Development of an Integrated Online System for Flood Forecasting and Warning by Integrating Numerical Weather, Hydrological, and Hydraulic Models: A Case Study of the Gorganrood Watershed

Document Type : Research Article

Authors

• Z. Alizadeh ¹
• M. Shahsavandi ¹
• M. Masoudi-Moghaddam ¹
• S. Talebi ²
• J. Yazdi ¹

¹ Department of Water Resources Engineering, Faculty of Civil, Water and Environmental Engineering, Shahid Beheshti University, Tehran, Iran

² GIS and Remote Sensing Department, Space Research Center, Iranian Space Research Center, Tehran, Iran

10.22067/jsw.2025.93995.1485

Abstract

Introduction
Floods rank among the most devastating natural disasters, causing significant loss of life and property each year. Floods are among the most destructive natural disasters, causing extensive loss of life and property annually. The Gorganrood watershed in northeastern Iran is particularly vulnerable to frequent and severe flooding due to its unique geographical and climatic characteristics. Factors contributing to this high flood risk include steep topographical gradients, impermeable soil, and degraded vegetation cover, which lead to rapid and devastating flash floods. The region has experienced a rise in both the frequency and intensity of floods in recent years, resulting in significant damages, such as the catastrophic flood of March 2019. Consequently, the development of effective flood forecasting and early warning systems (FFEWS) has become a critical priority for crisis management. Recent advancements in numerical modeling offer powerful tools for more accurate and timely flood prediction. Weather Research and Forecasting (WRF) models are widely used for their precision in simulating regional precipitation. Hydrological models like the Hydrologic Modeling System (HEC-HMS) are essential for converting predicted rainfall into surface runoff, while hydraulic models such as HEC-RAS excel at simulating river flow and mapping inundation zones. This research aims to develop and evaluate a sophisticated, integrated online system specifically for the Gorganrood watershed. The primary innovation of this study is the creation of a unified online platform that couples the WRF, HEC-HMS, and a 2D HEC-RAS model to forecast flood inundation up to 48 hours in advance. A further novelty lies in its software architecture, which utilizes React for the front-end and a Python-based Django framework for the back-end, a combination not previously applied in similar research for real-time visualization of flood forecasts.
 
Materials and Methods
 The research focused on the Gorganrood watershed, a major basin in northeastern Iran covering approximately 11,380 km². The system developed was an integrated Web-GIS software platform designed for end-to-end flood forecasting. The system's workflow began with the automated retrieval of meteorological data from the Global Forecast System (GFS). This data served as input for the regional WRF model to generate high-resolution precipitation forecasts. The rainfall predictions were then fed into a calibrated HEC-HMS model to simulate the rainfall-runoff process and generate flood hydrographs. In the final stage, a 2D HEC-RAS hydraulic model was executed for critical, populated river reaches to produce detailed flood inundation maps. The technological framework was built on modern software tools. The user interface (Front-End) was developed using React to create a dynamic user experience. The server-side logic (Back-End) was implemented in Python using the Django web framework. For data management, a PostgreSQL database with the PostGIS spatial extension was employed. GeoServer was used as the map server. The chosen models include:

WRF Model: Selected for precipitation forecasting due to its open-source nature, flexibility, and widespread use.
HEC-HMS Model: Chosen as the rainfall-runoff model for its suitability in a large basin with limited data. It was configured using the SCS-CN method for loss calculations, the SCS unit hydrograph method for runoff transformation, and the Muskingum method for channel routing.
HEC-RAS Model: The 2D version was selected for hydraulic modeling due to its ability to simulate complex, two-dimensional flow dynamics when floods overtop riverbanks and its free availability.

 
Results and Discussion
 The performance of each model component was rigorously evaluated. The WRF model was assessed using five historical storm events, comparing forecasts across five different lead times (6, 12, 18, 24, and 48 hours). Statistical analysis revealed that the 6-hour forecast horizon provided the optimal balance of accuracy and lead time, exhibiting the best performance metrics (R²=0.69, RMSE=12.25, NSE=0.0). Thus, a 6-hour lead time was adopted for the operational system. The HEC-HMS model was calibrated and validated against observed data from several hydrometric stations (Nodeh, Arazkuseh, etc.). The results demonstrated a good agreement between the simulated and observed hydrographs, particularly in capturing the peak discharge and timing of floods. Observed discrepancies in total flood volume were attributed to uncertainties in spatial rainfall data and potential measurement errors. For the numerous sub-basins lacking gauging stations, model parameters were regionalized using a clustering technique based on physiographic similarity to the calibrated sub-basins. The integrated online system allows users to run the entire forecast chain through a web interface. To manage the significant computational requirements, the 2D HEC-RAS model was implemented for two high-priority areas: the region downstream of the Golestan Dam and the flood-prone city of Aq-Qala. A key challenge was the high computational demand of the models, which was addressed by leveraging the High-Performance Computing (HPC) cluster at Shahid Beheshti University. Another challenge is the potential for model instability and limitations imposed by data quality, which can be mitigated by more detailed calibration.
 
Conclusion
This research successfully developed a comprehensive, integrated online system for flood forecasting in the Gorganrood watershed by coupling the WRF, HEC-HMS, and HEC-RAS models. The evaluation showed that the WRF model provided acceptable precipitation forecasts, the HEC-HMS model accurately simulated rainfall-runoff processes, and the 2D HEC-RAS model produced valuable, high-resolution flood inundation maps. The system's robust software architecture, utilizing React, Python/Django, and PostgreSQL/PostGIS, provides an efficient, scalable, and user-friendly platform for operational flood management. This work demonstrates that the integration of advanced numerical models into a single, automated platform is a highly effective approach to mitigating flood risk. The resulting system offers a powerful tool for crisis managers and serves as a replicable model for developing similar advanced warning systems in other flood-prone basins.
 
Acknowledgement
The authors would like to acknowledge the use of the High-Performance Computing (HPC) system at Shahid Beheshti University for the execution of the numerical models in this research.
 
Keywords: Django, Flood forecasting, Gorganrood watershed, HEC-HMS, HEC-RAS (2D), Integrated system, React, WRF
 
 
 

Keywords

• Flood Forecasting
• React
• WRF
• HEC-HMS
• HEC-RAS (2D)

Main Subjects

• Irrigation

©2025 The author(s). This is an open access article distributed under Creative Commons Attribution 4.0 International License (CC BY 4.0)

References

1. References

   1. Alizadeh, Z. (2022). Real-time adaptive operation of dam reservoir systems under flood conditions. Ph.D. Dissertation, Supervised by Dr. Jafar Yazdi, Shahid Beheshti University.
   2. Avand, M., Moradi, H.R., & Ramazanzadeh Lasboyee, M. (2021). Spatial prediction of future flood risk: an approach to the effects of climate change. Geosciences. https://doi.org/10.3390/geosciences11010025
   3. Bajracharya, S.R., Khanal, N.R., Nepal, P., Rai, S.K., Ghimire, P.K., & Pradhan, N.S. (2021). Community assessment of flood risks and early warning system in ratu watershed, Koshi basin, Nepal. Sustainability, 13(6), 3577. https://doi.org/10.3390/su13063577
   4. Brunner, G.W. (1995). HEC-RAS river analysis system. Hydraulic reference manual. Version 1.0.
   5. Byaruhanga, N., Kibirige, D., Gokool, S., & Mkhonta, G. (2024). Evolution of flood prediction and forecasting models for flood early warning systems: A scoping review. Water, 16(13), 1763. https://doi.org/10.3390/ w16131763
   6. Chen, W., Wu, H., Kimball, J.S., Alfieri, L., Nanding, N., Li, X., & Zhong, W. (2022). A coupled river basin urban hydrological model (DRIVE Urban) for real-time urban flood modeling. Water Resources Research, 58(11), e2021WR031709. https://doi.org/10.1029/2021wr031709
   7. Cheong, T.S., Choi, C., Ye, S.J., Shin, J., Kim, S., & Koo, K.M. (2023). Development of flood early warning frameworks for small streams in Korea. Water, 15(10), 1808. https://doi.org/10.3390/w15101808
   8. (2017). MIKE FLOOD: User Guide. DHI Water & Environment.
   9. Falck, A., Tomasella, J., & Papa, F. (2021). Assessing the potential of upcoming satellite altimeter missions in operational flood forecasting systems. Remote Sensing, 13(21), 4459. https://doi.org/10.3390/rs13214459
   10. GeoServer Project. (2024). GeoServer User Manual. Retrieved from https://docs.geoserver.org
   11. Ghahraman, K., Nagy, B., & Nooshin Nokhandan, F. (2023). Flood-prone zones of meandering rivers: Machine learning approach and considering the role of morphology (Kashkan River, Western Iran). Geosciences, 13(9), 267. https://doi.org/10.3390/geosciences13090267
   12. Jafarzadegan, K., Moradkhani, H., Pappenberger, F., Moftakhari, H., Bates, P., Abbaszadeh, P., & Duan, Q. (2023). Recent advances and new frontiers in riverine and coastal flood modeling. Reviews of Geophysics, 61(2), e2022RG000788. https://doi.org/10.1029/2022rg000788
   13. Kalimisetty, S., Singh, A., Korada Hari Venkata, D.R., & Mahammood, V. (2022). 1D and 2D model coupling approach for the development of operational spatial flood early warning system. Geocarto International, 37(15), 4390-4405. https://doi.org/10.1080/10106049.2021.1886335
   14. Karami, M., Abedi Koupai, J., & Gohari, S.A. (2024). Integration of SWAT, SDSM, AHP, and TOPSIS to detect flood-prone areas. Natural Hazards, 120(7), 6307-6325. https://doi.org/10.1007/s11069-024-06483-7
   15. Liu, Y., Wang, H., & Lei, X. (2021). Real-time forecasting of river water level in urban based on radar rainfall: A case study in Fuzhou City. Journal of Hydrology, 603, 126820. https://doi.org/10.1016/j.jhydrol.2021.126820
   16. Merino, A., García-Ortega, E., Navarro, A., Sánchez, J.L., & Tapiador, F.J. (2022). WRF hourly evaluation for extreme precipitation events. Atmospheric Research, 274, 106215. https://doi.org/10.1016/j.atmosres.2022.106215
   17. Mishra, A., Mukherjee, S., Merz, B., Singh, V.P., Wright, D.B., Villarini, G., & Stedinger, J.R. (2022). An overview of flood concepts, challenges, and future directions. Journal of Hydrologic Engineering, 27(6), 03122001. https://doi.org/10.1061/(asce)he.1943-5584.0002164
   18. Montgomery, D.R., Buffington, J.M., Smith, R.D., Schmidt, K.M., & Pess, G. (1995). Pool spacing in forest channels. Water Resources Research, 31(4), 1097-1105.
   19. Morshed, M.A., Mushwani, H., Sahak, K., & Hairan, M.H. (2024). The current state of early warning system in South Asia: A case study of Afghanistan. International Journal of Disaster Risk Reduction, 100, 104201. https://doi.org/10.1016/j.ijdrr.2023.104201
   20. Muñoz, P., Orellana-Alvear, J., Bendix, J., Feyen, J., & Célleri, R. (2021). Flood early warning systems using machine learning techniques: The case of the Tomebamba Catchment at the southern Andes of Ecuador. Hydrology, 8(4), 183. https://doi.org/10.20944/preprints202111.0510.v1
   21. Nanda, T., Sahoo, B., Beria, H., & Chatterjee, C. (2016). A wavelet-based non-linear autoregressive with exogenous inputs (WNARX) dynamic neural network model for real-time flood forecasting using satellite-based rainfall products. Journal of Hydrology, 539, 57-73. https://doi.org/10.1016/j.jhydrol.2016.05.014
   22. (2020). NCEP GFS: NOAA's Global Forecast System. National Centers for Environmental Prediction, National Oceanic and Atmospheric Administration.
   23. Obe, R.O., & Hsu, L.S. (2015). PostGIS in Action, 2nd Edition. Manning Publications.
   24. Patel, P., Thakur, P.K., Aggarwal, S.P., Garg, V., Dhote, P.R., Nikam, B.R., & Al-Ansari, N. (2022). Revisiting 2013 Uttarakhand flash floods through hydrological evaluation of precipitation data sources and morphometric prioritization. Geomatics, Natural Hazards and Risk, 13(1), 646-666. https://doi.org/10.1080/19475705.2022. 2038696
   25. Ren, D.F., & Cao, A.H. (2023). Precipitation–runoff simulation in Xiushui river basin using HEC–HMS hydrological model. Modeling Earth Systems and Environment, 9(2), 2845-2856. https://doi.org/10.1007/s40808-022-01679-x
   26. (1972). National Engineering Handbook, Section 4: Hydrology. Soil Conservation Service, U.S. Department of Agriculture, Washington, D.C.
   27. (1986). Urban hydrology for small watersheds (TR-55). Soil Conservation Service, U.S. Department of Agriculture, Washington, D.C.
   28. Shaikh, A.A., Pathan, A.I., Waikhom, S.I., Agnihotri, P.G., Islam, M.N., & Singh, S.K. (2023). Application of latest HEC-RAS version 6 for 2D hydrodynamic modeling through GIS framework: a case study from coastal urban floodplain in India. Modeling Earth Systems and Environment, 9(1), 1369-1385. https://doi.org/10.1007/s40808-022-01567-4
   29. USACE-HEC. (2018). HEC-HMS Hydrologic Modeling System User’s Manual, Version 4.3. U.S. Army Corps of Engineers, Hydrologic Engineering Center.
   30. Zang, Y., Meng, Y., Guan, X., Lv, H., & Yan, D. (2022). Study on urban flood early warning system considering flood loss. International Journal of Disaster Risk Reduction, 77, 103042. https://doi.org/10.1016/j.ijdrr.2022. 103042