Estimation of carbon storage in plantations is an alternative strategy to mitigate climate change. The carbon storage estimation model was developed by building a biomass allometric model with specific predictors. Utilization of cajuput plantations is harvesting the leave-twigs, which is different from the utilization of commons plantation forests that emphasize on timber forest products. Therefore, the objective of the research was to develop an allometric model of cajuput leave-twigs biomass. The study was conducted in KPH Yogyakarta on cajuput plants aged 23, 27, 31, 33, 40, 41, and 43 years. 210 samples were used to develop allometric models, and 60 samples were used as model validation materials. Leave-twigs biomass of cajuput (Y) was estimated with the predictors of the canopy height of cajuput (X1), mean quadratic of canopy diameter (X2), and mean quadratic of stem diameter (X3). The results showed that the allometric model of leave-twigs biomass of cajuput was Y = 3,597X1
0.961X2
0.145X3
0.247. The results of the model validation test resulted in R2
adj value of 0.428; the aggregate deviation was 9.76%; the mean deviation was 0.50%; the bias was - 2.16%, and RMSE was 0.47%. The allometric model was less accurate in predicting leave-twigs biomass of cajuput with the predictors of canopy height, canopy diameter, and stem diameter. Furthermore, it is important to build an allometric model of leave-twigs biomass by adding other predictors.