| Dahurian Larch(Larix gmelinii)is the main species of timber forests.By building the stem taper functions of Dahurian Larch(Larix gmelinii),we could understand the change rules of stem,and we also could predict the volumes of stem and bark.Building the additive and compatible systems of Dahurian Larch(Larix gmelinii)could provide the basis for consistently estimating the volumes of stem with or without bark and bark.Therefore,the research on stem taper and bark thickness and their volumes could provide the reliable methods for scientific operation and management.Based on the field measured data of 1281 Dahurian larch trees in Xing'an Mountains,in this paper we synthesized the building methods of additive models at home and abroad,and summarized four types of six methods about additive model building,namely,the single model estimation method,total amount control method(including three methods on the logical relationships about diameter out bark(dob),diameter inside bark(dib),and double-bark thickness(dbt)),total amount decomposition method and adjusted proportion method.And then,we also studied the building process of each method.On the basis of above methods,we studied the predicting approaches of volumes of integral and non-integral models,and built six methods of additive systems of stem taper and bark thickness.After that,we also built six methods of additive and compatible systems of stem taper,bark thickness,stem volumes and bark volumes,which realized the simultaneous estimation of stem taper,bark thickness,and their volumes.Finally,the results were compared and analyzed.The main conclusions are as followed:(1)Among the candidate models of stem taper and bark thickness,Kozak(2004)model was the optimal stem taper function and the optimal bark thickness equation,and their model fitting precision Ra2 were 0.9770 and 0.6704,respectively.Meanwhile,Max and Burkhart(1976)model was the optimal and integrable stel taper functions(Ra2=0.9729).The fitting results of two models were implemented paired-T test,and the result was insignificant.Additionally,bark thickness model(4-1)(Ra2=0.5371)was selected as the basic bark thickness model to build additive systems or additive and compatible systems with the sub-optimal stem taper-Max and Burkhart(1976).(2)Though analyzing Kozak(2004)model and using the multivariable regression technique,we removed the insignificant parameters and variables of the models,and added the dummy variables into the model to explain the difference of three regions and improve the fitting precision of the model.(3)On the basis of newly built Kozak(2004)models of stem taper and bark thickness with with dummy variables,we built six methods of additive systems.Among them,the additive system of stem taper and bark thickness with method five(total amount decomposition method)was the optimal.The fitting results of the additive system was fine,and Ra2s of dob,dib,and dbt were 0.9777,0.9685,and 0.6652,respectively.In addition,their residuals were randomly distributed,and the fitting results and test results were consistent.Especially,during the analysis of segmented test,the additive system was also fine,and all index of dob,dib,and dbt models were stable.(4)On the basis of Max and Burkhart(1976)stem taper function and bark thickness model(4-1)with dummy variables,we built six methods of additive systems.Among them,the additive system of stem taper and bark thickness with method four(total amount control method)was the optimal.The fitting results of the additive system was fine,and Ra2s of dob,dib,and dbt were 0.9740,0.9680,and 0.6532,respectively.In addition,their residuals were randomly distributed,and the fitting results and test results were consistent.Especially,during the analysis of segmented test,the additive system with method four was still optimal.(5)Based on the estimated parameters of six types of additive systems for Kozak(2004)stem taper function and bark thickness model,we shelled scripts with Python to realize the composite integration rule of Gauss-Legendre method,and then we used the parameters and scripts to calculate predicting volumes of dob,dib,and dbt,and built six methods of additive and compatible systems for simultaneously predicting dob,dib,dbt,and their volumes.Among them,the additive and compatible system with method five(total amount decomposition method)was the optimal.The fitting results of the additive and compatible system were fine,and Ra2s of volumes of dob,dib,and dbt were 0.9501,0.9382,and 0.7853,respectively.In addition,their residuals were randomly distributed,and the fitting results and test results were consistent.(6)Based on the additive systems built by Max and Burkhart(1976)stem taper function and bark thickness(4-1)equation,we obtained volume formulas of stem taper and bark thickness by calculating indefinite integral of stem taper and bark thickness equations,and then we also built the six methods of additive and compatible systems for simultaneously predicting dob,dib,dbt,and their volumes.Among them,the additive and compatible system with method four(total amount control method,dbt=dob-dib and dbt_v=dob_v-dib_v)was the optimal.The fitting results of the additive and compatible system were fine,and Ra2s of dob,dib,dbt,and their volumes were 0.9727,0.9667,0.6418,0.9942,0.9315,and 0.7575,respectively.In addition,their residuals were randomly distributed,and the fiting results and test results were consistent.(7)The predicting precisions of two additive and compatible systems had no significant difference,especially,the difference between stem taper and its volume was in 1%.Therefore,in practice,if the application of additive and compatible systems focused on predicting the diameter of stem top and bark thickness,we suggested the additive and compatible system based on the Kozak(2004)model.And yet,if the application of additive and compatible systems focused on predicting the stem volume,merchantable volume,or bark volume,the additive and compatible system based on the Max and Burkhart(1976)model was suggested to use. |
PDF Full Text Request