Gpu-Based Parallel Computations of the Hydrological Regime in the Kiliya Delta of the Danube River

M. Sorokin, M. Zheleznyak, L. Anishchenko, B. Sverdlov

Èlektron. model. 2025, 47(4):90-112

https://doi.org/10.15407/emodel.47.04.090

ABSTRACT

A mathematical model of the hydraulic regime of the Danube Kiliya Delta based on the two-dimensional parallel shallow-water hydrodynamic model COASTOX-UN is described. This model uses unstructured computational grids with triangular cells. The parallelization algorithm for solving the model equations on graphics processing units (GPUs) is presented using the FCT scheme of the model as an example. The high computational efficiency on GPUs is demonstrated: the modeling speed increases by more than 100 times compared to single-core computation and about five times compared to the parallel MPI version on a multi-core workstation. The use of GPUs makes it possible to calculate hydrodynamic processes in a large part of the delta on a detailed grid of high dimensionality. Verification of the model with hydrological station data from the Danube Hydrometeorological Observatory confirms its accuracy in reproducing water level and discharge dynamics in the delta branches. This is supported by a high correlation between calculations and field measurements as well as a high index of agreement. A prediction based on the modified delta model showed that the planned reconstruction of the Danube — Black Sea Deep Water Navigation Route (DWNR) will have a negligible impact on the hydraulic regime of the delta. The deviations of the water discharge do not exceed 3,5 % of the current runoff distribution, and the lowering of the water level in the branches of the delta does not exceed 6 cm on average in the section from Kiliya city to the sea.

KEYWORDS

river delta modeling, Danube Delta, shallow water equation, parallel computing, GPU computing, Danube‐Black Sea DWNR.

REFERENCES

  1. De Goede, E.D. (2020). Historical overview of 2D and 3D hydrodynamic modelling of shallow water flows in the Netherlands. Ocean Dynamics, 70(4), 521-539. https://doi.org/ 10.1007/s10236-019-01336-5
  2. Dutta, U., Singh, Y.K., Prabhu, T.S.M., Yendargaye, G., Kale, R.G., Kumar, B., Khare, M., Yadav, R., Khattar, R., & Samal, S.K. (2021). Flood forecasting in large river basins using FOSS tool and HPC. Water, 13(24), 3484. https://doi.org/10.3390/w13243484
  3. García-Feal, O., Cea, L., González-Cao, J., Domínguez, J.M., & Gómez-Gesteira, M. (2020). IberWQ: A GPU accelerated tool for 2D water quality modeling in rivers and estuaries. Water, 12(2), 413. https://doi.org/10.3390/w12020413
  4. Espoo Inquiry Commission. (2006). Report on the likely significant adverse transboundary impacts of the Danube-Black Sea navigation route at the border of Romania and Ukraine (67 pages). https://mtu.gov.ua/files/31110551_dod.pdf
  5. Anishchenko, L.Ya., & Sverdlov, B.S. (2008). Features of the assessment of the impact of dredging works on water ecosystems in water bodies of different classes. Environmental Protection and Life Safety, 3, 5-12(in Ukrainian).
  6. Demchenko, R.I., Zheleznyak, M.I., Kivva, S.L., Kolomiets, P.S., & Khomitsky, V.V. (2006). The hydrodynamics of Black Sea coastal zone near the branch Bystry mouth of the Delta of the Danube: 1. Waves transformation on the current and bottom inhomogeneities. Applied Hydromechanics, 8(4), 15-25 (in Russian).
  7. Zheleznyak, M.J., Demchenko, R.I., Khursin, S.L., Kuzmenko, Y.I., Tkalich, P.V., & Vitiuk, N.Y. (1992). Mathematical modeling of radionuclide dispersion in the Pripyat-Dnieper aquatic system after the Chernobyl accident. Science of the Total Environment, 112(1), 89-114. https://doi.org/10.1016/0048-9697(92)90241-j
  8. Summary and findings of the transboundary environmental impact assessments of Danube-Black Sea navigation route project implemented in the Ukrainian part of the Danube Delta. (2009). 115-120. https://rb.gy/iz6g8l
  9. Zheleznyak, M., Kivva, S., Pylypenko, O., & Sorokin, M. (2022). Modeling of behavior of Fukushima-derived radionuclides in freshwater systems. In K. Nanba, A. Konoplev, & T. Wada (Eds.), Behavior of radionuclides in the environment III (pp. 199-252). Springer. https://doi.org/10.1007/978-981-16-6799-2_11
  10. Sorokin, M.V. (2023). Parallelization of numerical solutions of shallow water equations by the finite volume method for implementation on multiprocessor systems and graphics processors. Environmental Safety and Natural Resources, 46(2), 163-193 (in Ukrainian). https://doi.org/10.32347/2411-4049.2023.2.163-193
  11. Sorokin, M.V. (2024). Algorithms for solving two-dimensional surface water dynamics equations for GPUs: Development and implementation for forecasting the consequences of the destruction of the Kakhovka Dam. In Proceedings of the XII ISPC "WATER FOR PEACE", 22-27 (in Ukrainian). http://waterday.iwpim.com.ua/Water_day_tezy2024.pdf
  12. Kotsyruba, V., Proshchyn, I., Sorokin, M., & Pylypenko, O. (2024). Improved method for forecasting the consequences of emergencies of a terrorist nature at hydraulic facilities.Modern Information Technologies in the Sphere of Security and Defence, 51(3), 5-14 (in Ukrainian). https://doi.org/10.33099/2311-7249/2024-51-3-5-14
  13. Zia, A., & Banihashemi, M.A. (2008). Simple efficient algorithm (SEA) for shallow water flows with shock wave on dry and irregular beds. International Journal for Numerical Methods in Fluids, 56(11), 2021-2043. https://doi.org/10.1002/fld.1585

Full text: PDF