The decision tree models have been successfully used for a wide range of classification problems, such as land cover mapping, agricultural studies, soil classification, vegetation, and identifying surface waters. A review of previous studies revealed a lack of evaluation of the J48 decision tree (JDT) model for detection of water bodies from ASTER imagery. Therefore, in this study, a JDT model was developed to identify river water area from the ASTER L1T data, and its performance was evaluated against three well-established classification methods: support vector machine (SVM), maximum likelihood (ML), and minimum distance (MD). The study area was a segment of the Pulai River in Johor State, southern Malaysia. Classification accuracies were statistically assessed using overall accuracy and kappa coefficient. The results indicated that while all four methods were suitable for river water mapping, SVM achieved the best performance, with an overall accuracy of 99.18% and a kappa coefficient of 0.982. The JDT model, applied using ASTER VNIR bands and SWIR band 4, yielded a comparable overall accuracy of 99.08% and a kappa value of 0.980, closely followed by ML (99.04%, 0.979) and MD (98.71%, 0.973). Given the marginal difference of only 0.10% between JDT and the top-performing SVM, combined with JDT's advantages in interpretability and rule-based transparency, the J48 decision tree represents a practical and robust alternative for operational water body mapping. Future efforts to further improve JDT accuracy could incorporate additional explanatory variables alongside ASTER bands or employ satellite imagery with higher spatial and spectral resolution.
Rokni, K. (2026). River Water Mapping from ASTER L1T Data Using a Decision Tree-Based Model. (e107622). Sustainable Earth Trends, (), e107622 https://doi.org/10.48308/set.2026.246510.1217
MLA
Rokni, K. "River Water Mapping from ASTER L1T Data Using a Decision Tree-Based Model" .e107622 , Sustainable Earth Trends, , 2026, e107622. doi: 10.48308/set.2026.246510.1217
HARVARD
Rokni, K. (2026). 'River Water Mapping from ASTER L1T Data Using a Decision Tree-Based Model', Sustainable Earth Trends, (), e107622. doi: 10.48308/set.2026.246510.1217
CHICAGO
Rokni, K., "River Water Mapping from ASTER L1T Data Using a Decision Tree-Based Model." Sustainable Earth Trends, (2026): e107622, doi: 10.48308/set.2026.246510.1217
VANCOUVER
Rokni K. River Water Mapping from ASTER L1T Data Using a Decision Tree-Based Model. Sustain Earth Trends. 2026;():e107622. doi: 10.48308/set.2026.246510.1217
