 UTC

Proposal for change to APSIM.docx Size: 13854

[ApsimBot]

Author: sno036 Date: 2010-10-06 13:33:00 +0000 UTC

I recommend that this proposal be accepted. 

 

The rationale for this includes:

-          The lack of any N2O prediction is a current limitation for APSIM given the current funding focus and the proposed scheme has already been accepted and reviewed (and re-used) but the modelling community

-          The proposed scheme for N2O from denitrification does not have any feedbacks into the N cycling system so cannot possibly degrade any existing simulations

-          The N2O losses from nitrification are sufficiently small that they should have negligible effect on existing simulations but that can readily be tested with the existing validation dataset

-          The validation for the proposed scheme has been published in the reviewed literature

 

The provisos I would like to add are:

-          Provided that the testing of the N2O from nitrification does not cause any issues with the existing validations then use the parameter set in the Thorburn paper as the defaults

-          Include in the documentation and release notes that the N2O prediction has had testing under a limited range of soil/climate and production systems and although that testing produced sensible results there should still be a higher degree of user caution attached to the predictions

-          Externalise the constants/parameters for all rate, temperature, water content, pH sensitivities into the XML to facilitate further testing

-          Clarify/redefine “dnit” as total denitrification and add parameters (depth arrays) of dnit_N2 and denit_N2O

-          [suggest this is discussed by Reference Panel as there is not a really obvious solution] Clarify/redefine “nit_tot” as nitrification flux leaving the nh4 pool, “nit_eff” as the flux arriving in the no3 pool and nit_N2O as the gas lost in the process – all these outputs as depth arrays

 

The comments I would like to add are:

-          There is a significant project over the next three years based out of the GHG Centre in NZ to test/improve our ability to predict N2O losses and this will involve testing a few different nitrification and denitrification algorithms.  My plan is that APSIM is used as a home base for this testing and improvement and this should result in a lot more testing of the whole C/N cycle under higher fluxes and different than much of the existing testing and validation of SoilN2.  I would include the proposed scheme in that testing and that might result in improved parameter values or better confidence in the existing values.

 

Val Snow, 6 Oct 2010

[ApsimBot]

Author: hut104 Date: 2010-11-29 14:39:00 +0000 UTC

The issues raised in the review for this task have been addressed as follows:

 

The provisos I would like to add are:

-          Provided that the testing of the N2O from nitrification does not cause any issues with the existing validations then use the parameter set in the Thorburn paper as the defaults

This has been parameterised in the constants file such that it will have no effect in user runs.

-          Include in the documentation and release notes that the N2O prediction has had testing under a limited range of soil/climate and production systems and although that testing produced sensible results there should still be a higher degree of user caution attached to the predictions

Done

-          Externalise the constants/parameters for all rate, temperature, water content, pH sensitivities into the XML to facilitate further testing

Done  Water content effect on N2:N2O ratios externalised as is the one parameter described by Thorburn et al.

-          Clarify/redefine “dnit” as total denitrification and add parameters (depth arrays) of dnit_N2 and denit_N2O

-          [suggest this is discussed by Reference Panel as there is not a really obvious solution] Clarify/redefine “nit_tot” as nitrification flux leaving the nh4 pool, “nit_eff” as the flux arriving in the no3 pool and nit_N2O as the gas lost in the process – all these outputs as depth arrays

 Done Dnit is total denitrification.  There is also outputs for no3 and nh4 losses from denitrification.  Also, the total and effective nitrification are also available as outputs.