Allometric Equation for Aboveground biomass estimation of selected indigenous tree species namely, Balanites aegyptiaca, Piliostigma thonningii, Terminalia macropetra, and Vitellaria paradoxa was conducted in Gambella National Park, Southwestern Ethiopia. The allometric equation that was obtained for the four species had significant (P< 0.000) fit for linear model. Selection of fourty-eight (48) tree sampled numbers based on their Diameter at Breast Height (DBH) and height distribution using preferential sampling. The selection of the best model was based on lowest AIC, RSE and P values and highest R2.The best model for predicting above ground biomass for Balanites aegyptiaca, model 6, Ln(AGB= -4.5429+2.7029 ln(DBH) + 0.7704ln(H)+0.2702ln(ρ ) having adjusted R2=0.97and p-value of 2.554X10-07.For Piliostigma thonningii, model 9,Ln (AGB)=-1.9544 + 2.8207ln(DBH) +0.2358ln(H)was best model with, adjusted R2=0.964 and p-value of 1.861X10-08.. Similarly, for Terminalia macroptera, model- 15, Ln(AGB)=-5.8474+ 3.4216ln(DBH)+0.4712ln(H)+ 0.2261ln(ρ), with adjusted R2=0.985 and p-value of 0.0006379 and for Vitellaria paradoxa, model 20, Ln(AGB= - 8.6865+5.1368ln(DBH) + 0.6006 ln(ρ)+ 0.9559ln(H) was the best model with ,adjusted R2=0.981and p-value of 0.0003608. The models fit regression of all selected species and R2 were > 80%. Comparison of these results with those obtained using generalized allometric model revealed differences with biomass estimation. The results ofthis studyor the newly proposed allometric models also significantly improved the estimation accuracy as compared to the previous developed equations.