Benchmarking the Indian National CartoDEM against SRTM for 1D hydraulic modelling. Issue 4 (2nd October 2019)

• Record Type: Journal Article
• Authors: Kumar, Ashish; Dasgupta, Antara; Lokhande, Sujit; Ramsankaran, RAAJ
• Is Part Of: International journal of river basin management. Volume 17:Issue 4(2019)
• Page Start: 479
• Page End: 488
• Publication Date: 2019-10-02
• Subjects: Index Terms – digital elevation models -- shuttle radar topography mission -- Cartosat digital elevation model -- flood modelling -- cross-section extraction -- 1D-hydrodynamic modelling; Watershed management -- Periodicals; Water resources development -- Periodicals; Hydraulic engineering -- Periodicals; Watershed hydrology -- Periodicals; Water resources development; Watershed management; Watersheds; Periodicals; 551.483
• DOI: 10.1080/15715124.2019.1606816
• Languages: English
• ISSNs: 1814-2060

Abstract

ABSTRACT: Accurate forecasts are a necessity in a world with frequent flood disasters and climatic anomalies. Studies have demonstrated that one of the primary sources of uncertainty in flood modelling is the topographic input data, the bathymetry in particular. As high-resolution elevation/bathymetry datasets are often unavailable in developing regions, the present study benchmarks the freely available Indian National Digital Elevation Model (DEM) CartoDEM, against the currently preferred global DEM, SRTM, for 1D-hydrodynamic modelling. Techniques for improving simulated Water Surface Elevation (WSE) profiles, while using cross-sections derived from these free DEMs are proposed. A distributed vertical error characterization using ground surveyed control points was performed and used to correct the DEM-extracted cross-sections. The approach was tested by simulating a 25-year return period flood in a reach of the Erai River, near Chandrapur city in Maharashtra, India. The results show an improvement of ∼84% in the root-mean-squared-error value for the WSE generated by the SRTM after correction, while the CartoDEM degraded by ∼41%. It was observed that while the uncorrected CartoDEM performs better than the uncorrected SRTM, the converse is true after the bias correction.