Cable pop trendy blaze

[har917]

CABLE

Description

This dev branch includes revisions to facilitate the use of the fire model BLAZE (in conjunction with POP). This requires 3 tasks

• updating the science code of BLAZE to that in the NESP2pt9_BLAZE and BLAZE_9184 branches
• updating coupling code between BLAZE and CABLE-CASA
• adding new code to read BLAZE inputs for the BIOS (and global) configurations which are in netCDF format.

This development work will be done based around the BIOS configuration/ACT9 test case and branches off the CABLE-POP_TRENDY branch. This configuration will use binary formats for many of the BIOS specific inputs as the work to convert these (gridinfo and meteorology read) remains to be completed.

You can link issues by using a supported keyword in the pull request's description or in a commit message:

Fixes #392

Relates/intersects with #232, #322, #391, #393

Type of change

• branch merge

Checklist

• The new content is accessible and located in the appropriate section.
• I have checked that links are valid and point to the intended content.
• I have checked my code/text and corrected any misspellings

Please add a reviewer when ready for review.

[har917]

Edits at 786b3b8 are science neutral. These remove many of the debugging write statements and reduced log files from ~190MB (for a 10 year, ACT9 case) to 27kB.

[har917]

edits to casa_inout achieve two components

1. correct for the dimensionality of variables fluxCtoCO2_plant_fire and fluxCtoCO2_litter_fire (which are npt variables not npt,mplant and npt,mlitter as assumed)
2. facilitate output of the above vars on the grid.

Edits will follow the template set out for the other 1D variables.

[har917]

Edits up to 18695c9 are bitwise comparable in states and fluxes when compared to runs from the branch without any edits. New variables added are

• to cable output: NEP - noting that the NBP is also adjusted (correctly) to include any fire and harvest fluxes
• to casa output: fluxCtoCO2_plant_fire and fluxCtoCO2_litter_fir

Note that any downstream analyses that depend on NBP will need to be re-assess &/or adjusted to use NEP.

[har917]

Science changes to BLAZE internals are required in

• interpolate_fire_mortality in the POP code
• a new section in adjust_POP_for_fire
• changes in the blaze and simfire routines

We also need to check

• how the variables from the casa_dump files make their way into model through casa_cable (line ~510)
• the evaluation of %FluxfromPtoCO2 in casa_cnp (fire flux from plant to atmosphere)
• treatment of fapar_ann_max in climate module

[har917]

71b773d contains changes in the science core of BLAZE and SIMFIRE.

Interestingly applying these changes reduces the impact of simulated fire on the carbon stocks for the ACT9 test case. This comes from both from changes in BLAZE and the changes to SIMFIRE (mainly a resolution of inputs issue).

There are remaining science questions to check with the development community.

EDIT: looking at the code changes - this reversion back towards the no_fire case is expected. The science changes involved need to be partnered with changes inside POP (related to the evaluation of %rkill) WIthout these changes the model does not kill/burn any biomass (but does continue to burn litter).

[har917]

Also interesting to note that the run with the adjustments to simfire was A LOT faster (3.5 minutes vs 8.5 minutes)

[har917]

49a1dce hopefully includes all code changes to bring the science of BLAZE/SIMFIRE in the CABLE-POP_TRNDY branches to be in sync with the NESP2pt9/BLAZE_9184 branch.

There are outstanding science questions that the community will need to resolve.

There are also additional changes around the i/o code that need to be addressed so that the module can be run using a landmask/netCDF for all inputs rather than the binary inputs as currently.