Automated Assessment of Inland Flooding

Reflection

Methodology

Firstly, the current approach to extracting major rivers from flowline datasets can be improved. Samuel Roy, a scientist from USGS, suggests that stream order selection is helpful in identifying major rivers. By adding this modification, the flowline mask in the optimization process might be more helpful.
Secondly, the optimization process can be improved by exploring various ways to combine the features (NDWI and flowline) and techniques (StandardScaler and PCA) currently in use. Each combination of these elements can produce different results, and selecting the optimal combination is crucial for enhancing the model's performance.
Thirdly, one of the initial datasets included in this proposal is the Digital Elevation Model (DEM). Because flood event observations often come with elevation values, integrating DEM data can provide valuable insights. However, high-water marks from the STN (Short-Term Network) dataset need to be carefully processed. Some locations may have multiple high-water marks with different elevation values. Discussion with a professional will be necessary to ensure accuracy in the analysis.