Construction of Carbon Storage Models for Pinus Yunnanensis and Pinus Armandii Based on Additive Models
The forest ecosystem plays a crucial role in the global carbon balance as a large carbon storage reser-voir in China's terrestrial natural ecosystem.Therefore,the study of carbon storage in forest systems is of great sig-nificance for addressing global climate change issues and ensuring global sustainable development.This study se-lected a field survey sample plot in Dayao County,Yunnan Province,with Pinus yunnanensis and Pinus armandii as the research objects for field investigation.Based on the field survey data,the biomass and carbon storage of Pinus yunnanensis and Pinus armandii were measured.Three carbon storage estimation models were combined with additive aggregation method to estimate the carbon storage of Pinus yunnanensis and Pinus armandii and compare them with the measured values,verify the applicability of the models used for predicting carbon reserves of Pinus yunnanensis and Pinus armandii,and compare the effectiveness and accuracy of different models.Through research,it has been concluded that in the prediction and evaluation of carbon storage in Pinus yun-nanensis,the carbon storage model established using biological conversion coefficients has strong predictive abili-ty:MAB is between 1.013 and 2.683,MPB is between 4.353 and 9.855,MAB is below 5,and MPB is below 10.The model has good predictive ability for carbon storage in various parts of Pinus yunnanensis;The average error(E)is within 4%between 1.225%and 2.272%,and the data is stable and close to the measured value.In the prediction and evaluation of carbon storage of Pinus armandii,the forest carbon storage model established u-sing stock volume has strong predictive ability:MAB is between 1.685 and 2.196,MPB is between 6.560 and 9.120,MAB is below 5,MPB is below 10,and the model has good predictive ability for carbon storage of various parts of Pinus armandii.The average error(E)is within 4%between-3.783%and 3.934%,and the data is stable and close to the measured value.
NETL
NSTL
万方数据
维普
