Amount of carbon fixed, transit time and fate of harvested wood products define the climate change mitigation potential of boreal forest management - A model analysis

by Holger Metzler, Samuli Launiainen, and Giulia Vico

Abstract

Boreal forests are often managed to maximize wood production, but other goals, among which climate change mitigation, are increasingly important. Examining synergies and trade-offs between forest production and its potential for carbon sequestration and climate change mitigation in forest stands requires explicitly accounting for how long forest ecosystems and wood products retain carbon from the atmosphere (i.e., the carbon transit time). We propose a novel mass-balanced process-based compartmental model that allows following the carbon path from its photosynthetical fixation until its return to the atmosphere by autotrophic or heterotrophic respiration, or by being burnt as wood product. We investigate four management scenarios: mixed-aged pine, even-aged pine, even-aged spruce, and even-aged mixed forest. The even-aged clear-cut based scenarios reduced the carbon amount in the system by one third in the first $18$ yr. Considering only the amount of carbon stored in the ecosystem, these initial losses are compensated after $42-45$ yr. At the end of an $80$ yr rotation, the even-aged forests hold up to $31\,\%$ more carbon than the the mixed-aged forest. However, mixed-aged forest management is superior to even-aged forest management during almost the entire rotation when factoring in the carbon retention time away from the atmosphere, i.e., in terms of climate change mitigation potential. Importantly, scenarios that maximize production or amount of carbon stored in the ecosystems are not necessarily the most beneficial for carbon retention away from the atmosphere. These results underline the importance of considering carbon transit time when evaluating forest management options for potential climate change mitigation and hence explicitly tracking carbon in the system, e.g. via models like the one developed here.

Reproducing the manuscript results

Note: This has been tested for WSL2 on Windows 10.

First, create a new conda environment:

conda create --name BFCPM python=3 conda activate BFCPM

Then, it is necessary to install bcg_md2. This package stores the sub-models for the MeanTree, soil and the wood products and allows the computation of transit times of the implemented model.

Installation of bgc_md2

``` git clone --recurse-submodules https://github.com/MPIBGC-TEE/bgcmd2.git cd bgcmd2 ./installdeveloperconda.sh cd ..

```

Install BFCPM

git clone https://github.com/goujou/BFCPM.git cd BFCPM ./install_developer_conda.sh

After installation

All the data and figures for the manuscript have can be reproduced by two notebooks: - notebooks/runsimsforpaperserial.ipynb - reproduces all the simulation data, can take hours to days - NOTE: If the variables pre_spinup_date and sim_data are not changed, then pre-computed (and provided) data as presented in the manuscript will be overwritten. - notebooks/figures_notebook.ipynb - reproduces the figures from simulation data - pre-computed simulation data can be found in - data/pre-spinups/2023-07-25/: Pre-spinup for all simulations - data/simulations/2023-07-26/

Model documentation