Changes in land use land cover includes residential development, farmland expansion and exploration of a terrestrial ecosystem that have detrimental effect on the natural environment. In order to detect spatiotemporal land use changes of the OGRB, this study looked into Random Forest (RF), Support Vector Machine (SVM), and K-Nearest Neighbor (KNN) machine learning algorithms. Among these optimal algorithm was assigned to further classification of all available datasets. For this purpose, cloud free Thematic mapper (TM) 1994 data, Enhanced Thematic Mapper Plus (ETM+) and Operational Landsat Imagery (OLI) 2014 and February month 2024 images were freely obtained from Google Earth Engine (GEE) platform. Classification was performed in open source QGIS 3.36.3 machine leaning Dzetsaka Classification Plugins (DCP). Ground truth, Google Earth, and supplementary data were used to assess each period classification through the analysis of confusion matrices in ArcGIS Pro 2.8 to determine Kappa Statistics and Overall accuracy. RF was shown to be the most effective and precise classifier with an overall accuracy of 0.91, outperforming SVM (0.88) and K-Nearest Neighbor (0.75) throughout the 2014 timeframe. Therefore, it was adopted to classify all the remaining 1994, 2004 and 2024 years datasets. In average over 0.91% accuracy was achieved in all dataset classification. The results of this study can be mostly attributed to the increases in agricultural land from 0.8% in 1994 to 30.8% in 2024, built up area from 0.08 in 1994 to 5.6% in 2024 and water body from 0.62 in 1994 to 1.3% in 2024. Mainly in expense of forest from 51% in 1994 to 29% in 2024 and shrub/scrub 37% in 1994 to 33% in 2024. The outcomes of this research can contribute to improving land policy, management and public understanding of land use dynamics in the study area.