Lateral transfer of C / N / P compounds

In the context of the ESM2025 project we started a new "task force" on the merge of existing developments around the lateral transfert of C / N / P compounds in ORCHIDEE with the objective to have all developments in a unique version (the Trunk) to allow ESM simulations (with the coupling between land - ocean - atmosphere)

• Persons involved: Pierre Regnier (UCL), Goulvain Laruelle (UCL), Haicheng Zhang (UCL), Nicolas Vuichard (LSCE), Philippe Ciais (LSCE), Philippe Peylin (LSCE), Rony Lauerwald (Inrae), Bertrand Guenet (ENS),
• Coordination (in charge of organising regular meeting on this topic): Pierre R. and Nicolas V.

First meeting 13 December 2021

Present: Pierre Regnier (UCL), Goulvain Laruelle (UCL), Haicheng Zhang (UCL), Nicolas Vuichard (LSCE), Philippe Ciais (LSCE), Philippe Peylin (LSCE), Rony Lauerwald (Inrae), Bertrand Guenet (ENS)

We mainly discussed what are the existing developments linked to lateral transfers of C / N / P compounds (organic and inorganic): Several branches / versions exists:

• ORC-PEAT: a peatland model version that is provides crucial input for the transfert of compounds in the arctic
• ORC-CN-Vertical (reference as ORC-V3, a tag of the TRUNK): Content the N cycle and the vertical discretisation of C in the soil (following that of MICT); this version is currently coupled to the ESM
• ORC-Vertical-SOC/DOC/POC/Priming/: used by Rony ???
• ORC-DIC/mangrove : developped by Simon Bowring and currently re-used for a Mangrove project by Bertrand
• ORC-Peat-MICT-SOC/DOC/DIC : developed by Chunjing and tested over a few sites

=> Several versions exists and given the need to have a "lateral transfer module" that can work with different ORCHIDEE version for different objectives the proposed strategy is :

• Develop a "transport routine" as independant as possible of the production in the sols of different compounds
• Need to see if the transport module can be "de-coupled" from the "water routing.f90 routine" ?
• Start the coupling with ORCHIDEE-CN (V3) so that it can first meet the needs of ESM2025
• Have soon a meeting with the groups developping the water routing scheme and the energy transfer (Temperature calculation)

=> First technical meeting will take place on the 16 December in Paris with Nicolas, Pierre, Haicheng => Pierre will organise the next teleconf. => We need to add few other people in the discussion (Daniel G., Chunjing, Elodie, ...)

Meeting 16 December 2021

Present: Ronny Lauerwald, Pierre Régnier, Nicolas Vuichard, Haicheng Zhang

Meeting about the ORCHIDEE version used for the river N model developed within the ESM2025 project.

Two presentations given by Ronny and Haicheng are attached

The ORCHIDEE version used within ESM2025 is ORCHIDEE-v3 (formerly ORCHIDEE-CN). For the development to be done in ESM2025, Nicolas will create a branch named ORCHIDEE-v3-lat, starting from a copy of the branch ORCHIDEE-v3. Rony, Haicheng, Nicolas will have access to this branch for possible development.

The minimal requirements in terms of N input for the river N model are : dissolved NO3, dissolved NH4, and DON. Within ESM2025, we will start with this minimal set. Modelling of PON might be envisaged later on.

• Dissolved NO3 and NH4

About dissolved NO3 and NH4, we will first use what is available from the nitrogen_dynamics routine, ie a non-vertically discretized amount of diss. NO3 and NH4. A transfer function will be developed to disentangle dissolved N from runoff and drainage. IPSL may develop within ESM2025 a vertically-discretized scheme of the mineral N dynamics within the soil that might be used later.

• Dissolved organic nitrogen

About DON, currently, there is no ORCHIDEE version which computes DON. A version further named ORCHIDEE-MICT-LEAK has been developed by Ronny and Bertrand to model a vertically discretized distribution of DOC (Ronny will check this again). This version does not model Nitrogen but only Carbon. Alternatively, the version of stomate_Soicarbon implemented by Bertrand within ORCHIDEE-v3 is inherited from ORCHIDEE-MICT and simulates the vertical distribution of SOM (Carbon and Nitrogen) but not dissolved forms.

The most suitable option seems to be to use ORCHIDEE-MICT-LEAK version of stomate_soilcarbon and to extend it to account for Soil organic N and DON. Nicolas will ask to Bertrand whether the replacement of the MICT version of stomate_soilcarbon by the MICT-LEAK version is relevant in ORCHIDEEv3 or if one should only add the MICT-LEAK version, keeping also the MICT version of stomate_soilcarbon. If needed, we will organize together with Bertrand, Rony, Haicheng and Nicolas a meeting early 2022 for agreeing on the use of the different stomat_soilcarbon versions. In general, Ronny is interested to contribute to the incorporation of dissolved forms in ORCHIDEE-v3 and Haicheng is interested in the development of vertically resolved N soil module.

• Issue with spinup

FORCESOIL routine is needed for all vertically discretized versions of stomate_soilcarbon. To check if FORCESOIL in ORCHIDEE-v3 has to be updated for working with the MICT-LEAK stomate_soilcarbon.

• Water and elements (C and N) routing

A version of routing exists which transports Carbon: version ORCHIDEE-Clat We will start the development required within ESM2025 from that version of routing. Haicheng will replace the routing scheme of ORCHIDEE-v3 by the one from ORCHIDEE-Clat and route the N species. Later on and thanks to the position hired at ULB, N processes will be simulated and a modular and offline version of the routing of elements will be developed. This needs some brainstorming with ULB, P. Peylin, Jan Polcher and Agnès Ducharne, to identify the best strategy, especially about the spatial resolution to which the module should work (basin level or grid-cell ?)

• Other action to be done

Nicolas will ask for opening an account for Haicheng at CCRT supercomputer

Meeting 16 February 2022

Meeting on DON modelling into ORCHIDEE

Present: Bertrand Guenet, Goulven Laruelle, Ronny Lauerwald, Pierre Regnier, Nicolas Vuichard, Haicheng Zhang

Several ORCHIDEE versions have been developed over the last years, which account for a vertical discretization of SOM pools.

• ORCHIDEE-MICT-LEAK developed by Ronny and Bertrand, accounts for DOC but does not model Nitrogen pools (only Carbon);
• ORCHIDEE-PEAT developed by Chunjing Qiu, which also accounts for PEAT was merged by Chunjing into ORCHIDEE-MICT-LEAK to come up with a version with PEAT and DOC but no routing of DOC called ORCHIDEE-MICT-PEAT-LEAK;
• ORCHIDEE-v3 developed by Bertrand based on ORCHIDEE-MICT, which simulates the vertical distribution of SOM (Carbon and Nitrogen) but not dissolved forms.

Summary of actions

We agree on:

• starting from the version ORCHIDEE-v3.
• Including the DOC modelling from ORCHIDEE-MICT-PEAT-LEAK into ORCHIDEE-v3
• Defining DON from a DON/DOC stoichiometry equals to the SON/SOC stoichiometry (as starting point)

Pierre proposes that Minna Ma, who will start a post-doc at ULB this Spring (April), may perform this merge / update of code, under the supervision of Nicolas, Bertrand, Ronny.

Haicheng will then help with the routing of the N species along the inland water network.

Later on, we will also implement Haicheng’s work on erosion and POC, which will then be extended to PON.