The present study was conducted in the Northern Ethiopia, the most ancient town, Axum to detect urban expansion of for the last 3 decades and predict the trend of urban expansion for the year of 2025. To do so, multi-temporal satellite images of 1985, 2000 and 2015 with other datasets were considered and analyzed to scrutinize level of urban expansion. Maximum likelihood algorithm of supervised classification was applied for each Landsat satellite images to generate land use and land cover map of the study area using ERDAS Imagine software. The levels of accuracy of these classifications were evaluated using confusion matrix to derive the overall accuracy. The finding of this study indicated that in the base year (1985), the largest LULC was occupied by agriculture covered, 60% of the entire area in 1985, 48% in 2000 and 38% in 2015, while the urban area increased by 10%, 19% and 39% in 1985, 2000 and 2015, respectively. The decline of the agricultural area was the fast conversion of the agricultural land into the urban areas. The vegetated area was 19% in 1985, 16% in 2000 and 20% in 2015, and also barren lands were possessed 11% in 1985, 17% in 2000 and 3% in 2015 of the entire study area. Even though many changes have observed among the LULC in the year between 1985 and 2015, the highest rate of changes was observed in urban land which increased by 9.9% in every year and vegetated areas by 0.1% per year. Besides, agricultural and barren lands are also decreased in size by 1.18% and 2.38%, respectively, in 2015 when compared with 1985. Therefore, the built-up area was under fast growing currently and some of the causative factors include population growth of the town and rural to urban migration. Finally, the forecast urban expansion of Axum Town for the year, 2025 was found 1497.9 ha Therefore, urban planning authorities and urban planners should think about the future impact of horizontal urban expansion and population growth of the town in line with possible infrastructures development.