On the Detection of Snow Cover Changes over the Australian Snowy Mountains Using a Dynamic OBIA Approach

This study detected the spatial changes in Snow Cover Area (SCA) over Snowy Mountains New South Wales, Australia. We applied a combination of Object-Based Image Analysis (OBIA) algorithms by segmentation, classification, and thresholding rules to extract snow, water, vegetation, non-vegetation land covers. For validation, Maximum Depths (MSDs) were collected at three local snow observation sites (namely Three Mile Dam, Spencer Creek, Deep Creek) from 1984 2020. Multiple Landsat 5, 7, 8 imageries extracted daily MSDs. The process was followed applying an Estimation Scale Parameter (ESP) tool build variance (LV) object heterogeneity for each satellite scene. By matching required segmentation parameters, optimal separation step image objects weighted bands Digital Elevation Model (DEM). In classification stage, few cover classes initially assigned, different indices—Normalized Differential Vegetation Index (NDVI), Surface Water (SWI), Normalized (NDSI)—were created. These indices used adjust thresholds ruleset functions. resulting MSDs all proves noticeable reduction trends during period. SCA classified maps, with overall accuracy nearly 0.96, reveal non-significant trends, although considerable fluctuations past 37 years. variations concentrate north south-east directions, some extent similar pattern year. Although long-term are not significant, since 2006, maximum values has decreased, fewer wet dry episodes. A preliminary analysis climate drivers’ influences on MSD variability also been performed. dynamic indexing OBIA indicated that continuous processing images is effective method obtaining accurate spatial–temporal information, which critical managing water resources other geo-environmental investigations.