I have a multi-year archive of satellite scenes covering the same AOI, and I need a clear, defensible yearly change detection study. The focus is strictly on satellite imagery analysis—no aerial photos or LiDAR—so please bring your expertise with tools such as Google Earth Engine, ArcGIS Pro, QGIS, or a Python stack (rasterio, GDAL, NumPy). Scope • Pre-process each yearly image (cloud masking, atmospheric or BRDF correction, and precise co-registration). • Run a consistent change-detection workflow that highlights where and how the land surface has altered from one calendar year to the next. • Generate intuitive visual outputs and quantitative summaries that a non-GIS stakeholder can understand. Deliverables 1. Cleaned, clipped yearly image composites (GeoTIFF). 2. Change-detection rasters for every year-to-year interval plus a cumulative change map. 3. Vector layers or CSV tables detailing change magnitude and area by class if you derive classes. 4. A concise PDF note explaining methods, algorithms, and any thresholds used. Acceptance criteria • All layers must be in the spatial reference I provide. • No missing pixels in the AOI after cloud removal. • Change results replicate when your script is re-run on the same data. Include a short note on the algorithm you plan to use (e.g., image differencing, post-classification comparison, or machine-learning-based approaches) and a link or screenshot from at least one similar project so I can gauge fit.