Two new methods applied to crown width additive models: a case study for three tree species in Northeastern China

Abstract Key message The non-linear seemingly unrelated regression mixed-effects model (NSURMEM) and generalized additive (GAM) were applied for the first time in crown width (CW) models of larch ( Larix gmelinii Rupr.), birch Betula platyphylla Suk.), poplar Populus davidiana Dode). radii four directions (CR) exhibited different growth trends responded differently to tree size competition variables. In absence calibration, GAM was more accurate than NSURMEM CR CW predictions. Context Crown are fundamental indicators used describe crowns. complexity species natural forests is often ignored. There logical additivity among that also overlooked. Furthermore, existing methods have some drawbacks. Aims We aim to: (i) evaluate utility two new developing Rupr . ), Dode) secondary Northeastern China; (ii) explore patterns gain important ecological insights. Methods develop patterns. predictive ability evaluated using leave-one-plot-out cross-validation (LOOCV). Results At a fair level without provided slightly better results NSURMEM. response variables may be non-uniform due complex stand conditions strategies. Conclusion newly available external datasets. recommended calibration. This study has implications understanding forest dynamics decision-making critical management.