This study aimed at testing different data gathering and adjustment strategies of traditional and generic hypsometric models in order to improve Eucalyptus sp. tree height estimates. Tree height and diameter pairs were colleted in 36 plots, belonging to Veracel Celulose S.A. The data were divided into different treatments according to age, site, and combination of age and site of the stands, totaling 14 treatments. The data gathering strategies consisted in using a different number of trees with height measurement to form the database used for the model adjustments. When considering the data gathering strategies where the first five trees per plot and the first five trees per plot plus the dominants were selected, it was concluded that the best strategies consisted in using the Curtis model adjusted per plot, and the two generic models from Campos & Leite (2006) and Scolforo (2005). When considering the data gathering strategy in which trees were added based on the absence of an individual on a determined diameter quartiles, only the Curtis model per plot was selected.
Height-diameter relationship; forest inventory; traditional and generic models