An-Najah National University 

Faculty of Graduate Studies 

 
MODELING RUNOFF OF WADI AL ZEIMAR 

CATCHMENT CONSIDERING CLIMATE 

CHANGE 

 

By 

Shahd Abdulhakim Yousef Bishawi 

 

 

 

Supervisor: 

Dr. Mohammad N. Al Masri 

 

 

This Thesis is Submitted in Partial Fulfillment of the Requirements for the Degree of 

Master of Water and Environmental Engineering, Faculty of Graduate Studies,  An-Najah 

National University, Nablus - Palestine. 

2024 





III 

 

Dedication 

To my parents, I would want to dedicate this thesis. Many thanks for everything! I really 

don't know how to express how much I appreciate and thank you. You have served as my 

inspiration, pillar of support, and mentor. I've learned from you to be bold, tenacious, and 

self-assured. Having you as my parents makes me incredibly grateful and humbled. Thank 

you for everything. To take a quote from Albert Schweitzer, “At times our own light goes 

out and is rekindled by a spark from another person. Each of us has cause to think with 

deep gratitude of those who have lighted the flame within us.” 

  



IV 

 

Acknowledgements 

"My Lord, increase me in knowledge." Quran-Verse (20:114) 

I am grateful, to Allah, for His blessings that have helped me complete my thesis. 

I also acknowledge the guidance of our Prophet Muhammad SAW, who has 

contributed to the betterment of our lives today.  

As I complete my thesis, I intend to send out my sincere thanks for all of the kind 

advice and help that many individuals have offered me. 

I express my gratitude to my supervisor Dr. Mohammad N. Almasri, for his aid and 

support. In addition, I would like to express my deepest gratitude to the members of my 

defense committee for their invaluable guidance, support, and encouragement throughout 

this thesis. 

I also thank my parents and brothers, for all the support and encouragement they have 

given me throughout my master’s enrollment.  

Special thanks and gratitude to my husband and son, who have been very helpful and 

understanding in my journey.  

I would also like to express our gratitude to the Middle East Desalination Research Center 

(MEDRC), represented by Dr. Subhi Samhan, for their financial assistance through the 

Palestinian Water Authority (PWA) scholarship program. 

 

 

 

 

 

 

 

 

 

 





VI 

 

Table of Contents  

Dedication ....................................................................................................................... III 

Acknowledgements ......................................................................................................... IV 

Declaration ....................................................................................................................... V 

Tables of Contents .......................................................................................................... VI 

List of Tables ................................................................................................................ VIII 

List of Figures ................................................................................................................. IX 

List of Appendices .......................................................................................................... IX 

Abstract ........................................................................................................................... XI 

Chapter One: Introduction and Theoretical Background .................................................. 1 

1.1 General Background .................................................................................................. 1 

1.2 Objectives .................................................................................................................. 3 

1.3 Research Questions .................................................................................................... 3 

1.4 Research Needs .......................................................................................................... 4 

1.5 Literature Review ...................................................................................................... 4 

Chapter Two: Materials and Methods ............................................................................... 8 

2.1 Study Area ................................................................................................................. 8 

2.1.1 Location ................................................................................................................... 8 

2.1.2 Population ................................................................................................................ 8 

2.1.3 Geology ................................................................................................................... 8 

2.1.4 Soil Cover ................................................................................................................ 9 

2.1.5 Land Use .................................................................................................................. 9 

2.1.6 Wadis ....................................................................................................................... 9 

2.1.7 Climate .................................................................................................................. 10 

2.1.8 Temperature ........................................................................................................... 10 

2.1.9 Rainfall .................................................................................................................. 11 

2.1.10 Climate Change .................................................................................................... 12 

2.2 The Methodology ..................................................................................................... 13 

2.2.1 Data Collection and Analysis ................................................................................ 14 

2.2.1.1 Ground Surface Contours.................................................................................... 14 

2.2.1.2 Lag Time ............................................................................................................. 14 

2.2.1.3 Land Use / Land Cover and Soil Maps ............................................................... 15 



VII 

 

2.2.1.4 Channels’ Average Cross Section and Properties ................................................ 17 

2.2.2 Model Development .............................................................................................. 19 

2.2.2.1 Terrain Data ......................................................................................................... 20 

2.2.2.2 Basin Model ........................................................................................................ 21 

2.2.2.3 Time-Series Data ................................................................................................. 25 

2.2.2.4 Meteorological Models ....................................................................................... 27 

2.2.2.5 Control Specifications ......................................................................................... 27 

2.2.3 Model Calibration and Validation ......................................................................... 27 

2.2.4 Projection of Extreme Precipitaion Indices to Simulate Climate Change ............. 29 

2.2.4.1 Extreme Precipitation Indices of the Study Area ................................................ 29 

2.2.4.2 Frequency Analysis of Nablus Station Meteorological Data .............................. 30 

2.2.4.3 Meteorological Climate Change Scenario Input for the Model .......................... 32 

2.2.5 Assessment of Anabta’s Culverts .......................................................................... 33 

2.2.5.1 Description of Anabta’s Culvert System ............................................................. 33 

2.2.5.2 HY-8 Culvert Hydraulic Analysis Program ........................................................ 34 

Chapter Three: Results .................................................................................................... 35 

3.1 Simulation Runs ....................................................................................................... 35 

3.2 Model Calibration .................................................................................................... 37 

3.3 Model Validation...................................................................................................... 39 

3.4 Climate Change Scenario Run ................................................................................. 42 

3.5 Culvert No.1 Redesign ............................................................................................. 43 

Chapter Four: Conclusions, Limitations and Recommendations ................................... 45 

4.1 Conclusions .............................................................................................................. 45 

4.2 Limitations ............................................................................................................... 45 

4.3 Recommendations .................................................................................................... 46 

List of Abbreviations ...................................................................................................... 47 

References ....................................................................................................................... 48 

Appendices ...................................................................................................................... 57 

 ب ................................................................................................................................. الملخص 

 

  



VIII 

 

List of Tables 

Table 1: Communities with each sub-basin ................................................................................. 23 

Table 2: Physical properties of each sub-basin ........................................................................... 23 

Table 3: Initial input parameters for each sub-basin ................................................................... 24 

Table 4: Initial input parameters for each reach .......................................................................... 24 

Table 5: Classification of model performance according to E .................................................... 29 

Table 6: Extreme Precipitation Indices for Nablus for 1997-2022 ............................................. 30 

Table 7: Extreme Precipitation Indices for Nablus for 2041-2070 ............................................. 30 

Table 8: Initial and Calibrated Lag Time .................................................................................... 38 

Table 9: Initial and Calibrated SCS Curve Number Parameters ................................................. 38 

Table 10: Initial and Calibrated Muskingum Parameters ............................................................ 39 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



IX 

 

List of Figures 

Figure 1: Al Zeimar Catchment Basin Model ............................................................................. 22 

Figure 2: Cumulative Distribution Function of maximum daily rainfall data of Nablus Station 32 

Figure 3: Hyetograph of the synthetic storm ............................................................................... 33 

Figure 4: Hydrograph of Run 1 ................................................................................................... 35 

Figure 5: Hydrograph of Run 2 ................................................................................................... 37 

Figure 6: Hyetograph of Storm 2 ................................................................................................ 40 

Figure 7: Hydrograph of storm 2 ................................................................................................ 40 

Figure 8: Hydrograph of Run 3 ................................................................................................... 41 

Figure 9: Hydrograph of Run 4 ................................................................................................... 42 

Figure 10: Side View of the Culvert ............................................................................................ 44 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



X 

 

List of Appendices 

Appendix A: Figures ................................................................................................................... 57 

Figure A.1: Al Zeimar Catchment Location and Digital Elevation Model ............................ 57 

Figure A.2: Al Zeimar Communities...................................................................................... 58 

Figure A.3: Al Zeimar Catchment Soil Cover ........................................................................ 58 

Figure A.4: Al Zeimar Land Use ............................................................................................ 59 

Figure A.5:  AL Zeimar Catchment Wadis ............................................................................. 59 

Figure A.6: Average Annual Temperatures of Summer and Winter for the Nablus Station ... 60 

 Figure A.7: Average Annual Rainfall of  Nablus .................................................................. 61 

Figure A.8: Number of Rainy Days of Nablus Statoin .......................................................... 61 

Figure A.9: Average Annual Precipitation of Tulkarem ......................................................... 62 

Figure A.10: Al Zeimar Rainfall Contours ............................................................................. 62 

Figure A.11: Methodology Flow Chart .................................................................................. 63 

Figure A.12: Al Zeimar Catchment Sink-Partial Flume ........................................................ 64 

Figure A.13: Hyetograph of Storm 1 ..................................................................................... 64 

Figure A.14: Storm 1 Hydrograph ......................................................................................... 65 

Figure A.15: Built Up Section of Al-Zeimar Stream in Anabta ............................................. 65 

Figure A.16: Alexa Storm Hyetograph .................................................................................. 66 

Figure A.17: Aerial View of Al Zeimar Stream ..................................................................... 66 

Figure A.18: Culvert in Anabta .............................................................................................. 67 

Appendix B: Tables ..................................................................................................................... 68 

Table B.1: Communities within Al Zeimar ............................................................................ 68 

Table B.2: Soil Types within Al Zeimar ................................................................................. 69 

Table B.3: Land Use Within Al Zeimar ................................................................................. 69 

Table B.4: Wadis within Al Zeimar ........................................................................................ 70 

Table B.5: Curve Number Tables ........................................................................................... 71 

Table B.6: Infiltration Rate for each Soil Group .................................................................... 74 

Table B.7: Anticedent Moisture Content  ............................................................................... 74 

Table B.8: Mannings Coefficent values ................................................................................. 74 

Table B.9: Maximum Daily Rainfall for Nablus Station (1997-2022) .................................. 75 

 

 

 



XI 

 

MODELING RUNOFF OF WADI AL ZEIMAR CATCHMENT 

CONSIDERING CLIMATE CHANGE 
 

By 

Shahd Abdulhakim Yousef Bishawi 

Supervisor 

Dr. Mohammed N. Almasri 

 

Abstract 

Water covers a big portion of Earth's surface, but the availability of freshwater 

suitable for human use is limited. Scientists are focusing on hydrology and water 

resources management, with hydrological modeling gaining popularity for 

predicting water resource changes. Modeling helps understand processes like 

rainfall, evapotranspiration, infiltration, runoff, and underground water 

movement in the hydrological cycle. In regions like Palestine, intense storms can 

lead to flash floods due to primary precipitation processes. Climate change poses 

a significant challenge, as rising global temperatures are expected to alter 

hydrological processes, increasing the frequency of droughts and floods. To 

address flood risks and adapt to these changes, understanding the hydrological 

response to future climate change is crucial. This study integrated a hydrological 

model of the Al Zeimar catchment in Palestine with climate projections to assess 

the impact of future climate trends on watershed hydrology. The study used HEC-

HMS 4.10 to model the rainfall-runoff relationship of the catchment, achieving 

reliable results with a Nash-Sutcliffe efficiency coefficient of 0.863. Climate 

change projections indicated rising temperatures, decreased precipitation, and 

increased maximum daily rainfall, heightening the risk of flash floods.  Frequency 

analysis on maximum daily precipitation at the Nablus rainfall station from 1997-

2022 forecasted a peak runoff of 92 m3/second for a 25-year return period event, 

necessitating the redesign of a culvert at the entrance of Anabta Town. The 

culvert, initially sized at 10 m2, was enlarged to 30 m2 to withstand the expected 

storm intensity. However, this solution is not feasible nor applicable. 

Recommendations included further assessment of other culverts in the town, and 

searching for solutions other than enlarging culverts. 

Keywords: Hydrologic Modeling; Run-Off; Wadi Al Zeimar; Climate Change;  



1 

 

Chapter One 

Introduction and Theoretical Background 

1.1 General Background 

During a rainstorm, the question “What happens to the rain?” might come to mind. In 

1961 Penman listed Hydrology as the branch of science that seeks to answer this question 

(Singh & Woolhiser, 2002). Rainfall is part of the hydrologic cycle. The term cycle 

suggests that water is generated from one source and ultimately finds its way back there. 

Oceans are the source and last destination of water. Ocean water evaporates and turns 

into atmospheric water vapor. A portion of the moisture present in the atmosphere is 

released as precipitation, which occasionally evaporates before it reaches the ground's 

surface. Some of the water brought to the Earth’s surface by precipitation may evaporate 

on site while some might seep into the ground and some could flow across the land to 

form streams. Once water has seeped into the earth, it can evaporate, seep below to 

groundwater reservoirs, or be absorbed by plant roots and then transpired by the plants. 

When water enters groundwater reservoirs, it can either flow straight into streams or the 

ocean, or it can migrate laterally until it gets near enough to the surface to be exposed to 

transpiration or evaporation. It can also reach the land surface and produce lakes, seeps, 

or springs. Flowing stream water can condense in lakes and surface reservoirs, seep into 

groundwater reservoirs, evaporate or be transpired by riparian plants, or return to the 

ocean, where the cycle starts all over again (ASCE, 2013). In this study, rainfall is the 

most significant process. The main element influencing the hydrology of wadi systems 

in this study is rainfall. In arid and semi-arid locations, wadi systems are frequently 

affected by irregular storms which fluctuate significantly in terms of both time and 

location. The part of rainfall that enters wadi channel systems after infiltration, detention 

storage, and other abstractions are met is known as direct runoff, or storm flow. Surface 

runoff, which begins as a thin layer of sheet flow known as overland flow, is frequently 

the predominant runoff pathway because of high rainfall intensities and sparse 

vegetation.  This overland flow becomes concentrated and changes into channel flow 

when it approaches a wadi or canal. Flood flows typically lose water by infiltration into 

the channel bed in dry and semi-arid regions. These losses are referred to as transmission 

losses. These losses may decrease channel flow while approaching downstream 



2 

 

(Weshah, 2010). Understanding all the above is essential and can be achieved through 

modelling. 

Simplified representations of real hydrologic systems, known as hydrologic models, help 

better understand hydrologic processes through studying the functioning of watersheds 

and their response to different inputs (Verma, Jha, & Mahana, 2009) . Watershed 

hydrological models are designed to answer Penman’s question at a specific level of 

detail depending on the problem at hand (Singh, Computer Models of Watershed 

Hydrology, 1995). The availability of adequate data to accurately represent each of the 

modeled processes is a fundamental need for any physically based model. Geospatial 

technologies, including geographic information systems (GIS), global positioning 

systems (GPS), and remote sensing, are enabling significant advancements in data 

collection, preparation, and administration (Daniel, et al., 2011). (Beckers, Smerdon, & 

Wilson, 2009) suggest that a crucial element to enhance the performance of hydrologic 

models is to properly quantify climate change, which includes potential future changes 

in temperature and precipitation, the occurrence of severe events, and changes in glacier 

mass balances. Climate change is widely acknowledged as the biggest environmental 

issues the world is now facing (Parry, Canziani, Palutikof, Linden, & Hanson, 2007). 

According to Eltahir and Pal (2022), future temperatures in the Middle East are predicted 

to rise, leading to increased evapotranspiration and potential changes in climatic patterns 

resulting in reduced overall rainfall in the region. On the other hand, recent studies 

indicate that the Eastern Mediterranean area is projected to experience an increase in 

extremely heavy daily rainfall and a decrease in annual precipitation as a result of global 

warming (Almazroui, Islam, & Jones, 2023). The Mediterranean climate is often 

characterized by its irregular rainfall patterns, which may aggravate water stress during 

specific times and lead to a period of water scarcity (Ceballos, Fernandez, & Ugidos, 

2003). A climate change study in Israel (Historic Palestine) concluded that by 2050, 

mean temperature will escalate by 0.7⁰-0.8⁰C and there will be a drop in precipitation by 

4 % to 2% (Pe'er & Safriel, 2000). 

Palestine is made up from 33 watersheds where one of them is the well-known Al-Zeimar 

watershed. Al Zeimar stream runs from the hills of Nablus in the West Bank to the coastal 

region of Tulkarm and discharges into the Mediterranean Sea. The watershed is made up 

of a variety of land use and land cover types including built-up areas (for instance the 



3 

 

western part of Nablus City). The built-up areas are responsible by large in the generation 

of the surface runoff at high rates. Due to that, and in addition to climate changes, the 

watershed encountered in the past many incidents of flooding that did lead to 

catastrophic consequences of infrastructure damage and death causalities. In January of 

2013, two young girls lost their lives due to the flooding of Al Zeimar in Anabta area in 

Tulkarm. This fatal event was a shock to the community, because nothing similar 

occurred in decades before. The culverts in the area are usually partially blocked because 

residents dump everything there. This, in addition to the powerful storm, knocked down 

trees into the path of the culvert, which lead to backflow and flooding of the area. This 

thesis aims at assessing the impact of climate change on flood hazard in Al Zeimar. To 

achieve this purpose, the hydrological model of the catchment was developed using 

HEC-HMS (Hydrologic Engineering Center-Hydrologic Modeling System). After the 

model was calibrated and validated, it was used to project the climate change shifts in 

the study area. Moreover, a culvert at the entrance of Anabta Town was tested for peak 

flow and re-designed to handle a 25-year return period storm.   

1.2  Objectives 

The following are the objectives of the research:  

• To model the rainfall-runoff generation of Wadi Al Zeimar using HEC-HMS 

• To simulate the effect of future climate changes on the runoff generation in the study 

area  

• To redesign the culvert at the entrance of Anabta using HY-8 based on the runoff 

value from HEC-HMS under future climate changes 

1.3 Research Questions 

In light of the identified objectives, this research aimed to answer the following 

questions: 

•   What are the key factors and methodologies necessary to develop a reliable and 

representative hydrologic model for Al Zeimar catchment? 

 

•   How do climate fluctuations contribute to the occurrence of extreme flood events 

in Al Zeimar catchment? 

 



4 

 

•  What are the operational capabilities of the culvert at the entrance of Anabta in Al 

Zeimar during extreme flood events? 

1.4 Research Motivations 

Several factors contributed to giving this research a great importance, listed from them 

are the following:  

• This study is the first to model the runoff of Al Zeimar catchment considering climate 

change 

• The HEC-HMS model of Al Zeimar catchment, is the only calibrated and validated 

model of the catchment based on real run-off data. 

• The study area has many features that led to choosing it, as in the following:  

a. The wadi runs by a good number of human communities. 

b. The wadi passes by several water sources, such as the Anabta Well. 

c. A flash flood occurred in this catchment in 2013 and caused a lot of damage and 

loss of life. 

d. The wadi of the catchment is transboundary. It starts from the hills of Nablus, 

passing by Tulkarm then the green line, and pour into the Mediterranean. 

1.5 Literature Review 

Al Zeimar catchment has always been a study area of interest for researchers. Previous 

studies, such as the study conducted by Ahmad (2014) , mainly focused on the quality of 

the discharge in Al Zeimar wadis because they suffered from several sources of pollution; 

effluents that enter the stream through the wadi beginning with deposition of raw sewage 

from the western side of Nablus wastewater addition to agricultural and industrial 

pollution (Darawosheh, 2014). In 2013, Nablus Waste Water Treatment Plant was 

launched to treat wastewater from the upstream wadis to protect surface and ground 

water sources from pollution, and to eliminate the diseases and odors due to the flowing 

wastewater in Al Zeimar wadi (Palestinian Water Authority (PWA), 2024). Few research 

outlined the hydrological assessment of the catchment’s wadi and the development of 

the rainfall-runoff relationship that is fundamental for catchment management and an 

essential tool in future flood prediction.  

Zayed, Qutob, and Almasri (2011) conducted a study in the West Bank focusing on Wadi 

Al Zeimar in  Anabta Their objective was to address the challenges within the Wadi, 



5 

 

engage stakeholders, and forecast potential flood-related issues. They measured the 

wastewater generated from western Nablus and nearby villages during summer using 

water level monitoring devices at wadi culverts. For the winter season when stormwater 

mixes with wastewater, they estimated the flow synthetically employing the SCS (Soil 

Conservation Service) Unit Hydrograph method. By developing a flow hydrograph of 

Al Zeimar, they constructed a flood map to pinpoint critical sections. Their 

recommendations included advocating for further research to expand the study area and 

implementing regulations to curb garbage dumping in the wadi. 

Hammad et al (2011) derived a synthetic unit hydrograph for Al Zeimar using the SCS 

Synthetic Unit Hydrograph method, chosen for its reliance on the physical characteristics 

of the watershed. Notably, at the time of the study, the catchment remained ungauged. 

Meteorological data were sourced from Nablus and Tulkarm metrological stations as 

well as municipalities. The synthetic unit hydrograph generated for a 1-cm rainfall event 

resulted in a flow peak of 21.65 m3/s, a substantial figure given the magnitude of rainfall. 

An alternative approach, Snyders Method, necessitating similar data as SCS, yielded a 

significantly higher flow peak of 184.15 m3/s. It was recommended that flow 

measurement instruments be installed in catchment wadis to acquire authentic runoff 

data, facilitating the development of a rainfall-runoff model for the catchment 

In 2013, a catastrophic event that changed the focus of the direction of research in the 

catchment occurred. It all started when the region woke up to a rainy day on the 6th  of 

January, nobody suspected anything. The water level raised as usual in Al Zeimar 

catchment. Suspicions started rising when the intensity of the rainstorm increased, and 

was continuous. On the 7th, the rain water started seeping into the houses around the 

wadis. The fatal event occurred on the 8th of January, the culverts in the Anabta section 

of the catchment were completely clogged and the area flash flooded. The flood 

destroyed everything on its way, from cars, trees, and infrastructure, but the biggest loss 

was the lives of two young ladies who were trapped in a taxi, and got swept by the flood 

(AQUA Consulting Center, 2013). The storm was called Alexa with substantial 

precipitation in the form of both rain and snow, it was the worst storm to hit the area in 

60 years (ICRC, 2021). In the summer of the same year, AQUA Consulting Group 

studied the rehabilitation of  Al Zeimar’ culverts. This project came as a direct response 

to the flood and its disastrous consequences. The implementation of studies related to 



6 

 

the project were proceeded in several steps: the first was an attempt to understand what 

happened in a scientific manner based on accurate data and special mathematical models, 

the second was to evaluate the status of the wadi and the existing culverts in terms of 

their ability to drain surface runoff from different rain storms, and the third was to 

develop proposals and solutions consistent with the reality of the valley, its 

characteristics, ownership boundaries, and the nature of the surrounding buildings, and 

fourthly, the structural design of culverts and any other facilities that could be affected 

by the new situation, such as: streets. After data collection and analysis, they developed 

a HEC-HMS model to study the rainfall-runoff relationship in the catchment. It is 

important to note that there was no real runoff data at that time, so the model was neither 

calibrated nor validated. The model simulated the storm that occurred between the 6th 

and the 8th of January 2013, and resulted in a simulated peak flow of 75 m3/s at the wadis 

outlet. Peak flows on Anabtas’ culverts were used in HY-8 (Culvert Hydraulic Analysis 

Program), to redesign the culverts, for the fatal storm, and storms with return periods of 

10 years.  They recommended that the wadis culverts be continuously cleaned, because 

trash partially clogged them and decreased their drainage capacity. Furthermore, they 

recommended that all the culverts should be enlarged (AQUA Consulting Center, 2013). 

According to the United Nations Development Program (2020), the Climate Change 

Profile for Palestinian Territories highlighted forecasts of climate change that showed a 

significant rise in temperature and a drop in the amount of precipitation that falls on a 

yearly basis. In addition, they predicted that by the end of this century, an increase in 

temperature ranging from 1.8° to 5.1° is probably to take place. Yearly rainfall rates are 

likely to drop by 10% in 2020, by 20% by 2050, and may reach a drop of 35% by 2100 

with a high risk of summer droughts. Additionally, a significant decrease in the length of 

cold spells and a significant rise in the duration of heat waves are predicted. Regarding 

flooding, flash floods are more likely to be caused by High Precipitation Events (HPEs) 

and rainfall variability. 

Shadeed performed a trend analysis of temperature and precipitation data for the Nablus 

Metrological Station (NMS) in the West Bank, Palestine, in 2013. It was determined 

through rainfall studies that the NMS has highly variable rainfall, with an average annual 

rainfall of 654 mm and annual variations of from 335 to 1,400 mm. Rainfall is generally 

trending toward a wetter environment, with an estimated 48 mm rise over the course of 



7 

 

the study period. With an average temperature of 18.4°C and a standard deviation of 

0.8°C, the temperature variability between years (1975-2011) was demonstrated through 

the analysis of temperature. It was concluded that the changes in climate of NMS are 

considerable and largely affect the sustainable yield of the water resources in the West 

Bank (Shadeed, 2013). 

Assaf et al.  (2018) obtained high-resolution climate projections over Israel with the 

regional model COSMO-CLM. For the years 2041–2070, the simulation offers high-

resolution spatial variability of total precipitation and intensity. The forecasts indicated 

that seasonal mean temperatures will generally rise over the domain, peaking at 2.5°C, 

particularly in the winter and fall. The extreme temperature index indicates a rise, with 

a higher lowest temperature than a higher maximum temperature. There were drops of 

up to 40% in the overall amount of seasonal precipitation in Israel's central 

Mediterranean climate. 

 

 

 

 

 

 

 

 

 

 

 



8 

 

Chapter Two 

Materials and Methods 

2.1 Study Area 

2.1.1 Location 

Al Zeimar catchment lies in the West Bank, Palestine. It is located in the North-West, 

and in one of the 33 catchments of the West Bank. The catchments’ streams are 

transboundary streams, where they originate in Nablus, and flow towards Tulkarm. The 

streams  cross the 1949 Armistice Agreement Line flowing through historic Palestine, 

then drain into the Mediterranean Sea. The total watershed area is 600 km2 where 172 

km2 are on the Palestinian land. The total length of Al Zeimar stream is 50 km where 27 

km flow in the Palestinian area. The average inclination is about 1.2% in the first 10 km 

and 0.9% in the remaining length (Dotan, et al., 2017) , (Becker, Friedler, & Haddad, 

2007) .  The altitude of the catchment varies from 75 m amsl (above measured sea level) 

in the plain of Tulkarm to 925 m amsl in the mountains of Nablus. Figure A.1 in 

Appendix A shows the map of the location of Al Zeimar catchment within the West Bank, 

and shows the digital elevation model of the catchment. 

2.1.2 Population 

The Zeimar catchment lies within two main cities, Nablus and Tulkarem.  Nablus has a 

population of 174387, and Tulkarm has a population of 71161 people according to 

population statistics by (Palestinian Central Bereau of Statistics PCBS, 2023). PCBS??? 

Other medium sized towns also contribute to the flow of the river. Figure A.2 shows the 

communities within the catchment. Table B.1 shows the communities within the 

catchment. 

2.1.3 Geology 

The Wadi Al Zeimar research region contains geological strata that span the Senonian to 

Quaternary ages. The catchment region is mostly covered by sedimentary carbonate 

rocks like chalk, dolomite, limestone, and marl (ARIJ A. , 1996).  



9 

 

2.1.4 Soil Cover  

The majority of land is covered by loose surface material called soil. Soil is made up of 

both organic and inorganic materials. In addition to being a source of water and nutrients, 

soil gives agricultural plants the structural support they need (Leeper & Uren, 1993). 

Table B.2 summarizes the soil cover of the catchment(Rofe and Raffety Consulting 

Engineers, 1965) (ARIJ A. , 1996). Figure A.3 in Appendix A shows the spatial 

distribution of soil over the catchment. 

2.1.5 Land Use 

As for the land use within Al Zeimar catchment, Table B3 in Appendix B shows the 

different land use types and their areas within the catchment. Figure A.4 in Appendix A 

shows the distribution of the land use types over the catchment.  

2.1.6 Wadis 

Wadis are naturally occurring waterways that remain dry for the majority of the year. 

They temporarily transform into conveyors of runoff during wet seasons, primarily 

during and after a heavy rainfall. Surface runoff does not generate with every storm. 

Permanently flowing water often exists in the gravels below the surface of a large wadi 

subsurface terrain (Sen, 2008). Figure A.5 in Appendix A shows the wadis in Al Zeimar 

Catchment. Table B.4 in Appendix B shows the wadi corresponding to each ID. 

In the dry season, the flow in the wadi is mainly treated wastewater. The Nablus WWTP 

shown in Figure A.5 is located near Beit Leid. The wastewater is collected from western 

Nablus and from five villages namely; Zawata, Beit Eba, Beit Wazan, Deir Sharaf and 

Qusin. One of its main purposes is to improve the environmental and health conditions 

in upper wadi Zeimar (ID 0 in Figure A.5). The treated wastewater discharge from the 

treatment plant varies from 10,000 m3/day and 14,000 m3/day (Abu Jafal, Hmeidan, 

Bitar, & Abu-Salameh, 2023). Wastewater from almost all communities in the wadi is 

conveyed in closed sewage systems, except for almost 100 residencies in Anabta that are 

not connected to the main sewage line, causing the treated wastewater to mix with 

untreated wastewater in the wadi. In addition, olive mill waste is dumped in the wadi 

especially in October and November. In the rainy season, the direct runoff in Al Zeimar 

wadi is approximately 23,800 m3/day (ARIJ A. R., 2015).  



10 

 

2.1.7 Climate 

There are 5 main climatic zones in the West Bank. Al Zeimar catchment is located in the 

Central highland’s region which lies 400 to 1000 m amsl, and has an annual rainfall that 

varies between 300 mm in the south to 600 mm in the north (Climate Change Profile - 

Palestinian Territories, 2020). The weather conditions of the basin are typical East 

Mediterranean. The East Mediterranean climate conditions are branded by moderate air 

temperatures during the rainy winter season and dry and stable hot weather conditions 

during summer (Hochman, Mercogliano, Alpert, Saaroni, & Bucchignani, 2018). The 

spatial distribution of the climatic features is affected by topographical and coastal 

effects. Such effects are immediately noticeable in the mountainous areas (Hochman, et 

al., 2017) . There are two main weather stations in Al Zeimar, one in Nablus and the 

other one in Tulkarm. Each station measures the following parameters:   

• Daily Rainfall 

• Evaporation 

• Relative Humidity 

• Wind Speed 

• Air Pressure 

As for the station’s locations, Nablus station is located at a longitude of 35°15' E and a 

latitude of 32°13' N, with an elevation of 570 meters above sea level. In contrast, 

Tulkarem station is situated at a longitude of 35°01' E and a latitude of 32°19' N, with a 

much lower elevation of 83 m amsl. (Palestinian Meteorological Department, 2024). 

2.1.8 Temperature 

• The coldest months of the year extend from October – April, and the hottest extend 

from May – September. Temperatures in winter range from 8 to 14 °C, while the 

summer temperatures range from 21 to 40 °C (Becker, Friedler, & Haddad, 2007). 

Figure A.6 in Appendix A shows graphs of the average annual temperatures for the 

coldest and hottest months of the years 2007 – 2018 for the Nablus station 

(Palestinian Meteorological Department, 2024) (The Nablus station was used 

because the Tulkarem station has data gaps. The following are the general comments 

about the hot months (2007-2018) : 



11 

 

• The average temperature during the hot months fluctuates between approximately 

21.5 and 23.5°C. 

• The lowest average temperature appears to be around 21.5°C, occurring early in the 

data range (likely in the late 2000s). 

• There is a noticeable peak in the middle of the chart, with the temperature reaching 

approximately 23.5°C, indicating a significant increase compared to the initial years. 

• Following the peak, there is a slight decline but temperatures remain higher than the 

initial years. 

• Towards the end of the range, there is another rise, with temperatures reaching near 

the peak values. 

As for the cold months (2007-2022), the following are the general comments: 

• The average temperature during the cold months fluctuates between approximately 

11.0°C and 16.5°C.  

• The lowest average temperature is around 11.0°C, appearing early in the data range 

(likely in the late 2000s). 

• There is a significant peak reaching around 16.5°C in the middle of the chart, 

showing a substantial increase compared to other years. 

• Post the peak, temperatures drop again but not as low as the initial years. 

• Towards the end of the range, temperatures rise again, indicating another significant 

increase. 

2.1.9 Rainfall 

The year in Palestine is classified into two seasons: the rainy season and the dry season. 

Rain casually begins early in November, and ends late April. The span flocculates greatly 

between years. There are plenty of sunny days during the rainy season as well. One or 

more days of rain are followed by a stretch of beautiful weather. Rainfall usually lasts 

for five days, although occasionally it might last for 10 or fifteen days (Rice, 1886). 

Precipitation in arid and semi-arid regions is generally identified by extremely high 

spatial and temporal variability (Whaeater & Al-Weshah, 2002). Figures A.7 and A.9 in 

Appendix A shows the graphs of the average annual rainfall in mm, and Figure A.8 shows 

the number of rainy days for the Nablus and Tulkarm stations from (2007 – 2018). The 

average annual rainfall in Nablus is 660 mm. From the chart of Figure A.7 in Appendix 



12 

 

A it can be seen that the average rainfall values are very flocculant, which is the general 

is a characteristic of semi-arid regions. After 2013, it is observed that the average annual 

rainfall decreased, while the number of rainy days increased, and this supports the 

predicted climate change scenarios for the study area. The graph in Figure A.9 in 

Appendix A shows average annual rainfall for Tulkarem station. The average annual 

rainfall in Tulkarm is 602 mm. It is notable that the amounts of rainfall in Nablus are 

generally higher than in Tulkarm. In addition, in 2013, there was an extreme rainfall 

event that led to fatal consequences, and this can be seen from the big rainfall values 

over the same average of rainfall days in 2013 for both Nablus and Tulkarem. The map 

labeled Figure A.10 Appendix A shows the map of the rainfall contours of the whole 

catchment, the rainfall contour shapefile was retrieved GEOMOLG spatial database 

(Government P. M., 2023).  

2.1.10 Climate Change 

Climate change may cause significant seasonal swings or changes in rainfall in some 

parts of the world. Thus, it is essential to do research into how various hydrologic 

components are affected by climate change (Abbas, Zhao, & Wang, 2022). This might 

lead to more frequent water shortages, flooding, and problems with food security. The 

hydrologic cycle is impacted by climate change in ways that affect soil moisture, 

evapotranspiration, and precipitation (Abbas, Zhao, & Wang, 2022). Understanding 

climate disorder commonly involves establishing a hydrological model set by taking into 

account various simulations through a series of global climate models (GCMs) 

(Christierson, Vidal, & Wade, 2012). To downscale climatic data from global scale to 

catchment scale, the chain “GCM–RCM (Regional Climate Models)–HM (Hydrologic 

Models)” is efficient for scenario-based data collected by downscaling (Fonseca & 

Santos, 2019). Despite the fact that climate change is not a pressing issue to the 

Palestinian, the climate risks are noteworthy due to the limited capacity to respond to the 

effects of climate change (Climate Change Profile - Palestinian Territories, 2020). 

Climate trends from recent and projected models reveal that temperatures in the area will 

increase, precipitation will decrease, and high precipitation events (HPE) will happen. 

Hochman et al. (2018) obtained high resolution climate projections over Israel for the 

period 2041-2070 with the regional model COSMO-CLM. The climate projections 

obtained from the previous study were used in this study to simulate the future climate 



13 

 

shifts using HEC-HMS. It is significant to remember that the use of global climate 

models to simulate the intricate climate regime of the Mediterranean region has drawn 

criticism (Lionello, Gacic, Planton, Trigo, & Ulbrich, 2014). To bridge the gap between 

GCMs and simulations of local climatic variables, the dynamic downscaling approach 

was used in the previous study. The temporal distribution of daily rainfall data is used to 

assess the environmental impacts of short-term phenomena climate change processes 

(Zhang, et al., 2011). The future projections of Hochman’s’ study assumed that 

greenhouse gases will increase and peak around 2040, and then will start to decline 

(Moss, et al., 2010). The extreme precipitation indices (EPI) (The unit is mm/day) are: 

• SDІІ: Mean Precipitation on wet days (> 1 mm) 

• Rx 1day: Maximum of daily precipitation 

• 99p: 99th percentile of daily precipitation 

• 90p: 90th percentile of daily precipitation 

The projections for the EPI for the period of 2041-2070 were made for the whole country 

(Palestine), the following results show projections specifically over the study area of Al 

Zeimar catchment according to (Assaf, Paola, Pinhas, & Hadas, 2018) ??? 

• SDІІ is projected to decrease in Al Zeimar catchment by almost 1 mm/day. 

• Rx 1 day is projected to increase by almost 12 mm/day.  

• The 99 p is projected to increase by almost 14 mm/day. 

• The 90 p is projected to decrease by almost 3 mm/day. 

2.2 The Methodology 

The methodology can be divided into five major tasks: 

• Data collection and analysis  

• HEC-HMS model development  

• HEC-HMS model calibration and validation  

• Projection of the EPI to simulate climate change 

• Redesign of the culvert at the entrance of Anabta using HY-8 



14 

 

2.2.1 Data Collection and Analysis 

In any system, the quality of the output is determined by the quality of input (Awati, 

2015). Wherefore for the HEC-HMS model to be able to generate reliable results, the 

input data should be precise and representative. The following are the data required to 

create the inputs for the model:  

2.2.1.1 Ground Surface Contours 

Contours are needed to develop the Digital Elevation Model (DEM) of the study area. 

DEM is the digital representation of the land surface elevation with respect to a reference 

datum (Balasubramanian, 2017). The 25 m contours of the study area were extracted 

from  (Government P. M., 2023) a GIS shapefile, and then were processed in ArcMap 

10.8 using the 3D Analyst.  

2.2.1.2  Lag Time  

According to the Hydrologic Modeling System HEC-HMS: Technical Reference Manual 

(2022) , a unit hydrograph is a discharge hydrograph resulting from 1 unit of direct runoff 

distributed uniformly over the catchment that result from a rainfall of a specified time 

span. The unit hydrograph procedure assumes that discharge at any time is directly 

proportional to the volume of runoff and that time factors affecting hydrograph shape are 

constant. The SCS unit hydrograph method makes use of a dimensionless, curvilinear 

unit hydrograph to route excess precipitation to the watershed outlet.  This 

dimensionless, curvilinear unit hydrograph expresses discharge, (q), as a ratio of the peak 

discharge (qp) for any time (t), as a fraction of the time of rise, (Tp). The following 

summarizes the related equations: 

• Tp can be related to the duration of excess precipitation as : 

𝑇𝑝 =
𝑡𝑟

2
+  𝑡𝑝                                                     (1) 

o tr: duration of excess precipitation (or computational time step) 

o tp: the basin lag time which is defined as the time difference between the center of 

mass of excess precipitation and the peak of the unit hydrograph 

• The unit peak discharge Qp for 1-cm of excess rainfall (Feldman & Hydrologic 

Engineering Center U.S, 2022):  



15 

 

𝑄𝑝=2.08 (
𝐴

𝑇𝑝
)                                                          (2)   

where A is in (km2,) Qp in (m3/s), Tp in (hr) 

• To utilize the SCS Unit Hydrograph method, the basin lag time is needed, it is 

estimated to be: 

𝑡𝑝 = 0.6𝑇𝑐                                                        (3) 

• Tc: Time of concentration, longest travel time that it takes surface runoff to reach the 

discharge point of a catchment 

The Kirpich Formula was used to find the time of concentration:  

𝑇𝑐 = 0.01947𝐿0.77𝑆−0.385                                      (4)   

• Tc: time of concentration in (min.) 

• L: Maximum length of travel water (m) 

ΔH: Difference in elevation between the most remote point on the catchment and the 

outlet (National Engineering Handbook, 2021). HEC-HMS automatically calculates L, 

S, and ΔH after the watershed is delineated. 

2.2.1.3 Land Use / Land Cover (LULC) and Soil Maps 

According to the National Engineering Handbook (2021), the terms "land cover" 

describes the surface cover of the earth, including bare soil, water, plant, and urban 

infrastructure. The term "land use" describes the use of the land, such as agriculture, 

wildlife habitat, or recreation. When coupled with the term "Land Use/Land Cover," 

(LULC) typically denotes the classification of both natural and human-made features on 

the terrain during a given period of time (SATPALDA, 2018). LULC maps are used with 

the soil map of the catchment, to identify the NRCS-CN (previously named SCS-CN) 

input parameters for the model. The NRCS-CN method, is a method used to estimate 

runoff volume resulting from a storm event. The NRCS runoff equation, according to the 

National Engineering Handbook (2021): 



16 

 

𝑄 =
(P−0.2S)2

(P+0.8S)
    For P > Ia                                          (5)   

𝑄 =  0              For P ≤ Ia                                           (6) 

• Q = Runoff (mm) 

• P = Rainfall (mm)  

• Ia  = Initial abstraction (mm) 

• S = Potential maximum retention after runoff begins (mm) 

Initial abstraction (Ia) includes all losses (water retained on the landscape) before runoff 

begins. It includes water retained in surface depressions, water intercepted by vegetation 

and other cover, and water lost to evaporation and infiltration. Ia is highly variable but 

is generally correlated with soil and cover parameters. Through studies of many small 

agricultural watersheds, researchers found Ia to be approximated by (United States 

Department of Agriculture, 2021): 

𝐼a =0.2𝑆                                                                (7)  

The potential maximum retention, S, can range from zero on a smooth, impervious 

surface to infinity in deep gravel  

𝑆 (𝑚𝑚) =
25400

𝐶𝑁
− 254                                                  (8) 

Runoff curve numbers (CN) can be any value from zero to 100, but for practical 

applications it is within a limited range of about 40 to 98. Researchers developed the 

runoff curve numbers in Table B.5 in Appendix B by examining rainfall runoff data from 

small agricultural watersheds. These values were sourced from the USDA's National 

Engineering Handbook (2004)   The runoff curve number for a given soil-cover type is 

not constant but varies from storm to storm, and even within the storm (United States 

Department of Agriculture, 2021). The NRCS-CN method assigns a representative 

runoff curve number for a given soil and cover type for design purposes. The NRCS 

defines four hydrologic soil groups that, along with land use, management practices, and 

hydrologic conditions, determine a soil’s associated runoff curve number. The 

classification shown in Table B.6 in Appendix B. These values were also sourced from 

the USDA's National Engineering Handbook (2004). When the model is used for event-



17 

 

based flood prediction, the AMC becomes a vital factor affecting runoff production 

(Bahremand & De Smedt, 2006). The AMC values in Table B.7 is determined according 

to the sum of all precipitation for the 5 days and the season (dormant or growing) 

(Mihalik, Levine, & Amatya, 2008). 

The CN are for AMC II. The CN values for dry soil (AMC I) or saturated soil (AMC II) 

are calculated using the following formulas (Luijten, Jones, & Knapp, 2002) : 

𝐶𝑁I =
(−75∗𝐶𝑁II) 

(−175+𝐶𝑁II)
 𝐹𝑜𝑟 𝐴𝑀𝐶I ( Dry Soil)                               (9) 

𝐶𝑁III =
(175∗𝐶𝑁II)

(75+𝐶𝑁II)
 , 𝐹𝑜𝑟 𝐴𝑀𝐶 III ( Saturated Soil)                     (10) 

For the sub-catchments within the watershed that have different soil types and land 

covers, a composite CNc is determined by weighting the curve number values for the 

different sub-areas in proportion to the land area associated with each (Shadeed, 

Hydrologic Analysis, 2019) : 

𝐶𝑁𝑐 =
𝐶𝑁1𝐴1+𝐶𝑁2𝐴2+⋯.+𝐶𝑁𝑖𝐴𝑖..+𝐶𝑁𝑛𝐴𝑛

∑ 𝐴𝑖𝑛
𝑖=1

                                (11) 

After the land use and soil type shapefiles are obtained, they can be compiled in a GIS 

based database. For land use, the area of each subclass was calculated. For soil type, the 

Hydrologic Soil Group (HSG) was assigned for each soil type. It is important to note 

that the study area is made up of clay loam (C) and clay (D). Both shapefiles were 

intersected, and the attribute table of the resulting shapefile was exported to MS Excel 

(Easier to process the data because GIS requires Python).  The CN for each sub-basin 

was assigned by the methodology shown in the flow chart in Figure A.11 in Appendix 

A.  

2.2.1.4 Channels’ Average Cross Section and Properties 

According to information on the HEC-HMS Tutorials and Guidelines website (U.S. 

Army Corps of Engineers, 2022) , the hydrologic network is made up of various 

elements, one of them is the reach. It simulates a part of a stream or river. Flood routing is 

a method to calculate flow depths, velocities, volumes, and discharges at a river section. 

When a flood wave enters a river upstream section, using flood routing one can detect 



18 

 

the movement of flood wave along a channel length and thereby can calculate 

flood hydrograph at any downstream section of the river. The conservation of mass is 

the main principle of the routing methods. They require important field data, such as 

cross-sectional surveying, roughness, flow depth and velocity measurements (Tayfur, 

2023). The Muskingum flood routing method is one of the most common hydrological 

approaches. Using the mass-conservation equation this method defines the channel 

storage in each time interval as a function of the inflow entering the channel and the 

outflow exiting the channel (Niazkar & Zakwan, 2022). The equation of the total storage 

in the channel reach is : 

𝑆 = 𝐾𝑄 + 𝐾𝑥(𝐼 − 𝑄)                                        (12) 

• S: Storage (m3/s) 

• I: Inflow Discharge (m3/s)  

• Q: Outflow Discharge (m3/s)  

• K: Time of travel of the flood wave through the channel reach (sec) 

• x: Value that depends on the shape of the wedge, in natural stream it varies from 0 

and 0.3 

To apply the Muskingum Routing method in the HEC-HMS model, k and x should be 

defined.  

• K is the travel time through the reach. Its initial value is estimated by comparing flow 

length to a flood wave velocity  (U.S. Army Corps of Engineers, 2022) : 

𝑇 =
𝐿

𝑉𝑤
                                                            (13) 

• T: Travel time of a flood wave (sec) 

• L: Length of the reach (m) 

• Vw: Flood wave velocity (m/s) 

The flood wave velocity is estimated using Mannings Equation. It is one of the most 

commonly used equations governing open channel flow. The Mannings equation is an 

empirical equation that applies to uniform flow in open channels and is a function of the 

https://www.sciencedirect.com/topics/earth-and-planetary-sciences/hydrograph
https://www.fsl.orst.edu/geowater/FX3/help/8_Hydraulic_Reference/Open_Channel_Flow.htm


19 

 

channel velocity, flow area and channel slope (Hydraulic Reference - Manning Equation, 

2004) : 

𝑄 = 𝑉𝐴 = (
1

𝑛
) 𝐴𝑅2/3√𝑆                                           (14) 

• Q = Flow Rate (m3/s)  

• V = Velocity (m/s)   

• A = Flow Area (m2) 

• n = Manning’s Roughness Coefficient 

• R = Hydraulic Radius (m) 

• S = Channel Slope (m/m) 

The Manning’s n is a coefficient which represents the roughness or friction applied to 

the flow by the channel. Manning’s n-values are selected from Table B.8 in appendix B. 

The roughness coefficient values were retrieved from the Handbook of Hydrology 

(2019) 

To find the initial value of K, the equation becomes  

𝑉 = (
1

𝑛
) 𝑅2/3√𝑆                                            (15)  

The initial estimation of the previous equations’ parameters was done through field visits, 

and using aerial photographs of the study area. For x, great accuracy in determining it 

may not be necessary because the results of the method are relatively insensitive to the 

value of this parameter (Shadeed, Hydrologic Analysis, 2019). 

2.2.2 Model Development 

The hydrologic model of the watershed was simulated using HEC-HMS 4.101. HEC-

HMS (Hydrologic Engineering Center-Hydrologic Modeling System) was developed by 

the United States Army Corps of Engineering (USACE). The software is a generalized 

modeling system that can simulate a wide range of watershed types. It is a 

mathematically based (conceptual) deterministic semi-distributed event-

 
1 https://www.hec.usace.army.mil/software/hec-hms/ 

https://www.hec.usace.army.mil/software/hec-hms/


20 

 

based/continuous model. By breaking down the hydrologic cycle into smaller, more 

manageable components and drawing boundaries around the watershed of interest, a 

model of the watershed is created  (Feldman & Hydrologic Engineering Center U.S, 

2022). The components of a HEC-HMS model are: 

• Terrain Data 

• Basin Model 

• Meteorological Model 

• Time-Series Data 

• Control Specifications 

HEC-HMS was used for the hydrologic modeling of Al Zeimar for various reasons. First 

it allows for the simulation of precipitation-runoff processes, which can be crucial for 

understanding flood forecasting, water availability, and watershed management (U.S. 

Army Corps of Engineers, 2022). The flexibility of HEC-HMS to incorporate various 

hydrologic processes and data inputs, such as soil moisture accounting, snowmelt, and 

reservoir operations, makes it a valuable tool for both researchers and practitioners in 

hydrology (Sharif, Mahmood, & Nasiri, 2021). Moreover, the software's continuous 

development and updates ensure that it remains relevant with the latest scientific 

advancements and computational techniques (U.S. Army Corps of Engineers, 2022). By 

integrating GIS data, HEC-HMS enhances the spatial analysis capabilities, providing a 

more accurate representation of the watershed's physical characteristics (Maidment D. , 

2019).  

2.2.2.1 Terrain Data 

The terrain elevations from ground sites sampled at regularly spaced horizontal intervals, 

including the elevations of the highest and lowest points, are contained in the DEM that was 

created using contour lines (Gandi & Sarkar, 2016). The DEM is added to HEC-HMS as 

Terrain Data, and is then linked to the Basin Model. It is the main input for the GIS extension, 

where it is processed to automatically delineate the watershed, and create the hydrologic 

network of the catchment (Subbasins and reach elements). It is important to note that the 

DEM should be projected, and have the same coordinate system as the basin model. The 

Projected Coordinate System used is “Palestine_1923_Palestine_Grid”. 



21 

 

2.2.2.2 Basin Model 

 A basin model allows for the physical visualization of a watershed. Automatic 

delineation of basins using HEC-HMS involves a systematic process facilitated by 

computational algorithms. Initially, digital elevation models (DEMs) are processed to 

identify watershed boundaries and flow directions. This step is crucial as it defines the 

spatial extent of the watershed under study. Subsequently, the delineated watersheds are 

characterized by their geometric properties such as area, perimeter, and shape indices, 

which provide insights into their hydrological behavior. The delineation process is often 

followed by parameterization, where hydrological parameters such as soil types, land 

use, and vegetation cover are assigned to the watershed elements. These parameters play 

a significant role in simulating the hydrological processes within the watershed using 

HEC-HMS. Once the delineation and parameterization are completed, the hydrologic 

model is calibrated and validated against observed hydrological data to ensure its 

accuracy and reliability in predicting watershed responses to various hydro-

meteorological inputs. This automated approach streamlines the watershed delineation 

process, enabling efficient hydrological modeling and management of water resources 

(Maidment D. , 2019). 

The following basin model is the result of the automatic delineation of the watershed 

using the GIS extension, it will be approved as the physical representation of Al Zeimar 

catchment. 

 

 

 

 

 

 

 

 



22 

 

Figure 1 

 Al Zeimar Catchment Basin Model 

 

 

 

 

 

 

 

 

 

 

 

 

Hydrologic elements are connected together in a network to simulate runoff processes. 

Available elements include subbasins, reaches, junctions, and sinks (Feldman & 

Hydrologic Engineering Center U.S, 2022) . 

Subbasins  

 It is worth considering segmenting the watershed into hydrologic units consisting of 

areas with consistent shapes. If achievable, these subareas ought to be around the same 

size, have a uniform drainage pattern, and uniform land use (Feldman & Hydrologic 

Engineering Center U.S, 2022). Table 1 shows the communities within each sub-basin. 

 

 

 



23 

 

Table 1 

Communities with each sub-basin 

Sub-basin Community 

1 
Bazaria, Anabta, Kfr Labad, Kfr Ruman, Ezbet Abu Khmesh, Noor Shams 

Camp, Bal’a, Iktaba, Shweikah, Zenaba, Tulkarem City 

2 Burqa, Ramin, Masodeya, Sebastya, Naqora 

3 Qusin, Beit Iba, Deir Sharaf, Beit Led 

4 Beit Imrin, Nisf Jbeil 

5 Ijnesenya 

6 Asira Shamaliya, Zawata 

7 Nablus City, Juneid, Beit Wazan, Al Ain Camp 
 

It is important to note that the communities within the same subbasin discharge to the 

same outlet (named junction in the figure 1).  

Table 2 shows the physical properties of each subbasin (automatically calculated from 

HEC-HMS): 

Table 2 

Physical properties of each sub-basin 

 

 

 

 

 

 

 

Table 3 shows the initial input parameters for each subbasin for the model 

development:  

 

 

Sub-basin Area km2 
Longest Flow 

Path km 
Basin Slope 

Drainage Density 

km/km2 

Sub-Basin1 41.055 17.61 14.52% 0.355 

Sub-Basin2 30.100 12.02 15.42% 0.243 

Sub-Basin3 16.683 10.31 11.8 % 0.535 

Sub-Basin4 10.888 7.25 17.8% 0.048 

Sub-Basin5 10.956 8.48 17.2% 0.078 

Sub-Basin6 13.595 9.90 19.1% 0.130 

Sub-Basin7 12.730 7.94 21 % 0.136 



24 

 

Table 3 

Initial input parameters for each sub-basin 

Sub-basin Lag Time (min) Initial Abstraction (mm) Curve Number 

1 46 14.32 78.7 

2 33 14.45 77.85 

3 33 16 76 

4 21 17.2 74.7 

5 24 16.93 75 

6 26 13.5 79 

7 21 12.7 80 

 

It is important to note that for the curve number and initial abstraction, any AMC 

condition can be used, but it should be the same condition used for both parameters. 

(This only works for HEC-HMS) 

a. Reaches 

The river or stream are conceptually presented as streams. Muskingum1 routing method 

was applied to route these channels using the collected data and equations mentioned in 

the methodology. The Table 4 shows the initial input model parameters:   

Table 4 

Initial input parameters for each reach 

Reach 
Length 

(km) 
Slope 

Channel 

Dimensions 

Width-Depth 

(m) 

Mannings’ 

n 

Muskingum 

K (Hr) 

Muskingum 

X 

1 14.59 0.01028 6 2 0.035 3.12 0.2 

2 8.95 0.01397 2 1 0.03 2.92 0.2 

3 7.34 0.02045 1.3 1 0.03 2.40 0.2 

 

b. Junctions 

Junctions are elements with one or more inflows and only one outflow. All the inflow is 

added together to produce the outflow by assuming zero storage at the junction. It is 

representing a river or stream confluence (Feldman & Hydrologic Engineering Center 

U.S, 2022). The subbasins upstream of the junction have a higher elevation, and their 

 
1 It is important to note that the Muskingum routing method implements full unsteady flow for open 

channels.  

 



25 

 

water drains into the junction by gravity. So, the junction at the end of each sub-basin is 

like the outlet of that subbasin, but the inlet of the sub-basin downstream of it.  

c. Sink 

The sink usually represents the lowest point of the watershed. It is the “plug” were the 

water sinks by gravity (Baths, 2023). The sink of the model was set at the location of the 

partial flume with projected coordinates of (152047,192247). It is important to note that 

the wadi flows beyond this point through the apartheid wall to the Yad Hana WWTP, 

then flows toward Netanya in historic Palestine until it discharges into the 

Mediterranean. Figure A.12 in Appendix A shows the partial flume at the outlet of the 

catchment. 

2.2.2.3 Time-Series Data 

The HEC-HMS model requires time-series data of:1 

1. Rainfall data for estimating basin-average rainfall.  

2. Flow data, or observed discharge, which is helpful for calibrating the model. 

• Precipitation Gauge 

There are several rainfall gauges in the catchment (mentioned in the study area section 

with details), but they measure daily rainfall. The model is event based, so data with 

a small-time interval is needed. The rainfall data used was obtained from the Qarni 

Shomron Station (IMS, 2023), which almost lies in the middle of the catchment, 

because it is the only rain station in the catchment that contains 10-minute rainfall 

data. Since only one precipitation gauge is used, the model considers that rainfall from 

this gauge is distributed uniformly all over the catchment. The storm that was used 

for Run 1 of the model started on 11 February 2020 at 22:00 and ended on 12 February 

2020 at 09:00. It resulted in 11 mm of rain over the 11-hour period. The hyetograph 

in Figure A.13 was drawn by HEC-HMS for Storm 1. It shows the temporal 

distribution of rainfall. It can be noted that from 22:00 to 02:00, the rain was light, 

but after 02:00, the rain started to increase at reached its highest amount at 05:00 

almost, then started to decrease and ended at 09:00.  

 
1 Time-series data are stored in the project as a gage. 

 



26 

 

• Discharge Gauge 

The data for the discharge gauge was obtained from the partial flume. As mentioned 

earlier, in winter, rainwater is mixed with treated wastewater in the wadis, so the 

partial flume reads both. The following methodology was used to quantify the 

discharge from rainwater only: 

1. Collection of the 24-hour readings of the partial flume for a dry winter day Qt 

(This represents amounts of treated wastewater only that the partial flume read) 

2. Calculate their sum, QD 

3. Plot Qt / QD vs. Time 

4. Check that ∑
𝑄𝑡

𝑄𝐷
= 1  

5. The plot represents the 24-hour wastewater flow pattern at the partial flume 

6. To quantify discharge from rainfall, on a rainy day, determine the amount of 

treated wastewater discharged from the Nablus Western WWTP (Qww) 

7. Multiply Qww with the wastewater flow daily pattern 

8. The resulting Qwwg is the wastewater amounts at the partial flume on the day of 

the event  

9. QR = QT – Qwwg, where QR is the runoff from effective rainfall 

Steps 1-5 were applied on 15 February 2020 (This day has zero rainfall), and the graph 

in Appendix A is the wastewater flow pattern at the partial flume. When treated 

wastewater is discharged from Western Nablus WWTP into the wadis, it does not reach 

the partial flume at the end of the catchment in the same amount as part of it evaporates 

and some infiltrates through the wadis bed. The pattern above is used to temporally 

quantify the 24-hr amount of treated wastewater that reaches the partial flume from the 

WWTP. For example, on 15 January of 2022, the treatment plant reported that 11,000 

m3 was discharged into the wadi. This number is multiplied by Qt/QD for 24-hours to 

obtain the amount that reached the outlet. For instance, at 23:20, 880 m3 reached the 

outlet (0.08x11,000 = 880 m3).  

The runoff “Runoff 1” resulting from Storm 1 starts at 12Feb2020 at 00:30 and ends at 

12Feb2020 at 20:00. Storm 1 started at 11Feb2020 at 22:00, but the runoff from the storm 

started to reach the outlet at 00:30 on 12Feb2020, after two and a half hours of the storm, 

which means that the catchments’ lag time is approximately two and a half hours. The 



27 

 

runoff hydrograph of Storm 1 is shown in Figure A.14 in Appendix A, this figure was 

drawn by HEC-HMS. The data was obtained from the partial flume, and wastewater flow 

was quantified and removed. The peak flow is 0.5 m3/s and occurs at 06:00. It is 

noticeable that the hydrograph has two peaks. Two peaks appear on a double peak 

hydrograph in response to a distinct rainfall pulse. The first peak, which represents a 

quick catchment response to precipitation, happens simultaneously with or soon after the 

precipitation begins. Usually, the delayed peak begins when the first peak is receding 

and the precipitation has stopped. In urban catchments, double peak hydrographs can 

arise for a variety of reasons. Direct surface runoff on impermeable land cover frequently 

causes the initial peak, and slower subsurface flow is responsible for the delayed peak 

(Martinez-Carreras, et al., 2016). The double peak can be explained in the Zeimar stream 

because in the Anabta section, 200 meters of the channel is built up concrete. In addition, 

the section ahead of the partial flume is made up of concrete, this facilitates the flow of 

water and thus it reaches faster to the outlet. Figure A.15 in Appendix A shows the built-

up part section in Anabta.  

2.2.2.4 Meteorological Models 

One of a HEC-HMS hydrologic model primary components is a meteorologic model. 

Preparing meteorologic border conditions for sub-basins is the main goal (Feldman & 

Hydrologic Engineering Center U.S, 2022). Based on the used event, it is assumed that 

there is one precipitation gauge uniformly covering the catchment.  

2.2.2.5 Control Specifications 

Despite having little in terms of parameter data, control specifications are a key part of 

the model. Their main function is to regulate the start and stop times of simulations as 

well as the time intervals that are used (Feldman & Hydrologic Engineering Center U.S, 

2022). 

2.2.3  Model Calibration and Validation 

Hydrological models are the mathematical models having some unknown coefficients 

known as parameters. The process of estimating those parameters from previous input-

output records is known as model calibration. Model validation is evaluating the 

calibrated model's performance across the historical record section that was not used for 



28 

 

the calibration. Calibration is typically done via trial and error. Different validation 

criteria are utilized for the model validation, and they are constructed based on the output 

records that are computed and observed (Lohani, Rainfall - Runoff Analysis and 

Modeling, 2016). 

Basically, four sources of uncertainty occur in deterministic simulation. The differences 

between recorded and simulated output result from:  

1. Error Source 1: Errors in the input data that are either random or systematic, such as 

temperature, precipitation, or evapotranspiration, and which are utilized to depict the 

catchment's input conditions in both space and time. 

2. Error Source 2: Errors in the recorded output data that are either random or 

systematic, such as discharge or water level data that are compared to the output of 

the simulation. 

3. Error Source 3: Errors due to non-optional parameter values, e.g., rain gage location 

(latitude and longitude) and the rain gage data file.   

4. Error Source 4: Errors due to incomplete or biased model structure. e.g., neglecting 

groundwater flow, and not considering all sources of runoff. 

As a result, only error source 3 is reduced during the calibration process, whereas all four 

error sources are to blame for the discrepancy between the simulated and recorded 

output. As "background noise," the measurement errors and error source 2 provided a 

minimal degree of disagreement below which changing any more parameters or 

tweaking the model won't make the findings better. The next step in the calibration 

process is to minimize error source 3 until it becomes negligible in comparison to data 

error sources 1 and 2 (Lohani, Rainfall - Runoff Analysis and Modeling, 2016). An 

accuracy criterion can be used to compare the measured and simulated outputs during 

the calibration process. This makes it possible to determine the ideal parameter values 

and to get an objective evaluation of the goodness of fit linked with each set of 

parameters. (Lohani, 2016)  The Nash-Sutcliffe model efficiency coefficient (NSE) is 

preferably used to assess the efficiency of hydrological models (Ikenberry, n.d.).  

 

𝐸 =
∑(𝑂𝑖−�̅�)2−∑(𝑀𝑖− 𝑂𝑖)2

∑(𝑂𝑖− �̅�)2                                                          (16) 



29 

 

O = Observed Values (m3/s) 

= Average of Observed Values (m3/s) 
 

M = Modeled Values (m3/s) 

The maximum value of E is 1.  A value of zero indicates that the model is only as good 

as using the mean of the observations. E can be less than zero (Ladson, 2008). The 

following table identifies the classification (Goodness) of the model depending on the 

value of E for the calibration and validation processes.  

Table 5 

Classification of model performance according to E 

Classification E (Calibration) E (Validation) 

Excellent E ≥ 0.93 E ≥ 0.93 

Good 0.8 ≤ E < 0.93 0.8 ≤ E < 0.93 

Satisfactory 0.7 ≤ E < 0.8 0.6 ≤ E < 0.8 

Passable 0.6 ≤ E < 0.7 0.3 ≤ E < 0.6 

Poor E < 0.6 E < 0.3 

 

 

Another calibration parameter is percent bias. Percent bias (PBIAS) calculates how likely it 

is on average for the simulated values to differ from the observed values. PBIAS should 

ideally be at 0.0, with low magnification values denoting precise model simulation. 

Overestimation bias is indicated by positive numbers, whereas underestimating bias in the 

model is indicated by negative values (Bigiarini, PBIAS HydroGOF, 2010).  

𝑃𝐵𝐼𝐴𝑆 = 100
∑ (𝑆𝑖−𝑂𝑖)𝑁

𝑖=1

∑ 𝑂𝑖
𝑁
𝑖−1

                                                  (17) 

S= Simulated Values (m3/s) 

O = Observed Values (m3/s) 

2.2.4 Projection of EPI to Simulate Climate Change  

2.2.4.1 Extreme Precipitation Indices of the Study Area 

Table 6 shows indices for the period from 1997-2022 for the Nablus station. (The data 

was obtained from manually processing rainfall data from (Palestinian Meteorological 

Department, 2024) using excel). 



30 

 

Table 6 

EPI for Nablus for 1997-2022  

 

 

Table 6 shows that the average precipitation on rainy days for the period between 1997-

2022 is about 12.5 mm. The maximum daily rainfall that occurred during this period was 

123 mm, and this occurred in 2013 in storm Alexa. When projecting the outputs of the 

study done by (Hochman, Mercogliano, Alpert, Saaroni, & Bucchignani, 2018)  on the 

Nablus station EPI’s, the extreme EPIs of the study area in the period from 2041-2070, 

according to the output of the climate projections become:  

Table 7 

 EPI for Nablus for 2041-2070 

Parameter Precipitation (mm) 

 SDІІ 11.5 

Rx1 day 135 

99p 46.2 

90p 13.4 

 

The mean daily rainfall during rainy days, decreased by 1 mm, to become 11.5 mm/day. 

This supports the theory that the total amounts of rainfall will decrease in the years 2041-

2070. As for the maximum amount of daily rainfall over the years 2041-2070, it is 135 

mm/day which also supports the claims of (Assaf, Paola, Pinhas, & Hadas, 2018) that 

the chances of flash flood will increase.  

2.2.4.2 Frequency Analysis of Nablus Station Meteorological Data 

Olofintoye et al. (2009) stated that even though the characteristics of precipitation are 

unpredictable and change over time, it is feasible to predict the amount of rainfall with 

extreme precision for various durations by employing a certain probability distribution. 

As a result, statistical frequency analysis of the maximum rainfall records completes the 

estimation of heavy rainfall. Since the frequency of extreme occurrences affects flood 

Parameter Precipitation (mm) 

 SDІІ 12.5  

Rx1 day 123  

99p 32.2  

90p 16.4 



31 

 

control systems' functioning and design, it is imperative to evaluate the statistics of 

extreme precipitation events  (Doan, et al., 2015). HEC-SSP1 is a software developed by 

the US Army Corps of Engineers, it stands for Hydrologic Engineering Center’s (HEC) 

Statistical Software Package (HEC-SSP). This software allows users to perform 

statistical analyses of hydrologic data. It was used to conduct the frequency analysis of 

the maximum daily rainfall data for the years 1997-2022 Table B.9 in Appendix B (Data 

retrieved from Palestinian Meteorological Department (2024)) of Nablus station. The 

Chi-Square was used to determine the best distribution that represents the maximum 

daily rainfall data. The chi-square (χ2) statistic is a tool used to assess how well a model 

fits real-world observations. It calculates the discrepancy between the actual and 

predicted frequencies of a given set of variables or occurrences (Hayes, 2023). The 

smaller the Chi-square value, the better the distribution fits the data. The normal 

distribution was the best distribution to fit the maximum rainfall data of Nablus for the 

years 1997-2022. The fitted distribution has been used to calculate the size of an event 

which correlates to a return period of 25 years, then the increase in the maximum daily 

precipitation from Table 7 was added.  

Figure 2 shows the cumulative distribution function – plotting position of the maximum 

rainfall data, and the normal distribution curve.  

 

 

 

 

 

 

 

 

 
1 https://www.hec.usace.army.mil/software/hec-ssp/download.aspx 



32 

 

Figure 2 

CDF of maximum daily rainfall data of Nablus Station 

 
 

According to (WSDOT, 2023), culverts are designed for the 25- and 100-years flow 

events. In this study, the 25 years flow event was used for the design because it is within 

the climate change projections period (2041-2070). The probability of exceedance of the 

25-year maximum daily rainfall event is 1/25 = 0.04 from Figure 2, the expected 

maximum daily rainfall after 25 years is 113 mm. According to the climate projections 

in Table 7 the maximum daily precipitation increase by 12 mm, which makes the 

maximum daily rainfall in that period 125 mm.  

2.2.4.3 Meteorological Climate Change Scenario Input for the Model 

The maximum daily rainfall from the previous section is 125 mm. To be able to use this 

as an input for the HEC-HMS model, it should be converted into a 10-min interval event. 

To achieve this, the 2013 Alexa hyetograph distribution was used to simulate a synthetic 

24-hour, 5-min interval storm.  Figure A.16 in Appendix A shows the hyetograph of 

Alexa storm, it was extracted from Weather Underground (n.d.) which is a commercial 

weather service that provides weather data online.  

Figure 3 shows the hyetograph of a synthetic storm with the same temporal distribution 

of rainfall as Alexa, which was obtained using through obtaining the ratio of each rainfall 



33 

 

pulse of each interval to the total amount of rainfall for Alexa, then multiplying this ratio 

by 125 mm for each rainfall pulse of the synthetic storm.  

Figure 3 

Hyetograph of the synthetic storm 

 

 

 

 

 

 

 

 

 

 

 

 

 

The storm starts at 08:00 am on 01Jan2048, and ends at 08:00 of 02Jan2048. From its 

start at 08:00 until 12:00, it rains and peaks its first peak at almost 10:00 am. From 12:00 

pm until 16:00, there is no rain, then it quickly begins to rain and peaks it’s second peak 

around 20:00, it decreases for a little while, and then starts to rain at 21:30, to peak it’s 

third peak around 22:30, and then peaks again at 01:30, then starts to decrease gradually 

and end. It is important to note that the last two peaks are very dangerous rainfall pulses, 

because they endure a large amount of rain in a small period of time, and this is what 

causes the disastrous floods and corresponding hazards. 

2.2.5 Assessment of Anabta’s Culverts 

2.2.5.1 Description of Anabta’s Culvert System 

Wadi Al Zeimar is the main wadi that runs in Anabta. It contains 13 culverts from the 

Bizzariya junction extending to the west of the town. Figure A.17 in Appendix A shows 

the aerial view, from the previous study conducted by Aqua Consulting Group (AQUA 

Consulting Center, 2013), shows the distribution of the culverts over the wadi. The 



34 

 

existing culvers are made up of concrete, and most of them are rectangular. The cross-

sectional area ranges from 9-18 m2.  

There are several houses close to the culverts, which makes them at risk in case of any 

flooding. The main culvert that was redesigned is culvert No. 1 in the aerial photo Figure 

A.17. It lays under Al Hejaz Gas Station. It is rectangular, with a span of 2 meters, and a 

height of 2 m, and has 2 barrels. Figure A.18 shows the culvert, and it is noticed that the 

wadi is very dirty, and as mentioned earlier, the uncontrolled dumping of garbage in the 

wadi reduces the capacity of the culverts and leads to environmental and health negative 

consequences. It is also noticeable that the culvert is close to the street and above it is a 

residential building and a parking lot.  

2.2.5.2 HY-8 Culvert Hydraulic Analysis Program 

The program that was used to redesign the culvert is HY-81. It is developed by the US 

Department of Transportation – Federal Highway Administration. HY-8 uses several key 

elements to automate culvert hydraulic simulations, simplifying culvert analysis and 

design. It assesses the culverts' effectiveness and determines whether or not the road 

overtops (Smemoe, 2022).  

Culvert No.1 was redesigned based on the peak flow at Anabta resulting from the climate 

change scenario (Run 4).  The culverts’ inlet elevation, outlet elevation and the roadway 

crest are respectively 166.6 m,166.5 m, 169.19m amsl. Overtopping of the roadway 

occurs when the elevation of water in the culvert exceeds 169.19 m. So, the proper design 

for the culvert considers that the water level must not exceed the roadway level, in 

addition, backflow of water must not occur. HY-8 automatically designs within these 

measures.  

 

 

 

 

 
1 https://www.fhwa.dot.gov/engineering/hydraulics/software/hy8/ 



35 

 

Chapter Three 

 Results 

3.1 Simulation Runs 

 The main way to compute the results is through simulation runs. One basin model, one 

meteorological model, and one set of control requirements make up each run (Feldman 

& Hydrologic Engineering Center U.S, 2022). Run 1 is made up of Al Zeimar catchment   

Model, Storm 1, and Runoff 1. Run 2 is used for calibration while Run 3 is for validation. 

Run 4 is used to simulate the future shift in climate.  

• For Run 1 the peak discharge at the outlet occurred at 08:30, with a value of 0.5 m3/s. 

The efficiency parameter (E) was 0.343, and the percent bias (PBias) was -24.75%. 

E is 0.343 so the models’ performance is passable. 

The model produced the same peak discharge as the observed data, but the difference is 

in the time of occurrence The hydrograph at the sink (outlet) is shown in Figure 4 below. 

Figure 4 

Hydrograph of Run 1 

 

 



36 

 

The following points discuss the behaviour of both observed and simulated 

hydrographs of Figure 4:  

2. At the first minute of the storm (time 00:00), the observed hydrograph had a reading 

of 0.18 m3/s, this reading is probably unquantified treated wastewater, while the 

model had a reading of zero. It is impossible for the model to give an instant reading 

because there are a number of losses. This error is due to error source 4 listed 

previously, where the model is uncomplete because HEC-HMS doesn’t have the 

option of defining sources of runoff other than storm water, and the wadi in reality 

has treated wastewater flowing in it. 

3. At 00:30, both hydrographs have the same behavior, but the modeled hydrograph 

underestimates the runoff, until at 02:20, both hydrographs intersect, and the 

simulated over estimates runoff, until 09:20 where they intersect again.  

4. In the observed hydrograph, the first peak occurred at 06:00 and the second occurred 

at 08:10, whereas in the simulated hydrograph, the first peak occurred at 07:00 and 

the second at 08:20, which shows a delay in the occurring of peaks and a slight 

underestimation of their values. The peak flow timing discrepancy occurs because 

the simulated hydrograph is generated using mathematical models, which makes the 

hydrograph sensitive to the model parameters, especially the lag time, where any 

under or over estimation in the lag time affects the time of the peak. Additionally, the 

precipitation data (limited stations that do not capture all sub catchments) and the 

assumption of uniform rainfall distribution all over the catchment (that may not 

perfectly match the reality) also cause the discrepancy.  

5. The real flow data has more pronounced peaks, suggesting it captures more detailed 

variations in flow. The modeled hydrograph is smoother and less variable. This 

smoothness might result from the simplifications in the model that don’t capture the 

steep slopes, valleys and impermeable surfaces which affect the flow. 

 

 

 

 

 



37 

 

3.2 Model Calibration 

In order to reduce the differences between both hydrographs (observed and simulated 

in Figure 4) to obtain reliable real results, calibration was manually done to the model 

parameters using the trial-and-error method. The peak discharge of 0.5 m3/s occurred at 

08:30, the NSE coefficient increased to 0.863, and percent bias decreased to 0.58%. 

The following is the  

 hydrograph of Run 2 at the sink of the catchment.  

Figure 5 

Hydrograph of Run 2 

 

By visually inspecting the hydrograph, the following comments can be concluded: 

1. Both hydrographs are similar in shape and magnitude. Both almost give a peak 

discharge of 0.5 m3/s, the difference is the time of occurrence of it. In the real storm, 

the peak occurred at 06:00, but in the model it occurred at 08:30.  

2. It is important to mention that the calibrated simulated hydrograph is smoother than 

the observed one. The simulation process involves averaging out short-term 



38 

 

flocculation, which means that rapid and sudden changes in the flow rate are 

smoothed over. The smoothed hydrograph is useful in long-term planning and 

infrastructure design, where it provides insights into overall trends without being 

overly influenced by short-term variations.  
 

Tables 8 ,9, and 10 show the differences between initial and calibrated parameters for 

each input parameter:  

• Transform – SCS UH  

Table 8 

Initial and Calibrated Lag Time 

Lag Time (min) Initial Calibrated 

Sub-basin 1 46 40 

Subbasin 2 33 29 

Subbasin 3 33 20 

Subbasin 4 21 20 

Subbasin 5 24 15 

Subbasin 6 26 15 

Subbasin 7 21 15 

 

The calibrated lag times are generally lower than the initial lag times. This indicates 

that the calibration process has identified a need for faster response times in the model, 

likely to better match observed data. 

• Loss – SCS CN 

Table 9 

Initial and Calibrated SCS CN Parameters 

Subbasin 
Initial Abstraction CN 

Initial Calibrated Initial Calibrated 

1 14.32 12 78.7 76 

2 14.45 13.14 77.85 78 

3 16 13.2 76 76 

4 17.2 11.55 74.7 70 

5 16.93 11.9 75 75 

6 13.5 14 79 79 

7 12.7 12.9 80 80 

 



39 

 

The CN values show minor adjustments, with some sub-basins experiencing slight 

increases or decreases. This indicates that the initial estimates were relatively close to 

the observed values, requiring only fine-tuning 

• Routing -Muskingum 

Table 10 

Initial and Calibrated Muskingum Parameters 

Reach 
K (hr) X 

Initial Calibrated Initial Calibrated 

1 3.11 3.6 0.25 0.3 

2 2.92 2.4 0.25 0.3 

3 2.39 2.3 0.25 0.38 

 

The K values show both increases and decreases across different sub-basins. This 

suggests that the calibration process identified a need for varied adjustments to better 

simulate the flood wave travel time through the river reach. The adjustments in K 

values improved the model's ability to simulate the timing and attenuation of flood 

waves more accurately. 

3.3 3.3 Model Validation 

The model needs to be evaluated to help make sure it can accurately replicate hydrologic 

processes under various settings. Comparing the model's output with observable data 

from a time period other than the calibration period is a common step in the validation 

process (Siegfried & Marti, 2021).The storm used for validation “Storm 2” was extracted 

from (IMS, 2023). It started on 20th March of 2020 at 14:40, and ended at 15:50 at the 

same day. It resulted in 1.4 mm on rainfall. Figures 6 and 7 show the hyetograph and 

hydrograph of Storm 2. 

 

 

 

 

 



40 

 

Figure 6 

Hyetograph of Storm 2 

 

 

 

 

 

 

  

 

 

 

 

 

 

 

 

 
 

 Figure 7 

Hydrograph of storm 2 

 

The key observations of the previous hydrograph are: 

1. At approximately 16:30, the hydrograph reached its peak discharge of 0.19 m3/s 

2. The sharp increase in discharge occurred between 15:00 and 16:30. Within this hour 

and a half, the flow escalated rapidly, then started to gradually decrease. 



41 

 

The peak discharge at the outlet of the catchment resulting from Run 3 that simulated the 

previous storm is 0.2 m3/s. It occurred at 16:10, and the Nasch-Sutcliff coefficient is 0.777, 

and PBIAS is 6.56%. The models’ performance is satisfying, and the amount and time of 

peak are very close to the real hydrograph. Figure 8 shows the hydrograph of Run 3. 

 

Figure 8 

Hydrograph of Run 3 

 

The model successfully captured the overall trend including the peak discharge. 

However, the initial rise is steeper in the simulated hydrograph, and the recession limb 

shows some deviation. 

 

 

 

 

 



42 

 

3.4 Climate Change Scenario Run 

The synthetic storm used for Run 4 is mentioned in section  2.2.4.3. Figure 9 shows the 

hydrograph of the run in Anabta. 

Figure 9 

Hydrograph of Run 4 

 

The flow starts in Anabta at almost 17:00 on 01 Jan., this is compatible with the 

hyetograph of the storm, where there is a small rainfall pulse at the beginning of the 

storm, then a bigger rainfall pulse occurs at 16:00 on 01 Jan., leading the runoff to start 

in Anabta.  

The hydrograph has two peaks, and this behavior is similar to the behavior at the outlet 

of the catchment. The first peak occurs at 21:00 on 01 Jan, and the second peak occurs 

at 01:40 on 02 Jan. It is important to note that the peak of 92 m3/s is a very high flow 

value. It can be noted that the huge peak discharge is at Sub-basin 1, Reach 1, and the 

outlet.  It occurred after the last two rainfall pulses mentioned earlier in section 2.2.4.3, 

and this supports the theory that rain pulses with high intensities (high precipitation over 

small time periods) are what cause flash floods.  



43 

 

3.5 Culvert No.1 Redesign 

The peak flow rate (92 m3/s) from Run 4 was used to redesign culvert No. 1 using HY-

8. If this flow rate occurs with the culverts’ current dimensions, it can drain only 24 m3/s. 

Where the head water elevation will reach 172.3 m amsl, which means that 3 meters of 

water will overtop to the road, and will cause catastrophic damage. To prevent this from 

happening, the culvert was redesigned using HY-8 as follows:  

• The outlet should be excavated to 166.2 m level. 

• The span of each opening should increase to 3 m. 

• The rise of each opening should increase to 2.5 m. 

• Two additional barrels are needed.  

The current area of the culvert is almost 10 m2, with the new design, it should be 

increased to 30 m2. It is important to note that the culvert cannot be enlarged this much 

because there are buildings on both sides, so this solution is not feasible nor applicable.  

Figure 10 shows a side view of the culvert. 

 

 

 

 

 

 

 

 

 

 

 



44 

 

 Figure 10 

Side View of the Culvert 

 

• Figure 10 shows the elevations of the inlet, outlet, and roadway. The blue line 

represents the level of water flowing in the culvert during the 92 m3/s peak flow rate. 

In the first 4 meters of the culvert, the flow level will be 168.9 m amsl, then it will 

flow fully with the culverts’ slope. This means that with the proposed design, the 

culvert can handle an extreme flow rate of 92 m3/s without any flooding of water. It 

is important to note that this design in not applicable nor feasible, because there are 

houses and buildings and other elements around the culvert on both sides. Thus, other 

solutions should be proposed rather than enlarging culverts. Other solutions include: 

 

• Flood mapping and zoning 

• Using green infrastructure 

• Detention Basins 

• Channelization and river restoration 

 

 

 



45 

 

Chapter Four 

Conclusions, Limitations and Recommendations 

4.1 Conclusions 

Even though climate change is not the most pressing issue for the people in the 

Palestinian Territories, the climate risks are significant and will affect current and future 

development challenges. The main objective of this study is to develop a rainfall-runoff 

model for the Al-Zeimar catchment, in order to be able to simulate and analyze the effects 

of climate change on the hydrology of the catchment, especially in the Anabta area, that 

witnessed catastrophic flooding in 2013. Data was collected, and then processed to build 

the HEC-HMS model. Then the future shifts in climate were projected on the study area, 

and used to assess the hydrological situation by testing the ability of a culvert at the 

entrance of Anabta to handle a 25-year storm, and it failed. A new design of the culvert 

was proposed using HY-8. In light of the previous steps, the following are the 

conclusions: 

1. The HEC-HMS model performed well in simulating the impact of average occurring 

rainfall events, and this is reflected in the calibration and validation runs. It is 

important to note that this model is the first model of the area which is based on real 

run off data. 

2. Due to the future climate shifts, the culvert at the entrance of Anabta failed to handle 

the 25-year extreme storm. This reflects true flood danger in the area, because Anabta 

has smaller and older culverts. 

4.2 Limitations 

The following are the limitations of the study:  

1. Data availability and reliability is a central point, especially in model building. 

Unfortunately, precise and reliable data are not available. For example, for 

precipitation data, the main available data is the daily and annual rainfall, and for 

models, smaller intervals are needed. There are a lot of data gaps, and a lot of 

uncollected data. All this makes the development of optimal solutions harder.  

2. There is no flow data measuring devices in the catchment. There is a partial flume 

that was installed at 2020 at the outlet, but its main use is to quantify treated 



46 

 

wastewater that flows all year in the wadis downstream of the treatment plant. It 

measures up to 90 cm of combined water, and this means small amounts of flow. This 

was a major limitation while calibrating and validating the model. 

3. There is no geodatabase for the rainfall and flow data. Data is disorganized and 

scattered and this made the collection of data take a very long time.  

4.3 Recommendations 

Regarding the research methods and ability to improve model results, the following 

recommendations can be taken into consideration: 

• Rainfall measuring devices for small intervals should be installed. Where the high 

density of rainfall gauges allows to capture localized variations in rainfall 

properties, therefore, enhance the accuracy and reliability of all the hydrologic 

studies In addition, it helps in investigating the response of the sub-catchments to 

different events. It is recommended that the rain gauges are installed downstream of 

potential flood-prone areas.  

 

• Rainfall and flow data within catchments should be systematically collected, 

cataloged, and maintained within a centralized database. This serves as a fundamental 

resource for hydrological analysis, watershed management, and informed decision-

making in water resource planning and flood mitigation efforts. By systematically 

documenting rainfall patterns and streamflow dynamics across catchment areas, this 

database facilitates a deeper understanding of hydrological processes, enabling 

researchers, policymakers, and water resource managers to identify potential risks. 

Moreover, the availability of historical data allows for the calibration and validation 

of hydrological models, enhancing their accuracy and reliability in predicting future 

scenarios and assessing the impacts of climate change on watershed hydrology. 

• Further study of the impacts of climate change on Anabtas culverts should be 

conducted.  

 

 

 



47 

 

List of Abbreviations 

Abbreviation Meaning 

AMC Antecedent Moisture Content 

Co Degrees Celsius 

CN Curve Number 

COSMO-

CLM 
Climate Limited-Area Modeling Community 

DEM Digital Elevation Model 

EPI Extreme Precipitation Indices 

GIS Geographic Information System 

GPS Global Positioning System 

GCM Global Climatic Model  

HEC-HMS 

HEC-SSP 

 

 

Hydrologic Engineering Center-Hydrologic Modeling 

System 

Hydrologic Engineering Center-Statistical Software 

Package 

HPE High