A new framework for prognosing forest resources under intensified disturbance impacts: Case of the Czech Republic

Abstract

Forest management decisions increasingly rely on modelling tools, which help identify future risks, optimize management decisions, and provide a suite of indicators beyond timber production. Here we developed and tested a novel simulation and upscaling framework (SUF) and used it for prognosing forest resources of the Czech Republic (Central Europe), which is currently one of Europe's hotspots of disturbance intensification.The SUF is based on an empirical forest model that simulates the development of 8240 forest stands representing forest conditions in 206 administrative districts of the Czech Republic. The effect of natural disturbances is considered via empirical species- and age-specific mortality probability (MP) functions parameterized based on the national forest damage reports and remote sensing data. An upscaling procedure was developed to obtain district- and country-wide estimates. We tested this framework for its ability to (i) reproduce the initial forest conditions from the year 2003, (ii) reproduce forest dynamics in 2003–2016 (i.e., before the recent disturbance wave), (iii) reproduce the recent mortality pulse in 2017–2020, and (iv) generate plausible and consistent outputs under several disturbance and management settings in 2003–2050.The SUF reliably reproduced forest dynamics in both testing periods. The country-wide growing stock (GS) simulated for 2004–2050 oscillated around the initial value of 661 mill. m3 if the reference MP and management were considered. Using the elevated MP (corresponding with the recent disturbance period) increased the mean annual mortality rate from 0.78 % to 1.19 % and caused GS to decrease by 21 % in 2050. The wave of elevated mortality lasted 16 years, ceasing in 2033 due to the depletion of vulnerable stands. Reducing the rotation length by 40 % increased the harvests temporarily and caused GS to decrease by 29 and 33 % in 2050 under reference and elevated MP, respectively. At the same time, mortality was reduced by up to 18 % due to the removal of potentially vulnerable stands.The presented SUF is able to accommodate diverse forestry data, reproduce real forest dynamics, and generate outputs that correspond with the national forestry statistics. Flexible adoption of different mortality and management regimes makes it a versatile tool for supporting management decisions and policies. The presented simulations highlighted the negative prospects of the regional forests and the need for a profound transformation of management practices and the regional forest-based sector.