A fast technique to delineate watershed boundaries (watershed marching algorithm)
A fast algorithm to delineate watershed boundaries
Researchers in the Department of Computer Science and the Academy of Natural Sciences at Drexel University have developed an algorithm and data model designed to delineate a watershed boundary using a 2D flow direction grid (D8) and data from digital elevation models from the USGS and NASA. The Haag Shokoufandeh Marching (HSM) algorithm creates watershed boundaries by marching along the perimeter of the underlying area being analyzed without entering internal or external area(s), the current modeling standard. The HSM algorithm’s retrieval complexity scales directly from the number of vertices needed to delineate a watershed, rather than the number of grid cells within the interior of the watershed, allowing the HSM algorithm to outperform existing techniques by an order of magnitude. Tests to retrieve the approximately 36,000 km2 Delaware River and 165,000 km2 Chesapeake Bay Watershed resulted in approximately 35 and 184 million read operations using traditional techniques compared to 45 and 150 thousand using the HSM algorithm.
Applications
- Hydrodynamic models
- Flooding, infrastructure, biological, and ecological modeling
- Nutrient and contaminate source tracking
- Insurance, engineering, and military sectors
Advantages
- Lower data storage needs
- Complexity of algorithm proportional to boundary length, rather than area, of underlying region
- Reduce operation complexity for faster read time
- Deploy as Restful API
Intellectual Property and Development Status
United States Patent Pending-11,954,410
References
Haag S. and Shokoufandeh A. A watershed delineation algorithm for 2D flow direction grids. ArXiv, 2017, arXiv:1708.00354.
Haag S. and Shokoufandeh A. Development of a data model to facilitate rapid watershed delineation. Environmental Modelling & Software, 2017, p. 1-13.