Hydraulic model for flood inundation in Diyala River Basin using HEC-RAS, PMP, and neural network

Faris Sahib Alrammahi; Ahmed Naseh Ahmed Hamdan

doi:10.1515/eng-2022-0530

Article Open Access

Hydraulic model for flood inundation in Diyala River Basin using HEC-RAS, PMP, and neural network

Faris Sahib Alrammahi and Ahmed Naseh Ahmed Hamdan

Published/Copyright: February 9, 2024

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information

From the journal Open Engineering Volume 14 Issue 1

Abstract

The Diyala River Basin in Iraq is vital for water supply to residential, agricultural, and the Tigris River (with approximately 4.5 billion cubic meters annually), but it faces frequent floods and droughts due to reliance on rainfall. This study aims to address these issues by simulating flood inundation using the hydrological engineering centre-river analysis system model and predicting high-intensity rainfall with artificial neural networks. ArcGIS and remote sensing tools aid model development with data from official sources and organizations such as national aeronautics and space administration and food and agriculture organization. The hydraulic model is calibrated using satellite imagery to depict a 2019 flood, and artificial intelligence predicts the precipitation patterns for the next 50 years based on historical data from 1981 to 2021. One of the challenges and difficulties encountered in the study is the scarcity of available data, as well as the absence of scientific research pertaining to the region regarding hydraulic modeling. The study identifies flood risks in March and April every year, notably for the Hemrin Dam, which may exceed permissible water levels (reach a level over 110 m where the Hemrin Crest level is 109.5 m). To mitigate this, an artificial canal is proposed to divert water annually, protecting the dam and downstream areas without disrupting operations. The diverted water could also augment the Tigris River in Kut Governorate during summer. The study demonstrates the value of integrating multiple modeling techniques and data sources for accurate hydraulic predictions. It offers insights for decision-makers in flood management and planning. This study contributes to efficient flood management strategies by adopting a multidisciplinary approach.

Keywords: hydraulic model; HEC-RAS; ANN; Diyala River Basin; probable maximum precipitation

1 Introduction

The Diyala River Basin (DRB) holds great strategic importance due to its impact on various aspects of life, agriculture, and industry. It contributes around 4.6 billion cubic meters (BCM) of water annually to the Tigris River in southern Baghdad, accounting for approximately 17% of the total flow [1,2]. Thus, it is recognized as a significant basin in Iraq, with 70% of its area located within the country, while the remainder lies in Iran [1].

Droughts are a recurring issue in the DRB, particularly during the summer when rainfall is scarce. Addressing these droughts requires international agreements with Iran to increase water quotas and optimize dam operations, specifically the Derbendikhan and Hemrin dams [1,3]. On the other hand, the basin is susceptible to floods, particularly in the spring season, often resulting from heavy rainfall and overflow from Iran. The floods in 2019 posed a significant threat to the Hemrin Dam and adjacent residential areas along the Diyala River [1], as shown in Figure 1.

Figure 1

Inundation maps in January and April 2019 as a part of DRB.

Developing an accurate and reliable hydraulic model is crucial to better understand and managing flood risks in the DRB. Such a model enables effective flood management strategies and provides valuable insights into the behavior of the basin’s water systems. This study employed a two-dimensional hydraulic model utilizing the hydrological engineering centre-river analysis system (HEC-RAS) software to simulate the floodplain in 2019. Satellite photographs from the corresponding period were used to validate the extent of the flood.

Artificial intelligence methods, such as artificial neural networks (ANNs), can be utilized for future rainfall forecasting [1]. These forecasts can then be used to estimate flood inundation maps by identifying the period of maximum rainfall. In conjunction with Arc-GIS, remote sensing technology is the primary tool for exploring, assessing, and managing the catchment.

One of the challenges and difficulties encountered in the study is the scarcity of available data, as well as the absence of scientific research pertaining to the region regarding hydraulic modeling.

This study aimed to develop an integrated framework for modeling flood inundation in the DRB, contributing to existing knowledge. It explored the feasibility of employing hydraulic (RAS) models, ANN, and geographic information system (GIS)-based remotely sensed photos within the DRB.

2 Previous related studies in DRB

In 2016, Fariba et al. [2] developed a conceptual model to assess equitable water distribution in multi-country rivers. The Sirwan River (a part of DRB), which flows through Iraq and Iran, was used as a case study. The study employed the water evaluation and planning system model to analyze the river’s flow lines and catchment, identifying five regional water management scenarios.

Abbas et al. [4] examined the impact of climate change on water resources in DRB. Using the soil and water assessment tool (SWAT) software and three releasing scenarios for 2046 to 2100, they found a continuous deterioration in the water sources of the Diyala River, indicating potential future dangers.

Madhloom et al. [5] conducted a study using GIS to analyze the spatial distribution of pollutants in the DRB. The study compared pollution standards with field measurements and generated a map of pollutant distribution across the Diyala watershed, revealing concentrations that exceeded acceptable standards.

In 2018, Madhloom et al. [6] focused on water resource management in the DRB using satellite images, remote sensing, and GIS technology. The study emphasized the effectiveness of GPS technology in providing significant results for decision-making in water management.

Thair et al. [7] analyzed sediments in the Hemrin Dam (a part of DRB) using the HEC HMS software from 1981 to 2014. Their hydrological model successfully simulated the sediment deposits in the dam.

Al-Musawi and Al-Rubaie [8] investigated the applicability of ANN and GIS models to predict and assess water quality in the DRB. Their study demonstrated that the neural network (NN) model accurately simulated water quality and distribution, generating colored maps using GIS to understand the river comprehensively.

Ibrahim and Saed [9] examined the effect of the construction of the Darbandikhan Dam on the Diyala River between 1962 and 2016, compared with the period from 1926 to 1961, using satellite images and GIS software. The dam’s construction increased water bodies by over 20% and expanded the reservoir area.

In 2020, Aziz et al. [10] analyzed the hydrological characteristics of the DRB using a digital elevation model (DEM) and GIS software. The study provided details on the morphological characteristics of the basin but did not compare the results with field data, limiting the conclusiveness of the findings.

Kadhim and Al-kubaisi [11] conducted a study on the classification of soil cover and its uses in the DRB using satellite images from 2018 to 2019. The study achieved acceptable values and quantities, with a high kappa coefficient indicating the result accuracy.

In 2020, Kadhim et al. [12] developed a hydrological model for the Darbandikhan Dam reservoir based on inflow data from 1979 to 2008, utilizing the SWAT model. The model’s accuracy was confirmed through statistical comparisons.

Jassam and Abed [13] studied the hydraulic characteristics and absorption of the Diyala River in Baqubah (a part of DRB) during flood events using the 1D HEC-RAS model. The study provided good results compared to recorded data, highlighting the need to raise river shoulders in the Kharnabat and Um Al-Atahm areas to mitigate flood risks.

In 2021, Karim et al. [14] and his colleagues used a two-dimensional HEC-RAS model to simulate and analyze the propagation of a flood wave caused by a hypothetical breakdown of the Al-Udhaim Dam in Iraq. The emphasis was on understanding flood wave propagation in order to estimate dangers and create emergency strategies. Derivation of breach hydrographs and propagation of downstream flood waves are part of the process. The findings show the serious potential effects of dam breaks and underline the importance of competent management and long-term flood management measures.

In 2021, Kadhim and his colleagues examined floods in the DRB using four models [12]. The dynamic-wave method yielded results closest to the recorded values, with variations observed based on changes in river bed slopes and friction coefficients.

Machine learning was used in a work to forecast flood susceptibility in Nigeria’s southeastern area by Ekwueme in 2022 [15]. The autoregressive integrated moving average model, which used hydrogeological data and remote sensing datasets, predicted a considerable increase in flooding patterns, with river discharge anticipated to grow by 15–150% between 2020 and 2024. This information assisted regional authorities in adjusting to and analyzing flood hazards caused by climate change.

3 Study area

The area of DRB covers 38,000 km² and is internally bordered by four Governorates: Baghdad and Salah Al-Din westward, Sulaymaniyah to the north, and Wasit to the south [13]. It is located between 33 ° 4 ′ 59 ″ − 35 ° 49 ′ 58 ″ N and 44 ° 30 ′ 00 ″ − 46 ° 49 ′ 58 ″ E .

As shown in Figure 2, the area is divided into two sections: the Upper DRB (UDRB) in Iraq between the Derbendikhan Dam and the Hemrin Dam, and the Lower DRB (LDRB) from the Diyala Barrage near the Hemrin Dam to the Tigris River south of Baghdad.

Figure 2

DRB satellite photograph.

DRB is located in Iraq and Iran, with an arid and semi-arid climate [16]. National aeronautics and space administration (NASA) Power Data Access Viewer [17] is a website that provides meteorological and sun parameters for renewable energy systems. It includes an extensive database used to characterize DRB’s climatic elements for many years, including Iraq.

Temperature is an essential factor in DRB water availability [18]. The average monthly maximum temperatures varied from 26.3 to 40°C, while the average monthly minimum temperatures ranged from −6.6 to 5.93°C. The lowest temperature recorded was in January 2008, while the highest was in July 2020 [19].

In Iraq, precipitation is seasonal, with the wet season extending from December to February [20]. The average daily rainfall at DRB varied from 4.88 to 6.87 mm/day, with the lowest readings ranging from 0.01 to 0.06 mm/day. All locations in the catchment had a minimum value of zero from June to September, with the maximum values occurring between January and March.

4 Research significance

The studies carried out on the DRB area were very few and took details related to the climate, the quality of water, and the spread of pollutants. Additionally, studies combining different models were restricted to specific hydrologic and hydraulic models or integrating hydrologic models with NNs using a few GIS tools.

Therefore, this research stands out in its originality and significance as the first study to comprehensively examine water movement in the entire DRB catchment. It introduces NNs to predict future droughts and potential floods in the study area.

5 Method and materials

5.1 River structures with HEC-GeoRAS

HEC-GeoRAS is an ArcGIS addon that makes it easier to prepare geometric data for export to HEC-RAS. It generates river geometry using DTM (Digital Terrain Model), including channels, banks, flood banks, and cross sections expressed as triangulated irregular network (TIN) or grid. HEC-GeoRAS also derives Manning’s n values from land-use data for cross sections. It also allows you to import HEC-RAS simulation data into ArcGIS for better display, including animated floodplain results [18].

GeoRAS was divided into two steps: RAS geometry and RAS mapping. Geometric data (stream centerlines, bank lines, flow route centerlines, and XS cut lines) were generated by the “RAS Geometry group” and sent to HEC-RAS through “Export RAS Data.” The RAS mapping group is responsible for importing RAS findings into ArcGIS for better display, as shown in Figure 3.

Figure 3

Flow chart of using HEC-GeoRAS and HEC-RAS.

5.2 Establishment of a hydraulic model

The HEC-RAS hydraulic simulation model is widely used for studying river systems, including water surface profiles, sediment transport, and water quality [21]. It supports 1D steady and unsteady flow and 2D analysis, making it appropriate for channel flow analysis and floodplain delineation. Because of its adequacy and an extensive collection of tools, the researcher used HEC-RAS version 6.1.

The 2D analysis in HEC-RAS was primarily used to examine the river system and probable floods. In HEC-RAS, water surface elevation is computed using continuity and the diffusion-wave version of the complete momentum equation, which approximates the Saint-Venant equation [22].

The diffusion wave equations and associated approximations were used to ensure the stability of flood wave simulations [14,23]. The Courant number (C) was used in the stability analysis, as indicated in equation (1), according to velocity (U), time step ∆ t , and cell size step ∆ X :

(1) C = U ∆ t ∆ X ≤ 2.0 .

Based on the mesh size and flow velocity, the Courant number is a non-dimensional parameter that dictates the time step requirements for transient simulations. The Courant number in multidimensional flow is calculated as the total velocities in all directions divided by the cell size.

To maintain stability and accuracy, the model must be refined by modifying the time step and cell size to keep the Courant number (C) within acceptable ranges (max C = 5 or ∆ t = 2 ∆ X U ) [23]. A Courant number of 1 or lower was taken to ensure steady and precise results.

Five main steps have been done for the river analysis modeling using HEC-RAS, as shown in Figure 4.

Figure 4

Steps of modelling by HEC-RAS.

Manning’s n value, a roughness parameter, was another critical parameter used to calibrate the two-dimensional model. RAS Mapper can create a land cover layer and associate it with terrain data for a 2D model.

5.3 Boundary conditions

The hydraulic model HEC-RAS was used to investigate the UDRB. Data on Derbendikhan Dam outflows and Hemrin Dam inflows were collected directly from the Iraqi Ministry of Water Resources, as shown in Figure 5.

Figure 5

Upstream and downstream of the HEC-RAS model.

5.4 Future prediction of rainfall runoff

5.4.1 Probable maximum precipitation (PMP)

Predicting the probable maximum flood (PMF) for the area is often required for hydraulic structure design, which depends on predicting the PMP. The recurrence likelihood of PMP in the studied region is also essential since it substantially influences PMF calculation (Figure 6).

Figure 6

Outflow from Diyala Barrage.

PMP was established by the World Meteorological Organization (WMO) in 1986 as the maximum depth of precipitation that can occur in a certain location [24]. The WMO specified that the estimated PMP should: (a) be specific to a specific area, not an entire region or country; (b) be physically possible, taking into account the local climatic conditions; (c) be specific to a specific time of year; and (d) not represent long-term trends.

Due to data constraints in the research region and the partition of DRB into subbasins, a statistical technique was selected in this dissertation to estimate the PMP for DRB. To estimate PMP using historical gauge data from the study watershed, Hershfield’s approach, supported by the WMO handbook, was used.

The PMP is calculated using the following equation [25]:

(2) PMP = X n + K m S n ,

where X _n is the mean of observed data, S _n is the standard deviation, and K _m is the frequency factor, which depends upon the number of observations and is calculated by the following equation [24]:

(3) K m = R max − R n − 1 S n − 1 ,

where R _max represents highest value in the series, R _n−1 represents the mean excluding the most considerable value, and S _n−1 denotes the standard deviation excluding the most significant value.

5.4.2 ANN

Researchers found that ANNs forecast the influence of climate change on river flows with a high accuracy rate of more than 95% [26]. The results also showed that ANN is well suited for mapping flood inundation, with an accuracy of over 96% [27]. Previous research has shown that the accuracy of ANN is affected by factors such as the amount of observed data, the network type used, and the activation functions used in each layer [28–30].

Deep-learning NNs were used in this study due to the difficulties and complexities involved in forecasting future precipitation and evaluating time series data. The “Adam” activation and optimization function, as well as 256 hidden layers, was used. The input was a sequential dataset representing yearly precipitations in DRB from 2011 to 2021, and the output was a sequential representation offering estimations for the same period as well as projections for future predicted years.

6 Results and discussions

The ArcMap software was used to produce the detailed data for the study region. Nine DEM images characterizing the DRB topography were obtained from the indicated US Geological Survey site and uploaded to the ArcMap, as shown in Figure 7. These DEMs of Tiff type are combined to form one mosaic Tiff picture.

Figure 7

DEM’s of DRB.

The downloaded DEM was clipped to the specific area covered by DRB, as shown in Figure 8 by some tools and steps, representing the studied area and initial information to begin the first step of DRB’s hydraulic model.

Figure 8

DEMs of UDRD and LDRB.

Because a detailed cross-sectional information about the river was not available for the study area, the parameters for the routing model had to be manually drawn. This was done using the basin model generated by HEC-GeoRAS, along with satellite imagery (DEM).

Overall, preparing Geo-RAS for HEC-RAS involves a combination of ArcGIS and HEC-GeoRAS tools to create the necessary input files for HEC-RAS modeling. The steps shown in Figure 9 were adopted to implement the HEC-GeoRAS.

Figure 9

Cross-section locations of the Diyala River channel.

The hydraulic model requires Manning’s values, which were calculated and input into the model in two steps; the first step involved obtaining a satellite image (TIF file type) of the “Land Cover” from the NASA database and adding it as an additional layer to the model as shown in Figure 10.

Figure 10

Manning’s roughness layer for UDRB and LDRB with related mesh.

For the water year covering October 2018 to May 2019, boundary conditions from multiple sources, including outputs from Derbendikhan Dam and subbasins near the Diyala River for UDRB and from Diyala Barrage to the Diyala River for LDRB, were integrated into the analysis. The outcomes are depicted in Figure 11.

Figure 11

Inundation map of UDRB and LDRB by HEC-RAS (April 26, 2019).

To match these satellite images, different sections were made along the Diyala River channel and compared with the same time of implementation on the hydraulic model after creating the inundation area, converting it into shapefiles, and comparing it to the ArcGIS software. The calibration and validation have been done, and the accepted results as shown in Figure 12 for UDRB and Figure 13 for LDRB.

Figure 12

Comparison at different locations along UDRB (April 26, 2019).

Figure 13

Comparison at different locations along LDRB (April 26, 2019).

According to the matching with authentic images, the researcher considered that the hydraulic model for the upper and lower parts of the DRB could be relied upon to an acceptable extent.

To apply these models for the next 50 years, finding the maximum possible amount of rainfall in the study area is necessary. Therefore, five subbasins in the upper part of the DRB were chosen to determine the potential amount of rainfall in them, based on the NASA recorded database, as shown in Figure 14. Using equations (2) and (3), the PMP depth in five selected areas could be calculated, as well as the years in which the maximum depths occur, as shown in Table 1. After observing the maximum depth achieved in all subbasins, it was found that the year in which the maximum depth occurred was 2019, during the study period of 1981–2021. The PMP was near the significant values of average annual precipitation in 2019 for all subbasins.

Figure 14

Chosen subbasins for calculating PMP of UDRB.

Table 1

PMP values of chosen subbasins of UDRB

Subbasin	PMP (mm/day)
1 (2019)	1.871
8 (2019)	1.474
9 (2019)	1.474
11 (2019)	1.474
18 (2019)	1.385

A code was used to implement the NN described in (3.4.2) for the selected subbasins. The rainfall data obtained from the NASA database and the annual rates in all selected subbasins were considered. The relationship between the recorded values and the values estimated by the NN model is shown in Figure 15, which recorded an R ² value of 0.9898, and this value indicates a strong relationship and high accuracy in representing the reality by the model.

Figure 15

Predicting model for the next 50 years by NN.

Returning to the hydraulic model (HEC-RAS) and using new critical boundary conditions, the results of extracting water levels and water surface profiles in the Hemrin reservoir are shown in Figure 16.

Figure 16

Simulated RAS model with critical case for UDRB and Hemrin Dam.

The water level of Hemrin Dam exceeded the design one, reaching over 110 m. So, a suggested spillway has been adopted as a solution. Through searching for elevations near the Hemrin Dam reservoir, the line of the spillway canal was selected, as shown in Figure 17.

Figure 17

Suggested line of spillway canal from Hemrin reservoir.

As depicted in the preceding figure, the proposed canal requires excavation and embankment activities, primarily in the first 12 km of its line.

The proposed canal, spanning approximately 100 km, is located on the left side of the Hemrin reservoir at an elevation of 105 m. The elevation gradually decreases to around 22 m throughout the first 10 km, with a slope of 0.0028. The remaining canal has a slope of approximately 0.00035. Manning’s values for the top and lower sections of the canal are 0.035 and 0.028, respectively. At the canal’s terminus is a low-lying marshland called Al-Shweijah Marsh, covering an area of about 450 km².

By choosing a weir with 400 m width and re-running the RAS model, the results are shown in Figure 18.

Figure 18

Water surface elevation in the critical case of Hemrin reservoir before and after adding the suggested canal.

Marsh works a maximum of 10 m, indicating a significant difference between the reservoir’s low level and its maximum design limit of 107.5 m (with a maximum depth of 21.8 m). In the studied scenario with the highest river flow and safe storage in the Hemrin reservoir, approximately 1.24 BCM of water could be stored in the proposed reservoir for about 43 days.

The researcher proposes linking the proposed reservoir (marsh) to the Tigris River via a canal in Kut Governorate, as shown in Figure 19. The canal is approximately 20 km long. It can transport significant amounts of water to the Tigris River during summer. This is important for raising water levels, reducing salinity, and other purposes.

Figure 19

Proposed new links with Tigris from proposed canal.

7 Conclusions

The results of this study revealed several significant findings:

ArcGIS-based remotely sensed images (DEM and TIN) provided a valuable data source for the analysis, especially in determining the levels, slopes, and lines for stormwater drainage and the course of Diyala River with its tributaries.

The use of ANN was essential and helpful in predicting the amount of precipitation in the next 50 years in the DRB.

HEC-RAS software provides the possibility of reducing the size of the grid along the river and enlarging its height in the river basin on both sides of the river, significantly saving operation time.

The study showed that there is a high probability of flooding and exceeding the allowable level of the Hemrin Dam reservoir, as it may reach an elevation of more than 110 m in Hemrin reservoir and thus threaten the region significantly, especially in the period between March and the end of April of each year. The proposed canal reduced the risk of this potential flood and kept the reservoir level within the design limits.

The study showed the possibility of disposing surplus water in times of floods by discharging it to a lowland through an artificial canal and linking it to the Tigris River when this quantity is needed in drought.

Considering the absence of similar studies in the mentioned region, the comparison was limited to satellite imagery of the site for the cases during the studied period.

The use of the conveyance channel will also provide the opportunity for utilization in the fields of electricity generation and irrigation in areas that were previously not reached by water. However, this requires significant effort and additional costs, and its operation becomes effective in cases where the water level in the Hemrin Dam reservoir rises to more than 105 m above sea level.

The article has introduced a unique and comprehensive hydraulic model design for the study area, which marks a new beginning for utilizing the outputs for decision-makers and researchers in the field of drought and floods in this region.

Funding information: We declare that the manuscript was done depending on the personal effort of the author, and there is no funding effort from any side or organization.
Conflict of interest: The authors state no conflict of interest.
Data availability statement: Most datasets generated and analyzed in this study are in this submitted manuscript. The other datasets are available on a reasonable request from the corresponding author with the attached information.

References

[1] Tanty R, Desmukh TS. Application of artificial neural network in hydrology – A review. Int J Eng Res. 2015;4(6):184–8.10.17577/IJERTV4IS060247Search in Google Scholar

[2] Fariba A, Jalal A, Ali M. Modelling equitable and reasonable water sharing in transboundary rivers: the case of Sirwan-Diyala river. Water Resour Manage. 2017;31:1191–107.Search in Google Scholar

[3] Chomba IC, Banda KE, Winsemius HC, Chomba MJ, Mataa M, Ngwenya V, et al. A review of coupled hydrologic-hydraulic models for floodplain assessments in Africa: Opportunities and challenges for floodplain wetland management. MDPI Hydrol. 2021;8(1):44.10.3390/hydrology8010044Search in Google Scholar

[4] Abbas N, Wasimi SA, Al-Ansari N. Impacts of Climate Change on Water Resources in Diyala River Basin, Iraq. J Civ Eng Archit. 2016;10(9):1059–74.10.17265/1934-7359/2016.09.009Search in Google Scholar

[5] Madhloom HM, Al-Ansari N, Laue J, Chabuk A. Modeling spatial distribution of some contamination within the lower reaches of Diyala river using IDW interpolation. MDPI Sustain. 2018;10(1):22.10.3390/su10010022Search in Google Scholar

[6] Madhloom HM, Al-Ansari N. Geographical information system and remote sensing for water resources management case study: The Diyala River, Iraq. Int J Civ Eng Technol. 2018;9(12):971–84.Search in Google Scholar

[7] Abdullah TH, Jalut QH, Ameen YW. Hydrologic modeling for sedimentation in Hemrin reservoir using HEC-HMS. Diyala J Eng Sci. 2018;11(4):67–72.10.24237/djes.2018.11410Search in Google Scholar

[8] Al-Musawi NO, Al-Rubaie FM. Application of artificial neural network and geographical information system models to predict and evaluate the quality of Diyala River water, Iraq. Assoc Arab Univ J Eng Sci. 2019;26(1):160–9.10.33261/jaaru.2019.26.1.021Search in Google Scholar

[9] Ibrahim A, Saed H. Derbendikhan and its effects on Diyala River using GIS. Diyala J. 2019;79:298.Search in Google Scholar

[10] Aziz NA, Abdulrazzaq ZT, Mansur MN. GIS-based watershed morphometric analysis using DEM data in Diyala River, Iraq. Iraqi Geol J. 2020;53(1):36–49.10.46717/igj.53.1C.3Rx-2020.04.03Search in Google Scholar

[11] Kadhim TH, Al-kubaisi MS. Evaluation land use -land cover classification of Diyala River area using Landsat 8 images. Iraqi J Sci Technol. 2020;11(2):70.Search in Google Scholar

[12] Kadhim MA, Al-Bedyry NK, Omran II. Evaluation of flood routing models and their relationship to the hydraulic properties of the Diyala River Bed. IOP Conf Ser Earth Environ Sci. 2022;961(1). 10.1088/1755-1315/961/1/012058.Search in Google Scholar

[13] Jassam WA, Abed BS. Hydraulic characteristics of the lower part of Diyala River. IOP Conf Ser Mater Sci Eng. 2021;1105(1):012107.10.1088/1757-899X/1105/1/012107Search in Google Scholar

[14] Karim IR, Hassan ZF, Abdullah HH, Alwan IA. 2D-Hec-Ras modeling of flood wave propagation in a semi-arid area due to dam overtopping failure. Civ Eng J. 2021;7(9):1501–14.10.28991/cej-2021-03091739Search in Google Scholar

[15] Ekwueme BN. Machine learning based prediction of urban flood susceptibility from selected rivers in a tropical catchment area. Civ Eng J. 2022;8(9):1857–78.10.28991/CEJ-2022-08-09-08Search in Google Scholar

[16] Al-Ansari NA, Al-Sinawi GT, Jamil AK. Suspended and solute loads on the Lower Diyala River. Vol. 159. Wallingford, UK: IAHS-AISH Publication: 1986; p. 225–235.Search in Google Scholar

[17] Ibrahim IA, Al-Dabbas M. Analysis of climate parameters as indicators of climate changes in central and eastern Iraq: Khanaqin climate conditions as a case study. Iraqi J Sci. 2021;62(12):4747–57.10.24996/ijs.2021.62.12.13Search in Google Scholar

[18] Subramahya K. Engineering Hydrology. New Delhi: McGraw-Hill; 2014. p. 35.Search in Google Scholar

[19] UN-ESCWA, Rohstoffe B für G und. Shared tributaries of the Euphrates River. Invent Shar Water Resour West Asia. IRAQ: Iraqi Ministry of Water Resources; 2013. p. 79–97.10.18356/24552f7e-enSearch in Google Scholar

[20] Hancock GR, Martinez C, Evans KG, Moliere DR. A comparison of SRTM and high-resolution digital elevation models and their use in catchment geomorphology and hydrology: Australian examples. Earth Surf Process Landf. 2006;34(March):613–28. http://doi.wiley.com/10.1002/esp.1730.Search in Google Scholar

[21] Martin O, Rugumayo A, Ovcharovichova J. Application of HEC-HMS/RAS and GIS tools in flood modeling: A case study for river Sironko – Uganda. Glob J Eng Des Technol. 2012;1(2):19–31.Search in Google Scholar

[22] Avarideh F, Attari J, Moridi A. Modelling equitable and reasonable water sharing in transboundary rivers: The case of Sirwan-Diyala river. Water Resour Manag. 2017;31(4):1191–207.10.1007/s11269-017-1570-4Search in Google Scholar

[23] US Department of Agriculture. Flood routing. Natl Eng Handb; 2014. p. 630.Search in Google Scholar

[24] World Meteorological Organization. Manual for Estimation of Probable Maximum Precipitation, Operational Hydrology Report 1. 2nd edn. Publication 332; 2009. p. 291.Search in Google Scholar

[25] Sarkar S, Maity R. Estimation of probable maximum precipitation in the context of climate change. MethodsX. 2020;7:100904.10.1016/j.mex.2020.100904Search in Google Scholar PubMed PubMed Central

[26] Majumder M, Kale GD. Water and energy management in India. e-book. Spinger; 2021. 10.1007/978-3-030-66683-5.Search in Google Scholar

[27] Tamiru H, Dinka MO. Application of ANN and HEC-RAS model for flood inundation mapping in lower Baro Akobo River Basin, Ethiopia. J Hydrol Reg Stud. 2021;36:100855. 10.1016/j.ejrh.2021.100855.Search in Google Scholar

[28] Pimenidis E, Angelov P, Jayne C, Papaleonidas A, Goos G. Artificial neural networks and machine learning – ICANN 2022. 31st Int Conf Artif Neural Networks Bristol, UK, Sept 6–9, 2022. 31st Inter. 2022.10.1007/978-3-031-15934-3Search in Google Scholar

[29] Cartwright H. Artificial neural network. Vol. 2190. Oxford University. Humana Press; 2021. p. 33–71.Search in Google Scholar

[30] Badano N, Valdés RE, Álvarez EÁ. Tidal current energy potential of Nalón river estuary assessment using a high precision flow model. Open Eng. 2018;8(1):118–23.10.1515/eng-2018-0015Search in Google Scholar

Received: 2023-06-13

Revised: 2023-08-12

Accepted: 2023-09-05

Published Online: 2024-02-09

This work is licensed under the Creative Commons Attribution 4.0 International License.

Articles in the same Issue

https://doi.org/10.1515/eng-2022-0530

Keywords for this article

hydraulic model; HEC-RAS; ANN; Diyala River Basin; probable maximum precipitation

Creative Commons

BY 4.0