|
Forest ecosystem management is facing increasing problems due to global climate change. Precise simulation and monitoring of forest biomass is a crucial method to elucidate the effects of climate change, understand the growth dynamics of forests, and promote the sustainable management of forests. This paper presents the optimization of the 3-PG stand growth process model using Morris sensitivity analysis and Bayesian parameter optimization algorithm, based on data from the Huitong Long-term Ecological Research Station for Chinese Fir Plantations. Additionally, a web-based fir biomass simulation and visualization system was developed using the optimized model. The results demonstrate that the optimized model has greatly enhanced the precision of the simulation in comparison to the default 3-PG model, achieving an average accuracy of 99.4%. This improvement is particularly relevant for simulating fir biomass. The biomass data for roots, stems, and foliage output by the model can be utilized for biomass visualization by considering the 5% quartile, median, and 95% quartile values to show the range of uncertainty in the model's predictions while highlighting the central tendency of the data. Through the system, users could easily access and analyze the biomass trends of various components of fir. By inputting fundamental information such as climate data and stand characteristics, users could obtain intuitive simulation results for fir biomass. This work involved the optimization of the 3-PG growth process model and the development of a fir growth simulation system using this model. The purpose of this system is to give data and technical assistance for the management of fir resources and the sustainable management of forest ecosystems. At the same time, in order to expand the application scope of the model, the system provides an interface for users to input data for model optimization by themselves, which realizes the simulation of the biomass of different tree species and responds to the dynamic changes of the forest ecosystem in a more comprehensive way
|
