Height-diameter models with mixed effects for Lysiloma latisiligutum (L) Benth. in Quintana Roo, Mexico

The aim was to determine the better allometric model with mixed effects that describes the total height-normal diameter (A-d) relation for Lysiloma latisiliquum trees in Quintana Roo, Mexico. The slenderness index (IE: A/d) and six local models were adjusted from 2214 pairs of A-d data obtained from the National Forest and Soil Inventory (2004-2009); to improve the fit and after the selection of the best model, the covariates conglomerate, canopy diameter (dc), clean stem height (Af), municipality (M), primary vegetation type (V-p) and altitude, were included under the mixed effects models approach. The IE decreases as the dimensions of the trees increase. The Hossfeld's model (1822) with the inclusion of the conglomerate covariate in the parameter ct o proved to be the best for describing the A-d, since it explains more than 70% of the sample variability and has a bias of 0.00013 m. Furthermore, when comparing the estimates with data from an independent sample, no significant differences were observed at a 99% level of reliability. The insertion of the different environmental and growth conditions modifies the allometric relationship A-d in Lysiloma latisiliquum in the tropical forests of Quintana Roo, Mexico, and the mixed models showed to be an option of statistical improvement and precision in the estimation.