Methods For Predicting Stand Structure,Competition And Growth Dynamics

Forest growth and yield models have been recognized as crucial tools in forest management and planning.As the objective of forest management changes from woody production to ecosystem multifunctionality,reliable simulations of uneven-aged or mixed forest become more urgent needed.However,the complexity of uneven-age or mixed forest has hampered the application of whole-stand models.An ideal option is to use size-class models or tree-level models,but the requirement for data and techniques is strictly regulated.For instance,diameter distribution models have been known as inaccurate,and the construction of tree-level models relies on long-term observations.In addition,the choices predictor variable,especially for describing competition,have been under debate.This study applied methods of biometrics to simulate the stand structure,competition and diameter growth.Various models and indices have been evaluated.The main content and results are as follows:(1)The diameter distributions of pine-oak forests in the Qinling Mountains were analyzed using the Weibull function in terms of forest types,which were classified based on stand density,average diameter and dominant height.Even under varying conditions,the diameter distributions always performed as reversed ‘J' curves,which means that the frequency decreased in larger size-classes.The stand density and average diameter significantly affected the dimeter distributions of pine-oak forests.For the stands of high density or low average diameter,the frequencies in medium and small size were larger.Nevertheless,the impact of dominant height that represents site fertility was negligible.Dominant tree species also affected diameter distributions.Pinus tabuliformis follows normal distributions,while the distibutions of Pinus armandii and Quercus aliena var.acuteserrata were skewed.(2)Diameter distribution models for each species group of pine-oak mixed forests were developed by use of the Weibull function.Both moment and hybrid estimation approaches were used to predict the Weibull parameters.For each approach,three fitting methods(maximum likelihood estimator regression(MLER),cumulative distribution function regression(CDFR)and modified CDFR)were employed to obtain estimates for coefficients of regression equations to predict Weibull parameters.Overall results indicated that the Moment Estimation approach was better than the Hybrid approach,and that the CDFR method was superior to the MLER and modified CDFR methods.The combination of Moment Estimation and CDFR is recommended.The models constructed in this study enable the prediction of the diameter distribution of uneven-aged pine-oak mixed forests in the Qinling Mountains based on common stand-level information.(3)Six distance independent competition indices were evaluated using permanent plots of loblolly pine(Pinus taeda).The competition indices were classified into three families:(1)size ratios,which include diameter ratio(DR)and basal area ratio(BR);(2)relative position indices,which include basal area of larger trees(BAL)and tree relative position based on the cumulative distribution function(CDF);and(3)partitioned stand density index and relative density.Results indicated that different families of competition indices were suitable for different tree survival or diameter growth prediction tasks.The diameter ratio was superior for predicting tree survival,whereas the relative position indices(BAL and CDF)performed best for predicting tree diameter growth,with CDF receiving the highest rank.(4)An individual-tree diameter growth model for Pinus tabuliformis was constructed based on one-time measurements and tree cores from temporary plots.The equivalence tests showed that the predictions matched observations without significant distinctions.The model considered tree size,site quality and competition conditions as the predictor variables.The results illustrated a positive correlation between site quality and diameter growth,and a negative correlation between competition and growth.Tree growth was suppressed for trees in small size classes.In addition,the tree-level competition index,BAL,showed stronger suppression of growth than the stand-level competition index BA.The modelling approach might fix the issue of lacking re-measurements,and provide a new perspective in forest growth forecasting.