Abstract

A rainfall-runoff model using Hydrologic Engineering Center–Hydrologic Modeling System (HEC-HMS) was carried out for the Kharkai River of Subarnarekha Basin, India to determine the magnitude of runoff from a known quantity of rainfall and the terrain inundation. In the present study, for the calibration and validation, the rainfall events of October 1–20, 2017 and July 12–31, 2019 were used. In the process of calibration, the computed discharge was found to be 1,674 cumecs against the observed discharge of 1,820 cumecs, with a Nash-Sutcliffe efficiency of 91.78%. Whereas in the process of validation, the computed discharge was found to be 3,266 cumecs against the observed discharge of 3,589 cumecs, and a Nash-Sutcliffe efficiency of 91.02%. Furthermore, to generate various inundation maps of the Kharkai River, an SRTM-DEM of 30 m x 30 m resolution, and a 30 m resolution DEM was used in the simulation process. The results of the inundation study showed that the 5, 20, 50, 100, and 500-year return period floods determined using Gumbel’s method are 3,647.791 m³/s, 5,797.717 m³/s, 7,160.188 m³/s, 8,181.168 m³/s, and 10,540.5 m³/s, respectively, and the corresponding inundation areas were 20.98 km², 33.65 km², 39.89 km², 44.07 km², and 56.65 km², respectively.

1 INTRODUCTION

Rainfall runoff modeling is an essential tool for flood inundation prediction, as it helps estimate water flow through rivers or drainage systems during heavy rainfall events (Peel and McMahon 2020). Recent studies emphasize advancements in flood modeling using machine learning, remote sensing, and real-time satellite data. This information is crucial for flood management and emergency response planning, enabling authorities to anticipate and mitigate the impact of floods on communities and infrastructure (Klijn et al. 2015). A rainfall-runoff simulation model is a process in which the quantity of runoff is estimated from the measured amount of rainfall in a watershed. Birkel and Barahona (2019) presented an overview on different tools used for rainfall runoff simulation models and described their advantages and challenges. The amount of runoff plays an important role for sustainable water resource planning and management (Hatamkhani et.al. 2022) as well as flood forecasting and flood plain inundation mapping. The quantity of floods, as well as flood inundation maps, are essential to predict the depth of inundation due to a flood event (Cohen et al. 2018) and its impact on the human population, as well as land use and land cover of a catchment. Many studies on rainfall-runoff and inundation models and associated hydrological modeling have been performed (Alaghmand et al. 2010; Arekhi 2012; Chu and Steinman 2009; Mendoza et al. 2019). Asadi and Boustani (2013) performed an HEC-HMS hydrologic model analysis for the simulation of lumped and semi-distributed storm flow in the Delibajak basin with the help of HEC-GeoHMS, using the SCS curve number method. Choudhari et al. (2014) used a HEC-HMS model for the simulation of the rainfall-runoff process in the Balijore Nala catchment of Odisha, India using an SCS curve number, SCS unit hydrograph, exponential recession, and Muskingum routing methods to compute runoff volume, peak runoff rate, base flow, and flow routing, respectively. Gebre (2015) simulated the runoff process of the Upper Blue Nile River Basin with the help of an HEC-HMS model using the Snyder unit hydrograph and exponential recession method. However, none of the above-mentioned researchers have done any work using both HEC-HMS and HEC-RAS to simulate rainfall-runoff modeling to assess the magnitude of floods. Using both HEC-HMS and HEC-RAS can provide a more accurate and comprehensive analysis of flood potential and inundation depths (Yu et al. 2022). By utilizing HEC-HMS to simulate the hydrological processes, the expected flow rates and volume of water that will be moving through the watershed can be estimated. This information can then be used as inputs to HEC-RAS to simulate the hydraulic behaviour of the river or stream (Mendoza et al. 2019) and to predict the water surface profiles and inundation depths. Using HEC-HMS and HEC-RAS in tandem allows the capture of the complex interactions between the hydrological and hydraulic processes, which can be important for predicting the extent of inundation during a flood event. Additionally, using both models can help to validate the results by comparing the simulated water levels and flows to observed data from previous floods or stream gauge measurements. In summary, using both HEC-HMS and HEC-RAS can provide a more comprehensive and accurate analysis of potential flooding and inundation depths in a watershed.

In view of the above facts, the present study was aimed at calibrating and validating the HEC-HMS model to estimate the runoff from a given amount of rainfall, and to develop a hydraulic model using HEC-RAS to derive the depth of flood and prepare flood inundation maps for floods of 10-, 25-, 50-, 100-, and 500-year return periods for the Kharkai River in the Subarnarekha basin to estimate the inundation area and the number of affected villages in which no contributory work has been reported to date.

2 MATERIALS AND METHODS

2.1 Study area

River Kharkai, which means “the fast-flowing river,” is one of the main tributaries of the river Subarnarekha. The Kharkai River originates in the Chota Nagpur Plateau in eastern India, and flows through the regions of West Singhbhum, Seraikela, East Singhbhum, and Bokaro, and finally reaches Sankchi, Jamshedpur in Jharkhand state, India. The boundary of the Kharkai River lies between the states of Orissa and Jharkhand (Mukherjee and Patel 2022). The surface and ground water potential of the Kharkai River basin is 1,085 million cubic meters, and 242 MCM, respectively (Mukherjee and Patel 2022). The Kharkai River basin has a total catchment area of 5,815 km², which is divided into the three sub-basins, namely Sub-basin 1, Sub-basin 2, and Sub-basin 3, covering an area of 155 km², 2,142.778 km², and 3,517.22 km², respectively. Figure 1 shows the Kharkai River basin location map.

Figure 1 Kharkai River basin location map.

2.2 Data used

In the present study, SRTM-DEM data of 30 m resolution was used. Daily rainfall data for the Kharkai basin was collected from the Central Water Commission, Odisha, Government of India (CWC 2019) for October 2013 and July 2017, and the daily flow data of the Adityapur discharge gauging site was collected from Central Water Commission, Odisha, Government of India (CWC 2019) for 2013 and 2017. A Triangulated Irregular Network (TIN) was used for the hydraulic analysis of the river system in HEC-RAS. Using a 30 m resolution SRTM-DEM, the TIN was derived for the study area, and the maximum annual discharge data of period 1992 to 2018 was used in the flood frequency analysis. ArcGIS 10.4 was used for generating various maps, and ERDAS Imagine 9.1 was used for generating the land use/land cover map. HEC-HMS 4.0 and HEC-GeoHMS 10.4 were used for hydrologic modeling.

2.3 Model development

A hydraulic model of the river network is developed with the help of HEC-GeoRAS using an extension in ArcGIS and HEC-RAS software (Yang et al. 2006). In the present study, a DEM is used to find the drainage pattern using ArcGIS and ArcHydro, which is further used to delineate the watershed and generate the stream network. Figure 2 represents the HEC-GeoHMS maps of the Kharkai basin.

Figure 2 HEC-GeoHMS maps of the Kharkai River basin.

The study area is divided into six categories: forest, water body, plantation, settlement, grass land, and barren land by using a supervised classification approach. The soils in the Kharkai basin have been divided as follows: clayey soil, loamy soil, loamy sand soil, and rocky soil, with 80% as fine, and 12% as rocky. Hydrological soil groups maps have been made by using the combined effects of land use/land cover features, soil characteristics, and texture, as Groups A, C, and D. Figure 3 shows the LULC map, soil map, and HSG map. Table 1 represents the curve number for each land cover type, and a weighted curve number is used in the calibration of the model. Table 2 represents the CN value of each sub-basin.

Figure 3 (a) LULC map; (b) Soil map; and (c) HSG map.

Table 1 Curve number for each land cover type.

Table 2 CN value and area of each sub-basin.

3 RESULTS AND DISCUSSION

3.1 Calibration of model

Considering the rain fall data from October 1–20, 2013, the model is calibrated by assigning the gauge weight using the Thiessen polygon method from the given three rain-gauge stations: Jamshedpur, Rairangpur, and Tiring. The input parameters initial abstraction, SCS-CN, impervious (%), and lag time have been used in such a way that observed and simulated discharges matched more accurately (SCS 1985). Figure 4 shows the HEC-HMS results during calibration. During the calibration process, the Nash-Sutcliffe Efficiency is computed between the observed and simulated results, which was 91.78% at the Adityapur junction. Further, the coefficient of determination (R²) of the scattered plot during calibration was 0.9868, which is shown in Figure 5.

Figure 4 Comparison of observed and simulated flow at Adityapur site (calibration).

Figure 5 Coefficient of correlation (R²) between observed and calibrated flow during calibration.

3.2 Validation of the model

Considering the rainfall data of event of July 12–31, 2017, the model was validated by assigning the gauge weight using the Thiessen polygon method at the same rain gauge stations. Figure 6 shows the outflow hydrograph at the Adityapur outlet during validation. The results of the computed and observed hydrographs during the validation process are shown in Figure 6. Further, the Nash-Sutcliffe Efficiency for the validated observed and computed hydrographs is 91.52%. The coefficient of determination (R²) of the scattered plot between the observed and computed hydrographs is 0.9504, as shown in Figure 7.

Figure 6 Comparison of observed and simulated flow at Adityapur site (validation).

Figure 7 Correlation between observed flow and simulated flow during validation.

3.3 Flood inundation mapping

The input parameters like discharge and Manning’s ‘n’ value were used in the HEC-RAS model to estimate the flood inundation depth, area, and velocity of flow during inundation. An SRTM-DEM at 90 m resolution was converted to TIN format using ArcGIS, a 3D analysis tool. Expected flood discharges for different return periods were estimated using Gumbel’s extreme value distribution method (Subramanya 2008) as shown in Table 3 and then put into steady flow to run the model. Finally, results of the HEC-RAS model were imported to HEC-GeoRAS, and flood inundation depth, area, and a flood velocity map for each return period (i.e., 5, 20, 50, 100, and 500) were prepared. Figure 8 shows the geometric data of the Kharkai River basin in HEC-RAS imported from HEC-GeoRAS.

Table 3 Computation of expected floods along the Kharkai River

Return period (T) in years	Reduced variate	Frequency factor K=(YT - Y̅n)/ Sn	Mean (X̅)	Standard deviation (𝜎𝑥)	Expected flood XT=X̅ + K𝜎𝑥
5	1.49994	0.878535	2,234.145	1,609.094	3,647.791
20	2.970195	2.214645	2,234.145	1,609.094	5,797.717
50	3.901939	3.061376	2,234.145	1,609.094	7,160.188
100	4.600149	3.695883	2,234.145	1,609.094	8,181.168
500	6.213607	5.162129	2,234.145	1,609.094	10,540.5

NOTE:
Y̅n = mean of reduced variate, and
Sn = standard deviation of reduced variate.

Figure 8 Geometric data in HEC-RAS imported from HEC-GeoRAS.

The inundation map for flood depth, area of inundation, and flood velocity is shown in Figure 9 (from a-j) for 5-, 20-, 50-, 100-, and 500-year return period floods. Inundation depth and velocity values vary between 0–20 m and 0–8.2 m/s for each return period flood. These maps provide valuable insights into flood likelihood and severity, with a 5-year return period, indicating a 20% annual flood chance, and a 100-year flood with a 1% chance, though more severe. Lower depth values represent less hazard zones, and higher depth values represent higher hazard zones. Table 4 represents the flood inundated area for various return period floods after calculating the submerged area in ArcGIS. The higher the return period of floods, the larger the area of submergence, and the probability of submergence becomes greater in the right-side settlement areas of the Kharkai River. Understanding these maps helps communities assess risks and plan accordingly. Communities can implement flood defences like levees and floodwalls, promote flood-resistant infrastructure, enhance early warning systems, preserve natural barriers such as wetlands, and adopt land use planning that restricts new developments in high-risk areas, ensuring safer and more resilient environments. Further, Figure 9 shows the cross-section of the river for floods of different return periods, corresponding depth, area, and velocity of flow. The profile plot and 3D-view of the flood map of the Kharkai River with river stations are shown in Figures 10 and 11, respectively.

Table 4 Flood inundated area for various return periods.

Figure 9 Inundated depth, area, and velocity for various return period floods.

Figure 10 Water profile plot of the Kharkai River for various return period floods.

Figure 11 3D-plan view of water surface profile with river stations.

4 CONCLUSIONS

The rainfall-runoff model for the Kharkai basin using HEC-HMS has been calibrated and validated for the Kharkai River in the Subarnarekha basin, India. The calibration and validation efficiency of the model has been determined using the Nash-Sutcliffe efficiency test, and the results are 91.78% and 91.52%, respectively at the outlet of the Adityapur gauging site. Further, the coefficient of determination values (R²) for the calibrated and validated model are 0.9868 and 0.9594, respectively. Furthermore, the HEC-RAS and HEC-GeoRAS models have been used for the assessment of flood inundation depth, area, and velocity for the 5-, 20-, 50-, 100-, and 500-year return periods using Gumbel’s method. The results of the inundation model show that the inundated areas are 20.98 km², 33.65 km², 39.89 km², 44.07 km², and 56.65 km² for the 5-, 20-, 50-, 100-, and 500-year return period floods. The depth of inundation and velocity of flow varies from 0–20 m and 0–8.2 m/s for floods of various return periods. The study utilizes SRTM-DEM data with a resolution of 30 m for terrain modeling, and 90 m for other analyses, where higher resolution data could improve flood inundation map accuracy. The Gumbel’s method used for estimating flood return periods assumes stationary conditions, leading to potential uncertainties as it does not account for future climate or land use changes. Additionally, model calibration and validation were based on limited rainfall events, and incorporating a broader dataset could enhance robustness. Future research could integrate climate change scenarios by incorporating projected rainfall patterns and intensities, assessing their impact on flood dynamics, and improving long-term flood risk predictions.

References

1. Alaghmand, S., R. Bin Abdullah, I. Abustan, and B. Vosoogh. 2010. “GIS-based river flood hazard mapping in urban area (A case study in Kayu Ara River Basin, Malaysia).” International Journal of Engineering Technology 2, 488–500.
2. Arekhi, S. 2012. “Runoff modeling by HEC-HMS model (Case study: Kan watershed, Iran).” International Journal of Agriculture and Crop Sciences (IJACS) 4, 1807–1811.
3. Asadi, A., and F. Boustani. 2013. “Performance evaluation of the HEC-HMS hydrologic model for lumped and semi-distributed stormflow simulation (Study area: Delibajak Basin).” American Journal of Engineering Research (AJER) 2, 115–121.
4. Birkel, C., and A.C. Barahona. 2019. “Rainfall-Runoff Modeling: A Brief Overview.” Reference Module in Earth Systems and Environmental Sciences. Elsevier. https://doi.org/10.1016/B978-0-12-409548-9.11595-7
5. Choudhari, K., B. Panigrahi, and J.C. Paul. 2014. “Simulation of rainfall-runoff process using HEC-HMS model for Balijore Nala watershed, Odisha, India.” International Journal of Geomatics and Geosciences 5, 253–265.
6. Chu, X., and A. Steinman. 2009. “Event and continuous hydrologic modeling with HEC-HMS.” Journal of Irrigation and Drainage Engineering 135, 119–124.
7. Cohen, S., G.R. Brakenridge, A. Kettner, B. Bates, J. Nelson, R. McDonald, Y.F. Huang, D. Munasinghe, and J. Zhang. 2018. “Estimating floodwater depths from flood inundation maps and topography.” Journal of the American Water Resources Association (JAWRA) 54 (4): 847–858.
8. CWC. 2019. “Water and related statistics.” Water Planning and Projects Wing, Central Water Commission (CWC), Odisha, Ministry of Water Resources, Government of India.
9. Gebre, S.L. 2015. “Application of the HEC-HMS model for runoff simulation of Upper Blue Nile River Basin.” Journal of Waste Water Treatment and Analysis 6, 1–8.
10. Hatamkhani, A., A. Khazaie Poul, and A. Moridi. 2022. “Sustainable water resource planning at the basin scale with simultaneous goals of agricultural development and wetland conservation.” AQUA – Water Infrastructure, Ecosystems and Society 71 (6): 768–781. https://doi.org/10.2166/aqua.2022.025
11. Klijn, F., H. Kreibich, H. De Moel, and E. Penning-Rowsell. 2015. “Adaptive flood risk management planning based on a comprehensive flood risk conceptualization.” Mitigation and Adaptation Strategies for Global Change 20, 845–864.
12. Mendoza, A., G. Soto-Cortes, G. Priego-Hernandez, and F. Rivera-Trejo. 2019. “Historical description of the morphology and hydraulic behavior of a bifurcation in the lowlands of the Grijalva River Basin, Mexico.” Catena 176, 343–351.
13. Mukherjee, J., and P.P. Patel. 2022. “Landscape characterization using geomorphometric parameters for a small sub-humid river basin of the Chota Nagpur Plateau, Eastern India.” In Drainage Basin Dynamics: An Introduction to Morphology, Landscape and Modeling 127–152. Geography of the Physical Environment, Cham: Springer International Publishing.
14. Peel, M.C., and T.A. McMahon. 2020. “Historical development of rainfall‐runoff modeling.” Wiley Interdisciplinary Reviews: Water 7 (5): e1471.
15. SCS. 1985. National Engineering Handbook, Section 4: Hydrology. Soil Conservation Service, USDA, Washington, DC.
16. Subramanya, K. 2008. Engineering Hydrology. Tata McGrawHill Publishing Company Limited, New Delhi.
17. Yang, J., R.D. Townsend, and B. Daneshfar. 2006. “Applying the HEC-RAS model and GIS techniques in river network floodplain delineation.” Canadian Journal of Civil Engineering 33 (1): 19–28.
18. Yu, Q., Y. Wang, and N. Li. 2022. “Extreme Flood Disasters: Comprehensive Impact and Assessment.” Water 14 (8): 1211.