Structural model of sandalwood (Santalum album) regeneration in the forest and community plantation in Timor Island, Indonesia

Seran YN, Sudarto, Hakim L, Arisoesilaningsih E. 2018. Structural model of sandalwood (Santalum album) regeneration in the forest and community plantation in Timor Island, Indonesia. Trop Drylands 2: 41-47. Sandalwood (Santalum album L.) is a very important forest product in NTT, an endemic species in the world with a high economic value.. This study aimed to identify and produce a structural model of sandalwood regeneration in both the forests and the community plantation in the Regency of Timor Tengah Selatan (TTS) and Timor Tengah Utara (TTU). The method used in this research was vegetation analysis by purposive sampling method on 8 observation stations with 87 plots. The plot size was 20x20 m2 (trees), 10x10 m2 (poles), 5x5 m2 (saplings), and 2x2 m2 (seedlings). Data observed in the field included the mean sandalwood population size in the forms of trees, poles, saplings and seedlings phase, vegetation data in sandalwood habitat which included tree wealth index, diversity index, number of individuals and sandalwood host diversity index data. Geographical factors such as altitude and slope, and abiotic factors such as soil organic matter, soil pH and soil conductivity were also recorded. Climate data included the number of dry months and rainfall. Sandalwood regeneration data included sandalwood vitality, pests and diseases and the number of seeds. Secondary data included climate data (ten years time) obtained from BMKG of NTT Province in Kupang. These data were used as the indicators of the latent variables (six variables) which consisted of geography, soil, climate, population, vegetation, and regeneration. Obtained data were subjected to both descriptive analysis and multivariate statistics with structural modeling of Warp Partial Least Square (WarpPLS 6.0). The results showed that most of the proposed indicators significantly influenced the compiled six latent variables except the host diversity. Some indicators significantly or highly significantly affected the latent variable with 15 indicators that significantly composed the latent variable. The resulting structural model is very relevant and has a relevance value of Q2 prediction of 96,65% so that the structural model proposed in this study has very relevant and high predictive value on factors that influence sandalwood regeneration. Therefore, this model is feasible or appropriate to be used as recommendations in the framework of sandalwood development in the forest and the community plantation in the West part of Timor Island, Nusa Tenggara Timur.