As a result of record rainfall and floods in the 2006/07 wet season, a number of landslides occurred in the upper reaches of the Magela Creek and East Alligator River, Northern Territory. These landslides occurred on well vegetated, exhumed Oenpelli Dolerite surfaces surrounded by Mamadwerre Sandstone, and had the potential to supply sediment to both Magela Creek and the East Alligator River. Currently the Supervising Scientist Division monitors the Magela Catchment to assess the impact on stream sediment loads by mine-related activities at the Ranger mine. These landslides have the potential to increase the baseline load of fine suspended sediment to Magela Creek during future wet seasons and are not mine related. It is therefore important to quantify the impact and extent of natural events affecting stream sediment loads, to enable them to be differentiated from mine impacts. This study had two objectives: firstly, to evaluate the use of remotely sensed data in detecting landslides; and secondly to determine whether the data extracted can be used to accurately quantify the area of each landslide. Employing an object based classification approach, two ALOS AVNIR-2 multispectral satellite images and a 1 sec DEM were used to identify 42 possible landslides. Assessment of these results was undertaken using data derived from visual assessment of the ALOS AVNIR-2 data combined with aerial surveys (by helicopter) of the study area. This assessment identified a total of 56 landslides across the 88 km2 study area. Thirty four landslides were correctly classified in the study area, while 22 landslides remained undetected and 8 areas were incorrectly classified as landslides. Correctly classifying landslides using an object-based approach was found to be related to the size of the landslide and its associated debri flow. In addition, a GPS was used to measure the areal extents of 11 landslides in the field to assess if the final classified product could be used to accurately determine the area of each landslide. These results showed that while 94 % of the areas covered by the 11 landslides on the ground were captured in the classification there were significant commission errors, due to classification of debri flow as landslides. The results of this study show that object-based classification methods can be used to detect the occurrence of landslides, however it was found that the spatial resolution of ALOS AVNIR- 2 and DEM data was not sufficient to discriminate between landslides and their associated debri flows to enable accurate measurement of landslide area.