Growth and Productivity Modeling of Seven Eucalyptus Species in Souiniet’s Arboretum in the Northwestern of Tunisia

This study consists of the development of tree growth models to deduce stands productivity and determine the highest productive species in the conditions of the concerned plantation. Seven Eucalyptus, introduced in the arboretum of Souiniet (north-west of Tunisia, wet Mediterranean bioclimate) in a Cork Oak natural forest, were studied. Stem analysis and non-linear growth modeling regression equations were used to predict wood productivities. Gompertz and Chapman–Richards growth function appeared as being great numerical models to estimate the Eucalyptus tree diameter and height evolutions, respectively. Moreover, an adapted Chapman–Richards model allowed predicting the volume of trees in an efficient manner. The values of mean annual volume productivity of the Eucalyptus spp.studied, allow us to classify them in order of increasing annual productivity, as follows: E. sideroxylon, E. cinerea, E. maidenii, E. macrorhyncha, E. tereticornis, E. viminalis and E. bicostata. The first three Eucalyptus spp. appeared as the best-adapted and most suitable Eucalyptus trees for new plantations in this area. These species had the highest mean annual increments, ranged from 5 to 10 m3.ha−1.year−1 with 15 to 20 years of rotation. E. bicostata is the most promising, with annual average production exceeding 10 m3.ha−1.year−1 after 25 years, and reaching 20 m3.ha−1.year−1 at 40 years old. These modeling approaches provide additional knowledge on the productivity of the different Eucalyptus species, thus enabling forestry operators to simulate the development of forest stands in order to optimize timber production and harvesting.