Estimation of structural forest attributes, such as volume, basal area, and tree density using a combination of remote sensing and field data, is currently considered a favored option compared to only using field survey data. In a comparative study, multiple linear regression (MLR) and classification and regression tree (CART) models were used to estimate volume, basal area, and tree density using advanced space-borne thermal emission and reflection radiometer (ASTER) and satellite poure I’observation de la terre (SPOT)-high resolution grounding (HRG) imagery in the Darabkola forests, located at the Hyrcanian region of Iran. Results showed that the CART model using SPOT-HRG data achieved the best overall performance for all three forest structural attributes, with adjusted and for volume, adjusted and for basal area, and adjusted and for tree density. In general, the CART model, using both ASTER and SPOT-HRG data, produced better estimates of forest attributes compared to the MLR model. In addition, results showed that forest attribute estimations using SPOT-HRG were better than those obtained from ASTER data.