This study investigated biomass allocation in young stands of European beech(Fagus sylvatica L.)and Norway spruce(Picea abies(L.)Karst.)across 31 forest sites in the Western Carpathians,Slovakia.A total of 541 trees a...This study investigated biomass allocation in young stands of European beech(Fagus sylvatica L.)and Norway spruce(Picea abies(L.)Karst.)across 31 forest sites in the Western Carpathians,Slovakia.A total of 541 trees aged 2–10 years,originating from natural regeneration and planting,were destructively sampled to quantify biomass in four components:foliage,branches,stems,and roots.Generalized non-linear least squares(GNLS)models with a weighing variance function outperformed log-transformed seemingly unrelated regression(SUR)models in terms of accuracy and robustness,especially for foliage and branch biomass.When using height as the predictor,SUR models tended to underestimate biomass in planted beech,leading to notable underprediction of aboveground and total biomass.Biomass allocation patterns varied significantly by species and regeneration origin.Using a non-linear system of equations and component ratio modelling,we found out that planted spruce displayed low variability and a consistent dominance of needle biomass,while naturally regenerated beech showed greater variability and a higher proportion of stem biomass,reflecting stronger competition-driven vertical growth.Interspecific differences in total biomass were more pronounced when using tree height,with spruce generally exhibiting greater biomass than beech at equivalent heights.Overall,stem base diameter marginally outperformed tree height as a predictor of biomass.However,tree height-based models showed strong performance and are particularly suitable for integration with remote sensing applications.These findings can directly support forest managers and modellers in comparing regeneration methods and biomass estimation approaches for early-stage stand development,carbon accounting,and remote sensing calibration.展开更多
基金funded by the grant“EVA4.0”,No.Z.02.1.01/0.0/0.0/16_019/0000803 supported by OP RDE as well as by the projects APVV-19-0387,APVV-22-0056,and APVV-23-0293 from the Slovak Research and Development Agencyco-funded by the European Commission under the Horizon Europe Teaming for Excellence action+1 种基金project Ligno Silvagrant agreement No.101059552。
文摘This study investigated biomass allocation in young stands of European beech(Fagus sylvatica L.)and Norway spruce(Picea abies(L.)Karst.)across 31 forest sites in the Western Carpathians,Slovakia.A total of 541 trees aged 2–10 years,originating from natural regeneration and planting,were destructively sampled to quantify biomass in four components:foliage,branches,stems,and roots.Generalized non-linear least squares(GNLS)models with a weighing variance function outperformed log-transformed seemingly unrelated regression(SUR)models in terms of accuracy and robustness,especially for foliage and branch biomass.When using height as the predictor,SUR models tended to underestimate biomass in planted beech,leading to notable underprediction of aboveground and total biomass.Biomass allocation patterns varied significantly by species and regeneration origin.Using a non-linear system of equations and component ratio modelling,we found out that planted spruce displayed low variability and a consistent dominance of needle biomass,while naturally regenerated beech showed greater variability and a higher proportion of stem biomass,reflecting stronger competition-driven vertical growth.Interspecific differences in total biomass were more pronounced when using tree height,with spruce generally exhibiting greater biomass than beech at equivalent heights.Overall,stem base diameter marginally outperformed tree height as a predictor of biomass.However,tree height-based models showed strong performance and are particularly suitable for integration with remote sensing applications.These findings can directly support forest managers and modellers in comparing regeneration methods and biomass estimation approaches for early-stage stand development,carbon accounting,and remote sensing calibration.
