Flood hazard zoning using geographic information system (GIS) and HEC-RAS model (Case study: Rasht City)

[Research]

Flood hazard zoning using geographic information system (GIS) and    HEC-RAS model (Case study: Rasht City)

V. Gholami*¹, A. Asghari² , E. Taghvaye  Salimi¹

1 - Department of Range and Watershed Management, Faculty of Natural Resources, University of Guilan, Sowmeh Sara, Iran.       

2- Department of Geography and Urban Planning, Payam Noor University of Shahryar, Shahryar, Iran

* Corresponding author’s E-mail: Gholami.vahid@gmail.com

ABSTRACT

Rivers are important water resources for human life, but sometimes cause irreparable damages. The flood plains are fertile terrains which are endangered by flood. Flood hazard mapping is one of the basic methods in flood fighting. In order to decline flood damages, the simulation of the hydraulic behavior of the rivers during flood occurrence is very important. In this study, areas that are flooded were zoned along Siahrood and Goharrood rivers (Rasht City, northern Iran). The rivers bed and banks terrains of Goharrood and Siahrood were simulated using HEC-GeoRAS extension and digital map (scale: 1000). Pick discharges with different return periods were estimated using stochastic analysis. HEC-RAS software and geographical information system were applied for simulating the hydraulic behavior of the rivers and providing flood zoning map. The GIS was an efficient tool for data-processing and mapping stages. Finally, the flood zones associated with 2, 10, 25, 50, 100 and 200 years return periods were mapped and necessary analysis were conducted during the present research. The results showed that some parts of Rasht City (river bank terrains) are endangered flood hazard.

INTRODUCTION

Hydraulic models and geographic information system (GIS) can be effectively applied in flood hazard zoning. For flood studies, terrain geometry is one of the most important factors. Using GIS can provide higher speed and accuracy in simulating terrain geometry. The best strategy to avoid flood damage is certainly to avoid flood prone areas (Hooijer et al. 2004; Roy et al. 2003; Kreibich et al. 2005). Islam and Sado (2000) used GIS to simulate hydraulic behavior of floods in several rivers of Bangladesh. Their study showed that GIS increased the accuracy of the results in simulation.

Also, in flood management, estimating the maximum flood discharge is necessary for predicting watershed hydrological behavior (Gholami et al. 2010). A flood is an overflow of an expanse of water that submerges the land. Flood hazards are the most common and destructive of all natural disasters and are a constant threat to life and property. Each year, flood disasters result in tremendous losses and social disruption worldwide (Uddin et al. 2013). Flood management in a basin will not be successful unless the watershed’s hydrological behaviors are predicted (Bhadra et al. 2008; Esmaeeli Gholzom & Gholami 2012; Khaleghi et al. 2014). We can estimate the maximum flood discharge using stochastic distributions. In recent years, agencies and other groups engaged in disaster mitigation have placed much emphasis on the objective of achieving disaster-resilient communities (Kreibich et al. 2005). Recently, there has been an increase of interest in using the program HEC-RAS to evaluate the river hydraulic simulation (Ortega & Graz 2009). Ortega et al. (2009) used hydraulic model based on the SWD-PSI by employing the program HEC-RAS, which was adjusted to the model with a calibration of the rating curve of recent floods. Balasch et al. (2011), calculated peak flows by utilizing the HEC-RAS hydraulic modeling software on one-dimensional, gradual, varied, steady, sub-critical flow. For flood hazard zoning, supply and processing of flood zoning maps in Vadu watershed, Austin, Texas, Anagram et al. (1999) used the HEC-RAS hydraulic model with the aid of aerial photographs. Hill (2001) mentioned various benefits in using HEC-GeoRAS (ArcView-GIS) extension and HEC-RAS software for flood hazard zoning. Santos et al. (2011), used the HEC-GeoRAS to compare historical-hydro geomorphological reconstitution and hydrological-hydraulic modeling in the estimation of flood-prone areas. Hyalmarson (1988) investigated the river bed erosion and deposition while flooding. Jonson et al. (1999), used HEC-RAS model in predicting and determining the boundary of wetlands along 10 km of Vioming Griol in US. Carson (2006) simulated the river hydraulic behavior and investigated the flood and bank erosion hazard in Utah, US. Different studies have been done in terms of flood hazard zoning such as: Gogoi et al. (2010) and Carrasco et al. (2013). The present study was carried out to map flood hazard zones in Goharrood and Siahrood rivers in Rasht City (Northern Iran).

MATERIALS AND METHODS

Rasht City is located in northern Iran which is the capital of Guilan Province (Fig. 1). Goharrood and Siahrood rivers after passing from the Alborz Mountains and then Rasht City fall into the Caspian Sea. The rivers flow from Sefidrood River watershed. Sefidrood River is the biggest river in Guilan Province. Urban flood is one of the main problems in the Rasht City management (Fig. 2). This study has been done in the urban part of the rivers in Rasht City. Rasht is one of the most vulnerable cities of Iran in terms of crisis management planning and dealing with disasters such as earthquakes, floods, and snow. Rasht City has an area of about 10, 2 hectares and 104 urban districts. Regarding the items mentioned above and considering Rasht as one of the large cities in the north of Iran, with high population density and old texture, and one of Iran most vulnerable cities in Iran, comprehensive planning in crisis management is necessary in this city and the preparedness of governmental organizations as supportive and protective forces for citizens is more important (Hosseini et al. 2013). Past floods caused damages to the trains of rivers banks. Urban flood is one of the main problems in the Rasht City, especially for traffic.

To simulate the hydraulic behavior of the river, geographic information system (HEC-GeoRAS extension and ArcView software) and HEC-RAS software were used, due to their very high ability of mentioned software (Kang 2010). At first, the annual statistical data of maximum instantaneous discharges were analyzed in terms of homogeneity, relevancy and adequacy then verified with 95% of confident limit via Run test method. Then, the best statistical distribution was selected using Smada software (Log Pearson III) and the maximum discharges were estimated by applying suitable statistical distribution. In this method, statistical distribution which can cover most of the calculated and observed data was selected as the best distribution. With the aid of the HEC-GeoRAS extension in ArcView software, the data of hydrometric stations in Lakan and Siahrood were used according to the river plan maps with the scale of 1:1000. The bed conditions such as the center stream line, banks, flow paths, cutlines and river floodplain in the objective section, were also simulated. A view of urban map with a scale of 1:1000 is presented in Fig. 3. Cross-sections 134 for Goharrood and cross-sections 63 for Siahrood were introduced as the general conditions of the rivers based on their bed situation. In next step, then, Manning coefficient of these cross-sections were separately determined for bed, right and left banks by field studies, and river path monitoring.  Cawen method and GPS set were also employed. According to Cawen method, to estimate the Manning roughness coefficient, n = (n₀+ n₁ + n₂ + n₃ + n₄) m₅ in which n is Manning roughness coefficient, n₀ is bed roughness coefficient that depends on channel materials, which is selected for homogeny, smooth and direct channels and other roughness coefficients which include   n₁ , n₂ , n₃ , n₄ and m_5. They are coefficients of the effects of cross section irregularity, variation of cross section, existence of barriers in channel path, vegetation and meandering degree in the path (Kang 2010). The simulated bed conditions were imported from GIS to HEC-RAS medium to present a model and simulation of the hydraulic behavior of Goharrood and Siahrood rivers. Flow regime was considered mixed and normal depth was used. Finally, hydraulic behavior (water level and depth, water velocity and so on) of these rivers in different return periods were simulated and flood hazard zones were identified in Rasht City.

Fig. 1. Location of Rasht City in the north of Iran.

Fig. 2. Urban flood due to rainfall in Rasht. Flood results from river overflowing and the inefficiency of drainage channels in the City.

Fig. 3. The map of Rasht City, scale 1:1000.

RESULTS

Physical model of beds and adjacent   areas of the rivers (geometry) were simulated using GIS (HEC-GeoRAS extension) and digital maps are given in Fig. 4. The hydrometric stations data of Lakan (Goharrood River) and Siahrood were applied in estimating the maximum instantaneous discharges and were analyzed by means of Smada software. According to Figs. 5 and 6, the Log Pearson Type III distribution is the best for estimating the maximum instantaneous discharges in different return periods, the results are demonstrated in Table 1.

In order to study the rivers, a comparison between observed and estimated discharge values have been exhibited in Figs. 5 and 6 using the Log Pearson type III distribution. After presenting the hydraulic model of Siahrood and Goharrood rivers and simulating hydraulic behavior in discharges with different return periods, length-section of the rivers and water level in different return

periods are given in Figs. 7 and 8. A cross-section in Goharrood River and water level in different return periods are demonstrated in Fig. 9. The results revealed that the Goharrood and Siahrood rivers have   not been flooded in most of their sections in time return (Tr) 2 to 50 years. In the current geometric conditions of the rivers, we observed the river floods in T_r 200 year. So, considering floods with time return of 200 years is essential for management against disaster and planning the flood control projects. After modeling the hydraulic behavior of the river in HEC-RAS model, we imported the hydraulic model results in GIS medium. We evaluated the zoning of flood in 200 years because the other times flood hazards were not notable. In fact, A 200-year flood zoning was investigated because other flood time return zones were not extensive. Some examples of a 200-year flood hazard zoning map for the rivers in Rasht City are shown in Figs. 9 - 11.

Table 1.  Maximum instantaneous discharges in Siahrood and Goharrood rivers.

Fig. 4. 3D model (TIN), center stream line, banks and 3D cutlines in a part of study path, Goharrood River.

Fig. 5. Comparison between observed and estimated values using Log Pearson type III distribution (Lakan, Goharrood station).

Fig. 6. A comparison between observed and estimated values using Log Pearson type III distribution (Siahrood Station).

Fig. 7. Length-section of Siahrood and water level in different return periods.

Fig. 8. Length-section of Goharrood and water level in the different return periods.

Fig. 9. Flood zone with Tr200 of Goharrood River on the surface of Rasht City.

Fig. 10. Flood zone with Tr200 of Goharrood River on  the surface of Rasht City.

Fig. 11. Flood zone with Tr200 of Siahrood River on the surface of Rasht City.