基于气候敏感因子的黑龙江省红松人工林松塔产量模型构建OA

[Objective]This study aimed to develop a climate-sensitive cone yield model for Pinus koraiensis plantations,clarify the spatial variation patterns of cone yield across Heilongjiang Province,and elucidate the climatic response mechanisms governing cone production,thereby providing a scientific basis for precision management of timber-nut multifunctional Korean pine forests.[Method]Based on cone yield data from 24 permanent plots of Korean pine plantations in the Linkou Forestry Bureau(2004-2012)and Mengjiagang Forest Farm(2008,2013-2023),along with corresponding regional climate data,correlation analysis was conducted to identify climate-sensitive periods and key climatic indicators influencing cone production of Korean pine.A zero-inflated negative binomial mixed-effects model was subsequently developed to estimate individual-tree cone yield,with year included as a dummy variable,and graphical analysis was employed to validate the effects of explanatory variables and the reliability of the model.[Result](1)The reproductive development of Korean pine cones can be divided into three consecutive stages:differentiation and overwintering of floral primordia,initiation and overwintering of young cones,and subsequent cone enlargement leading to physiological maturation.Correlation analysis indicated that site index,stand basal area,diameter at breast height,mean maximum temperature during the growing season two years prior,and precipitation in August of the same growing season were the principal factors influencing individual-tree cone yield.(2)The zero-inflated negative binomial model effectively resolved the issues of excessive zeros and overdispersion in the cone production data.By including year as a dummy variable,the model successfully captured interannual variation.After incorporating climatic factors and random effects at the individual-tree level,the model's Akaike Information Criterion(AIC)decreased by 94 and the Bayesian Information Criterion(BIC)decreased by 109,indicating enhanced model performance.Moreover,significant differences were observed in cone production probabilities among mast,poor,and normal years.[Conclusion]The zero-inflated negative binomial model incorporating year as a dummy variable provides reliable estimates of individual-tree cone yield in Korean pine plantations.Site index,stand density,diameter at breast height,mean maximum temperature during the growing season two years prior,and precipitation in August of the same growing season were identified as the dominant factors influencing cone production.The proposed model can be integrated with individual-tree diameter growth models to dynamically predict regional-scale cone yield.