Hi Forqan Khalid Al-Daraji,

To estimate the shape of water bodies using remote sensing, several important indicators and algorithms can be employed. Key indices like the Normalized Difference Water Index (NDWI), Modified NDWI (MNDWI), and the Automated Water Extraction Index (AWEI) are effective for detecting and delineating water bodies in satellite imagery. These indices enhance the contrast between water and land features, making it easier to identify the boundaries of rivers and streams.

For algorithmic approaches, object-based image analysis (OBIA) can be particularly useful, especially when dealing with high-resolution imagery. Additionally, machine learning algorithms such as random forests or support vector machines have been applied successfully to classify water bodies and extract river morphology from satellite data.

However, when analyzing the morphology of rivers and narrow streams, spatial resolution is a critical factor. Landsat imagery provides a 30 m resolution, while Sentinel-2 offers a 10 m resolution. This may still be insufficient for capturing narrow water bodies accurately, especially when land adjacency effects become significant.

For more detailed comparisons of these methods, you might find the following papers helpful:

https://www.mdpi.com/2072-4292/13/1/137

https://www.sciencedirect.com/science/article/abs/pii/S0034425721000158

I hope this information assists you in your research.