Modelling of the Surface Water Salinity in the Southwest Region of Bangladesh

MODELLING OF THE SURFACE WATER SALINITY IN THE SOUTHWEST REGION OF BANGLADESH

MOHAMMAD ZIAUR RAHMAN

DEPARTMENT OF WATER RESOURCES ENGINEERING BANGLADESH UNIVERSITY OF ENGINEERING AND TECHNOLOGY DHAKA-1000, BANGLADESH

FEBRUARY 2015

MODELLING OF THE SURFACE WATER SALINITY IN THE SOUTHWEST REGION OF BANGLADESH

MOHAMMAD ZIAUR RAHMAN

DEPARTMENT OF WATER RESOURCES ENGINEERING BANGLADESH UNIVERSITY OF ENGINEERING AND TECHNOLOGY DHAKA-1000, BANGLADESH

FEBRUARY 2015

MODELLING OF THE SURFACE WATER SALINITY IN THE SOUTHWEST REGION OF BANGLADESH

Submitted by

Mohammad Ziaur Rahman

(Roll No. 0412162005 P)

In partial fulfillment of the requirement for the degree of

MASTER OF SCIENCE IN WATER RESOURCES ENGINEERING

DEPARTMENT OF WATER RESOURCES ENGINEERING BANGLADESH UNIVERSITY OF ENGINEERING AND TECHNOLOGY Dhaka-1000, Bangladesh

February 2015

CERTIFICATION OF APPROVAL

The thesis titled “Modelling of the Surface Water Salinity in the Southwest Region of Bangladesh” submitted by Mohammad Ziaur Rahman, Roll No. 0412162005 P, Session: April 2012, has been accepted as satisfactory in partial fulfillment of the requirement for the degree of Master of Science in Water Resources Engineering on 11th February, 2015.

...... Chairman Dr. Umme Kulsum Navera (Supervisor) Professor Department of WRE, BUET, Dhaka.

...... Member Dr. Md. Sabbir Mostafa Khan (Ex-Officio) Head Department of WRE, BUET, Dhaka.

…………………………….. Member Dr. Md. Mirjahan Miah Professor Department of WRE, BUET, Dhaka.

…………………………….. Member Mr. Abu Saleh Khan (External) Deputy Executive Director (Operation) Institute of Water Modelling House# 496, Road# 32 New DOHS, Mohakhali, Dhaka-1206.

iii

CANDIDATE’S DECLARATION

It is hereby declared that this thesis or any part of it has not been submitted elsewhere for the award of any degree or diploma.

Signature of the Candidate

Mohammad Ziaur Rahman

TABLE OF CONTENTS

Page No.

TABLE OF CONTENTS v LIST OF FIGURES viii LIST OF TABLES xiv LIST OF SYMBOLS xvi LIST OF ABBREVIATIONS xvii ACKNOWLEDGEMENT xviii ABSTRACT xix

CHAPTER ONE INTRODUCTION 1.1 Background of the Study 1 1.2 Problems and Scope of the Study 4 1.3 Objectives of the Study 5 1.4 Organization of the Thesis 6

CHAPTER TWO LITERATURE REVIEW 2.1 General 8 2.2 Major River System of Bangladesh 8 2.3 Salinity Intrusion in Bangladesh 14 2.4 Study of Surface Water Salinity in Bangladesh 16 2.5 Study of Surface Water Salinity around the World 21 2.6 Study of Surface Water Salinity in Mathematical Modelling 24 2.6.1 Study in Bangladesh 24 2.6.2 Study around the world 28 2.7 Summary 29

CHAPTER THREE THEORY AND METHODOLOGY 3.1 General 30 3.2 Theory of Salinity 30 3.2.1 Salinity 30

Page No.

3.2.2 Causes of Salinity 30 3.2.3 Composition and Measurement of Salinity of Sea Water 31 3.2.4 Salinity Distribution in the Ocean 33 3.2.5 Freshwater-Saltwater Interactions 35 3.2.6 Effect of Salinity 38 3.2.7 Governing Equations of Hydrodynamics & Model 39 3.3 Mathematical Modelling 43 3.3.1 MIKE 11 47 3.3.2 MIKE 21FM 53

3.4 Methodology 54 3.5 Summary 56

CHAPTER FOUR STUDY AREA & MATHEMATICAL MODEL SETUP 4.1 General 57 4.2 Study Area 57 4.2.1 River Systems of Southwest region of Bangladesh 57 4.2.2 Meteorology 59 4.2.3 Hydrology 63 4.2.4 Upstream Withdrawl and Impacts on Bangladesh 64 4.2.5 Activities in GDA 71 4.2.6 Flow Augmentation without Structural Intervention 80 4.3 Data Collection 84 4.3.1 Data Analysis 92 4.4 Mathematical Model Setup 107 4.4.1 Model Development 107 4.4.2 Salinity Modelling 117 4.5 Summary 124

CHAPTER FIVE RESULTS AND DISCUSSIONS 5.1 General 125 5.2 Description of the Flow Scenarios 125 5.3 Model Calibration 127

Page No.

5.4 Model Validation 139 5.5 Analysis of Different Flow Scenario 144 5.5.1 Temporal and Spatial Variation of Salinity 144 5.5.2 Salinity Intrusion Line for different flow Scenarios 157 5.5.3 Comparison of Salinity Intrusion Line 157 5.5.4 Changes in Salinity for different flow Scenarios 170 5.5.5 Estimation of Spatial area due to different Scenario 174 5.5.6 Effect of Trans-boundary Flow on Crop Production 179 5.5.7 Sundarban Salinity Map 179 5.5.8 Establishment of Hydrograph under Increased Flow Scenario 189

CHAPTER SIX CONCLUSIONS AND RECOMMENDATIONS 6.1 General 194 6.2 Conclusions 194 6.3 Recommendations 196 REFERENCES 197

vii

LIST OF FIGURES

Figure No. Title Page No.

Figure 1.1: Map Showing River System of Southwest Region of Bangladesh (Source: IWM, 2012) 3 Figure 2.1: Major River System of Bangladesh (Source: IWM, 2012) 9 Figure 3.1: Annual mean of the sea surface salinity distribution (Source: World Ocean Atlas, 2005) 33 Figure 3.2: Annual mean of the sea surface temperature (Source: World Ocean Atlas, 2005) 34 Figure 3.3: Relationships between surface salinity and Evaporation minus Precipitation (E-P) patterns (Source: World Ocean Atlas, 2005) 34 Figure 3.4: The Ghyben-Herzberg relation 36 Figure 3.5: Classification of Mathematical Modelling 44 Figure 3.6: Mathematical Background of HD Modelling 50 Figure 3.7: Reach section with h- and q-grid points, on which the Saint Veannt Equations are solved (Source: DHI, 2014) 51 Figure 3.8: Centered 6-point Abbot Scheme (Source: DHI, 2014) 52 Figure 3.9: Model Hierarchy of Salinity Model 53 Figure 3.10: Flow diagram of Methodology to assess the Scenario & impact of Salinity Intrussion of Southwest Region of Bangladesh 56 Figure 4.1: Map Showing Study Area 60 Figure 4.2: Annual Rainfall Distribution in the Southwest Hydrological Regions of Bangladesh (Source: BMD, 2008) 61 Figure 4.3: The Hydrlogical Regions of Bangladesh 65 Figure 4.4: Ganges River Discharge at Hardinge Bridge (Source: BWDB, 2009) 66 Figure 4.5: Observed minimum flow of the Ganges River at Hardinge Bridge (Source: BWDB, 2012) 67 Figure 4.6: Maximum, Minimum and Average Water Level of the Ganges at Hardringe Bridge from 1909 to 2009 (Source: BWDB, 2009) 67

viii

Figure No. Title Page No.

Figure 4.7: Superimposed Daily Average Flow (m3/s) curve of the Ganges for the three periods (1973-74, 1975-96, 1997- 2010) (Source: BWDB, 2010) 69 Figure 4.8: Superimposed Daily Average Flow (m3/s) curve (dry season) for the three periods (1973-74, 1975-96, 1997- 2010) (Source: BWDB, 2010) 69 Figure 4.9: Layout of the proposed Ganges Barrage (Source: BWDB, 2012) 75 Figure 4.10: Schematic Diagram of Diversion System (Source: BWDB, 2012) 78 Figure 4.11: Link Channel of the Ganges Diversion System (Source: BWDB, 2012) 79 Figure 4.12: Schematic Diagram of the Ganges River Distributaries 81 Figure 4.13: Location of Discharge and Water Level measurement Stations (Source: IWM, 2012) 86 Figure 4.14: Location of Salinity measurement Stations (Source: IWM, 2012) 88 Figure 4.15: Seasonal variation of Tidal Rivers (Source: IWM, 2014) 93 Figure 4.16: Maximum, minimum and average water level of Ganges at Hardinge Bridge from 1912 to 2012 (Source: BWDB) 95 Figure 4.17: Maximum, minimum and average water level of Gorai at GRB from 1960 to 2009 (Source: BWDB/IWM) 95 Figure 4.18: Pussur River Water Level at Hiron Point (1984 - 2013) (Source: IWM) 96 Figure 4.19: Lower Meghna River Water Level at Chitol Khali (2011- 12) (Source: IWM) 96 Figure 4.20: Upper Meghna River Water Level at Bhairab Bazar (2011-14) (Source: IWM) 96 Figure 4.21: Minimum flow of Ganges at Hardinge Bridge from 1934 to 2012 (Source: BWDB) 98 Figure 4.22: Monthly Maximum, Minimum and average flow variation of Ganges at Hardinge Bridge (Source: BWDB) 98 Figure 4.23: Monthly variation of discharge in Gorai Railway Bridge during 1964-2013 (Source: IWM, 2014) 99 Figure 4.24: Monthly variation of water level in Gorai Railway Bridge during 1964-2013 (Source: IWM, 2014) 99

Figure No. Title Page No.

Figure 4.25: Annual flow volume in the Ganges and the Gorai River (Source: IWM, 2014) 100 Figure 4.26: Seasonal variation of salinity in the three peripheral rivers from West to East 101 Figure 4.27: Salinity variation with upstream freshwater flow during dry season 102 Figure 4.28: Observed monthly average discharge in Ganges River from 2011-2014 103 Figure 4.29: Observed monthly average discharge in Gorai River from 2011-2014 104 Figure 4.30: Flow diversion through the Gorai during monsoon (top) and dry (bottom) period in different years 104 Figure 4.31: Monthly Salinity variation with upstream freshwater flow 105 Figure 4.32: Observed salinity at different location for with and without dredging 106 Figure 4.33: Catchment Boundaries of Rainfall Runoff Model (Source: IWM, 2012) 109 Figure 4.34: Map showing Existing Southwest Regional Model Domain and Boundary (Source: IWM, 2012) 110 Figure 4.35: Map showing Development of Extended Southwest Regional Model Network 112 Figure 4.36: Map showing Developed Model Network and Boundary 113 Figure 4.37: Methodology of the salinity modelling 117 Figure 4.38: Bathymetry of newly developed Bay of Bengal model (Source: IWM) 121 Figure 4.39: Water level variation at the south boundary 122 Figure 4.40: Manning map of newly developed Bay of Bengal Model (Source: IWM) 123 Figure 5.1: Gorai River Hydrograph at different flow scenario 127 Figure 5.2: Calibration of Southwest Regional Hydrodynamic Model against Water level (2012) 130 Figure 5.3: Calibration of Southwest Regional Hydrodynamic Model against Water level (2012) 131 Figure 5.4: Calibration of Southwest Regional Hydrodynamic Model against Flow (2012) 132

Figure No. Title Page No.

Figure 5.5: Calibration of Southwest Regional Hydrodynamic Model against Flow (2012) 133 Figure 5.6: Calibration of Southwest Regional Hydrodynamic Model against Flow (2012) 134 Figure 5.7: Calibration of Southwest Regional Salinity Model against Salinity (2012) 135 Figure 5.8: Calibration of Southwest Regional Salinity Model against Salinity (2012) 136 Figure 5.9: Calibration of Southwest Regional Salinity Model against Salinity (2012) 137 Figure 5.10: Calibration of Southwest Regional Salinity Model against Salinity (2012) 138 Figure 5.11: Validation of Southwest Regional Hydrodynamic Model against Water (2011) 140 Figure 5.12: Validation of Southwest Regional Hydrodynamic Model against Water (2011) 141 Figure 5.13: Validation of Southwest Regional Hydrodynamic Model against Flow (2011) 142 Figure 5.14: Validation of Southwest Regional Salinity Model against Salinity (2011) 143 Figure 5.15: The Salinity Zoning Map for Scenario-1(November, 2011) 145 Figure 5.16: The Salinity Zoning Map for Scenario-2 (November) 146 Figure 5.17: The Salinity Zoning Map for Scenario-3A (November) 147 Figure 5.18: The Salinity Zoning Map for Scenario-1 (March, 2012) 148 Figure 5.19: The Salinity Zoning Map for Scenario-2 (March) 149 Figure 5.20: The Salinity Zoning Map for Scenario-3A (March) 150 Figure 5.21: The Salinity Zoning Map for Scenario-1 (April, 2012) 151 Figure 5.22: The Salinity Zoning Map for Scenario-2 (April) 152 Figure 5.23: The Salinity Zoning Map for Scenario-3A (April) 153 Figure 5.24: The Salinity Zoning Map for Scenario-1 (May, 2012) 154 Figure 5.25: The Salinity Zoning Map for Scenario-2 (May) 155 Figure 5.26: The Salinity Zoning Map for Scenario-3A (May) 156 Figure 5.27: 1 PPT Salinity Intrusion Map for Scenario-1 (Dry, 2011) 158 Figure 5.28: 1 PPT Salinity Intrusion Map for Scenario-2 (Dry) 159

Figure No. Title Page No.

Figure 5.29: 1 PPT Salinity Intrusion Map for Scenario-3A (Dry) 160 Figure 5.30: 5 PPT Salinity Intrusion Map for Scenario-1 (Dry, 2012) 161 Figure 5.31: 5 PPT Salinity Intrusion Map for Scenario-2 (Dry) 162 Figure 5.32: 5 PPT Salinity Intrusion Map for Scenario-3A (Dry) 163 Figure 5.33: 1 PPT Salinity Intrusion Map for Scenario-1 (Nov 2011 & May 2012) 164 Figure 5.34: 1 PPT Salinity Intrusion Map for Scenario-2 (Nov & May) 165 Figure 5.35: 1 PPT Salinity Intrusion Map for Scenario-3A (Nov & May) 166 Figure 5.36: 5 PPT Salinity Intrusion Map for Scenario-1 (Nov 2011 & May 2012) 167 Figure 5.37: 5 PPT Salinity Intrusion Map for Scenario-2 (Nov & May) 168 Figure 5.38: 5 PPT Salinity Intrusion Map for Scenario-3A (Nov & May) 169 Figure 5.39: Changes of 1PPT Salinity Intrusion Line for different Scenario (Dry, 2012) 171 Figure 5.40: Changes of 5PPT Salinity Intrusion Line for different Scenario (Dry, 2012) 172 Figure 5.41: Relation between salinity intrusion against Ganges connected river Flow 173 Figure 5.42: Pushdown of 1 PPT Salinity Intrusion for different flow Scenario (Dry, 2012) 177 Figure 5.43: Pushdown of 5 PPT Salinity Intrusion for different flow Scenario (Dry, 2012) 178 Figure 5.44: 2 PPT Salinity Intrusion for different flow Scenario (Dry, 2012) 180 Figure 5.45: Increased Crop production from Scenario-3A to Scenario-1 (May, 2012) 181 Figure 5.46: Increased Crop production from Scenario-3A to Scenario-2 (May, 2012) 182 Figure 5.47: Spatial Distribution of Salinity of Sundarban Mangrove Forest during May, 2012 (Scenario-1) 185

xii

Figure No. Title Page No.

Figure 5.48: Spatial Distribution of Salinity of Sundarban Mangrove Forest during May (Scenario-2) 186 Figure 5.49: Spatial Distribution of Salinity of Sundarban Mangrove Forest during May (Scenario-3A) 187 Figure 5.50: Impact of Increasing Fresh Water Flown in Sundarban Mangrove Forest during May, 2012 188 Figure 5.51: Map showing flow hydrograph for increased flow Scenario (Hisna-Mathabhanga-Bhairab-Kobadak System) 191 Figure 5.52: Map showing flow hydrograph for increased flow Scenario (Gorai-Madhumati System) 192 Figure 5.53: Map showing flow hydrograph for increased flow Scenario (Chandana-Barsia-Kumar System) 193

xiii

LIST OF TABELS

Table No. Title Page No.

Table 2.1: Catchment Areas of the Brahmaputra 10 Table 2.2: Peak Discharge at Different Points in Catchment Areas of the Brahmaputra 10 Table 2.3: Monthly Flows (m3/s) of the Brahmaputra-Jamuna at Bhahadurabad (1956-2006) 11 Table 2.4: Key hydrological characteristics of Ganges 12 Table 2.5: Monthly Flows (m3/s) of the Surma-Meghna at Bhairab Bazar (1964-2006) 13 Table 2.6: Location of Estuarine Water Bodies and Upstream Rivers in the Study Area 14 Table 3.1: Composition of Salts in Sea Water 31 Table 3.2: Temperature and Salinity variation in the Ocean Water 35 Table 3.3: Irrigation water parameters 39 Table 4.1: Monthly Mean Rainfall Distribution in the Southwest Hydrological Regions for 2008 61 Table 4.2: Meteorological Parameters in Southwest Hydrological Region 63 Table 4.3: Evapo-transpiration (mm/day) for each Southwest Hydrological Region for 2008 63 Table 4.4: Historical Flow (m3/s) of the Ganges at Hardinge Bridge68 Table 4.5: Features of the dredging 2010-2013 72 Table 4.6: Details of river links in the distribution system 77 Table 4.7: Main Features of Major Distributaries of the Ganges 80 Table 4.8: Inventory of Discharge and Water level measurement Stations 85 Table 4.9: Time duration and locations for salinity measurement 87 Table 4.10: Details of river cross-section survey 89 Table 4.11: Tidal ranges at different location in different river 92 Table 4.12: List of Water level data collection stations 94 Table 4.13: Comparison between Existing and Extended Southwest Regional Model 111 Table 4.14: Comparison of upstream and downstream Boundaries used in the Existing and Extended Southwest Regional one-dimensional Model 114

xiv

Table No. Title Page No.

Table 4.15: List of rivers and corresponding length used in the 1-D model 115 Table 4.16: List of boundaries used in the extended salinity model 119

Table 4.17: Values of Kmix 120 Table 4.18: List of boundaries at the upstream side of Bay of Bengal Model 123 Table 5.1: Monthly Flow Diversions (m3/s) 127 Table 5.2: Model Calibration Location and Correlation factor “R” 129 Table 5.3: Validation Location and Correlation factor “R” 139 Table 5.4: Description of flow boundary for Scenario-3 170 Table 5.5: Relation between Ganges Connected River Flow vs. Salinity 173 Table 5.6: Typical variation of salinity levels in April 175 Table 5.7: Typical variation of salinity levels in May 176

LIST OF SYMBOLS q Discharge Afl Flow area qin Lateral inflow h Water level/water depth α Momentum distribution coefficient If Flow resistance f Momentum forcing ρw, ρf Density of Water ρa Density of air C Concentration D Dispersion coefficient a dispersion factor b dispersion exponent V mean flow velocity ∆x Grid spacing ∆t Time step u Local current speed K1, K2, K3 Calibration parameter for dispersive effects A Cross-sectional area K Linear decay coefficient CL Source/sink concentration qL Lateral inflow t Time x, y X and y are Cartesian Co –ordinate (s) g Acceleration due to gravity ε Sea surface elevation (m) Cw Wind friction factor W Wind speed (m/s) Ω Coriolis parameter in the Bay of Bengal Pa Atmospheric pressure (kg/m/s2) T Temperature S Salinity Dv Vertical turbulent (eddy) diffusion coefficient. Dh Horizontal diffusion coefficient Ĥ Source term due to heat exchange with the atmosphere F Horizontal diffusion σT Prandtl number

xvi

LIST OF ABBREVIATIONS

AD Advection-dispersion BMD Bangladesh Meteorological Department BoBM Bay of Bengal Model BUET Bangladesh University of Engineering and Technology BIWTA Bangladesh Inland Water Transport Authority BWDB Bangladesh Water Development Board CEGIS Center for Environmental and Geographic Information Services DHI Danish Hydraulic Institute FAP Flood Action Plan FAO Food and Agricultural Organisation FM Flexible Mesh GDP Gross Domestic Product GPS Global Positioning System GK Ganges-Kobadak GRB Gorai Railway Bridge GIS Geographic Information System GRRP Gorai River Restoration Project GWT Ganges Water Treaty ICZMP Integrated Coastal Zone Management Project IWM Institute of Water Modelling IPCC Intergovernmental Panel on Climate Change MoWR Ministry of Water Resources MIKE 11 One-dimensional One Layer Model Develop by MIKE ABBOTT MIKE 21 Two-dimensional One Layer Flexible Mesh Model Develop by MSL MIKEMean Sea ABBOTT Level NAM Nedbor Afstromnings Model NGO Non-Governmental Organisation NWMP National Water Management Plan NWRD National Water Resources Database SRF Sundarban Reserve Forest SRDI Soil Resources Development Institute SWMC Surface Water Modelling Centre (now renamed as IWM) SWRM South West Regional Model SLR Sea Level Rise WARPO Water Resources Planning Organisation GDA Ganges Dependent Area UNESCO United Nations Educational, Scientific and Cultural Organization

xvii

ACKNOWLEDGEMENT

The author would like to mention with gratitude Almighty Allah for giving him the ability to complete this research work.

The author expressed his sincere gratitude and thanks to his honorable supervisor, Dr. Umme Kulsum Navera, Professor, Department of Water Resources Engineering (WRE), Bangladesh University of Engineering and Technology (BUET), Dhaka, for her continuous guidance, constant support, supervision, inspiration, advice, infinite patience and enthusiastic encouragement throughout this research work.

The author is also indebted to the member of the Board of Examination namely Dr. Md. Sabbir Mostafa Khan, Professor and Head, Department of WRE, BUET, Dr. M. Mirjahan Miah, Professor, Department of WRE, BUET and Mr. Abu Saleh Khan, Deputy Executive Director (Operations), Institute of Water Modelling (IWM) for their valuable comments and constructive suggestions regarding this study.

The author is highly gratitude to Dr. M. Monowar Hossain, Executive Director, IWM for allowing him to avail all kind of support from IWM and providing him all types of logistic support including data & model. The author expressing his intense thanks to Md. Zahirul Haque Khan, Director & Division Head, Coast, Port & Estuary Management Division, (IWM), for his motivation, constants encouragement and appreciation.

The author would like to express a very special indebtedness to his parents and his wife whose encouragement and support was a continuous source of inspiration for this work.

Finally, the author likes to express his sincere gratitude to all other teachers and members of the Water Resources Engineering Department, BUET, for their cooperation and help in the successful completion of the work.

xviii

ABSTRACT

Salinities in the Bangladesh coast are dependent on the annual rainfall, evaporation, freshwater flows discharging from upstream and the impact of climate change. Average salinity concentrations at the coast are higher in the dry season than in the monsoon, due to reduction in freshwater flow from the upstream. The flow of the Ganges in Bangladesh reduced significantly due to withdrawal of water in the upstream boundary. The reduction of dry season flow in the Ganges has led to various water quality, ecological, hydrological and hydraulic problems in southwestern Bangladesh. The source and quality of surface water becoming an important issue for the safe drinking water and also for crop production in the Southwest region of Bangladesh.

The simulation of the calibrated salinity model has been carried out in this study to investigate the baseline condition of salinity in the Southwest region of Bangladesh. Several scenario run have been simulated with the calibrated and validated hydrodynamic and salinity model. The scenarios are mainly based on upstream discharge condition. The minimum flow in Gorai River has been considered as the worst condition for salinity intrusion from Bay of Bengal. Another scenario with increase in upstream flow through Ganges connected rivers has been simulated to identify the saline free zone at the most south end zone. The present condition and different flow scenarios have been assessed by MIKE One-Dimensional and MIKE Two-Dimensional Modelling system. The study has been simulated from November 2011 to June 2012.

It was observed that, the some of the major rivers of Southwest would be saline free and all other rivers will have significant reduction of salinity due to increase in flow of fresh water through Ganges and its distributaries. Augmentation in the Gorai River flow reduces the salinity in the Pussur River and Sibsa River. The major finding of this study is that 4200 sq. km land will be saline free respectively by increasing the upstream flow. The saline free river water could be used directly for agriculture, domestic water supply (with low cost water treatment) and industrial purposes. It is also observed that, crop production will increase by 4023 sq km area with increasing upstream flow condition.

xix

CHAPTER ONE

INTRODUCTION

1.1 Background of the Study

Bangladesh is a low-lying, riverine country located in South Asia with a largely marshy jungle coastline of 710 km (441 mi) on the northern littoral of the Bay of Bengal, formed by a delta plain at the confluence of the Ganges (Padma), Brahmaputra (Jamuna), Meghna Rivers and their tributaries. About 700 rivers including tributaries flow through the country constituting a waterway of total length around 24,140 kilometres (15,000 mi). The country has an area of 147,570 square kilometers and extends 820 kilometers north to south and 600 kilometers east to west. Bangladesh has a tropical monsoon climate characterized by heavy seasonal rainfall, high temperatures, and high humidity. Water is the life-line for teeming millions of people living in rural and coastal areas. Most of the people of the country are dependent on land and water for their life and livelihoods. The country‟s economy is basically agrarian and is critically dependent on the waters of the rivers and their distributaries. Now a days the use of ground water has become very popular for irrigation but there is no control over its use. Agriculture production accounts for about 24 percent of the GDP and 24 percent of exports and this sector alone sustains over 60 percent of the labour force (BWDB, 2012).

Three Major River Systems are prominent in the country: viz, The Ganges-Padma, the Brahmaputra-Jamuna and the Surma-Meghna System. The Ganges rises from the Gongotri glacier on the southern slope of the Himalyas at an elevation of above 7000 m. west of Nanda Devi range in Himachal Pradesh and northernmost Uttar Pradesh, west of Nepal. The river comes out of the Himalyan and Siwalik range near Dehradun and enters into the plains at Hardwar. The Ganges is the only source of fresh surface water for a vast area in the greater districts of Rajshahi, Pabna, Kushtia, Jessore, Khulna, Faridpur and Barisal. Sustainable water resource management in the Region is of paramount importance for raising productivity of agriculture, forestry and fisheries, safeguarding bio-diversity and promoting balanced economic growth (MoWR, 2005).

The study area encompasses the entire Southwest (SW) Area of Bangladesh bounded by the Ganges and the Padma in the north and in the east extending into the Bay of Bengal to the south and the international border to the west. The Gross Area of the Southwest (including South central) region is 41,500 sq. km. The SW region comprises two distinct zones extending from the Ganges and Padma rivers south to around Khulna and the coastal zone. It has an area of 26,100 km2. Total net cultivable area (NCA) is 1.38Mha. The South central (SC) region comprises the entire Bhola, Patuakhali, Jhalokhati, Borguna, Barishal, Shariatpur and Madaripur districts and includes an area about 15,400 sq. km (IWM, 2003).

The area is endowed with a large variety of land and marine resources. The world‟s largest mangrove forest, the Sundarbans occupies about 10% of the project area along the sea coast. About 77% of the project area is mainly under cultivation or under human settlement. Rivers, beels and water bodies account for the remaining 13% of the project area. Lack of safe drinking water has been identified as the number one issue in the daily life of the SW region of Bangladesh. In the recent years, groundwater based water supply in coastal area is suffering from a number of major problems mainly arsenic contamination, lowering of the water table, salinity and non-availability of suitable aquifers (PDO-ICZMP, 2004). In this way the source of surface water becoming the important issue for the safe drinking water and also for crop production in the SW region.

Salinity in the river system of southwest coastal region increases steadily from December through February, reaching maximum in the late March and early April (EGIS, 2001). About 20% of the net cultivable land of Bangladesh coastal region is affected by different degrees of salinity (Karim et al.,1990). The impact of salinity on crop production as well as aquatic environment is well documented (SRDI, 2003; Uddin, 2005). Farmers grow mostly low-yielding, traditional rice varieties during the wet season. Most of the lands remain fallow in the dry season (January–May) because of soil salinity (Mondal, 1997). The increase of salinity in the Ganges distributaries has also led to ecological impacts on the world's largest mangrove forest, the Sundarbans (Siddiqi, 2001; Lacerda, 2001). The river system of Southwest region of Bangladesh presented in Figure 1.1.

Figure 1.1: Map Showing River System of Southwest Region of Bangladesh

(Source: IWM, 2014)

1.2 Problems and Scope of the Study

The flow of the Ganges in Bangladesh reduced significantly due to withdrawal of water in the upstream at the Farakka Barrage. India commissioned the Farakka Barrage in West Bengal in 1975 to divert 40,000 cusec water of the Ganges River into the Bhagirathi-Hooghly Rivers for flushing silt and improve navigability of Kolkata Port connected to the Bay of Bangal on the south (BWDB, 2012). The reduction of dry season flow in the Ganges has led to various water quality, ecological, hydrological and hydraulic problems in southwestern Bangladesh. The main impact of reduced low flow values has been the drop in hydraulic head of the Ganges River system, and the consequent increase in salinity in southwestern Bangladesh Rivers (Rahman and Ahsan, 2001).

The coastal areas of Bangladesh have already been facing salinity problem which is expected to be exacerbated by climate change and sea level rise, as sea level rise is causing unusual height of tidal water. In dry season, when the flows of upstream water reduce drastically, the saline water goes up to 240 kilometers inside the country and reaches to Magura district. Presently around 31 Upazilas of Jessore, Satkhira, Khulna, Narail, Bagerhat and Gopalganj districts are facing severe salinity problem. Agricultural activities as well as cropping intensities in those Upazilas have been changing; as a result farmers cannot grow multiple crops in a year (Shamsuddoha and Chowdhury, 2007). More than 30% of the cultivable land in Bangladesh is in the coastal area. About 1.0 million ha of arable lands are affected by varying degrees of salinity. Farmers grow mostly low-yielding, traditional rice varieties during the wet season. Most of the lands remain fallow in the dry season (January–May) because of soil salinity and the lack of good-quality irrigation water (Mondal, 1997). In general, soil salinity is believed to be mainly responsible for low land use as well as cropping intensity in the area (Rahman and Ahsan, 2001).

The increase of salinity in the Ganges distributaries has also led to ecological impacts on the world's largest mangrove forest, the Sundarbans, a UNESCO World Heritage Site. It is located at the extreme end of the southern Ganges delta and it is about 10,000 km2 in southwest Bangladesh and West Bengal of India. A total area of 62% lies in the Khulna region of the south western part of Bangladesh, while the

remaining 38% is in India (Siddiqi, 2001; Lacerda, 2001).The Bangladesh portion of Sundarbans covers an area of 6017 km² of mangrove forests, wildlife sanctuaries and sand bars, out of this 1905 km² are made up rivers, creeks and canals (Wahid, 1995). The land area of Bangladesh Sundarbans is about 4112 km² (Katebi, 2001). Salinity levels increased in the Sundarbans when intake-mouths of the Mathabhanga, Kobadak and other rivers that used to bring fresh water from the Ganges to the south were silted up and thus lost their connection with the Ganges. Therefore the result of increase salinity and alkalinity has damaged vegetation, agricultural cropping pattern and changing the landscapes in the Sundarbans region. A salinity level of 10 ppt (parts per thousand) in the water inundating the shores of the canals and the rivers of the Sundarbans area have led to the "top dying;" a disease of the prevalent native Sundari trees (Hoque et al., 2006).

Salinity in the country received very little attention in the past. Based on the above situation, a research study has been taken to measure the present level of surface water salinity in the Southwest area also to identify the solution to address the problem associated with salinity ingression.

1.3 Objectives of the Study

The main objective of this study is to apply one-dimensional hydrodynamic model for the selected reach of the Southwest River System of Bangladesh. Secondly, establish flow hydrograph under different hydrological scenario to assess the distribution and characteristic of surface water salinity by Mathematical Modelling technique. Finally evaluate the trans-boundary flow that will sustain the major rivers in the Southwestern region of Bangladesh.

The specific objectives of the study are:

1. To assess the scenarios which will lead to increase in flow of rivers and channels dependent on the Ganges River;

2. Assessing the present level of surface water salinity;

3. To assess the hydrodynamic condition which will reduce the salinity level at downstream portion of Southwest region of Bangladesh.

Based on the above objectives of the study the possible outcomes of the research work can be summarised as follows:

(i) Base line information of the existing river system of Southwest Region of Bangladesh like water level, discharge rainfall, salinity etc.;

(ii) Ascertain the level of salinity intrusion in coastal rivers with an identification of its influence zone under the base condition and different scenarios;

(iii) Identification of the saline free zone by limiting salinity level (< 1ppt) with different flow scenarios.

1.4 Organization of the Thesis

Considering literature review, location of the study area, theories related to the salinity intrusion, mathematical modeling, data analysis, model calibration, results and discussions the thesis has been organized under six chapters which are described below:

Chapter 1 describes the background, highlights the objectives of the study and contains organization of the thesis.

Chapter 2 describes previous the major river system of Bangladesh, salinity intrusion in Bangladesh and studies related to this surface water salinity in Bangladesh and around the world.

Chapter 3 describes the theories regarding this research work. It also describes the equations developed for hydrodynamic and salinity modeling. It contains the mathematical modeling tools used in the study. Finally it describes the brief description of the methodology.

Chapter 4 describes the study area which includes the meteorology, Hydrology, Impacts upstream withdrawal of surface water in Bangladesh. It also contains the data collection and data analysis of the study area. Finally it describes briefly mathematical model setup.

Chapter 5 contains the model calibration and validation for hydrodynamic and salinity modeling. It also presents the model results and their analysis regarding salinity intrusion.

Chapter 6 provides the overall conclusions of the study and also some recommendations for further study.

CHAPTER TWO

LITERATURE REVIEW

2.1 General

This chapter contains the river system and salinity intrusion in Bangladesh specially in Southwest region area. A few Journal paper, Conference/Workshop paper and study reports/research papers relevant to the current study have been collected from different sources to understand the salinity related work inside Bangladesh and around the world is discussed in the following articles.

2.2 Major River System of Bangladesh

The rivers of Bangladesh mark both the physiography of the nation and the life of the people. About 700 in number, these rivers generally flow south. The larger rivers serve as the main source of water for cultivation and as the principal arteries of commercial transportation. Rivers also provide fish, an important source of protein. Flooding of the rivers during the monsoon season causes enormous hardship and hinders development, but fresh deposits of rich silt replenish the fertile but overworked soil. The rivers also drain excess monsoon rainfall into the Bay of Bengal. Thus, the great river system is at the same time the country‟s principal resource and its greatest hazard. The system can be divided into four major networks:

a) Brahmaputra-Jamuna river system, b) Ganges-Padma river system, c) Surma-Meghna river system, and d) Chittagong region river system

The major river system and geographical location of Bangladesh presented in Figure 2.1.

(a) The Brahmaputra-Jamuna System

The Brahmaputra rises in the northern slope of the Himalayas in the Kailash range and flows 1127 km straight to the east parallel to the Himalyan range. In Tibet the

Figure 2.1: Major River System of Bangladesh (Source: IWM, 2012)

river is called Tsanpo. The Tsanpo is a sluggish river in the south eastern part of Lasha where the river is possibly highest navigable river in the world. The river makes a hairpin bend in the eastern edge of the Himalyan range where the Himalya also makes a right angle bend to form the Arakan-Yoma of Burma.The Tsanpo cuts a number of deep gorges here at Namcha Barwa (7755 m) and enters into the Assam valley at Sadiya (135 m) in Northeast Assam. The Brahmaputra in Assam is called Dihang. Enriched by a number of large tributaries in the Assam valley, the river enters into Bangladesh some 12 km upstream of Noonkhawa in Kurigram district. The distribution of catchment areas of the Brahmaputra among different countries in the region is given in Table 2.1.

Table 2.1: Catchment Areas of the Brahmaputra

Country Area km2 % of total Area China (Tibet) 292670 48 Bhutan 54290 9 India 186480 31 Bangladesh 72520 12 Total 605960 100 (Source: BWDB, 2012)

Distribution of peak discharge at different points of the Tsanpo-Dihang- Brahmaputra-Jamuna River system are shown in Table 2.2.

Table 2.2: Peak Discharge at Different Points in Catchment Areas of the Brahmaputra

Station Country Peak discharge m3/s % of Bahadurabad Discharge Tsela Dzong Tibet 5663 7 Pandu Assam(India) 52386 69 Dhubri Assam(India) 62580 82 Bahadurabad Bangladesh 76455 100 (Source: BWDB, 2012)

The latest record of peak discharge at Bahadurabad is 1,03,129 m3/s on 8/9/1998. Since updates of the upstream stations are not available, the old records have been used to avoid distortion in the percentages.

In Bangladesh, the left bank tributaries of the river are: Dudh Kumar, Dharla, Teesta and Hurasagar River. The last one carries the combined flow of the Karatoa-Atrai and the Jamuneswari-Karatoa.

On the right bank, the river throws out a number of distributaries such as the Old Brahmaputra, Chital-Jhenai, Pungli, Lohajong, Dhaleswari and Ichamati Rivers etc. These distributaries pass through the flood plains in the greater districts of Mymensigh, Tangail and Dhaka and ultimately discharge into the lower Meghna at Bhairab Bazar through the Old-Brahmaputra and near Munshiganj through the Dhaleswari. Monthly flow of the Brahmaputra-Jamuna is shown in Table 2.3.

Table 2.3: Monthly Flows (m3/s) of the Brahmaputra-Jamuna at Bhahadurabad (1956-2006)

Statistical Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Indicator max 10135 9094 11179 16496 25361 47773 65202 69167 54710 46684 32709 13380 avg 5291 4456 5095 8416 15868 31591 46999 43296 38069 24555 12186 7273 min 3284 3225 3195 4813 7976 17547 27970 29848 22439 13512 7388 5087 (Source: BWDB, 2012)

(b) The Ganges-Padma System

The Ganges rises from the Gongotri glacier on the southern slope of the Himalyas at an elevation of above 7000 m. west of Nanda Devi range in Himachal Pradesh and northernmost Uttar Pradesh, west of Nepal. The river comes out of the Himalyan and Siwalik range near Dehradun and enters into the plains at Hardwar. Flowing Table 2.4 shows the key hydrological characteristics of the Ganges River (Sarker et al, 2003).

In the plains, the Ganges has an easterly course and receives tributaries from the north in Nepal (Mahakali, Karnali, Gandak, Kosi etc.) and from the south in Rajasthan and northern slope of Vindhya parbat (Tons, Sone, Punpun etc.). The river enters into Bangladesh from the west some 18 km east of the Farakka Barrage in India and flows further about 95 km between India-Bangladesh border before

entering fully into Bangladesh. From this point the river flows in a south eastern direction for another 120 km and confluences with the Jamuna River upstream of Aricha. Total length of the river up to Aricha is 2200 km.

Table 2.4: Key hydrological characteristics of Ganges

Parameters Ganges (Hardinge Bridge) Catchment Area (103 sq. Km) 1000 Average annual rainfall (mm) 1200 Average Annual Discharge (m3/s) 11300 Average maximum Water level (m PWD) 13.7 Slope (cm/km) 5 Total Sediment Transport (M tons/y) 550 Bed Material Transport (M tons /y) 195

Bed material size (D50) (mm) 0.15 Planform Wandering (Source: BWDB, 2012)

The combined flow in the name of Padma confluences with the Meghna Upper River, upstream of Chandpur and is now called Lower Meghna which after travelling for another 80 km falls into the Bay of Bengal through many estuaries. The greater districts of Kushtia, Jessore, Khulna, Faridpur and Barisal on the right bank of the course of Ganges-Padma-lower Meghna isknown as the Gangetic delta in Bangladesh. In Bangladesh,the Mohananda, the left bank tributary flowing by the side of Chapai-Nawabganj confluences with the Ganges near Godagari. In its further down-stream course, the river throws out on the left bank the Baral River in Natore and the right bank the Mathabhanga, Kumar, Gorai and Chandana-Arakandi Rivers before confluencing with the Jamuna. In the Padma reach of the river, the important right bank distributaries are Arial Khan River and Dubaldia River and further down-stream in the Lower Meghna reach, the right bank distributaries are Dharmaganj, Naya Bhangani, Babuganj and Tentulia Rivers.

The average flood discharge of the river is approximately 50,000 m3/s. The discharge is mainly contributed by the snowmelt of the Himalayas and the monsoon rainfall. In general, the peak flood occurs between the mid-August and mid-September. The bed

material of the river consists of fine sand. The river is very dynamic and the channel of the river shifts between meandering and braided.

The Meghna River drains an area of 77,000 km2, of which about 46,500 km2 (60%) lies in Bangladesh. The Barak River is the principal headwaters of the Meghna. The Barak rises at an elevation of 2900 m on the south side of Mount Javpo on the Nagaland-Manipur border. The Barak in India has a catchment of 25,265 km2 (33% of Meghna basin) and on entering into Bangladesh at Amalshid bifurcates into the Kushiyara River and the Surma River. At the point of bifurcation, the larger portion of flow enters into the Kushiyara while the smaller portion of the flow enters into the Surma. The Surma in its western course carries the flow from Meghalya of India and the Kushiyara in its south western course carries the flow from Tripura hills and meet at Markuli and the combined flow is known as upper Meghna. The Boulai system draining the Garo hills in the north Mymensigh confluences with the upper Meghna at Dilalpur not far from Bhairab Bazar.

The Lower Meghna receives the left bank spill flow of the Brahmaputra-Jamuna through the old Brahmaputra at Bhairab Bazar and the Dhaleswari at Munshiganj. The Meghna also receives the left bank tributary the Gumti at Daudkandi. Monthly flows of the Surma-Meghna at Bhairab Bazar are shown in Table 2.5.

Table 2.5: Monthly Flows (m3/s) of the Surma-Meghna at Bhairab Bazar (1964- 2006)

Statistical Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Indicator max 2349 2296 3146 4621 10619 13327 17245 16187 16483 11561 8885 3125 avg 548 441 699 1602 3817 7016 10764 11359 10310 7940 3964 1112 min 75 81 162 665 1521 3197 6018 7959 7116 4237 1606 353 (Source: BWDB, 2012)

(d) The Chittagong Region system

The rivers of Chittagong and Chittagong hill tracts are not connected to the other river systems of the country. The main river of this region is karnafuli. It flows

through the region of Chittagong and the Chittagong Hills. It cuts across the hills and runs rapidly downhill to the west and southwest and finally to the Bay of Bengal. Chittagong port is located on the bank of Karnafuli. The river has been dammed upstream at Kaptai to create a water reservoir for hydroelectric power generation. Other important rivers of the region are the Feni, Muhuri, Sangu, Matamuhuri.

(e) Estuaries

Estuaries are transition areas of the upland rivers and the sea. Tide and tidal flow of saline sea water from the sea mixes in the estuary and propagates inland. However, the extent is constrained by upland river flow and slope. In the process, tidal effect and salinity concentration decreases in the upstream direction, depending on various hydraulic factors. Cyclonic Storm Surges also propagates upstream. Estuarine water bodies are listed in Table 2.6. In the above table location of each estuary by upazila, water level station at the upstream end of the estuary and the name of upland river joining the estuary has been mentioned.

Table 2.6: Location of Estuarine Water Bodies and Upstream Rivers in the Study Area

Name of Location Name of Up-stream SI Estuary Upazila Station ID Station River 1 Raimongol Shyamnagar SW130 Kaikhali Ichamati 2 Malancha Shyamnagar-Koyra SW165 Kobadak FO Kobadak 3 Sibsa-Pussur Koyra-Mongla SW244 Mongla Rupsa-Pussur 4 Haringhata Sarankhola SW107.2 Rayenda Gorai -Haringhata 5 Bishkhali Patharghata- SW39 Patharghata Bishkhali Barguna 6 Buriswar Barguna-Amtali SW20 Amtali Barisal-Buriswar 7 Lohalia Kalapara-Galachipa SW183 Kaitpara Lohalia 8 Tentulia Galachipa SW290 Charmontaz 9 Shahbazpur Tozumuddin SW279 Tajumuddin Surma-Meghna (Source: BWDB, 2012)

2.3 Salinity Intrusion in Bangladesh

Saltwater intrusion is the movement of saline water into freshwater aquifers, which can lead to contamination of drinking water sources and other consequences.

Saltwater intrusion occurs naturally to some degree in most coastal aquifers, owing to the hydraulic connection between groundwater and seawater. Because saltwater has a higher mineral content than freshwater, it is denser and has a higher water pressure. As a result, saltwater can push inland beneath the freshwater. Certain human activities, especially groundwater pumping from coastal freshwater wells, have increased saltwater intrusion in many coastal areas. Water extraction drops the level of fresh groundwater, reducing its water pressure and allowing saltwater to flow further inland. Other contributors to saltwater intrusion include navigation channels or agricultural and drainage channels, which provide conduits for saltwater to move inland, and sea level rise. Saltwater intrusion can also be worsened by extreme events like hurricane storm surges (Barlow, 2003). .

It has already been stated that Bangladesh has a coastline of over 700 km. A large part of this is in the sea face of the Gangetic Delta in Bangladesh. The coast line in the Gangetic delta is intersected by large estuaries connected to upland rivers. Tide and salinity propagates through these estuaries to the upland.

Like many countries in the world, the coastline of Bangladesh is dominated by tides and salinity from the bay. Of the total coastline, the southwest region bounded by Ganges-Padma River on the north, the Gorai-Madhumati and Baleswar-Haringhata River on the east, Indian border on the west and the Bay of Bengal on the south, is the most severely affected area by salinity intrusion. High salinity, associated with sedimentation of rivers, rendered the SW region into a challenging hydro- morphological situation. The region consists of a very intricate river system where strong tidal effects appear even about 150 km upstream of the coast. The only significant upstream freshwater to the region is the Ganges water, which flows into the Gorai river. However, during the dry season, the mouth of the Gorai is almost dry. The salinity levels at sea during that period are comparatively high; consequently, the region is severely affected by salinity intrusion. Many studies and projects were carried out, but the adverse effects could not be removed. On the other hand, the situation in the adjacent south central region seems to be much better, as the salinity levels remain low in that region throughout the year. This is due to the fact that a considerable fraction of the freshwater discharge from the Padma River is

diverted into the region through different branches of the Meghna River and then flow through large rivers in the south namely the Bishkhali, the Buriswar and the Baleswar (Islam, 2007).

The huge freshwater outflow from the Ganges, the Jamuna and the Meghna induce a large zone of brackish water in the coastal region of Bangladesh. The salinity conditions in the northern-most part of the Bay of Bengal are governed by seasonal movements of the front between sea water and the brackish water. These movements are predominantly governed by the variations in freshwater discharge, coastal currents and mixing process. In the wet season the zone with brackish water may extend 200-300 km into the Bay of Bengal, whereas in the dry season the front between sea water and brackish water is located considerably closer to the coast. The general ocean currents in the Bay show a clockwise circulation of water in the dry season and anti-clockwise circulation in the wet season (Rahman, 2006).

2.4 Study of Surface Water Salinity in Bangladesh

Uddin et al., 2012 investigated the seasonal variation of soil salinity in coastal areas of Bangladesh. Bangladesh is likely to be one of the most vulnerable countries in the world to salinity problem. This paper discusses seasonal variation of soil salinity in coastal areas (Cox‟s Bazar, Bhola and Khulna) of Bangladesh. Three coastal districts from three coastal regions (eastern, central and western) were selected purposively for the collection of soil samples during four different seasons (winter, pre-monsoon, monsoon and post-monsoon). Soil samples from nine coastal areas (3 in each district) were collected (during December, 2009 to November, 2010) to determine seasonal variation of soil salinity of coastal districts. From the findings it can be said that soil salinity depend on annual rainfall and evaporation. In monsoon, soil gets enough water and soil salinity decreases as rain water dilutes the concentration of salt in the soil. In post-monsoon, soil salinity starts to increase because of lower rainfall and higher evaporation of moisture from soil surface. Increasing soil salinity continues up to pre-monsoon when soil becomes water stressed.

Uddin et al., 2011 had studied the Constraints and Management Strategy for crop production for climate change and salinity in Bangladesh. Climate change is an

important issue now-a-days. Global warming i.e. climate change causes sea level rise and that affect the coastal areas of Bangladesh. Salinity problem received very little attention in the past. It has become imperative to explore the possibilities of increasing potential of these (saline) lands for increased production of crops. Thus, it is necessary to have an appraisal of the present state of land areas affected by salinity. Salinity ingress causes an increase in soil salinity, especially when farmers irrigate their lands with slightly saline surface water at the beginning of the low flow period. SRDI (1997) reported that, soil salinity levels south of Khulna and Bagerhat towns ranged between 8 to 15 dS/m during the low flow season. It is also reported that, several sub districts (such as Kachua, Mollahat, and Fultali) south of the Sundarbans „“ known to be non-saline in the pre-Farakka period „“ have began to develop soil salinity during the low flow seasons of 1980s. The anticipated results of salinity ingress will be, at a minimum, of the same order for climate change induced low flow regime compared to similar effects shown by deliberate withdrawal of flows at Farakka barrage. The anticipated sea level rise would produce salinity impacts in three fronts: surface water, groundwater and soil. Increased soil salinity due to climate change would significantly reduce food grain production.

Islam and Albrecht, 2011 carried out a study of the water salinity in the Sundarbans rivers in Bangladesh. The objectives of this paper are to investigate the water salinity approximation in the Sundarbans rivers, which will be considered as a tool for decision making. The continuous reduction and deterioration of quality of the Ganges fresh water in the catchment is the root cause of salinity intrusion and damage of the Sundarbans ecosystems. Considering the present salinity intrusion trends in different ecological zones in the Sundarbans and management condition, an applied research and awareness education programme should be included as a potential environmental development agenda and should introduce for the stakeholders in different stages. To protect the mangrove wetland ecosystems in the Sundarbans region the alternative approach of proposed upstream water reservoir in Nepal and fresh water supply in the downstream should be ensured. Water salinity approximation on the Sundarbans rivers will support planning by decision makers to protect the special natural heritage site and world largest mangrove wetland ecosystems in Bangladesh.

Uddin and Haque , 2010 studied about the salinity response in southwest coastal region of Bangladesh due to hydraulic and hydrologic Parameters. This study assessed how salinity is influenced by the upstream fresh water discharge, local rainfall and mean tide level in the Sibsa and Pussur River of Paikgacha and Rampal, Bangladesh. Furthermore, the study aims to determine the relative influence of the different factors on salinity and identify key factors that control the salinity level in those rivers. Hence it was hypothesized that the resulting salinity levels reflect the combined result of multi-components. Paikgacha and Rampal experienced high and moderate salinity through the Sibsa and Pussur River respectively. Linear regression showed a significant correlation within salinity-river discharge as well as salinity- rainfall in the Pussur and Sibsa Rivers respectively. Again multivariate analysis showed that rainfall and river discharge are the key factors influencing salinity in the Sibsa and Pussur Rivers in Paikgacha and Rampal regions respectively.

Rahman and Bhattacharya, 2006 investigated the salinity intrusion and its management aaspects in Bangladesh. In the coastal area of Bangladesh salinity increases during minimum river discharges but never exceeds seawater salinity (34 ppt). Intrusion of saline water during the dry season is up to Char Gazaria where salinities are less than 1 ppt. Salinity intrusion can increase either due to a decrease of fresh water flow in the Lower Meghna River during the dry season, or due to further penetration of tide into the river system. The salinity front moves inland from the month of November due to the reduction of fresh water flows and intrudes up to 150 km inland in the lower Meghna in the southeast and up to 290 km up the Pussur River in the southwest of the country. Maximum salinity levels occur during March- April. The total trans-boundary sources of potable water in groundwater aquifers will be affected because agriculture is the main livelihood of a highly dense population which requires huge groundwater abstractions for irrigation purposes. In this situation, management of salinity intrusion is a vital issue for Bangladesh. A vision of a sustainable livelihood and environment for Bangladesh includes saline water proofing by structural management like coastal embankment projects, dams, sluices, etc., as well as coastal area zoning as a non-structural management strategy to change land use.

Alam et al., 2011 assessed the salinization of the coastal areas of Bangladesh corresponding to climate change using secondary data. The present situation of salinity of surface water in southern part of coastal zone of Bangladesh has been analyzed for selected stations. The analysis of rainfall for specific season & month shows a more reliable approach to identify change in climatic factor pattern. From this study, it is evident that increase of salinity level in surface water is gradually becoming more significant. The present situation analysis and available historical data indicates that the salinization trend is sharp in interior coastal zone. The exterior coast has a more high level of salinity and the trend is fluctuating. The study also suggests that the salinity is gradually intruding more towards inland and rainfall has a clear and significant relationship with salinity as a climatic factor. Historical rainfall analysis gives evidence that over the period rainfall pattern has changed both in magnitude and distribution, providing a possible evidence of climate change which influences salinity. In this study, only selected portion of the coastal region were analyzed based on available data. The coastal region of Bangladesh has a very complex hydro-geological nature and to understand the scenario of the entire coastal zone requires all hydrological system and climatic parameters to be taken into account.

Islam and Gnauck, 2007 studied the salinity Intrusion due to fresh water scarcity in the Ganges Catchment and the challenges for urban drinking water and Mangrove Wetland Ecosystem in the Sundarbans Region. Freshwater is the sustaining force for all life on this earth. It is integral not only the sustenance of our ecosystems but also to the survival of humans (Rao, 2006). The reduction of freshwater in the Ganges catchment has created environmental problems in urban drinking water supply at 12 small towns in the Sundarbans region in southwestern Bangladesh; and it is also one of the main threats for mangrove ecosystems. Since the diversion of Ganges water at Farakka Barrage in India from early 1975, salinity level has increased drastically in the south western part of Bangladesh. Due to reduction of fresh water flow urban drinking water supply, industrial production, agriculture, fisheries, navigation, hydromorphology and mangrove wetlands ecosystems have been affected. Urban area both surface and groundwater have become unfit for human consumption. In the consequences about 0.170 million hectares (20.4%) of new land, and almost 27 small

towns have been affected by various degrees of salinity during the last three decades. The saline front defined by 6 dS/m iso-haline has penetrated up to 173 km north from the coast. The salinity has exceeded the recommended level 1 dS/m for potable water. It is a new threat to supply quality drinking water to the small towns in the southwestern region in Bangladesh. The dominant mangrove Heritiera fomes and Ceriops decendra species are affected by top dying disease which is recognised as key management concern. The findings of this study a potential contribution for making comprehensive management plan are for urban water supply, and protection of the mangrove wetlands ecosystems in the Sundarbans region.

Seal and Baten, 2012 studied the Salinity Intrusion in Interior Coast in South Central part of Bangladesh. The main objective of the study are exploring the causes of salinity intrusion in the area adjacent to the lower Meghna, measuring the present level of surface water salinity, assessing and quantifying the impact on agriculture, identifying the solution to address the problem associated with salinity ingression. The study is intended to explore the level of salinity in surface water in Gosairhat upazila, an interior coast, and the causes behind the salinity intrusion. This study also intends to quantify the impact of salinity in agriculture and gives possible solutions to address the problem. Within IPCC scenarios, A2 and A1FI warn for a bleak future where the study area is likely to experience drastic decline in rice and wheat production by 2099 against the base year 2010. Present salinity concentration has already put a threat to the crop production and a significant yield loss has already been observed in the dry season. In the changing scenario of sea level rise, it has been predicted that the increasing concentration of salinity will create more pressure to the farmer by reducing yield on one hand and threatening livelihood, income generation and food security on the other hand. Therefore, to reduce the future loss and to prevent the present loss, the study recommended some adaptive techniques to manage salinity. Among them, fulfilling leaching requirements and selecting salinity tolerant crop varieties are important ones.

WARPO, 2001 studied that the Options for Ganges Dependent Area to assess the surface water demand in GDA and to select the alternative cost effective development strategies by identifying various improvement options. A wide range of

development options was considered in OGDA study involving interventions both with and without augmentation from the Ganges. Some options that were considered for diversion and distribution of Ganges flow are: restoration of Gorai River as pilot priority works, construction of offtake structure at Gorai, diversion through pumping, construction of Barrage in the Ganges and different water management options. From these findings OGDA study formulates some key strategy to relieve drainage congestion within the polder area through development of a sustainable river and drainage system, control salinity intrusion and relieve water shortages in the area. This is to be achieved by a combination of river and drainage improvement programmes, augmentation of dry season upland flows and improved management of trans-regional wet season flood flows. In line with these implementation of Gorai River Restoration Project and rehabilitations of the GK scheme were recommended.

2.5 Study of Surface Water Salinity around the World

Terray et al., 2012 studied near-surface salinity as nature‟s rain gauge to detect human influence on the tropical water cycle. The changes in the global water cycle are expected as a result of anthropogenic climate change, but large uncertainties exist in how these changes will be manifest regionally. This is especially the case over the tropical oceans, where observed estimates of precipitation and evaporation disagree considerably. An alternative approach is to examine changes in near-surface salinity. Datasets of observed tropical Pacific and Atlantic near-surface salinity combined with climate model simulations are used to assess the possible causes and significance of salinity changes over the late twentieth century. Basin-averaged observed changes are shown to enhance salinity geographical contrasts between the two basins: the Pacific is getting fresher and the Atlantic saltier. While the observed Pacific and interbasin-averaged salinity changes exceed the range of internal variability provided from control climate simulations, Atlantic changes are within the model estimates. Spatial patterns of salinity change, including a fresher western Pacific warm pool and a saltier subtropical North Atlantic, are not consistent with internal climate variability. They are similar to anthropogenic response patterns obtained from transient twentieth and twenty-first-century integrations, therefore suggesting a discernible human influence on the late twentieth-century evolution of

the tropical marine water cycle. Changes in the tropical and midlatitudes Atlantic salinity levels are not found to be significant compared to internal variability. Implications of the results for understanding of the recent and future marine tropical water cycle changes are discussed.

Yu, 2011 assessed the global relationship between the ocean water cycle and near- surface salinity, Ocean evaporation (E) and precipitation (P) are the fundamental components of the global water cycle. They are also the freshwater flux forcing (i.e., E‐P) for the open ocean salinity. To gain a better understanding of the ocean rain gauge concept, here they address a fundamental issue as to how E‐P and salinity are related on the seasonal timescales. A global map that outlines the dominant process for the mixed‐layer salinity (MLS) in different regions is thus derived, using a lower‐ order MLS dynamics that allows key balance terms (i.e., E‐P, the Ekman and geostrophic advection, vertical entrainment, and horizontal diffusion) to be computed from satellite‐derived data sets and a salinity climatology. The study suggests that the E‐P regime could serve as a window of opportunity for testing the ocean rain gauge concept once satellite salinity observations are available.

Barlow and Reichard, 2010 assessed the saltwater intrusion in coastal regions of North America. Saltwater has intruded into many of the coastal aquifers of the United States, Mexico, and Canada, but the extent of saltwater intrusion varies widely among localities and hydrogeologic settings. In many instances, the area contaminated by saltwater is limited to small parts of an aquifer and to specific wells and has had little or no effect on overall groundwater supplies; in other instances, saltwater contamination is of regional extent and has resulted in the closure of many groundwater supply wells. The variability of hydrogeologic settings, three- dimensional distribution of saline water, and history of groundwater withdrawals and freshwater drainage has resulted in a variety of modes of saltwater intrusion into coastal aquifers. These include lateral intrusion from the ocean; upward intrusion from deeper, more saline zones of a groundwater system; and downward intrusion from coastal waters. Communities within the coastal regions of North America are taking actions to manage and prevent saltwater intrusion to ensure a sustainable source of groundwater for the future. These actions can be grouped broadly into

scientific monitoring and assessment, engineering techniques, and regulatory approaches.

Woods, 2010 studied the surface water discharge and salinity monitoring of coastal estuaries in Everglades National Park, USA. Discharge and salinity were measured along the southwest and the southeast coast of Florida in Everglades National Park (ENP) within several rivers and creeks from 1996 through 2008. Tropical weather activity and wet-season duration have a profound affect on salinity and flow conditions within these estuaries. A short wet season in 2004 caused dry season salinities along the southeast coast of ENP to be greater than the 13-year mean conditions by almost 7 ppt. While annual discharge to the coast during 2004 was the lowest recorded; discharge during the following year (2005) was the highest recorded due to Hurricanes Katrina and Wilma. During Hurricane Katrina, 7 inches of rain fell on ENP, which substantially reduced the high salinities observed during the dry season of 2005 (Woods and Zucker, 2007).

Xinfeng and Jiaquan, 2010 studied that the rivers in the Pearl River Delta were seriously affected by the strong salinity intrusion and the drinking water of residents in this area was continuously threatened in recent years. To solve the salinity intrusion problem, the Pearl River Water Resources Commission (PRWRC) has organized fresh water transfers from upstream several times since 2005. Based on the analysis, this paper proposes some measures for preventing the salinity intrusion to ensure the safety of water supply and the sustainable development of water resources. From the analysis of factors affecting the salinity intrusion, the decrease of upstream runoff in dry period has a great influence on the fresh water supply, but the runoff can be adjusted by the upstream key reservoirs to a certain extent. For the topography change of estuary and rivers, some measures can be taken, such as prohibiting from dredging river sand, limiting dredging channels, and so on, in order to prevent riverbed recession. The influencing factors, such as the rise of sea level resulted from global warming and the effect of wind and its wind direction on salinity intrusion can be adjusted by human beings to a very limited extent, but the influence of the tide activity can be controlled by the engineering construction. In order to resist salinity intrusion effectively and protect the water supply in the Pearl

River Delta, some effective measures must be taken to control the increase of water consumption and water pollution.

Kijne, 1996 studied in a report on water and salinity balances for irrigated agriculture in Pakistan. Water and salt balances are calculated for three irrigated areas in Pakistan, which differ in water availability, amounts of water pumped for irrigation from groundwater, and salt content of the irrigation water. One of the sample areas is the Chasma Right Bank Canal (CRBC) command area in the North-West Frontier Province and the other two are in the Punjab, in the command areas of the Gugera Branch Canal and the Fordwah/Eastern Sadiqia Irrigation System. It is concluded from the analysis that current irrigation and agronomic practices are not sustainable. In the CRBC site, considerable groundwater recharge occurs, which in the absence of groundwater pumping leads to rising water tables, waterlogging, and salinity. The only feasible solution appears to be to limit the irrigation supply to farmers and to reduce the area under rice. In the sample irrigation areas of the Punjab, groundwater is mined, water tables drop, and salt continues to be added to the root zone because of the relatively high proportion of irrigation water derived from pumped groundwater. If the current high crop intensities are maintained, further degradation of land and water resources is inevitable.

2.6 Study of Surface Water Salinity in Mathematical Modelling

2.6.1 Study in Bangladesh

Khan and Kamal, 2014 assessed the current condition, future projections on Water Availability in the Ganges coastal zone. The water resources in the coastal zone of the Ganges delta are vital for crop production, fisheries, ecosystem sustenance and livelihoods. The present and future agriculture-aquaculture systems in the coastal zone of Ganges Delta depend to a large extent on the availability of fresh water which in turn is governed by a number of external drivers. The present condition of salinity intrusions and availability of water are analysed based historical data and field measurements on salinity, water flow, water level and applying numerical modelling technique (MIKE 11). Field measurement shows that salinity remains below 2ppt (3.5 DS/m) over the whole year in the major parts of Barisal divisions

that is water is available for irrigation. It implies that water is available for irrigation for rabi, Kharif-1 and Kharif-2 crop seasons. Simulation of sea level rise is carried out to see the effect on salinity intrusion, it is seen that in the changing climate with 22cm sea level rise, salinity remains within 2ppt in the low saline zone since this area receives plenty of freshwater from Lower Meghna river. In moderately saline area i.e. parts of Khulna Division, there is fresh river water from mid-June to mid- February i.e. salinity remains less than 2ppt. The level of river salinity and its temporal and spatial variation in the southwest coastal zone of Bangladesh are largely affected by the trans-boundary flow in the Ganges River and its distributaries. An immediate impact of the change in the Ganges River flow was observed in the level of river salinity in Khulna. The level of river salinity was below 1ppt at Khulna prior to1975. At present the level of river salinity at Khulna remains above 12ppt during the dry season.

Akhter et al., 2012 studied the impact of climate change on saltwater intrusion in the coastal area of Bangladesh. This paper presents the potential impacts of salinity intrusion due to climate change affecting fresh water pocket, drinking water, fish habitat, echo-system of the coastal area. Mathematical modelling systems MIKE 11 and MIKE21FM have been selected. The 5ppt isohaline intrudes 55 km to landwards for 60cm SLR and about 95 km for 120 cm SLR to landwards. The study shows that about 0 to 2 ppt salinity zone pocket becomes 7 to 11 ppt due to 120 cm SLR in Pussur, Baleswar, Tentulia River system. 200 cumec flow of water in the Gorai River, the distributary of the Ganges River can push the 1 ppt saline line by 20 km seawards in the southwest region. Sea level rise and decrease of upstream fresh water flow causing significant temporal and spatial change in salinity level which may affect the growth of agriculture, aqua-culture and drinking water supply. The analysis of the result shows that the existing freshwater pockets which are used for agriculture in the coastal areas are to be lost with the rise of 1m SLR. Drinking water and fish habitat are to be affected by salinity intrusion due to climate change. In the Sundarbans there are three species; Sundari, Gewana and Goran that grow in three salinity zones depending on salinity level; low, medium and high . The low salinity zones may become medium salinity zone and medium salinity zone may become high salinity zone due to climate change. This change is likely to affect the growth of

wood and bio-diversity of the Sundarbans. Fresh water flow from Ganges during dry season is crucial to reduce the salinity intrusion and maintaining fresh water zone in the coastal area.

BWDB, 2012 studied the Ganges Barrage Study Project. The pre-feasibility study carried out detailed analysis of the river morphology and review of satellite imageries between 1973 and 1999 to identify the most favourable barrage site in terms of future stability of the river channel. Pre-feasibility study recommended two sites for the barrage at the Hardinge Bridge area: one at Tagorbari, 10 km downstream of the Gorai River off-take and the other at Pangsha, 36 km downstream of the Gorai River off-take. The proposed Ganges Barrage would fulfill the long-felt demand for utilizing the Ganges water for agriculture, fisheries, ecosystems (including the Sundarbans mangrove ecosystem), navigation, domestic water etc. leading to socio-economic development of Bangladesh in general and the SW region in particular. The increased flow of fresh water through the Hisna-Mathabhanga- Kopotaksha River system, the Gorai-Modhumati-Nabaganga River system and the Chandana-Barasia River system would meet the demand of water for the ecosystem and arrest the under process environmental degradation in the SW Region including the Sundarbans. One dimensional model (MIKE 11) applied to develop OGDA model for the assessment of irrigation, drainage and flow augmentation and salinity intrusion of GDA area. The outputs have been compared with before Ganges Barrage and after implementing Ganges Barrage. The main function of the Ganges Barrage would be to create a reservoir for flow augmentation and equitable distribution of water both in the dry season and in the wet season in the GDA.

IWM, 2013 carried out a study of salinity zoning map for coastal zone of Bangladesh. A comprehensive assessment on salinity and storm surge was carried out through extensive and continuous salinity measurement and field investigation under this study. A synopsis on the salinity and storm surge issue in this study is given below Mathematical modelling systems MIKE 11 and MIKE21 have been selected. The study shows that more saline water intrusion is likely to occur during dry season with the increased sea level rise of 50cm in the year 2050. It is also evident from the study that more saline water intrusion through Baleswar-Kaliganga-

Madhumati River system under climate change condition and eventually freshwater zone becomes saline. In the present, salinity level remains from 0 to 2ppt in Nazirpur and Nesarabad upzilla of Pirojpur districts during dry period and river water is used for agriculture and aquaculture and other domestic and industrial uses. However, in the same area salinity level is likely to increase up to 5ppt in the changing climate with the sea level rise of 50cm in the year 2050 causing scarcity of freshwater for agriculture and other domestic uses during dry season. The vulnerability of coastal area to storm surge flooding considerably high, since it is predominantly low-lying and characterized by numerous tidal rivers and polders. Cyclones hit the country‟s coastal regions in the early summer (April–May) or late rainy season (October– November).Most of the cyclones hit the coasts of Bangladesh with north-eastward approaching angle. Over the last 52 years (1960-2012) about 19 severe cyclones hit the coast of Bangladesh. The high risk area is defined as the area that experiences 1m or more inundation depth since such depth can cause human casualties and damage of crops and infrastructure.

IWM, 2010 carried out the study „Strengthening the Resilience of Water Sector in Khulna to Climate Change‟. Bangladesh is considered as one of the most vulnerable countries in the world to climate change. The city of Khulna being located in the coastal area of Bangladesh and influence by tides from the Bay of Bengal is highly vulnerable to climate change. The main objective of the study is to assess the impacts of climate change on drainage, water availability, and the salinity situation in the Khulna city and to make recommendations to make the system more resilient to climate change and proper adaptation options based on social, economic, public health and urban planning aspects. The study summarizes that the city of Khulna would be vulnerable to climate change as established by available data and various analyses during the study. Use of sophisticate tools like mathematical modeling (MIKE 11 & MIKE21FM) and climate modeling are essential for assessing the impact of climate change and effectiveness of adaptations. The impact of Khulna Water Supply project with respect to dependable river flow and maximum river salinity and duration of high saline conditions were analyses. Model results indicate that with climate change in 2050 about 54% of the city area will be experience water

logging during a 1 in 10 year event. The study also found lack of awareness about climate change among stakeholder organizations.

IWM, 2003 studied the Sundarban Biodiversity Conservation by surface water Modelling tools. The Sundarbans Reserve Forest (SRF) is a complex ecosystem comprising the largest diversified mangrove forest of the world located in the southwest region of Bangladesh. The SRF provides habitat for a large number of flora and fauna, including various endangered species of mammals, reptiles, birds and fish. The sustainability of the Mangrove Forest and survival of the ecosystem largely depends upon the circulation of fresh water from the upland. But it is endangered with the increase of salinity and other harmful ingredients including organic and inorganic pollutants. In this study MIKE 11, a one dimensional (1D) modeling system developed by the DHI, Denmark is used by IWM. The South West Region Model (SWRM), developed at IWM has been adopted for surface water modeling of the Sundarban. The model results show that increase of flow in the western rivers may reduce the salinity in the western part (Kobadak - Betna system) of Sundarbans. Even increase in wet season flow through the said system reduces salinity in the impact zone. Fresh water flow in the Sibsa during the wet season has generally decreased after 1960. Augmentation in the Gorai flow reduces the salinity in the Pussur and Sibsa, but has got minimum effect on the channels in the west.

2.6.2 Study around the world

Gillibrand and Balls, 1998 did a one-dimensional salt intrusion model to investigate the hydrography of the Ythan estuary, a small shallow macro tidal estuary in the north-east of Scotland. The model employed in this study is the one-dimensional, transient, salt intrusion model developed by Thatcher & Harleman (1972, 1981). The Ythan is largely vertically well-mixed and the principal gradients are along the longitudinal axis of the estuary. Therefore, a one-dimensional model was sufficient to simulate the features of interest in this study without introducing the complexities and data requirements of two- and three-dimensional models. The model simulates the longitudinal distributions of water level, salinity and total oxidized nitrogen (TON) in the estuary. The principal aim of this study was to examine levels of nitrate in the estuary under various river flows, and to determine the importance of

variations in the riverine concentration to the estuary. The model showed that estuary nutrient concentrations are predominantly controlled by the river end-member. During strong rainfall events, such as occurred in February 1995, TON concentrations could be significantly raised throughout the estuary. The results suggest that the increase in riverine TON affects concentrations right to the mouth of the estuary at neap tides. During spring tides, when seawater intrudes further up the estuary, the increase in TON concentrations in the lower estuary is less.

TWDB, 2011 Compared of two hydrology datasets by TxBLEND Model, on Salinity Condition in Nueces Bay. This technical memo documented the calibration and validation of the Nueces Estuary TxBLEND hydrodynamic and salinity transport model in which inflows to Nueces Bay, via the Nueces River Inflow Point, were based on the USGS stream gage on the Nueces River at Calallen plus return flows from the Allison Wastewater Treatment Plant. The model domain, parameters, and model inputs (except for inflows) are consistent with that used for calibration and validation of the Nueces Estuary TxBLEND model. TxBLEND daily salinity output at the mid-Nueces Bay site for the period 2000-2009 was compared to observed measurements of salinity obtained from the SALT01 station maintained by the Division of Nearshore Research. The Alternate Hydrology differed from the Calibration Hydrology by an average of 21,165 acre-feet or 17%, with a minimum difference of 1,442 acre-feet in 1996 and a maximum difference of 50,359 acre-feet in 1992. Overall, these differences were not sufficient to dramatically alter the salinity predictions modeled by TxBLEND in Nueces Bay.

2.7 Summary

The literature review discussed the present river system, upstream flow condition and surface water salinity condition of the Southwest region of Bangladesh. Based on the previous study reports, Journals, research papers relevant to the salinity intrusion and salinity modelling it can observe that the effect of salinity intrusion with the changes of trans-boundary flow without any intervention at upstream boundary will be the vital issue at present scenario of Southwest region of Bangladesh. The present study will be useful for exploring the demand of upstream flow to pushdown the salinity line for safe drinking water and enhance crop production.

CHAPTER THREE

THEORY AND METHODOLOGY

3.1 General

The purpose of this chapter is to give a brief description of the related theories regarding Salinity in the Southwest region of Bangladesh. Moreover, various mathematical equations and formulas of MIKE 11 & MIKE21FM are also explained in the following section.

3.2 Theory of Salinity

3.2.1 Salinity

Salinity is a measure of the concentration of dissolved salts in water. Until recently, a common way to define salinity values has been parts per thousand (ppt), or kilograms of salt in 1,000 kilograms of water. Today, salinity is usually described in practical salinity units (psu), a more accurate but more complex definition. Nonetheless, values of salinity in ppt and psu are nearly equivalent. The average salinity of the ocean typically varies from 32 to 37 psu, but in polar regions, it may be less than 30 psu. Sodium chloride (table salt) is the most abundant of the many salts found in the ocean.

Salinity (S) conceptually ≡ grams of dissolved (<0.5 µm) inorganic ions per kg of Sea water. The average salinity of seawater is S = 35 which means that Sea water is

3.5% salt and 96.5% H2O by weight.

3.2.2 Causes of Salinity

Primary salinity is produced by natural processes such as weathering of rocks and wind and rain depositing salt over thousands of years. Salt deposits are unevenly distributed throughout Australia and the patterns and impacts of salinity vary in eastern and western parts of the country because of different topography and the age of the landscapes: salinity in the west tends to be more pervasively spread across the landscape, whereas salinity in eastern regions is more localised.

Secondary salinity has occurred with widespread land clearing and altered land use, and may take the form of "dry land salinity" or "irrigation-induced salinity". Dry land salinity occurs when deep-rooted native plants are removed or replaced with shallow- rooted plants that use less water. As a result of this vegetation imbalance, more water passes through soil to groundwater, raising the water table and bringing salt to the surface where it can be left behind as the water evaporates. Irrigation-induced salinity occurs when excess water applied to crops travels past the root zone to groundwater, raising the water table and salt to the surface. Salt may also be transported across groundwater systems.

3.2.3 Composition and Measurement of Salinity of Sea Water

Composition: Salt plays an important role in ocean circulation. In cold, polar regions, changes in salinity affect ocean density more than changes in temperature. The water salinity increases when salt is ejected into the ocean as sea ice forms. Because salt water is heavier, the density of the water increases and the water sinks. The exchange of salt between sea ice and the ocean influences ocean circulation across hundreds of kilometers. Salts constitute about 99.975% of total solids in typical seawater. The dissolved salts are almost completely ionised. The typical sea water (S=35) has the following proportions of the major ions which are shown in Table 3.1. Table 3.1: Composition of Salts in Sea Water

Ion Formula g/Kg mmol/Kg Weight% of Major Ions Sodium (Na+) 10.781 468.96 30.61% Magnesium (Mg2+) 1.284 52.83 3.69% Calcium (Ca2+) 0.4119 10.28 1.16% Potassium (K+) 0.399 10.21 1.10% Strontium (Sr2+) 0.00794 0.0906 0.04% Chloride (Cl - ) 19.353 545.88 55.04% 2- Sulfate (SO4 ) 2.712 28.23 7.68% - Bicarbonate (HCO3 ) 0.126 2.06 0.41% Bromide (Br - ) 0.067 0.844 0.19% - Borate (HBO3 ) 0.0257 0.416 0.07% Fluoride (F - ) 0.0013 0.068 0.01% Totals 11 35.169 1119.87 100% (Source: Pilson, 1998)

The "Law of Constant Relative Proportions” The ratios of the major ions in seawater - 2+ 2+ are constant - with slight exceptions for HBO3 (±<20%), Ca (±<1%) and Sr (±<2%).

(i) Thus, [Na+]/[Cl-] is the same for all seawater (Atlantic or Pacific, surface or deep). (ii) The ratio of any major ion to salinity (e.g. [Cl-]/S) is also constant, or nearly so. Ion concentrations in different waters are often normalized to salinity for comparisons.

The “Marcet Principle” determined that the major elements in seawater from six different areas are present in constant proportions to each other. This is now called the Marcet Principle (1819).

Measurement of Salinity: Sorensen's Titration Method utilize the "law of constant proportions" or "Marcet's Principle"; Chlorinity approach (Cl% ). Measure Cl % (which actually gives the sum of all halides; Cl- + F- + Br- + I-) by titration with

AgNO3 to precipitate AgCl, AgBr etc.. The titration gives the grams of Cl equivalent + - in 1 kg seawater. Ag + Cl AgCl(s).

Chlorinity (Cl) ≡ Mass of chlorine equivalent to total mass of halogen in 1 Kg. The relation between salinity and chlorinity can be expressed by Knudsen formula (3.1): Salinity (S) = 0.03+1.805 X Chlorinity (3.1)

The modern approach is to measure salinity in a Salinometer by conductance. (i) Sample conductivity is determined as a ratio to a standard of KCl in pure water (at 15oC and 1 atm pressure). Accordingly, salinity has no units using this method. (ii) "Practical salinity" is calculated for a seawater sample using a polynomial expression in terms of the conductivity ratio (R), the actual temperature (t) of the sample, and a flock of fitting constants (a, b and k):

(3.2)

3.2.4 Salinity Distribution in the Ocean

The spatial distribution of sea surface salinity in the ocean (Figure 3.1) is significantly different from the sea surface temperature one (Figure 3.2). Salinity affects seawater density, which in turn governs ocean circulation and climate. The salinity of surface seawater is controlled primarily by the balance between evaporation and precipitation. As a result the highest salinities are found in the so- called sub-tropical central gyre regions centered at about 20° to 3° North and South, where evaporation is extensive but rainfall is minimal. The highest surface salinities, other than evaporite basins, are found in the Red Sea.

Salinity variations are driven by precipitation, evaporation, runoff and ice freezing and melting. This is mainly because the surface sources of variability for temperature are different than for salinity: the ocean is indeed heated up in the tropics and cooled at high latitudes while salinity is dominantly modified by concentration-dilution related to the evaporation-precipitation- river runoff flux (E-P-R) . If it rains more than evaporated (E-P<0), for example in area of strong atmospheric convection (e.g., equatorial) or at moderate latitudes, the salinity diminishes at the ocean surface. In subtropical zones, evaporation dominates over precipitation (E-P>0) and salinity increases.

Figure 3.1: Annual mean of the sea surface salinity distribution

(Source: World Ocean Atlas, 2005)

Figure 3.2: Annual mean of the sea surface temperature

(Source: World Ocean Atlas, 2005)

The surface water salinity distributions are closely tied to E-P patterns. The Relationships between surface salinity and Evaporation minus Precipitation (E-P) patterns is shown in Figure 3.3.

Figure 3.3: Relationships between surface salinity and Evaporation minus

Precipitation (E-P) patterns (Source: World Ocean Atlas, 2005)

Because the seawater signatures of temperature and salinity are acquired by processes occurring at the air-sea interface, we can also state that the density characteristics of a parcel of seawater are determined when it is at the sea surface. Temperatures of seawater vary widely (-1 to 30oC), whereas the salinity range is small (35.0±2.0). The North Atlantic contains the warmest and saltiest water of the major oceans, the Southern Ocean (the region around Antarctica) is the coldest, and the North Pacific has the lowest average salinity. The temperature and salinity distribution of ocean water are presented in Table 3.2.

Table 3.2: Temperature and Salinity variation in the Ocean Water

Water Mass Temperature (oC) Salinity (ppt) North Atlantic Central Water 8-19 35.1-36.5 Antarctic Circumpolar Water 0-2 34.6-34.7 Antarctic Intermediate Water 3-7 33.8-34.7 North Pacific Intermediate Water 4-10 34.0-34.5 North Atlantic Deep Water 2-4 34.8-35.1 Antarctic Bottom Water -0.4 34.7 (Source: World Ocean Atlas, 2005)

3.2.5 Freshwater-Saltwater Interactions

Saltwater intrusion is a major concern commonly found in coastal aquifers around the world. Saltwater intrusion is the induced flow of seawater into freshwater aquifers primarily caused by groundwater development near the coast. Where groundwater is being pumped from aquifers that are in hydraulic connection with the sea, induced gradients may cause the migration of salt water from the sea toward a well, making the freshwater well unusable.

Because fresh water is less dense than salt water it floats on top. The boundary between salt water and fresh water is not distinct; the zone of dispersion, transition zone, or salt-water interface is brackish with salt water and fresh water mixing.

Under normal conditions fresh water flows from inland aquifers and recharge areas to coastal discharge areas to the sea. In general, groundwater flows from areas with higher groundwater levels (hydraulic head) to areas with lower groundwater

levels. This natural movement of fresh water towards the sea prevents salt water from entering freshwater coastal aquifers (Barlow, 2003).

Groundwater pumping/development can decrease the amount of fresh water flowing towards the coastal discharge areas, allowing salt water to be drawn into the fresh water zones of coastal aquifers. Therefore, the amount of fresh water stored in the aquifers is decreased (Barlow, 2003).

The first physical formulations of saltwater intrusion were made by W. Badon- Ghijben (1888, 1889) and A. Herzberg (1901), thus called the Ghyben-Herzberg relation. They derived analytical solutions to approximate the intrusion behavior, which are based on a number of assumptions that do not hold in all field cases.

The Ghyben-Herzberg Relation assumes, under hydrostatic conditions, the weight of a unit column of freshwater extending from the water table to the salt-water interface is balanced by a unit column of salt water extending from sea level to that same point on the interface. Also, for every unit of groundwater above sea level there are 40 units of fresh water below sea level. The following figure (Figure 3.4) shows the Ghyben-Herzberg relation.

Figure 3.4: The Ghyben-Herzberg relation

The Ghyben-Herzberg relation described in Equation ( 3.3):

(3.3) ( )

The thickness of the freshwater zone above sea level is represented as and that below sea level is represented as . The two thicknesses and , are related by and where is the density of freshwater and is the density of saltwater. Freshwater has a density of about 1.000 grams per cubic centimeter (g/cm3) at 20 °C, whereas that of seawater is about 1.025 g/cm3. The equation can be simplified to Equation (3.4):

(3.4)

This analysis assumes hydrostatic conditions in a homogeneous, unconfined coastal aquifer. According to this relation, if the water table in an unconfined coastal aquifer is lowered by 1 m, the salt-water interface will rise 40 m.

Generally, saltwater intrusion into coastal aquifers is caused by two mechanisms:

(i) Lateral encroachment from the ocean due to excessive water withdrawals from coastal aquifers, or (ii) Upward movement from deeper saline zones due to upconing near coastal discharge/pumping wells.

Saltwater intrusion into freshwater aquifers is also influenced by factors such as tidal fluctuations, long-term climate and sea level changes, fractures in coastal rock formations and seasonal changes in evaporation and recharge rates. Recharge rates can also be lowered in areas with increased urbanization and thus impervious surfaces. Intrusion has also occurred in areas because of water levels being lowered by the construction of drainage canals (Barlow, 2003).

Most incidents of saltwater intrusion occur in coastal regions, as has been the focus of discussion thus far, but inland areas can also be affected. Salinity issues in some regions surrounding the Rio Grande in New Mexico and Texas have been attributed to upwelling of deep-circulating groundwater, which is more saline due to natural underlying geologic formations (Doremus, 2008). The more saline groundwater is brought to the surface through pumping for irrigation and other uses. Similar occurrences have been noted in the Mississippi River Valley Alluvial Aquifer in

Arkansas, where in response to pumping, there is also upward movement of saline water from deeper formations (Reed, 2002).

3.2.6 Effect of Salinity

High concentrations of salt pose hazards for the environment as well as affecting agriculture and infrastructure and therefore, the wider economy. High levels of salinity in water and soil may cause native vegetation to become unhealthy or die and lead to a decline in biodiversity through dominance of salt-resistant species, potentially altering ecosystem structures. Reduced groundcover also makes soil more prone to erosion, which can pollute water with increased sediment, making it unsuitable for both human and animal consumption and threatening high value ecosystems and the plant and animal species they support. Despite the negative effects of salinity, some aquatic environments have adapted to a range of salt concentrations.

Increased salinity can reduce crop yields when it impairs the growth and health of salt-intolerant crops and may result in corrosion of machinery and infrastructure such as fences, roads and bridges. These impacts of salinity can be extremely costly - ranging from impaired agricultural production and additional water treatment costs to the replacement of corroded civil and agricultural infrastructure.

Salinity affected irrigation water has profound impact on crop production. Bauder, et. al., (2007) stated that the impact of irrigation water on crop production depends upon the presence of salient factors:

(i) Salinity or concentration of total dissolved salts (ii) The disproportion of sodium ion with other cations like calcium and magnesium (iii) Alkalinity or presence of excessive carbonate or bicarbonate more than the desired level (iv) pH (v) Some toxic ions like: Sodium and Chloride ion, if present more than the desired level

The Soil First Consulting, US based soil conditioners supplier, set desired level of irrigation water parameters which is presented in Table 3.3.

Table 3.3: Irrigation water parameters

SI Parameter Desired Level pH 5.5-6.0 is the ideal condition for irrigation; 1 > 7.0 can cause problems 2 Salinity or concentration of salt 1.5 dS/m 3 Alkalinity (carbonate) < 50 ppm 4 Chloride ion < 140 ppm 5 Sodium ion 0 – 50 ppm Source: Soil first Consulting (www.soilfirst.com)

3.2.7 Governing Equations of Hydrodynamics & Model

(i) Basic Hydrodynamic Equations

The basic governing equations for incompressible fluid flow are time independent and two dimensional. Equations in Cartesian coordinates (x,y) are as follows:

Continuity Equation:

(3.5)

Momentum Equation:

( ) ( ) (3.6)

( ) ( ) (3.7)

Where,

V= (u,v) is the velocity field. The first term on the RHS in Equations (3.6) & (3.7) refers to pressure forces, . P. The rest of the RHS describe viscous forces, The LHS is the momentum change that any element experiences as it moves between regions of different velocity in the flow field. This has the dimensions of a force, and is referred to as the inertia force,

(ii) MIKE 11 Hydrodynamic Equations

The equations which are solved for the flow simulations are called Saint Venant equation. These are derived from the Navier Stokes equation. Saint Venant equations are:

q Afl  qin (3.8) x t

 q 2      q  Afl  h f  gA gA I  (3.9)   fl  fl f  t x x w

Where, q Discharge

Afl Flow area qin lateral inflow h water level α momentum distribution coefficent If Flow resistance f Momentum forcing

ρw Density of Water

Equation (3.8) is called the mass equation or continuity equation and expresses conservation of mass. Equation (3.9) is called the momentum equation and expresses conservation of momentum.

(iii) MIKE 11 Advection-dispersion Equations

The advection-dispersion equation is solved numerically using an implicit finite difference scheme, which, in principle, is unconditionally stable and has negligible numerical dispersion. It solves the following advection-dispersion formula: AC QC   C  (3.6) (3.10)    AD   ... CqCAK LL t  xx  x 

Where,

C concentration D dispersion coefficient A cross-sectional area K linear decay coefficient

CL source/sink concentration qL lateral inflow

The Equation (3.10) is presented above briefly described in Reference Manual, DHI (2014).

(iv) MIKE 21 FM Hydrodynamic Equations

MIKE21 HD solves the vertically integrated equations of conservation of volume and momentum (the Saint Venant equations):

Conservation of mass Equation

 p q    0 (3.11) t x y

Conservation of momentum Equation

The momentum equation in the x-direction is given by:

2 22 p   p    pq  f  qp 1  xy h  h p         a  a      gh 2 p Cq WWxw 0  xt   yh  h  x 2 h  y   x (3.12)

The momentum equation in the y-direction is given by:

q   q2    pq   f  qp 22 1  h  ph        yx  a  a      gh 2 q Cp WWyw 0  yt   xh  h  y 2 h  x   y (3.13) Where, p and q flux in x and y directions respectively (m3/s/m) t time (s), x and y (m) are Cartesian Co-ordinate (s)

h water depth (m) g acceleration due to gravity (9.81 m2/s) ε sea surface elevation (m) 3 w and a air and water density respectively (kg /m ) Cw wind friction factor = 0.0008 + 0.000065W (Wu, 1982) W wind speed (m/s) Ω Coriolis parameter =5.2*10-5 s-1 in the Bay of Bengal Pa atmospheric pressure (kg/m/s2)

(v) MIKE 21 FM Transport Equations

The transports of temperature (T) and salinity (S) follow the general transport- diffusion equation as:

T uT vT wT   T     F   D  ˆ  STH (3.14) t x y z T z  v z  s

s us vs ws   s     F   D  ˆ  SsH (3.15) t x y z s z  v z  s

Where,

Dv vertical turbulent (eddy) diffusion coefficient Hˆ source term due to heat exchange with the atmosphere

Ts and Ss temperature and the salinity of the source F horizontal diffusion terms

The horizontal diffusion terms defined by:

           (3.16) FF sT ),(    Dh   Dh  sT ),( x    yx  y 

Where,

Dh horizontal diffusion coefficient T and s temperature and the salinity

The diffusion coefficients can be related to eddy viscosity defined by:

A Dh  (3.17)  T

Vt Dv  (3.18)  T

Where,

Dh horizontal diffusion coefficient

Dv vertical turbulent (eddy) diffusion coefficient

σT a constant Prandtl number

The Equation (3.11) to (3.18) are presented above briefly described in MIKE 21 FLOW MODEL FM , DHI (2014).

3.3 Mathematical Modelling

A mathematical model can be consist of simple to complex mathematical equations with linear and/or non-linear terms, and ordinary or partial differential equation terms. Mathematical Modelling is a simplified representation of a complex system in which the behaviour of the system is represented by a set of equations, together with logical statements, expressing relations between variables and parameters (Clarke).

Since studying the real world processes could be extremely time consuming, expensive and even dangerous, models are constructed to study pertinent system responses. Therefore, a model can be viewed as a decision making tool which provides approximate but reasonable description of the behavior of a complex system. Because of the approximations involved, a model could never be considered as an one-to-one representation of the reality. As noted by, a model will never describe every aspect of the real world and will contain aspects that have no corresponding counterpart in the real world. In general, a model is constructed in such a way that it is as simple as possible and as complex as needed to produce results of required resolution and accuracy.

Mathematical models could be classified in many different ways based on the specific technique used, nature of input and output, discretization of the decision space and so on. These classification are obviously overlapping and not mutually exclusive. Because of this, all the possible groups cannot be represented in a simple tree format. Different types of mathematical models which are mainly used in water resources are shown in Figure 3.5.

MATHEMATICAL MODELS

Optimum Search Deterministic Statistical Models

Systems Empirical Conceptual Correlation Analysis

Compotent Integrated Decision Stochastic Processes Processes Theory

Linear or Non Linear

Lumped or Distributed

Discrete or Continuous

Figure 3.5: Classification of Mathematical Modelling

Many modelling software are available in the market for the water flow and salinity modelling. The list of the software packages are MIKE, ISIS, SOBEK, HEC-RAS, RiverWare, WEAP, MODSIM, RIBASIM, WaterWare etc. All those models are capable of dealing with hydrodynamic and water quality modelling. Different types of Mathematical model used in water resources are described is below:

SaltMod is a mathematical, numerical computer program for the simulation and prediction of the salinity of soil moisture, ground and drainage water, the depth of

the water table (water table), drain discharge and leaching of salts in irrigated agricultural lands under different geohydrologic conditions, varying water management (water management) options, including the (re)use of groundwater (ground water) for irrigation by pumping from wells (conjunctive use), and several crop rotation schedules. It uses salt and water balances (water balance, budget). The model aims at sustainable land use and environmentally sound optimal water management for sustainability and can be used for the modeling (modelling) of reclamation (remediation, rehabilitation, restoration) of saline soils.

HEC-RAS is a computer program that models the hydraulics of water flow through natural rivers and other channels. The program is one-dimensional, meaning that there is no direct modeling of the hydraulic effect of cross section shape changes, bends, and other two- and three-dimensional aspects of flow. HEC-RAS finds particular commercial application in floodplain management and flood insurance studies to evaluate floodway encroachments. Some of the additional uses are: bridge and culvert design and analysis, levee studies, and channel modification studies.

IQQM is a hydrologic modelling tool developed by the Department of Land & Water Conservation, NSW (DLWC) for use in planning and evaluating water resource management policies. The main components of IQQM are: User interface shell, In-stream water quantity, In-stream water quality, Rainfall-runoff, Gated spillway operation, Graphical tools, Statistical tools, and Climate data tools.

SOBEK is a powerful modelling suite for flood modelling, flood forecasting, optimisation of drainage systems, control of irrigation systems, sewer overflow design, river morphology, salt intrusion and surface water quality. The hydrodynamic 1D/2D simulation engine is the computational core of SOBEK. SOBEK is the ideal tool for analysing the effects of urban flooding, river floods and dam break.

RiverWare is a general river and reservoir modeling tool widely used in the US due to its interpreted language for expression of multi-objective operating policies. RiverWare applications include operational scheduling and forecasting, planning, policy evaluation, and other operational analysis and decision processes (Zagona et

al., 2001). RiverWare has the capability to model: (I) Hydrology and hydrologic processes of reservoirs, river reaches, diversions, distribution canals, consumptive uses, groundwater interaction and conjunctive use (II) Hydropower production and energy uses and (III) Water rights, water ownership, and water accounting transactions.

MODSIM is a generic river basin management decision support system for analysis of long term planning, medium term management, and short term operations on desktop computers operating under MS Windows 2000/XP. The basic solver in MODSIM is a state-of-the-art network flow optimization algorithm up to two orders of magnitude faster than solvers in other river basin modeling packages and capable of simulating complex, large-scale networks.

TxBLEND is a computer model designed to simulate water levels, water circulation, and salinity condition in estuaries. The model is based on the finite-element method, employs triangular elements with linear basis functions, and simulates movements in two horizontal dimensions (hence vertically averaged). TxBLEND is an expanded version of the BLEND model developed by William Gray of Notre Dame University to which additional input routines for tides, river inflows, winds, evaporation, and salinity concentrations were added along with other utility routines to facilitate simulation runs specific to TWDB‟s needs (Gray 1987, TWDB 1999).

Now a day's technology such as mathematical modelling and respective tools are incredibly available and improved for water resources engineering and DHI is exceptional than any other organization in the world for their tools for mathematical modelling. MIKE by DHI software is the result of years of experience and dedicated development. It transforms science into practice and gives the competitive edge. MIKE by DHI truly models the world of water - from mountain streams to the ocean and from drinking water to sewage. In this study MIKE was used for One- Dimensional and Two-Dimensional Modelling. For One-Dimensional model MIKE 11 was used and for Two-Dimensional MIKE 21 FM was used.

3.3.1 MIKE 11

MIKE 11 is a software package for simulating flows, water quality and sediment transport in estuaries, rivers, irrigation channels and other water bodies. The word MIKE comes from the name of the Danish scientist Michael Barry Abbott MIKE 11 (1= unidirectional, 1 = one layer) developed by the Danish Hydraulic Institute (DHI) of Denmark. MIKE 11 is a user-friendly, fully dynamic, one-dimensional modeling tool for the detailed analysis, design, management and operation of both simple and complex river and channel systems. The model has been designed to perform detailed modeling of rivers, including special treatment of floodplains, road overtopping, culverts, gate openings and weirs. An add-on geographic information system (GIS) module provides an interface for display of river modeling results for floodplain management.

The core of MIKE 11 system consist of the hydrodynamic (HD) module that is capable of simulating unsteady flows in a network of open channels. The results of HD simulations consist of time series of water levels and discharges. The computational scheme is applicable for vertically homogeneous flow conditions extending from steep river flows to tidal influenced estuaries. The system has been used in numerous engineering studies around the world. The Basic Module, Model input data, output data and types of model used in MIKE 11 modelling system are listed in below:

(a) Basic Modules

(i) Hydrology (NAM) (ii) Hydrodynamic (HD) (iii) Advection-Dispersion (AD) (iv) Cohesive Sediment Transport (CST) (v) Non-cohesive Sediment Transport (NST) (vi) Water Quality (WQ)

(b) Model Input Data

(i) Hydrometric (WL, Q)

(ii) Hydrometeorological (Rainfall, Evaporation) (iii) Topographic (River cross-section, Land level) (iv) Concentrations (Sediments, Salinity, BOD, DO) (v) Temperature

(i) Water Level, Discharge (ii) Concentrations (Sediment, Salinity, BOD, DO) (iii) Temperature (iv) Flood-depth Map (with the help of MIKE-GIS)

(d) Types of Model Development using MIKE 11

(i) Hydrological Modelling - Rainfall-Runoff (NAM) Modelling (ii) Hydrodynamic Modelling - River and Flood Plain Modelling (iii) Sediment Transport, Salinity, Water Quality Modelling

The one-dimensional salinity model consists of three different modules: rainfall runoff (NAM), hydrodynamic (HD) and salinity model (AD).

(i) Rainfall Runoff Modelling (NAM)

Rainfall-runoff process has to be simulated in the modelling system before the hydrodynamic modelling is applied, and for this purpose, the model shall employ time series in the same format as the other models to be used in the system, and be able to exchange data with these models in a common data format. Specifically, the output files from the rainfall-runoff model should be directly readable by the selected river model. Such models should, preferably be lumped and conceptual, and should include the total hydrological cycle.

Module Name : NAM (Nedbor Afstromnings Model) Module Type : “Deterministic, lumped, conceptual” hydrological model Developer : Technical University of Denmark. Modifier : Danish Hydraulic Institute (DHI) Dimension Coverage : 0-D (dimensionless)

NAM is the abbreviation of the Danish “Nedbor-Afstromnings-Model” meaning precipitation-runoff-model. The NAM hydrological module models the rainfall- runoff processes occurring at the catchment scale. NAM forms part of the hydrological or rainfall-runoff of the MIKE-11 river modelling system. The rainfall- runoff (RR) modules can either be applied independently or used to represent one or more contributing catchments that generate lateral inflows to a river network. In this manner it is possible to treat a single catchment or a large river basin containing numerous catchments and a complex network of rivers and channels within the same modelling framework.

The NAM model can be characterized as a “DETERMINISTIC, LUMPED, CONCEPTUAL” model with moderate input data requirements. NAM can be used either for continuous hydrological modelling over a range of flows or for simulation of single events. NAM gives the runoff (Q) and water level. These data are further used in Hydrodynamic models and so on (Reference Manual, DHI, 2014).

(ii) Hydrodynamic Modelling (HD)

The MIKE11 hydrodynamic module is an implicit, finite difference model for the computation of unsteady flow in rivers. This model can describe subcritical as well as supercritical flow conditions through a numerical scheme that adapts according to the local flow conditions. The mathematical background of Hydrodynamic modelling is presented in Figure 3.6.

The MIKE 11 1D engine solves the problem of water flowing through a network of reaches, nodes, and structures. The problem is fully specified by boundary conditions at the network boundaries and initial conditions. The equations (3.8) and (3.9) which are solved for the flow simulations are called Saint Venant equation for conservation of mass and conservation of momentum.

To solve the Saint Venant equations, a finite difference scheme is implemented. Finally, it should be noted that the solution scheme does not include a detailed description of hydraulic jumps. However, flow conditions both upstream and downstream of a hydraulic jump are accurately described. The schemes used to solve

the Saint Venant equations are based on an implicit finite difference scheme developed by Abbott and Ionescu (1967).

Figure 3.6: Mathematical Background of HD Modelling

There are some assumptions regarding the Saint Venant equation which are shortcomings of MIKE 11 HD Model:

i. Flow is assumed one-dimensional and velocity is considered uniform over the cross-section. ii. Water across the cross-section is horizontal iii. The effect of boundary friction and turbulence can be accounted for through resistance law analogue to those of steady state flow. iv. Streamline curvature is small and vertical acceleration is negligible, e.g. pressure is hydrostatic. v. Average channel bed slope is small. vi. Fluid is homogeneous and incompressible, density variation over the cross- section is neglected.

A computational grid of alternating q-grid points (discharge) and h-grid points (water level) points is used as illustrated in Figure 3.7. The computational grid is automatically generated on the basis of the user requirements. q-grid points are placed midway between neighboring h-grid points and at structures, while h-grid points are located at cross sections, or at equidistant intervals in between if the distance between cross sections is greater than a user-specified maximum distance. The sign of the discharge is defined by convention as positive in the positive x- direction (increasing chainage).

The adopted numerical scheme is the implicit 6-point scheme by Abbott and Ionescu (Abbott and Ionescu, 1967), as shown in Figure 3.8. The basic idea is to use values adjacent in time and space to write down the derivative of q and h and thereby convert the two Saint Venant equations to a set of coupled implicit finite difference equations. Details are given below, first for the continuity equation and then for the momentum conservation equation.

Figure 3.7: Reach section with h- and q-grid points, on which the Saint Veannt

Equations are solved (Source: DHI, 2014)

Figure 3.8: Centered 6-point Abbot Scheme (Source: DHI, 2014)

(iii) Salinity Modelling

MIKE 11 advection-dispersion (AD) module computes the salt transport in a river system. The module is based on the one-dimensional equation of conservation of mass of a dissolved or suspended material, i.e. the advection-dispersion equation. The module requires output from the hydrodynamic module, in time and space, in terms of discharge and water level, cross-sectional area and hydraulic radius. The advection-dispersion equation is solved numerically using an implicit finite difference scheme, which, in principle, is unconditionally stable and has negligible numerical dispersion. It solves the advection-dispersion formula which is presented in Equation (3.10).

This model follows the one-dimensional (vertically and laterally integrated) equation for the conservation of mass of a substance in a solution, i.e. the one-dimensional advection-dispersion equation. The equation reflects two transport mechanisms:

(i) Advective (or convective) transport with the mean flow. (ii) Dispersive transport due to concentrations gradients.

The main assumptions underlying the advection-dispersion equation are:

(i) The considered substance is completely mixed over the cross-section, implying that a source/sink term is considered to mix instantaneously over the cross- section. (ii) The substance is conservative or subject to a first order reaction (linear decay).

The sequence of MIKE 11 salinity modelling is outlined in Figure 3.9.

MODEL HIERARCHY OF SALINITY MODEL WL/Q at U/S Boundary Q Input from NAM WL/Q at D/S Boundary

Hydrodynamic Module: MIKE 11 HD

Advection-Dispersion Transport Module: MIKE 11 AD

Water Quality Module: MIKE 11 WQ (Salinity)

Figure 3.9: Model Hierarchy of Salinity Model

3.3.2 MIKE 21FM

MIKE 21 FM model is based on a two-dimensional flexible mesh approach and it has been developed for applications within oceanographic, coastal and estuarine environments. The modelling system can also be used in overland flooding studies. The system is based on the numerical solution of the two-dimensional incompressible Reynolds averaged Navier-Stokes equations with assumptions of Boussinesq and of hydrostatic pressure. Thus the model consists of continuity, momentum, temperature, salinity and density equations. The spatial discretization of the equations is performed using a cell-centered finite volume method. The spatial domain is discretized by subdivision of the continuum into non-overlapping

elements. In the horizontal plane an unstructured grid of triangular or quadrilateral elements is applied.

The model simulates the water level variations and flows in response to a variety of forcing functions. The water levels and flows are resolved on a rectangular grid covering the area of interest when provided with the bathymetry, bed resistance coefficients, wind field, and hydrographic boundary conditions. The modeling tool is capable of handling convective and cross momentum, bottom shear stress, wind shear stress at the surface, barometric pressure gradients and Coriolis forces. Momentum dispersion is handled using the Smagorinsky formulation. Different sources and sinks (mass and momentum) can be described in the model and the model is also capable of handling flooding and drying (MIKE 21 FLOW MODEL FM , DHI, 2014). The governing equations used in MIKE21 FM in solving hydraulic problems in coastal areas are presented in Equation (3.11) to (3.13). The governing equations of general transport-diffusions are Equation (3.14) and (3.15).

3.4 Methodology

The approach of the study is to investigate the hydrodynamic condition which will reduce the salinity level at downstream portion of Southwest region of Bangladesh.

(i) Data collection on precipitation, evaporation, water level, discharge, salinity and river cross-section has been collected from different sources to understand the baseline condition of hydrology and salinity of Southwest region. Under this study the literature and theories on salinity has been reviewed briefly.

(ii) NAM Rainfall Runoff Model will be applied to estimate the runoff generated from rainfall occurring in the catchment. Rainfall data forms the basic input to the Rainfall Runoff Model (NAM). Potential evapo-transpiration data is also an input to the Rainfall Runoff Model (NAM). Due to non-availability of potential evapo-transpiration data from any direct measurement, pan evaporation data collected from BWDB stations are used in the model after some processing. The rainfall runoff model (NAM) contains 44 catchments. The catchment SW- 27 represents the entire Sundarban area. Rainfall-runoff process has to be

simulated in the modelling system before the hydrodynamic modelling is applied.

(iii) The principal modelling tools used in the study was the one-dimensional and two-dimensional (1-D and 2-D) modelling systems, MIKE11 and MIKE 21FM of DHI. The existing calibrated (Year 2012) and validated (Year 2011) Southwest Region Model (SWRM) developed at Institute of Water Modelling (IWM) will be applied to assess the base line salinity condition of the area. The modelling systems have been applied extensively in Bangladesh over the last decade. The salinity developments of southwest region of Bangladesh take place during the dry period. Therefore, all model simulations (calibration/verification, applications and sensitivity tests under 2D & 1D modelling) mainly concentrated for a six-month dry period from January to June.

(iv) The existing one-dimensional Southwest Regional Model (SWRM) developed by MIKE 11 will be extended and updated with the dredged X-sections to restore the flow thorough Ganges connected River of Southwest Regional rivers.

(v) The two-dimensional HD model of Bay of Bengal model (BoB) developed by MIKE 21FM covers the Meghna estuary and Bay of Bengal area. The calibrated and validated Bay of Bengal model will be used to generate the d/s boundary condition (Salinity) for southwest regional model. The developed SWRM under this study contains total 39 boundaries, of which 27 are upstream and 12 are downstream boundaries. The 12 downstream salinity boundary for 1-D model will be used from the output of Bay of Bengal salinity model simulation.

(vi) Finally the Salinity model will be simulated with the scenarios which will lead to increase in flow of rivers and channels dependent on the Ganges River thus will reduce the salinity intrusion of Southwest region of Bangladesh. A step by step procedure to assess the Scenario & impact of Salinity Intrusion of Southwest Region of Bangladesh is furnished in the Figure 3.10.

Data Collection Literature (Water Level, Flow, Salinity Review & Cross Section)

Simulation of existing calibrated and validated Hydrodynamic & Salinity Model (SWRM and BoBM) by MIKE 11 and MIKE 21 FM

(Scenario-1)

Development of the extended SWRM Model for the study

Development of scenario with worst flow condition & with different Ganges flow condition

Simulation of South Western Regional Salinity Model for the flow Scenarios

Impact assessment and Comparison of salinity intrusion for different flow scenarios

Figure 3.10: Flow diagram of Methodology to assess the Scenario & impact of

Salinity Intrusion of Southwest Region of Bangladesh

3.5 Summary

This chapter deals with the theories involve in this study such as Composition of Seawater, units of salinity, Impact of salinity on crop production & irrigation water standard parameter, Basic equations of hydrodynamics for river, Saint Venant equations and governing equations of Advection-dispersion formula for 1-D and 2-D mathematical modelling of salinity. A brief description of the methodology and approaches are also provided in this chapter.

CHAPTER FOUR

STUDY AREA & MATHEMATICAL MODEL SETUP

4.1 General

Various kinds of data have been collected in order to update & development of Hydrodynamic and Salinity Model. Different data such as salinity level, water level, river cross-section and discharge have been collected from different sources. Those data have been used to develop and calibrate the mathematical model. This chapter describes a brief discussion about the collected data and mathematical model setup.

4.2 Study Area

The Southwest region has a complex drainage pattern of rivers, cross channels, estuarine and ephemeral water courses driven by heavy seasonal rainfall and inflow of sediment from the Ganges. This region comprises an area of about 41,500 sq. km. (27% of total area of Bangladesh), 10% of which is Sundarbans (Total area 601,700 ha out of which app. 411,200 ha is land, remaining area is rivers, canals and creeks) and 13% is surface water and beels (IWM, 2014). Ground elevation varies from (+) 6.00 to (+) 14.00 m PWD. The Indian border bound it on the west, by the Ganges- Padma and Lower Meghna Rivers in the east and north and by the Bay of Bengal to the south. Mean annual rainfall ranging from 1500 mm in the north-west to 2900 mm in the south-east. Population in this region is about 28.6 million (2001 Census), which is 23% of the country‟s total 123 million. Annual population growth rate is 1.1% (vis a vis 1.5% nationally), density of population is 770 people/sq. km. Whereas national level density is 835 people/sq. km. and total no. house hold is 5.95 million. It is expected that by the year 2020 population will be 34 million.

4.2.1 River Systems of Southwest region of Bangladesh

Water resources of the area originate from three separate, but largely interlinked sources such as surface water flows from Transboundary Rivers, surface water originating from rainfall over the area and groundwater. Boundary rivers the Ganges, the Padma and the Lower Meghna feed a number of regional rivers and channels. The annual average flow of the Ganges in the pre-Farakka period was 11,690 m3/sec,

during post-Farakka period (1975-88) flow declined to 11,300 m3/sec and further (1989-96) to 9,500 m3/sec. After Ganges water sharing treaty with India in 1996, mean flow for March increased to 1183 m3/sec from 526 m3/sec as was during Pre- Treaty period. The flows in the major distributaries of the Ganges, the Gorai- Modhumati River area dependent on the Ganges discharge, morphological conditions of the Gorai itself and offtake. The annual flow volume of the Gorai during the post- Farakka period has shown a decreasing trend, not entirely due to the closing of the offtake during the dry season, but wet season flows have been decreasing also due to the changing offtake morphology. From the 1988 dry season, the river began to close completely, with the only flows taking place from May-June to December. Emergency capital dredging works in the Gorai commenced in 1998 and continued for two years. The only perennial stream that supplies fresh water into the Southwest area is the Arial Khan. It discharges water into the Modhumati-Nabaganga system through the Madaripur beel route with an average dry season fresh water flow of about 20 m3/sec. A further contribution from the Arial Khan enters the Swarupkati and Balesware estuary system.

The Kobadak is an important river with respect to the tidal area, providing freshwater to maintain the saline front at bay. Indications are that this river is deteriorated rapidly with heavy siltation occurring during the dry season that is not fully flushed out by the monsoon flows. The Kumar (Magura) is the central drainage chennel in the Ganges-Kobadak project.

Within and around the Sundarbans there are numerous rivers some of which are deteriorating due to siltation. Rivers in the north-eastern part of Sundarbans such as Bhola Nadi, Khorma, Arua, Sela Gang, Mongla Nala are reported to getting excessive siltation causing damage to flora and fauna, navigational route and overall mangrove forest eco system due to reduction of brackish water flow. Bhola Nadi that is surrounded the north-eastern corner of the Sundarbans under Bagerhat district, once was completely silted up, river bed level and adjacent ground level was same, in 1997 the river bed was excavated by BWDB under FFW program to restore flow but it is again getting silted up rapidly. The Pussur-Sibsa estuary receives a major part of its freshwater discharge from the Gorai-Modhumati during the wet season.

Condition of the Pussur is deteriorating slowly due to siltation on the other hand the Sibsa is becoming gradually deeper.

The rivers Mathabhanga and Hisna conveyed dry season flow to the Southwest area in the past but their offtakes remained dry during dry season for a few years. The mouth of the Mathabhanga is just west of Ramkrishnapur, upper reach of which flows almost along the international boundary and heavy siltation takes place on the riverbed. The Hisna River takes off from the Ganges from just east of Ramkrishnapur and is now disconnected totally. The main offtake of the Hisna deteriorated in the same manner as that of the Mathabanga. Options for restoration of these two rivers can be considered in addition to the Gorai. The Kobadak and Bhairab Rivers will be benefited if dry season flows in the Mathabhanga and Hisna Rivers can be restored.

Lack of safe drinking water has been identified as the number one issue in the daily life of the SW region of Bangladesh. In the recent years, groundwater based water supply in coastal area is suffering from a number of major problems mainly arsenic contamination, lowering of the water table, salinity and non-availability of suitable aquifers (PDO-ICZMP,2004). In this way the source of surface water becoming the important issue for the safe drinking water and also for crop production in the SW region. The study area map is presented in Figure 4.1.

4.2.2 Meteorology

For baseline description of this study, meteorological data have been collected and analysed from Ganges Barrage study project considering specific parameters that are directly related to water resources such as rainfall, evapo-transpiration, temperature, sunshine hours, humidity and wind speed. Meteorological data for different hydrological regions of the SW and SC have been collected from the National Water Resources Database (NWRD) of WARPO, which contains long series temporal data showing daily values for meteorological stations maintained by the Bangladesh Meteorological Department (BMD).

Figure 4.1: Map Showing Study Area

(i) Rainfall

In Bangladesh, almost 80% of the annual rainfall is observed during the wet season. Monthly rainfall of the SW & SC has been sub-grouped by the dry season and the wet season statistics of the year 2008 as shown in Table 4.1. November to January are the driest months with almost negligible rainfall and July-September are the wettest months with the highest rainfall intensity. The SW region receives 1020.5 mm of annual rainfall of which 800.0 mm occurs during the wet season.

Table 4.1: Monthly Mean Rainfall Distribution in the Southwest Hydrological Regions for 2008

Hydrological Dry season Wet Season Annual Region N D J F M A M J J A S O Dry Wet Total

SC 0 0 32.5 44.5 27 52 151 285 389.5 390 161 17 306.5 1242.5 1549.0

SW 11 4.5 32 38 11 80 44 158 190 288 156 8 220.5 800 1020.5 Mean Rainfall 5.5 2.25 32.2 41.2 19 66 97.2 221.5 289.7 339 158.5 12.5 263.5 1021.2 1284.7 (mm)

(Source: NWRD, 2008.)

Annual rainfall distribution in the Southwest Regional area for the hydrologic year 2008-09 is presented in Figure 4.2 which shows that mean rainfall is 1284.75 mm.

Figure 4.2: Annual Rainfall Distribution in the Southwest Hydrological Regions of

Bangladesh (Source: BMD, 2008)

(ii) Temperature

During the year 2008 annual maximum and minimum temperature for the Southwest has been found to be 340C and 180C respectively. Mean maximum dry and wet seasonal temperatures have been found to be 330C and 360C in the SW, while mean 0 minimum temperature for the same season has been recorded at 13.35 C in the SW.

(iii) Humidity

In 2008, the maximum average humidity was found to be 83% in the SC region and the minimum was found to be 79% in the SW region. Wet season is more humid than the dry season, and it varies from 79% to 84% in the SW and 80% to 87 % in the SC. Maximum and minimum values of humidity have been found to be 100% in the month of July at the Bhola station in the SC region and 51% in March at the Satkhira station in the SW region.

(iv) Wind speed

In 2008, mean wind speed was found maximum at 100.6 km/ day for the SC region during the dry season and 172.0 km/day for the SW region during the wet season. The pre-monsoon season is the transition period when the northerly or northwesterly winds of the winter season gradually change to the southerly or southwesterly winds of the summer monsoon (June-September). During the early part of the rainy season, the winds are neither strong nor persistent. During the early summer month (March/April) and late monsoon (October/November), a series of thunderstorms occur especially during the evening hours with great intensity.

(v) Sunshine hours

Sunshine data is available as hour/day. Mean annual, dry and wet seasonal daily sunshine hour statistics for the Southwest was found maximum at 7.0 hr/day in the SC and 5.8 hr/day in the SW region. Region-wise climatic variability is shown in Table 4.2.

Table 4.2: Meteorological Parameters in Southwest Hydrological Region

Maximum Minimum Wind Speed Sunshine Humidity Temperature Temperature Hour Hydrologic al Region (°C) (Km/day) (hr/day) (%) A D W A D W A D W A D W A D W

SW 34 33.0 36.0 18 13.35 22.39 127.7 83.0 172.0 5.0 5.8 4.2 81 79 84

SC 33 32.4 34.6 19 14.45 22.92 124.7 100.6 148.8 6.05 7.0 5.1 83 80 87

(Source: BMD, 2008) (A= Annual; D= Dry Season; W= Wet Season.)

(vi) Evapo-transpiration

The Food and Agricultural Organisation (FAO) recommend the Penman-Monteith method for the estimation of evapo-transpiration (ET0). Hydrologic region wise mean annual, dry and wet seasonal evapo-transpiration in the Southwest area is presented in Table 4.3.

Table 4.3: Evapo-transpiration (mm/day) for each Southwest Hydrological Region for 2008

Annual ETo Dry Seasonal ETo Wet Seasonal ETo Hydrologic Region Avg Min Max Avg Min Max Avg Min Max

SW 3.87 3.35 4.50 3.60 2.85 4.70 4.10 3.64 4.60

SC 3.44 3.07 3.70 3.20 2.82 3.60 3.70 3.30 4.00

(Source: BMD, 2008) Evapo-transpiration was estimated to be 3.87 mm/day in 2008-09 for the SW region with mean ET0 of the dry and the wet season varying from 3.60 to 4.10 mm/day. The maximum ET0 was recorded at 4.70 mm/day in the SW region in the dry period.

4.2.3 Hydrology

Hydrology in Bangladesh is dominated by the monsoon. Based on this, the Hydrological Seasons are defined as Monsoon Season: June-September, Post Monsoon Season: October-November, Winter Season: December-February and Summer or Pre-monsoon Season: March-May.

Bangla calendar however uses six seasons: These are Summer(Grishma kal) season – months of Baishakh and Jaistha, Rainy (Barsha kal) season – months of Ashar and Sraban, Autumn (Sarat kal) – months of Bhadra and Aswin, Late Autumn (Hemanta kal) – months of Kartik and Agrahayan, Winter (Sheet kal) months of Paush and Magh, Spring (Basanta kal) – months of Falgoon and Chaitra. Bangla calendar months are based on agricultural practices in the country.

In the context of the Ganges Water Treaty 1996, two defined seasons have received prominence. These are the Dry Season: (January – May) and the Critical period: (March 11 to May 10). Obviously, the period from June to December is treated as the Wet Season.

The Water Resources Planning Organisation (WARPO) of MoWR has delineated eight hydrological regions in Bangladesh by natural boundaries of principal river systems, while the hilly areas of the eastern part of the country also forms one hydrological unit. Based on these principles, eight regions of the country are: (i) Northwest (NW), (ii) the North Central (NC), (iii) the Northeast (NE), (iv) the Southeast (SE), (v) the South Central (SC), (vi) the Southwest (SW), (vii) the Eastern Hills (EH), and (viii) the active floodplains and char lands of the Main Rivers and Estuaries (RE). The hydrological region of Bangladesh is shown in the following figure (Figure 4.3).

4.2.4 Upstream Withdrawl and Impacts on Bangladesh

Farakka Barrage is a barrage across the Ganges River, located in the Indian state of West Bengal, roughly 16.5 kilometers (10.3 mi) from the border with Bangladesh near Chapai Nawabganj District. Construction was started in 1961 and completed in 1975. Operations began on April 21, 1975. The barrage is about 2,240 meters (7,350 ft) long. The feeder canal from the barrage to the Bhagirathi-Hooghly River is about 25 miles (40 km) long. The barrage was constructed by Hindustan Construction Company. It has 123 gates and it serves water to the Farakka Super Thermal Power Station. There are also sixty canals which can divert the water to other destinations.

Figure 4.3: The Hydrlogical Regions of Bangladesh

The barrage was built to divert up to 40,000 cu ft/s (1,100 m3/s) of water from the Ganges River into the Hooghly River during the dry season, from January to June, in order to flush out the accumulating silt which in the 1950s and 1960s was a problem at the Port of Kolkata (Calcutta) on the Hooghly River. The Hooghly River divides Murshidabad and Malda districts of West Bengal.

During the post-Farakka period major changes have occurred in the dry season flow, especially between January and May. During the pre-Farakka period the minimum monthly average flow was 1,500 m3/s. The recorded minimum monthly average flow was 170 m3/s in April, 1997. However, the post-Farakka flood flows were of the same as the pre-Farakka flood flows. Historical Discharge and yearly minimum flow the Ganges River at the Hardinge Bridge are shown in Figure 4.4 and Figure 4.5. Maximum, Minimum and Average Water Level of the Ganges at Hardinge Bridge from 1909 to 2009 are shown in Figure 4.6.

80000 Year: 1932 to 2009 70000

60000 ) s / 50000 3 m (

e g r 40000 a h c s i

D 30000

20000

10000

0 Nov-32 Jun-42 Jan-52 Aug-61 Mar-71 Oct-80 May-90 Dec-99 Jul-09 Year

Figure 4.4: Ganges River Discharge at Hardinge Bridge (Source: BWDB, 2009)

Year: 1934 to 2012

Figure 4.5: Observed minimum flow of the Ganges River at Hardinge Bridge

(Source: BWDB, 2012)

16 maximum minimum Average

12 ) WD P , 10 m ( l e v e L r 8 e t a W

4 1909 1919 1929 1939 1949 1959 1969 1979 1989 1999 2009 Year

Figure 4.6: Maximum, Minimum and Average Water Level of the Ganges at

Hardinge Bridge from 1909 to 2009 (Source: BWDB, 2009)

Discharge of the Ganges in Bangladesh is measured at the Hardinge Bridge. Discharge data at this station is available since 1934 up to 2010. The Farakka Barrage was put into operation in 1975. Data for the period 1934 to 1974 is considered as the historical flow. The flow during the period 1975-1996 was controlled by various short period agreements or absence of any agreement between India and Bangladesh. Flow during the period 1997 and onward is controlled by the GWT, 1996. Summary of average monthly flows for these three periods are presented in Table 4.4 and Figure 4.7. The dry season (Jan-May) flow of the river is presented separately in Figure 4.8.

From the figure (Figure 4.8) and Table (Table 4.4) it is clear that during the dry season particularly in March and April, the flow has been reduced to about less than half of its pre-Farakka period.

Table 4.4: Historical Flow (m3/s) of the Ganges at Hardinge Bridge

1934-74 (Pre-Farakka Period)

Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Max 5910 5040 4080 3260 4330 20300 50800 72000 73200 57800 26500 11200

Average 3088 2676 2305 2034 2151 4381 17817 38201 35990 17798 7076 4188

Min 1900 1800 1390 1190 1190 1680 3620 13600 12500 5190 3480 2410

1975-96 (Post-Farakka Period)

Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Max 3800 2690 2240 2180 5990 17300 46500 69200 75800 50300 23600 6570

Average 1693 1254 975 932 1335 3433 18664 37290 37718 15823 5401 2745

Min 842 374 261 267 393 622 2620 11600 7300 4190 2010 1520

1997-2010 (Post-GWT, 1996 Period)

Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Max 5950 2571 1606 1632 3539 22126 45384 67370 74278 52046 17753 28467

Average 1939 1346 945 934 1313 4101 21674 33793 31667 19076 7554 4824

Min 1040 739 212 330 581 713 1911 14034 16488 6340 854 1705 (Source: JRC (B) and Processing & FFWC, BWDB compiled by the Consultants)

Daily Flow Regime - Ganges at Hardinge Bridge 100000 1934-1974 Average Daily Flow 1975-1996 Average Daily Flow 1997-2010 Average Daily Flow

10000

Discharge in m3/s Discharge 1000

100 Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Date

Figure 4.7: Superimposed Daily Average Flow (m3/s) curve of the Ganges for the

three periods (1934-74, 1975-96, 1997-2010) (Source: BWDB, 2010)

Daily Flow Regime - Ganges at Hardinge Bridge 10000 1934-1974 Average Daily Flow 1975-1996 Average Daily Flow 1997-2010 Average Daily Flow

1000 Discharge in m3/s Discharge

100 Dec Jan Feb Mar Apr May Jun Jul Date

Figure 4.8: Superimposed Daily Average Flow (m3/s) curve (dry season) for the

Three periods (1934-74, 1975-96, 1997-2010) (Source: BWDB, 2010)

Impact on Bangladesh

In 1974 India built a barrage on the Ganges at Farakka in order to divert water for its own use. The water is diverted to the Hoogley River via a 26-mile long feeder canal. During the last 19 years only between 1977 and 1982 has there been an agreement between Bangladesh and India to address water sharing during the summer months (Singh, 1987b). The unilateral and disproportionate diversion of the Ganges since that time has caused a dangerous reduction in the amount of sediment and water flow of the Ganges in Bangladesh. Now Bangladesh's delta receives less sediment and inadequate water flow for navigation and irrigation during the summer months. The lack of a workable management plan for water allocation to Bangladesh has created a situation where irrigation of crops and navigation are both impossible during the summer months. The summer of 1993 was characterized by almost completely dry riverbeds across the country as reported by all major newspapers in Bangladesh. Groundwater also dropped below the level of existing pumping capacity. Such conditions lead to significant decreases in food production and curtailment of industrial activities. As a result of a lack of adequate freshwater inflow, the coastal rivers experienced saltwater intrusion 100 miles farther inland than normal during the summer months, affecting drinking water in these areas (Zaman, 1983).

The reduction in sediment supply to Bangladesh caused by the dam at Farakka has curtailed delta growth and has led to increased coastal erosion. At the present time the rates of sediment accumulation in the coastal areas are not sufficient to keep pace with the rate of relative sea level rise in the Bay of Bengal (Khalequzzaman, 1989). While the average sedimentation rates range between 4 and 5mm/year, the rate of sea level rise is 7mm/year (Emery and Aubrey, 1989). In addition, the reduced summer flows allow sediment to be deposited on the riverbeds downstream of the dam, resulting in decreased water carrying capacity during the rainy seasons (Alexander, 1989). This reduction in carrying capacity due to river bed aggradation has increased the frequency of severe floods over the last decade, causing enormous property damage and loss of life. If the amount of sediment influx in the coastal areas is further reduced by damming the rivers, then a relative sea level rise in the Bay of Bengal by 1 meter will severely curtail the delta growth, resulting submergence of

about one-third of Bangladesh (Broadus and others, 1986; Milliman and others, 1989).

The unilateral diversion of the Ganges water by India at Farakka Barrage has caused a series of adverse environmental and ecological problems in Bangladesh. A long- term solution to water sharing problems between Bangladesh and India is imperative. Without regional cooperation between the co-riparian nations, any major inter-basin development activity is almost impossible.

4.2.5 Activities in GDA

(a) Gorai River Restoration Project, Phase-II

The Gorai is the major right bank offtake from the River Ganges within Bangladesh and the principal source of dry season freshwater flow to the South West Region that covers approximately 41,500 km2 (27 percent of the total area of the country) and has a population of about 26 million (23 percent of the country total). For several reasons many spill channels including the Gorai to the area are now disconnected from the parent river, the Ganges. For the last 20 years or so, the dry season flow (November - May) in the Gorai River has been decreasing, and ceased to flow in one period of time (1988-98). The environmental impact of this decrease is very serious in terms of increased salt-water intrusion in the coastal area, around Khulna, and on the World‟s largest mangrove forest area, the Sundarbans (IWM, 2014).

To get the river flowing again, a number of solutions were considered. One of the better solutions was found to be dredging a deeper channel at the bifurcation where the Gorai splits off from the Ganges including 30 km downstream. After implementation of three dredging seasons, 1998, 1999 and 2000, the dredging increased the water flow in the Gorai River, restoring the fish population and allowing year-round navigation. Meanwhile no step was taken as maintenance dredging, which could keep the Gorai sustainable and allow substantial amount of flow during the dry season. In view of such observation, Bangladesh Water Development Board (BWDB) has again taken up steps to carry out dredging from the off-take, for a length of 30 km of the river. The Government approved a DPP under

the name “Gorai River Restoration Project Phase-II (GRRP-II)” for this purpose whereas during 1999 to 2000, GRRP, Phase I was conducted. After a long gap and considering the morphological changes that already took place during this period, present study has been formulated, and is being conducted by IWM with application of its past working experience and involvement in the Phase I.

In order to restore the Gorai offtake and to ensure substantial amount of fresh water flow towards the south-west region of Bangladesh, Bangladesh Water Development Board (BWDB) has taken up steps to dredge the Gorai from its offtake up to 30km downstream under a project titled Gorai River Restoration Project, Phase-II (GRRP- II). BWDB accomplished the capital dredging during the dry season of 2010-11, then subsequent maintenance dredging during the dry seasons of 2011-12 and 2012-13. It was decided that BWDB should continue maintenance dredging in the Gorai during the dry season of 2013-14.

The impact of the dredging in terms of enhancement of flow and conveyance area during lean flow period, the offtake condition and navigability during the last three years have been analysed. It has been observed that after three years, dredging has improved the overall situation. During this phase (GRRP Phase-II), dredging has been done in four stages. Features of these dredging are given in Table 4.5.

Table 4.5: Features of the dredging 2010-2013

Dredging Period of SI Type of Dredging Length volume in lakh dredging (m3) 11/11/2009 1 Priority Dredging 3km 430 m 9.03 to 7/6/2009 2/12/2010 2 Capital Dredging 25.51 km 78.95 to 3/6/2011 3/11/2011 3 1st year maintenance Dredging 18.26 km 54.36 to 21/5/2012 30/11/2012 4 2nd year maintenance dredging 11.95 km 50.32 to 25/6/2013 (Source: IWM, 2014)

Tidal prism was calculated at five different locations in Rupsa-Pussur-Sibsa River system and one location in Baleswar River system. It is evident from the calculation that minor change in tidal prism is observed in Rupsa and upstream part of Pussur River system whereas almost no change is observed in Sibsa, Baleswar and downstream part of Pussur River system. It indicates that Gorai river flow does not have significant impact on Baleswar, Sibsa and downstream part of Pussur River system.

Salinity data was collected both for 2011 and 2012 dry season to assess the change in salinity level due to dredging in Gorai River. Salinity data was collected at different location of Pussur-Sibsa and Baleswar River system both for before and after dredging of Gorai River. Salinity data from eight different stations has been analysed. Significant changes have been taken place in the Pussur-Sibsa River system along with its upstream rivers and the most significant change has been occurred at Bardia. It is found that the salinity level reduces significantly from 10 ppt to zero ppt at Bardia whereas at Khulna it reduces from 17 ppt to 8 ppt due to increase of flow in Gorai river which is helpful for farmer, fishermen and community consumers in that area. The decreasing trend is also found in Shalta, Badurgasa, Mongla and Nalian. But no change has been observed in Baleswar River system.

(b) Ganges Water Treaty

The Ganges Water Treaty (GWT) was signed between the Government of Bangladesh and India on 12 December, 1996 for the sharing of the Ganges waters at Farakka. The GWT 1996 has the following main features

The sharing between India and Bangladesh of the Ganga/Ganges water at Farakka by ten-day periods from 1 January to 31 May every year will be as follows:

Availability at Farakka Share of India Share of Bangladesh 70000 cusec or less 50 % 50 % 70000 – 75000 cusec Balance of flow 35000 cusecs 75000 cusecs or more 40000 cusecs Balance of flow

Subject to the condition that India and Bangladesh each shall receive guaranteed 35000 cusec of water in alternate three ten day periods during the period 11 March to 10 May. Every effort would be made by the upper riparian to protect flow of water at Farakka as in the 40 years average availability. In the event flow at Farakka falls below 50000 cusecs in any ten days period, the two governments will enter into immediate consultations to make adjustments on an emergency basis, in accordance with the principles of equity, fair play and no harm to either party. The treaty ensured minimum flow in the Ganges River in Bangladesh during dry season which improved the salinity condition in the south-west of Bangladesh.

The propose Ganges barrage (GBR) would be one of the largest barrage of the world. Pansha has been selected as the suitable location for barrage construction. Pangsha is situated on the River Ganges 37 Km upstream of the confluence of the two large rivers, the Ganges and the Jamuna at Goalundo in Bangladesh. Size of the water way, arrangements and number of gates, sizing of spillways and under sluices and the created reservoir due to barrage operation has been fixed during the feasibility study phase of the Ganges barrage study.

Size of the waterway and sill level within the afflux limitation is inter related in terms of satisfying the discharge requirements. The Lacy‟s regime parameter is a standard basis utilized historically for the design in the case of the majority of major barrages contracted in the alluvial basins of the Indus and Ganges River. The Lacey regime parameter has been estimated for the design flood of 80000 cumec and also for the dominant discharge which is around 43000 cumec. For the design flood, 1341 m width is needed and for dominant flow 985 m waterway is needed.

The proposed Ganges Barrage at Pangsha includes; 78 spillways, 18 under sluices, fish passes, one navigation lock and a hydropower plant. With all these facilities the total length of the barrage is about 2100 m. 12.5 m PWD pond level has been fixed based on the afflux and diversion flow requirements criteria. The sill level of the under sluices has been taken as 1.5 m lower that the spillways sill level of the barrage to help maintain a clear waterway.

The sill levels are: Spillway Sill level + 0.00 mPWD Under Sluice Sill level -1.50 m PWD

18 m width with 13.1 m height for spillway gates and 18m wide with 14.9 m height for undersluice gates has been adopted. Radial gates have been selected for the operation of barrage.

The Ganges barrage at Pangsha will create a reservoir of more than 100 Km long in the upstream with a pond level of 12.5 mPWD. With a pond level of 12.5 m PWD the Ganges Barrage reservoir will be a capacity of about 2890 million m3 of water. Figure 4.9 shows the layout of proposed Ganges Barrage.

Figure 4.9: Layout of the proposed Ganges Barrage (Source: BWDB, 2012)

Diversions System & River Links of Ganges Distributaries

(i) Hisna Diversion System

Under Hisna distribution system includes the following river network systems;  Hisna-Mathabhanga-Nabaganga-Chitra-Fatki-Chitra-Afra system  Mathabhanga-Upper Bhairab-Kobadak-Betna system  Kobadak-Harihar-Gengrail system  Upper Bhairab-Bhairab-Mukteswari-Hari system

In the Ganges Barrage study report it is seen that flows in the above river systems will be augmented by Ganges flow diversion from the barrage reservoir under gravity flow condition. The situation of Hisna River has been so deteriorated that whole offtake area is silted up. At present there is no linkage with the Ganges as the main channel has already been shifted further to north-eastern side. A new link channel of 4.85 km long is required for making connection in between Ganges and existing Hisna in Daulatpur Upazila of Kushtia district.

(ii) Gorai Diversion System

Gorai diversion system includes the following river network systems;  Gorai-Modhumati-Katakhali-Atharabanki-Poylahara-Bishnu system  Modhumati-Nabaganga-Rupsa-Pussur system  Modhumati-Baleswar system

Flows in above river systems will be augmented by the diversion of Ganges flow through the main distributory Gorai River. Major flow diversion shall be done through the Gorai for irrigation, navigation, salinity and sedimentation control in the Gorai and its distribution system.

(iii) Chandana Diversion System

The following river systems are in the Chandana diversion system:  Chandana-Barasia-Modhumati River system  Chandan-Kumar-Kumar Nadi-Modhumati system

A new link channel of 15.0 km long is necessary for connectivity of the Chandana with the barrage reservoir. The Ganges flow diversion will be done under gravity flow condition through an off-take structure for controlling the inflows. Kumar River has no more linkage with the Ganges. This will be augmented by the Ganges flow from Chandana through existing Naopara Khal, Horai River, Rajapur Khal (total length 38 km). Major improvement of these rivers will be required to pass the design discharges.

Almost all the rivers of Southwest region have no dry season flows due to severe morphological changes of Ganges at the offtakes of Hisna, Gorai and Chandona. Many rivers have already been silted up. Some links in between the internal river system will be required to establish linkages for conveying Ganges flow in the area. The border Rivers Mathabhanga and Kobadak need diversion for conveying the desired flows downstream. Under GBP eight river links in the distribution systems are proposed. Details of these linkage are shown in Table 4.6.

Table 4.6: Details of river links in the distribution system

Link Name of Links/River Diversion Location (Upazila) Length (km) 1. Ganges to Hisna Daulatpur 4.00 2. Mathabhanga to Naboganga Chuadanga Sadar 8.20 3. Naboganga to Chitra Chuadanga Sadar 3.30 4. Mathabhanga to U. Bhairab Darsana Chuadanga 4.70 5. Kobadak Diversion Jhikorgachha 5.80 6. Kobadak to Betna Jhikorgachha 3.00 7. Ganges to Chandona Pangsha 15.00 8. Chandona to Kumar Baliakandi, Rajbari 31.70 75.70 (Source: BWDB, 2012)

The Schematic diagrams of diversion systems are shown in Figure 4.10. The alignments of the above new links/diversions have been shown in Figure 4.11. From the Table 4.6 it is found that Hisna diversion system requires six (1-6) links and Chandona system requires two links (7 & 8).

Gorai Diversion system

Hisna Diversion system

Chandana Diversion system

Figure 4.10: Schematic Diagram of Diversion System (Source: BWDB, 2012)

Figure 4.11: Link Channel of the Ganges Diversion System (Source: BWDB, 2012)

4.2.6 Flow Augmentation without Structural Intervention

A restoration of the river system will lead to the restoration of the aqua economy i.e. water related economy such as trading, fishery, agriculture & river transportation. In view of above, planning might be centered for enhancing the up-stream flow augmentation during dry season, which will be beneficial for agricultures, fisheries and flushing of incoming sediment by high tide.

Starting from the upstream, the important distributaries of the Ganges-Padma-Lower Meghna system are: Baral (takes off from the left bank at Charghat in Natore district), Mathabhanga (takes off from the right bank at Daulatpur), Hisna, (on the right bank at Majhdier, offtake needs a new cut for about 4 km), Gorai-Madhumati (the most important distributary takes off from the right bank at Kushtia), Chandana- Barasia (on the right bank at Sengram), Arial Khan (on right bank at Chowdhury Char), Tentulia (on the Lower Meghna). Table 4.7 below presents main features of the distributaries of the Ganges River in Southwest region of Bangladesh. Figure 4.12 shows the distributaries of Ganges River.

Table 4.7: Main Features of Major Distributaries of the Ganges

Mathabhanga Chandana Name Gorai River Hisna River River Barasia River Length (Km) 121 86 125 55

Source Gangs Ganges Ganges Ganges Daulatpur, Kushtia Sadar, Rajbari Sadar, Daulatpur, Source Location Kushtia Kushtia Rajbari Kushtia Ichamoti- Outfall Madhumati Madhumati Mathabhanga Kalindi Damurhuda, Kasiani, Gangni, Outfall Location Sripur, Magura Chuadanga Gopalganj Meharpur Avg width (m) 29 280 40 30

River Type Meandering Meandering Meandering Meandering

Flow type Seasonal Perennial Seasonal Seasonal Maxmum Flood July-Sep 8880 (Sep) 52.60 (August) July-Sep Flow (m3/s) Source: Bangladesher Nad-Nadi (Rivers of Bangladesh), BWDB, August-2011

Figure 4.12: Schematic Diagram of the Ganges River Distributaries It is strategy which is to restore flows in the distributaries from the Ganges by re- excavation of the upstream river and providing the link channel for the connectivity of the rivers. In order to ensure the optimal inflow, regular maintenance program has to be adapted. There are three river system exist on the Ganges right bank and those are: Hisna-Mathabhanga-Bhairab-Kobadak System, Gorai-Madhumati System and Chandana-Barasia-Kumar System.

(i) Hisna-Mathabhanga-Bhairab-Kobadak System

Currently, there are discontinuities between different river reaches at several locations in Hisna-Mathabhanga-Bhairab-Kobadak River system. Mathabhanga- Nabaganga, Mathabhanga-Upper Bhairab and Kobadak-Bhairab have no existing connection. Presently, Hisna River is totally cutoff from the Ganges River. Mathabhanga River receives flood spills from Ganges only during high stages in the Ganges. In rest of the year Mathabhanga River remains dry or carries rainfall runoff.

The Hisna River is influencing the water regime that are being used for the G-K Project. It will travel and interface the Rivers Kumar and the River Bhairab. At Dharsona, Bhairab is the river which was the river once that was connected to the River Kobadak but early 20th century due to the construction of Cross-dam and loopcut with a view to straightening the river, the River Kobadak had been disconnected and becomes perennial to semi perennial river and further it was aggravated after reduction of water flow at Ganges after Farakka operation.

Hisna near Bharamara Upzila Silted up upstream of Hisna

Restoring the flow of the rivers of Mathabhanga- Hisna including Bhairab system & augmentation at offtake of rivers‟ might be the vital way of long term sustainable solution.

(ii) Gorai-Madhumati System

The Gorai, important river of the Ganges-Padma River system, has developed from the combined flow of the three large offshoots of the Ganges north of Kushtia town. The Kaliganga, its first split-channel, takes off south of Kushtia and joins the Kumar River near Shailkupa. On its southeastern course it has changed its name several

times, viz, Madhumati, Baleshwar and Haringhata. It receives flows from the Kumar, the Nabaganga and the Chitra Rivers through different channels. At present the Rupsa-Pussur receives much of its flow from the Gorai through its spill-channel, the Nabaganga. The Rupsa-Pussur receives the Mongla Channel and the Mirgamari Cross-Channel, a spill from the Bhola on the left bank on its downward flow to the Sundarbans at Mongla Port and Chandpai Forest outpost respectively. On the right bank it receives the flows of the Dhaki, the Manki and the Bhadra - three affluent spill-channels of the Sibsa system. About 32 km north of the sea, the Rupsa-Pussur joins the Sibsa system and results in the formation of a 7.5 km wide Morzal River and debouches into the Bay of Bengal through the estuary of the Marjhata and the Pussur Rivers (Islam, 1978; BWDB, 1988).

To get the river flowing again, a number of solutions were considered. One of the better solutions was found to be dredging a deeper channel at the bifurcation where the Gorai splits off from the Ganges including 30 km downstream. After implementation of three dredging seasons, 1998, 1999 and 2000, the dredging increased the water flow in the Gorai River, restoring the fish population and allowing year-round navigation. In order to restore, dry season flows from the Ganges to the Gorai and to gain insight in the contributions that dredging can give on the long-term restoration of the Gorai River offtake and contributing to the control of salinity intrusion into the southwest region of Bangladesh. BWDB arranged to make a dry season low flow channel to maintain perennial flow in the Gorai through dredging starting in December 2010 to June 2013 in the 30 km of the Gorai.

(iii) Chandana-Barasia-Kumar System

Chandana River an offtake of the Ganges River, branches out at the village Dahuka of pangsha upazila in rajbari district. The river follows a 30 km long meander course from Pangsha to Kalukhali. Then it continues south for about 20 km through a straight course and joins the Barasia at Arkandi village between Kamarkhali and madhukhali. From this point the river is renamed as the Chandana-Barasia or Chandana-Arkandi. The joint course of the Chandana and the Barasia flows further

south through boalmari and kashiani upazilas and debouches into the Gorai- Madhumati north of Bhatia Bazar. The length of this part is about 50 km.

In the past, the Chandana was connected with the Gorai and Kumar Rivers. Actually the Kumar received part of the Gorai flow through the Chandana. But at present, the Chandana is almost disconnected from the Gorai and Kumar due to the shifting of the course of the river, construction of various flood-proofing structures, changing of landscape, etc. The river overspills during the monsoon, but its mouth remains dry in winter.

Overall the river is not navigable and shows no erosion or flood-causing tendency. The Bangladesh water development board has taken up various renovation works to maintain the flow of the river. To revive the Chandana and the river system (Chandana-Barasia-Modhumati River system and Chandan-Kumar-Kumar Nadi- Modhumati system) further downstream dredging of dry channel and link channel is required for the connectivity of the River system. The execution of the Offtake Management, re-sectioning the Rivers/Canal section and regular maintenances and link channel for the connectivity of the River system would be imperative to achieve the objective.

4.3 Data Collection

Different data on salinity level, water level, river cross-section and discharge were collected for consecutive two years considering monsoon and dry season under GRRP project by IWM. The data has been used in this study as secondary data and other data were collected from IWM, BWDB and BMD. The discharge measurement has been conducted by using Acoustic Doppler Current Profiler (ADCP). The water level data were collected from BWDB and IWM. Water level (WL), flow (Q) and salinity measurement locations are described in the Table 4.8 and Table 4.9.

The bathymetric survey data were collected by IWM from different study. Historical hydrometric data such as water level and discharge were also collected from IWM, BWDB. Rainfall data were collected from BMD.

(i) Water Level & Discharge Data

A detailed Water level and discharge measurement for 21 different stations was carried out under GRRP and CPWF study to validate the Southwest Regional Salinity Model and Bay of Bengal Salinity Model. An inventory is furnished in the Table 4.8.

Table 4.8: Inventory of Discharge and Water level measurement Stations

Easting Northing Data Period of SI Station Name River Name (BTM) (BTM) Type Collection 1 457341 518494 Rupsa Rupsa Flow 2 480558 535641 Mollar Hat (up) Madhumati Flow Jan-March 2011; Aug- Sep 2011; Feb- 3 459014 484706 Mongla Nala Mongla_Nulla Flow Apr 2012 4 449858 515414 Batiaghata Lower Shoilmari Flow 5 446195 518110 Batiaghata (up) Upper Shoilmari Flow Akram Point 453669 431665 Pussur Flow 6 (Pussur) Akram Point 449731 436450 Sibsa Flow 7 (Sibsa) Mollar Hat 8 478646 541631 Madhumati WL (down) 9 499661 514762 Kaliganga Kaliganga_sw WL Jan-March 2011; Aug- Sep 2011 501373 499464 Pirojpur Kocha WL 10 11 441737 489678 Nalianala Sibsa WL Borhan Uddin 569348 485755 Tentulia River WL 12 (down) 391756 496876 Town Sripur Ichamoti WL 13 14 405073 493869 Sanssasir Chalk Habra WL WL 15 458275 485585 Mongla Pussur Flow WL 16 521372 448482.4 Amtali Buriswar River Jan-March 2011; Flow Aug- Sep 2011; Feb- WL Apr 2012 17 490373 442227.9 Char Doani Baleswar River Flow WL 18 466552 551766 Bardia Nabaganga_M Flow WL 19 451684 508913 Ancharia Kazibacha Flow WL Aug- Sep 2011; Feb- 20 525068 454836 Dalachara Bighai river Apr 2012 Flow WL 21 528104 462481 Khasher Hat Bighai Flow

Figure 4.13: Location of Discharge and Water Level measurement Stations

(Source: IWM, 2012)

(ii) Salinity Measurement Data

A detailed salinity measurement for 33 different stations was carried out under GRRP study. An inventory is furnished in the Table 4.9. The salinity data was collected from December 2010 to May 2011 & from December 2011 to May 2012 in two times of an alternate day (High water and Low water).

Table 4.9: Time duration and locations for salinity measurement

SI Easting (BTM) Northing Station Name River Name 1 524544 429026(BTM) Khepupara Kolapara Adhanmanik 2 521662 562931 Madaripur Arialkhan 3 441878 499285 Darunmollik Badurgacha 4 526546 470249 Pirojpur Baleswar 5 490891 446769 Char Doani Baleswar 6 450845 532106 Hospital Ghat Bhairab 7 479624 460270 Sharankhola Bhola 8 499079 436429 Bishkhali DS Bishkhali 9 523314 447413 Amtali Buriswar 10 450692 498833 Chalna Chunkuri 11 438339 507306 Shundor Mohol Gangrail 12 440310 500106 Habour Khali Habour Khali 13 564093 522293 Ganeshpura Hilsha 14 398460 484940 Bashantapur Ichamoti 15 405391 453669 Kaikhali Jamuna 16 511168 416432 Mohipur Khaprabhanga 17 429112 457305 Kobadak Kobadak 18 445937 516892 Thanibuina L_Salta 19 485165 551415 Haridashpur MBR 20 479502 541515 Chapailghat Madhumati 21 489537 530228 Patgati Madhumati 22 580448 529659 Moju Chowdhury Hat Meghna 23 602434 496692 Jarirdona Regulator Meghna 24 573010 502623 Daulatkhan Meghna 25 565543 569112 Chandpur Meghna 26 466720 553003 Bardia Nabaganga 27 459437 484855 Mongla Pussur 28 446661 410366 Hiron Point Pussur 29 457314 521335 Khulna Rupsha 30 440785 484463 Nalian Sibsa 31 509045 510288 Swarupkathi Swarupkathi 32 532187 430676 Madhupara Tentulia 33 566410 492721 Burhanuddin Tentulia

Figure 4.14: Location of Salinity measurement Stations (Source: IWM)

(iii) Cross Sectional Data

Cross-section survey has been collected from IWM. The collected bathymetry data was surveyed under different projects in IWM which are carried out for the updating of previous model and development of new dedicated models. The detail of the survey specification is furnished in the Table 4.10.

Table 4.10: Details of river cross-section survey

Sl. Length River Interval No of Cross- Data Name of the River No Stretch (km) (m) section Source 1 Haringhata 16 500 32 2 Pussur 100 1000 100 3 Sibsa 70 1000 70 4 Aura Sibsa 20 1000 20 5 Dhaki 20 1000 20

6 Sutakhali 30 1000 30 IWM 7 Baleswar 55 1000 55 (2011) 8 Kocha 15 500 30 9 Sarupkathi 15 1000 15 10 Kaliganga and Madhumati 90 2000 45 11 Nabaganga, Atai and Rupsa 60 2000 30 12 Gorai 30 500 60 13 Betna 48 500 96

14 Marirchap 38 500 76 IWM 15 Parulia Sapmara 24 500 48 (2012) 16 Kholpetua 8 500 16 17 Afrakhal 40 5000 8 18 Agarpur 15 5000 3 19 Amanatganj 30 5000 6

20 Arialkhan 75 5000 15 GBS 21 Badugacha 20 5000 4 (2010) 22 Baleshwar 85 5000 17 23 Bansana 15 5000 3 24 Betna 105 5000 21

Sl. Length River Interval No of Cross- Data Name of the River No Stretch (km) (m) section Source 25 Bhadra 30 5000 6 26 Bhairab_Lower 15 5000 3 27 Bishnu 30 5000 6 28 Chitra 45 5000 9 29 Daudkhali 25 5000 5 30 Deluti 25 5000 5 31 Ghagor 30 5000 6 32 Gobra 10 5000 2 33 Gashiakhali 30 5000 6 34 Gunkhali 15 5000 3 35 Katakhali 5 5000 1 36 Kazibacha 20 5000 4 37 Kobadak 70 5000 14 38 Kaliganga-SW 10 5000 2 39 Kumar Nadi 45 5000 9 40 Kumarkhali 15 5000 3 GBS 41 Kumar River 45 5000 9 (2010) 42 MBR 35 5000 7 43 M.G.Canal 10 5000 2 44 Mongla-Nulia 20 5000 4 45 Marirchap 30 5000 6 46 Nalua-Nullah 10 5000 2 47 Naria 15 5000 3 48 Nabaganga-M 35 5000 7 49 Old Pussur 30 5000 6 50 Otra River 20 5000 4 51 Palong 35 5000 7 52 Polyhara 35 5000 7 53 Pussur 40 5000 8 54 Rupsa 35 5000 7 55 Salta-W 10 5000 2

Co-ordinate System

All the data and presented maps given in this study are in Bangladesh Transverse Mercator (BTM) co-ordinate system. The position (x, y) received from the GPS is in WGS-84 ellipsoid were later converted to Bangladesh Transverse Mercator (BTM) under Everest 1830 ellipsoid by using Hydro Pro Software. The parameters used in BTM projection system is given below:

Conversion parameter from WGS-84 Ellipsoid to Local Ellipsoid (Everest 1830)

Everest-1830 ellipsoid

Semi-major axis a = 6,377,276.34518 m Semi-minor axis b = 6,356,075.41511 m Inverse flattening 1/f = 300.8017

Datum Transformation Parameters

Method : Seven Parameters Rotation X : 0 Rotation Y : 0 Rotation Z : 0 Translation X : -283.729 m Translation Y : -735.942 m Translation Z : -261.143 m Scale : 0 ppm

Projection parameter

Projection method : Transverse Mercator Latitude of origin : 0° N Central meridian : 90° E False Northing : -2,000,000 m False Easting : 500,000 m Scale factor : 0.9996

4.3.1 Data Analysis

(a) Seasonal variation (Moving Average)

The study area consist of a number of major rivers namely Madhumati, Atai, Rupsa, Kazibacha, Pussur, Sibsa, Nabaganga, Kaliganga, Kacha and Baleswar, Kobadak, Kholpetuain the downstream. All these rivers are tidal in nature and tidal characteristics are also different. Tidal range in the Madhumati River is very less which varies from 0.5m to 1.9m whereas in Kacha River at Pirojpur it varies from 1.2m to 3.4m and in Baleswar River at Sarankhola it varies from 1.1m to 3.2m. Again at Khulna in Rupsa River tidal range varies from 1.4m to 3.3m but at Mongla it varies from 1.5m to 3.7m. Table 4.11 shows the tidal ranges at different location in the model area.

Table 4.11: Tidal ranges at different location in different river

SI Location River Tidal range (m) 1 Mollahat Madhumati River 0.50-1.90 2 Pirojpur Kacha River 1.20-3.40 3 Sarakhola Baleswar River 1.10-3.20 4 Khulna Rupsa River 1.40-3.30 5 Mongla Pussur River 0.94-4.13 6 Akram Point Pussur River 1.00-3.40 7 Hiron Point Pussur River 0.95-3.20 8 Noapara Betna River 1.50-5.40 9 Paikgacha Kobadak River 1.50-4.40 10 Rupsa Bhairab River 1.75-3.00

Seasonal variation has also been observed in all the rivers in the study area and it is also different in different river. It is seen that seasonal variation is about 1.5m at the downstream of the Bhairab River and at the middle it is about 1.8m. It has been found that seasonal variation is 0.7m at Mongla in Pussur River. Baleswar River at Sarankhola the seasonal variation is almost same like Mongla which is 0.75m whereas at Mollahat in Modhumati river seasonal variation is higher than Mongla and Sarankhola and the value is 1.25m. It is clear from the Figure 4.15 that the seasonal variations are 1.25m, 0.9m and 0.8m at Khulna, Mongla and Hiron Point respectively in Rupsa-Pussur River system whereas at Sharankhola in Baleswar River system it is 0.8m. Again it is found that seasonal variations are 1.5m & 1.8m

near Rupsa and Afra khal in Bhairab River. The tidal fluctuation and seasonal variation of Tidal River are shown in Figure 4.15.

Location: marked as (7) in Figure 4.14

Location: marked as (29) in Figure 4.14

Location: marked as (27) in Figure 4.14

Location: marked as (28) in Figure 4.14

Figure 4.15: Seasonal variation of Tidal Rivers (Source: IWM, 2014)

(b) Maximum & Minimum Water Level

Historical water level data from 1960 to 2009 of Ganges at Hardinge Bridge and Gorai at Gorai Railway Bridge have been collected and analyzed to get an idea about the amount of flow through Ganges and Gorai. For the One and two dimensional mathematical model water level data have been collected for Ganges and its major distributaries from IWM for the 2009-2010 hydrological years. Table 4.12 shows the list of locations for which data is collected.

Table 4.12: List of Water level data collection stations

SI Station Name River Name Time Period Data Source

1 Hardinge Bridge Ganges 1960-2010 BWDB 2 Gorai Railway Bridge Gorai 1960-2010

3 Hiron Point Pussur 1984-2013

4 Baruria Padma 1984-2013 BIWTA 5 Bhairab Bazar Upper Meghna 1984-2013

6 Chital Khali Lower Meghna 1984-2013

By analyzing the collected historical water level data of Ganges at Hardinge Bridge and Gorai at GRB it is found that the water level are in declining trend since Farakka Barrage is in operation. And average water level has increased slightly after the Ganges Water Treaty in 1996 though maximum water level at Hardinge Bridge is in decreasing trend. Figure 4.16 and Figure 4.17 shows the maximum, minimum average water level of Ganges River from 1912 to 2012 and Gorai River from 1960 to 2009. Figure 4.18 shows the historical water level of Pussur River at Hiron Point from 1984 to 2013. Figure 4.19 and Figure 4.20 shows the measured water level of Lower Meghna River at Chitol Khali (2011-12) and Upper Meghna River at Bhairab Bazar (2011-14).

Location: Shown in Figure 4.11

Figure 4.16: Maximum, minimum and average water level of Ganges at Hardinge

Bridge from 1912 to 2012 (Source: BWDB)

Location: Shown in Figure 4.11

Figure 4.17: Maximum, minimum and average water level of Gorai at GRB from

1960 to 2009 (Source: BWDB/IWM)

Location: Shown in Figure 4.11

Figure 4.18: Pussur River Water Level at Hiron Point (1984 - 2013) (Source: IWM)

Location: Shown in Figure 4.11

Figure 4.19: Lower Meghna River Water Level at Chitol Khali (2011-12) (Source: IWM)

Location Shown in Figure 4.11

Figure 4.20: Upper Meghna River Water Level at Bhairab Bazar (2011-14) (Source: IWM)

Discharge data of Ganges at Hardinge Bridge have been collected from 1934 to 2012 from BWDB. Moreover, discharge data of Gorai measured in GRB have been collected from IWM for 2009-2013 hydrological year for the mathematical model development.

Ganges River: The average flood discharge of the Ganges River is approximately 50,000 m3/s. The discharge is mainly contributed by the snowmelt of the Himalayas and monsoon rainfall. In general, the flood peak occurs between the end of August and early September.

Before mid 1970s the river flowed without any major intervention along its course. Through the implementation of the Farakka Barrage in 1975, India started to divert a part of dry season flow of the Ganges River through the Hoogly River. During the post-Farakka period major changes have occurred in the dry season flow, especially between January and May. During the pre-Farakka period the minimum monthly average flow was 1,500 m3/s. The recorded minimum monthly average flow was 170 m3/s in April, 1997 (Figure 4.21). However, the post-Farakka flood flows found to be of the same as the pre-Farakka flood flows.

After the GWT of 1996 the dry season flow is increased but not so significant that can mitigate the demand of GDA. The minimum flow of Ganges River near Hardinge Bridge from 1934 -2012 is shown in Figure 4.21. The monthly variation of maximum, minimum and average discharge of Ganges River near Hardinge Bridge is shown in Figure 4.22.

Gorai River: The Gorai River is mainly dependent on the Ganges flows both in the dry and wet seasons. The decline in the post-Farakka dry season Ganges flow resulted in a reduction in the dry season water levels in the Ganges. It caused an increase in the bed level at the Gorai mouth from about 4 m PWD in 1964 to 7 m PWD in 1989. Flow through the Gorai River as measured at the Gorai Railway Bridge varies from 0 to 8800 m3/s. It is revealed from the feasibility study of GRRP that peak water levels along the Gorai River vary from 13.5 m PWD at Gorai

Railway Bridge to 4.8 m PWD at Bardia. The monthly variation of flow and water level at Gorai Railway Bridge in Gorai River are shown in Figure 4.23 and Figure 4.24, respectively.

Figure 4.21: Minimum flow of Ganges at Hardinge Bridge from 1934 to 2012

(Source: BWDB)

Figure 4.22: Monthly Maximum, Minimum and average flow variation of Ganges at

Hardinge Bridge (Source: BWBB)

The annual flow volume of Ganges and Gorai during the post- Farakka period showed a clear declining trend, Figure 4.25. It is also reported that the Gorai has stopped flowing within the dry season every year since 1988. The Gorai offtake was seen to have opened sometime between mid-May and mid-June and opened between end December to end January. The reduction in the mean monthly discharges compared to the pre-Farakka period from November to May is also quite significant.

Figure 4.23: Monthly variation of discharge in Gorai Railway Bridge during 1964-

2013 (Source: IWM, 2014)

Figure 4.24: Monthly variation of water level in Gorai Railway Bridge during 1964-

2013 (Source: IWM, 2014)

Figure 4.25: Annual flow volume in the Ganges and the Gorai River

(Source: IWM, 2014)

(d) The Trend of Salinity in the Southwest Region

The daily range of salinity concentrations at the river entrances varies with the spring-neap tide as well as with the season. Tidal amplitudes during spring tides are around 2.5 to 3 times higher than the neap tides. The higher water levels occurring at the coastal boundaries cause greater volume of saline water to enter inland during neap tides. The dilution effect of any freshwater flows in the inland rivers is consequently weaker during spring tides. As a result, maximum salinities occurring during spring tides are generally higher than neap tide concentrations. Salinity levels remain higher in the western part of the study region than the eastern part. This is due to the fact that Gorai, a distributary from the Ganges, is the only significant upstream freshwater source in this region. The eastern part of southwest region remains less saline since it receives freshwater flow from Padma and lower Meghna River through Arial Khan, Bishkhali and Buriswar Rivers. As a result, salinity levels in the region decreases from west to east as well as from south (the Bay of Bengal) to north. The salinity levels in this area exhibit a distinct seasonal variation. Average

100

salinity concentrations at the coast are higher in the dry season than in the monsoon, due to lack of freshwater flow from the upstream. The salinity generally builds up from October to the late May. The seasonal variation of salinity in the three peripheral Rivers Ichamoti represents high saline zone, Pussur shown moderate saline zone and Guishkhali is in low saline zone. The seasonal variation of the three river are shown in Figure 4.26.

High Saline: Ichamoti River

Moderate Saline: Pussur River

Low Saline: Guiskhali River

Figure 4.26: Seasonal variation of salinity in the three peripheral rivers from West to

East (Source: IWM)

101

(e) Effect of Trans-boundary flow on Salinity Intrusion

Salinities in the Bangladesh coast are dependent on the volumes of freshwater flows discharging from upstream. The salinity conditions have further deteriorated in the last few decades because of decrease of flow in the Ganges and empoldering effect. In 1975, India commissioned Farakka Barrage on the Ganges at about 17km upstream of the Indo-Bangladesh border to divert about 40,000 m3/s of flow into Bhagirathi-Hoogly river system. As a consequence of such a large-reduction of the available flow, the Ganges dependent area in Bangladesh was exposed to serious fresh water shortage. The withdrawal of freshwater flow has resulted in landward movement of salinity front in the Ganges dependent coastal area of Bangladesh. In 1996, Bangladesh and India signed Ganges Water Treaty (GWT) for Ganges water sharing between the two countries. The treaty ensured minimum flow in the Ganges River in Bangladesh during dry season which improved the salinity condition in the south-west of Bangladesh. Salinity variation at Khulna in the Rupsa River with upstream freshwater flow during dry season is presented in Figure 4.27.

Figure 4.27: Salinity variation with upstream freshwater flow during dry season

(Source: BWDB & IWM)

102

The recorded maximum discharge was around 43000 m3/s in the Ganges River whereas it increased in the following year and the recorded peak discharge was almost 58000 m3/s on 7 Sep 2013, see Figure 4.28. However, within 18 days the discharge reduced and became 13000 m3/s. It can be said that the drastic fall of discharge did not allow scouring in the mouth of the Gorai. The monthly average discharges near Gorai offtake are also shown in Figure 4.29.

Figure 4.28: Observed monthly average discharge in Ganges River from 2011-2014

(Source: IWM, 2014) During (GRRP Phase-II) dredging has been done in four stages. The calculated flow diversion in Gorai during monsoon and dry period have been plotted and shown in Figure 4.30. The flow through Gorai during monsoon period of 2011 to 2013 was satisfactory but it is seen that in the pre-monsoon of 2014, the flow diversion in Gorai is relatively less than in the other years. For instance, in the month of May and June of 2014, the flow diversion was only 1.33% and 4.92%, respectively (Figure 4.30, top). Otherwise, in the month of May to July, the percentage of flow was found maximum in Gorai in the previous years. Similarly, in dry season of 2012 the flow diversion increased than the previous year. However, in the following year the flow diversion has been reduced and in December 2013 it was almost 2%. At the end of the dry period of 2013-14, the situation became worst. The percentage of flow reduced to less than 1%.

103

Figure 4.29: Observed monthly average discharge in Gorai River from 2011-2014

(Source: IWM, 2014)

Figure 4.30: Flow diversion through the Gorai during monsoon (top) and dry

(bottom) period in different years (Source: IWM, 2014)

104

In 2012, Gorai dredging restored the dry season flow temporarily into the area decreasing the salinity level slightly. The salinity levels along with dry season flow are shown in the following figure (Figure 4.31). Effect of trans-boundary flow on salinity intrusions in Gorai River at Khulna station is evident in the following figure.

Figure 4.31: Monthly Salinity variation with upstream freshwater flow

(Source: IWM) Salinity data was collected both for 2011 and 2012 dry season to assess the change in salinity level due to dredging in Gorai River. Salinity data was collected at different location of Pussur-Sibsa and Baleswar River system both for before and after dredging of Gorai River. Salinity data from eight different stations are furnished in the Figure 4.32. It is evident that significant changes have been taken place in the Pussur-Sibsa River system along with its upstream rivers and the most significant change has been occurred at Bardia. It is found that the salinity level reduces significantly from 10 ppt to zero ppt at Bardia whereas at Khulna it reduces from 17 ppt to 8 ppt due to increase of flow in Gorai River which is helpful for farmer, fishermen and community consumers in that area. The decreasing trend is also found in Shalta, Badurgasa, Mongla and Nalian. But no change has been observed in Baleswar River system.

105

Figure 4.32: Observed salinity at different location for with and without dredging

condition (Source: IWM, 2012)

106

4.4 Mathematical Model Setup

In this study MIKE is used for One-Dimensional and Two-Dimensional Modelling. MIKE11 was used for One-Dimensional model. Subsequently, MIKE21 FM was used for Two-Dimensional model. The calibrated & validated Southwest Regional Model (SWRM) by IWM was used for the simulation of base line scenario (Scenario-1) and worst flow condition (Scenario-2). In this study the existing SWRM was further extended by dredged cross section from Ganges Barrage study. The developed model under this study was simulated for different flow scenarios (Scenario 3A, 3B and 3C).

4.4.1 Model Development

(a) The Existing Southwest Regional Model (SWRM)

The existing South West Region Model (SWRM) covers the entire area lying to the south of the Ganges and west of the Meghna estuary. Total catchment area and length of rivers/channels of the SWRM are around 37,300 km2 and 5,600 km, respectively. The Bay of Bengal and the international border with India form the southern and western boundaries respectively. The rivers of the southwest region of Bangladesh are dominated by tide. Many rivers, particularly those in the southern part, carry very little fresh water flow, but instead act as tidal channels for tides originating in the Bay of Bengal. Freshwater inflows originate from the Gorai, an offtake of the Ganges, and from numerous smaller off-takes from the Lower Meghna.

In the northern part of the model, the main non-tidal river systems comprise the Gorai, Arial Khan, Jayanti and Upper Meghna. The southern rivers mainly comprise tidal estuary systems, the largest being the Jamuna, Malancha, Pussur-Sibsa, Baleswar, Tentulia and Lower Meghna. Interconnected with these larger rivers are a myriad of smaller tidal channels and drainage canals. The tidal channel network is particularly complex in the Sundarbans Mangrove Forest in the far southwest corner of the region.

107

The existing SWR Model has 230 river branches and 32 boundaries of which 12 are directly connected to the sea at the downstream. The cross-sections of most of the river branches have been updated with recent data surveyed in 2009-2012. Among the upstream boundaries three are dominant with freshwater flow from the Ganges- Brahmaputra-Meghna basin. These three boundaries are at Gorai Railway Bridge on the Gorai River, Baruria on the Padma River and Bhairab bazar on the Upper Meghna River. At Bhairab bazar, satisfactory rating curves cannot be generated due to scattered data and tidal influence during dry period; as a result water level time series has been used as upstream boundary of Upper Meghna River. At Gorai Railway Bridge and Baruria, rating curves have been updated and the rated discharges have been used for Gorai and Padma boundaries. The downstream water level and salinity boundaries have been generated based on observed data and the Bay of Bengal model results.

The Southwest Region Model is developed using two separate module of MIKE11 modelling system; rainfall-runoff modelling (hydrological modelling, MIKE11- NAM) and hydrodynamic modelling (MIKE11-HD).

Hydrological (Rainfall-Runoff) Modelling & Hydrodynamic Modelling (HD)

NAM Rainfall Runoff Model is applied to estimate the runoff generated from rainfall occurring in the catchment. The model takes into consideration the basin characteristics including specific yield, initial soil moisture contents and initial ground water level and irrigation/abstraction from the surface or ground water sources. The catchments of the rainfall runoff model are delineated according to the topographic barriers/water shed boundaries, roads and river networks. The rainfall runoff model (NAM) contains 44 catchments. The catchment SW-27 represents the entire Sundarban area. Due to the absence of hydro-meteorological stations inside the Sundarbans, rainfall and evaporation data have been used from neighbouring stations for this catchment. However, input from NAM model for this catchment is ignored in HD model, since the runoff generated in SW-27 is insignificant compared to the volume of flow of the Sundarban river system. Figure 4.33 shows the catchment boundaries of south-west regional model.

108

Figure 4.33: Catchment Boundaries of Rainfall Runoff Model (Source: IWM, 2012)

109

Figure 4.34: Map showing Existing Southwest Regional Model Domain and Boundary (Source: IWM, 2012)

110

(b) The Extended Southwest Regional Model (SWRM)

The extended hydrodynamic/salinity model of the southwest region contains total 39 boundaries, of which 26 are upstream and 13 are downstream boundaries. Additional six flow boundaries were included for the extended SWRM The comparison upstream and downstream boundary between existing Southwest Regional Model by IWM and Extended Southwest Regional Model developed under this study is described in the Table 4.14. The Tidal Boundaries remains same for both of the Model.

The river network of the one dimensional model, MIKE11 includes important distributaries which have significant impact on overall Ganges flow through the Hisna, Gorai & Chandana River system. A list of river and corresponding length is furnished in the Table 4.15. The model network and boundaries are shown in Figure 4.36.

The additional dredged rivers were connected with link channel. Dredged sections were incorporated from different study projects by IWM. Total length of river network of the developed extended Southwest Regional Model is 6269.39 km. The length of dredged section incorporated in the extended Southwest Regional Model is 1658.87 km. The comparison between Existing and Extended Southwest Regional Model is described in Table 4.13.

Table 4.13: Comparison between Existing and Extended Southwest Regional Model

Boundary River Length Dredged Section Model Domain Water (km) Length (km) Flow Level Existing SWRM 15 18 5246.62 - Extended SWRM 15 24 6269.39 1658.87

111

Figure 4.35: Map showing Development of Extended Southwest Regional Model Network

112

Figure 4.36: Map showing Developed Model Network and Boundary

113

Table 4.14: Comparison of upstream and downstream Boundaries used in the Existing and Extended Southwest Regional one-dimensional Model

Boundary Existing SWRM Extended SWRM SI SI Type River Name Ch. River Name Ch. 1 WL Andharmanik 39700 1 2 WL B_Ichamoti 0 2 3 WL Betmar_Gang 36890 3 4 WL Buriswar 57000 4 5 WL Haringhata 17000 5 6 WL Jamuna 61740 6 7 WL Khaprabhanga 0 7 8 WL Malancha 62700 8 9 WL Pussur 98210 9 10 WL Pussur_Khal 30830 10 11 WL Sela_Gang 73290 11 12 WL Shahabaz-1 25800 12 13 WL Supoti_Khal 28630 13 Same Same 14 WL Tentulia 90000 14 15 WL Upper_Meghna 29100 15 16 Inflow Buri-Bhadra 0 16 17 Inflow Daudkhali 0 17 18 Inflow Harihar 0 18 19 Inflow Hatia 0 19 20 Inflow Ichamoti 0 20 21 Inflow Kobadak 180000 21 22 Inflow Kumar 0 22 23 Inflow Marirchap 0 23 24 Inflow Padma 12000 24 25 Inflow Sitalakhya 0 25 26 Inflow U Solmari 0 26 27 Inflow Betna 60000 27 Betna-Ext 0 28 Inflow Begabati 48000 28 Chandana 0 29 Inflow Chitra 41000 29 Chitra 0 30 Inflow Chitra 120000 30 Fatki 0 31 Inflow Gorai 11901 31 Gorai 0 32 Inflow Kobadak 0 32 Hisna 0 33 Inflow Teka-Hari-Teli-Geng 0 33 Kaliganga_U 5000 34 Kumarf2 0 35 Kumark 22000 36 Mathabhanga 0 37 Mukteswari 0 38 Nabaganga_U 0 39 Upper Bhairab 0

114

Table 4.15: List of rivers and corresponding length used in the 1-D model

Length Length SI River Name SI River Name (km) (km) 1 AFRAKHAL 36.0 51 DHAKI 14.0 2 AGARGOTTA 28.4 52 DHANSAGAR 8.0 3 AMTALI 13.5 53 DHARMAGANJ 15.0 4 ANDHARMANIK 39.7 54 DHUJI KHAL 7.2 5 ARIALKHAN 157.0 55 DHULIA 23.0 6 ARIALKHAN-L 43.5 56 DUMKOLI 16.0 7 ARIALKHAN-OFFTAKE 11.0 57 ESAMATI 18.9 8 ARMAL_KHAL 22.3 58 FATKI_RIVER 48.0 9 ARPANGASIA 60.6 59 FIRINGI 7.6 10 ATAI 14.5 60 GALGHASIA 20.5 11 ATHAROBANKI 55.1 61 GASHIAKHALI 23.5 12 AURA_SIBSA 20.0 62 GHAGOR 57.5 13 B_ICHAMOTI 12.0 63 GOBRAKHAL 9.0 14 BADAMTOLA 16.0 64 GORAI 190.0 15 BADURGACHA 9.5 65 GUNKHALI 8.5 16 BAGIDUNIA 4.9 66 GURRANI 3.6 17 BAL 16.6 67 HABARKHALI 8.3 18 BALESWAR 135.0 68 HABRA 12.3 19 BARAPANGA 11.5 69 HADDA 6.0 20 BARASIA_ARBT 51.0 70 HANSRAJ 36.1 21 BARASIALA_G 36.0 71 HARIA 18.5 22 BEGABATI 26.0 72 HARIHAR 39.0 23 BELUA 21.5 73 HARINGHATA 17.0 24 BETMAR_GANG 36.9 74 HARINTANA_K 3.6 25 BETNA 91.0 75 HARTA 9.0 26 BETNA-EXT 18.3 76 HATIA 12.0 27 BHAIRAB 17.0 77 HISNA 44.0 28 BHAIRAB_L 10.5 78 ICHAMOTI 130.0 29 BHAIRAB_U 133.3 79 ILSHA 32.0 30 BHOLA 67.2 80 JAMUNA 61.7 31 BIGHAI 23.4 81 JATA_GANG 12.4 32 BISHKANDIA 51.5 82 JHALOKATI 10.8 33 BISHKHALI 96.0 83 JHAPJHAPIA 9.5 34 BISHNU 21.0 84 JHAPJH-MANGA 2.5 35 BOGI-LINK 2.8 85 JOYANTI-1 32.5 36 BURI-BHADRA 20.0 86 KALABADAR-1 19.0 37 BURISWAR 57.0 87 KALABADAR-2 10.0 38 CHALKI_GANG 17.0 88 KALAGACHI 20.0 39 CHANDANA 54.0 89 KALIGANGA_U 37.0 40 CHANNEL_X 13.8 90 KALIGANGA1 10.0 41 CHARPUTIA 4.8 91 KALIGANGA-SW 30.5 42 CHITRA 110.0 92 KANKSIALI 17.0 43 CHITRA-NARAIL 40.0 93 KATAKHALI 28.2 44 CHORA_BETMAR 13.1 94 KATAKHALI-K 3.5 45 CHUNKURI 6.0 95 KATAKHALI-SC 14.0 46 CHURKUNI_G 30.3 96 KAZIBACHA 13.5 47 CONNECTION 8.0 97 KESONKHALI 11.5 48 DALACHARA 20.0 98 KHAIRABAD 40.5 49 DAUDKHALI 23.0 99 KHAPRABHANGA 20.4 50 DELUTI 8.5 100 KHASITANA 3.4

115

Length Length SI River Name SI River Name (km) (km) 101 KHOLPETUA 55.5 151 NAYABHANGANI 32.0 102 KIRTONKHOLA 40.5 152 PADMA 71.4 103 KOBADAK 228.0 153 PADMA-North Channel 42.0 104 KOCHA 15.0 154 PADMA-South Channel 32.5 105 KOGADUNIA 14.6 155 PAIRA 24.0 106 KOYRA 12.0 156 PALANG 39.5 107 KOYRA_KHAL 20.5 157 PANDAB-1 19.0 108 KUMAR 115.6 158 PANDAB-2 6.0 109 KUMAR_NADI 46.1 159 PARULIA SAPMARA-1 15.8 110 KUMAR-1 20.5 160 PARULIA SAPMARA-2 7.3 111 KUMAR-2 27.5 161 PATAKATA_K 7.0 112 KUMARF1 26.1 162 PATHURIA_G 26.0 113 KUMARF2 16.0 163 PATKUSATA 7.8 114 KUMARK 93.9 164 PATUAKHALI 10.5 115 KUMARKHALI 10.3 165 POYLAHARA 29.0 116 L-BHADRA 12.0 166 PUSSUR 98.2 117 LINK-BHAIRAB-MUKTESWARI 2.5 167 PUSSUR_KHAL 30.8 118 LINK-CHANDANA-KUMAR 38.0 168 PUTIA_KHAL 26.4 119 LINK-KK 1.0 169 RANGAMATIA 25.0 120 LINK-KOBADAK-BETNA 15.3 170 RUPSA 17.4 121 LINK-KOBADAK-LOOPCUT 5.1 171 SAKBARIA 36.8 122 LINK-MATHABHANGA-NABAGANGA 8.0 172 SALTA(W) 6.7 123 LINK-MATHBHANGA-BHAIRAB 4.4 173 SELA_CUT 0.7 124 LINK-NABAGANGA-CHITRA 3.3 174 SELA_GANG 73.3 125 LOHALIA 71.5 175 SHAHABAZ-1 25.8 126 LOWERMEGHNA 39.1 176 SHANDHA 12.5 127 LOWERMEGHNA-EAST 16.8 177 SHATLA 8.0 128 LOWERMEGHNA-WEST 23.7 178 SHIKARPUR 17.0 129 L-SALTA 9.0 179 SIBSA 81.5 130 L-SOLMARI 7.0 180 SITALAKHYA 69.4 131 M.G.CANAL 6.0 181 S-MEGHNA 19.0 132 MADHUMATI 59.0 182 S-MEGHNA1 5.0 133 MAGURA-DIV 0.5 183 S-MEGHNA2 7.5 134 MAJABHANGA 8.2 184 SONA_KHAL 10.2 135 MAJUDKHALI 6.5 185 SONATOLA 17.5 136 MALANCHA 62.7 186 SUNDARIKOTA 6.5 137 MANGA 6.3 187 SUND-JHAJHAP 8.5 138 MARA_GANG 29.6 188 SUPOTI_KHAL 28.6 139 MARDAT 12.5 189 SUTARKHALI 28.0 140 MARIRCHAP 37.6 190 SWARUPKATI 22.0 141 MATHABHANGA 162.5 191 TALDUP 19.4 142 MBR 30.0 192 TEKA-HARI-TELI-GENG 44.1 143 MEGNA_KHAL 27.5 193 TENTULIA 90.0 144 MINAJNADI 17.0 194 TORKI-1 43.0 145 MONGLA_NULLA 14.3 195 TORKI-2 5.0 146 MRIGAMARI 12.8 196 U-BHADRA 25.5 147 MUKTESWARI 32.0 197 UPPER_MEGHNA 80.9 148 NABAGANGA_L 3.0 198 UPPER-BHAIRAB 83.3 149 NABAGANGA_M 26.0 199 U-SOLMARI 14.6 150 NABAGANGA_U 167.0 200 UZIRPUR 10.5

116

4.4.2 Salinity Modelling

Salinity modelling aids to assess the salinity intrusion under known boundary conditions without detailed measurements throughout the entire coastal region. To model the salinity variation in the estuary it is very important to have a well calibrated water flow model. The salinity model, based on this hydrodynamic model, will describe the transport and advection of salinity. Figure 4.37 shows the overall methodology of the salinity model.

Figure 4.37: Methodology of the salinity modelling

117

(a) One dimensional Salinity model (MIKE-11)

The one-dimensional salinity model consists of three different modules: rainfall runoff (NAM), hydrodynamic (HD) and salinity model (AD). The salinity intrusion in the river system is very much dependent on the freshwater flow from the upstream, tidal dynamics of the coastal river system and surface water runoff as a result of rainfall events. This is why the salinity model needs the results from the rainfall-runoff model and the hydrodynamic model.

The rainfall-runoff (NAM) model is applied to estimate the runoff generated from rainfall occurring in the catchments of the model area. The model considers the basin characteristics including specific yield, initial soil moisture contents and irrigation/water extraction from the surface or ground water sources in the catchments. Rainfall and evaporation data from secondary sources (BWDB and BMD) have been incorporated into the model. The model computes evaporation, percolation and other losses and gives the catchment runoff as outputs.

The 1-D hydrodynamic model calculates water flow and water level using the runoff generated from the catchments (output of NAM Model) as well as taking input of flow from the upstream rivers. It incorporates water flow data at upstream boundaries and water level data at downstream boundaries.

Model Boundaries

The locations of the boundaries of the salinity model are same as that of HD model. However there are minor changes in the nature of the boundaries. Few boundaries are closed, that is no net transport of salt is assigned and few boundaries are open where salinity concentration changes with the changes in the volume of flow. The closed boundaries in the HD model (where contribution of rainfall runoff were considered only) are treated as closed in the salinity model. All the tidal boundaries of HD model are considered as open, where salinity concentrations have been specified from the Two Dimensional Bay of Bengal Salinity Model. A complete list of salinity boundaries has been presented in Table 4.16.

118

Table 4.16: List of boundaries used in the extended salinity model

SI No. River Name Chainage (m) Type of Boundary 1 Betna-Ext 0 2 Buri-Bhadra 0 3 Chandana 0 4 Chitra 0 5 Fatki 0 6 Gorai 0 7 Harihar 0 8 Hatia 0 9 Hisna 0 10 Kaliganga_U 5000 Closed Boundary 11 Kumar 0 12 Kumarf2 0 13 Kumark 22000 14 Mathabhanga 0 15 Mukteswari 0 16 Nabaganga_U 0 17 Padma 12000 18 Sitalakhya 0 19 Upper Bhairab 0 20 Upper Meghna 29100 21 Andharmanik 39700 22 B_Ichamoti 0 23 Betmar Gang 36890 24 Buriswar 57000 25 Daudkhali 0 26 Haringhata 17000 27 Ichamoti 0 28 Jamuna 61740 29 Kobadak 180000 Open Concentration 30 Khaprabhanga 0

31 Malancha 62700 32 Marirchap 0 33 Pussur 98210 34 Pussur_Khal 30830 35 Sela_Gang 73290 36 Shahabaz-1 25800 37 Supoti_Khal 28630 38 Tentulia 90000 39 U Solmari 0

119

Calibration Parameters

Calibration parameter of the salinity model is mainly the dispersion coefficient that is determined by the function of flow velocity and channel conveyance, which in another turn lead to the volume of flow passing through the channel section. Dispersion coefficient in the model is defined through the following formulation:

D  aV b (4.1)

Where, a is the dispersion factor and b is the dispersion exponent and V is the mean flow velocity. For the salinity model of the Southwest, the parameters for different rivers have been selected from previous salinity study of IWM. Dispersion coefficient in those studies varies from 20 m2/s to 2200 m2/s in different river branches. In the deepest rivers, comparatively high dispersion coefficients have to be used to compensate for the effects of pressure gradients induced by longitudinal density gradients. The density induced contribution to the pressure gradients increases linearly with the water depth, thus the effect is larger in deeper rivers, where density driven flows develop during slack water with a negligible surface slope and affect longitudinal mixing.

In MIKE11 AD concentration at open boundaries depends on the calibration

Parameter Kmix. This parameter is used to ensure a smooth transition between calculated and specified boundary in the case of flow reversal. A list of Kmix values used in different open boundaries has been presented in Table 4.17. In the Southwest

Salinity Model the values of Kmix vary from 0.04 to 1.00.

Table 4.17: Values of Kmix

SI River Name Chainage Kmix SI River Name Chainage Kmix 1 Andharmanik 39700 0.20 8 Malancha 62700 0.04 2 B_Ichamoti 0 1.00 9 Pussur 98210 0.04 3 Betmar_Gang 36890 0.04 10 Pussur_Khal 30830 0.50 4 Buriswar 57000 0.06 11 Sela_Gang 73290 0.50 5 Haringhata 17000 0.10 12 Shahabaz-1 25800 0.10 6 Jamuna 61740 0.40 13 Supoti_Khal 28630 0.10 7 Khaprabhanga 0 0.20 14 Tentulia 90000 0.50

120

(b) Two Dimensional Salinity Model (MIKE 21FM)

The development of the Bay of Bengal Model (BoBM) is based on the MIKE21 FM module of DHI Water and Environment (FM stands for Flexible Mesh); the MIKE 21 FM modelling system, in turn, is based on an unstructured flexible mesh consisting of linear triangular elements.

Bathymetry

The Model set-up is based on several sources. In the regional area the bathymetry is based on Etopo2 and near the coastline on local surveys. In between these areas nautical sea-chart from C-Map has been used. Etopo2 are very accurate in deeper area. In the shallow area C-Map normally is more accurate than Etopo2 due to finer resolution. As C-Map is used for navigational purposes it‟s not very accurate in deep water. In the rivers and the inner part of the estuary local measurement has been used where possible. Bathymetry of newly developed Bay of Bengal model is shown in the Figure 4.38.

Chandpur

Mayanmar India

Bay of Bengal

West Boundary East Boundary

Figure 4.38: Bathymetry of newly developed Bay of Bengal model (Source: IWM)

121

Model Boundaries

The model has two types of boundary, at the south a wide open boundary is known as south boundary and at the north there are number of boundaries in the rivers known as north boundary. At the south boundary water level has been used whereas at the north boundary flow has been used. The Bay of Bengal is quite deep and the maximum depth along the southern open boundary is more than 2000 meter. At the southern boundary, water level time series is generated from global tide model of DHI and the salinity is kept constant at 32ppt. The salinity timeseries at the upstream boundaries are taken from measurements at Nalian (Sibsa River), Mongla (Pussur River), Arpangasia (Kobadak River), Pirozpur (Baleswar River) and Swarupkathi (Kaliganga River).

Downstream boundary: Water level has been used at the south boundary. This water level along the southern boundary has been extracted from the Global Tide Model. Figure 4.39 shows the variation of water level in time at the western, eastern and in the middle of the southern boundary.

Location Shown in Figure 4.38

Figure 4.39: Water level variation at the south boundary

Upstream/north Boundary: There are eighteen (17) open boundaries in the upstream side of the model. Flow from calibrated and validated southwest and eastern hilly regional model has been used at all these boundaries. A list of the boundaries is shown in Table 4.18.

122

Table 4.18: List of boundaries at the upstream side of Bay of Bengal Model

Sl. Sl. Name of the boundary Name of the boundary No. No.

1 Jamuna 10 Arail Khan-2 2 Arpangasia 11 Tentulia 3 Sibsa 12 Kalabadar 4 Pussur 13 Bamni 5 Gashiakhali 14 Little Feni 6 Baleswar 15 Feni 7 Kaliganga 16 Karnaphuli 8 Swarupkathi 17 Chandpur 9 Arial Khan-1

Calibration Parameters

To calibrate the model it is necessary to adjust the bed resistance. In this study a map with different bed resistance has been used which is shown in the Figure 4.40. All the rivers inside SW have been assigned a manning number of 67 for Pussur and Lower Meghna, 30 for Baleswar-Buriswar-Bishkhali and Tentulia. The Bay of Bengal has been assigned manning number of 45.

Figure 4.40: Manning map of newly developed Bay of Bengal Model (Source: IWM)

123

Dispersion coefficient is the main calibration parameter. Dispersion coefficient is defined in the model through the following different formulations as:

.xK 2 D  1 (4.2) t

2  .. uxKD (4.3)

2 3  .. utKD (4.4)

Where, ∆x is the grid spacing, ∆t is the time step and u is local current speed. The coefficient K1, K2, K3 is treated as a calibration parameter for dispersive effects due to geometric features of the considered river and estuary, including dispersion due to channel irregularities and due to the transverse velocity and concentration. The dispersion coefficient used in the model is varying from 1000 to 3000 m2/s and has been set through calibration process.

4.5 Summary

Data collection and model set up have been discussed in this chapter. Data have been collected from BWDB & IWM and used in mathematical model simulations for updating, calibrating and validating with measured water level, discharge and salinity data. The available salinity models for the coastal area of Bangladesh has been developed based on MIKE11 and MIKE 21 FM modelling system and applied in this study to find the spatial and temporal variation of salinity level over a year. The latest version of MIKE (2014) has been applied in this study for calibration and validation of hydrodynamic model and salinity model. ArcGis 10.1 has been applied for preparation of study area map, survey specification map, catchment/rainfall- evaporation distribution map, mathematical model network/boundary map and salinity map for different scenarios under this study.

124

CHAPTER FIVE

RESULTS AND DISCUSSIONS

5.1 General

To achieve the objectives a calibrated and validated salinity model for the southwest region of Bangladesh has been used. Three scenarios has been simulated by the calibrated salinity model.

Different types of analyses have been done to understand the outcomes of different simulations. The comparative results are discussed below in terms of spatial and temporal distribution of salinity, Salinity intrusion line and effect of trans-boundary flow on salinity intrusion.

5.2 Description of the Flow Scenarios

To achieve the objectives a salinity model for the southwest region of Bangladesh has been used. During the last two decades the mouth of the Gorai River is silted up due to morphological changes. As a result the mouth of the Gorai River is almost dry during the dry season (January to May). Consequently, the south-west region is seriously affected by salinity intrusion during that period. It is because of the salinity level at the sea (the downstream boundary) are comparatively high during January to May.

The simulation of the calibrated and validated salinity model has been carried out by increasing upstream flows which are connected with Ganges River for limiting salinity of Southwest region of Bangladesh. The model simulation has been conducted from November 2011 to June 2012. Several runs have been simulated with the calibrated and validated hydrodynamic Model and Salinity Model. Scenario- 1 (base line condition, 2011-2012) has been simulated to explore the present salinity intrusion. Scenario-2 (Minimum flow through Gorai River) has been simulated to understand the worst condition for salinity intrusion. Finally, Scenario-3A, Scenario- 3B and Scenario-3C (flow through Ganges connected Rivers) has been conducted to identify the saline free zone with improved upstream flow and dredged river

125

bathymetry condition. Scenario-3B and Scenario-3C has been conducted for the sensitivity of Ganges flow in salinity intrusion. The three simulations have been done under this study which is described below:

Scenario Flow Condition  Simulation of calibrated and validated Southwest Regional (Nov 2011 to June Scenario 1 Base line Scenario 2012)  No upstream connection with Ganges River  Apply existing upstream and downstream flow condition (2011-12)  Baseline Model (Scenario-1) simulated

Minimum flow with minimum flow through Gorai River Scenario 2 through Gorai  All other river (Hisna, Mathabhanga and River Chandana) are disconnected at dry season with Ganges River  Flow through Gorai, Hisna and Chandana based on Ganges 3A barrage study

Flow through  Restoration of flow through the Scenario 3 Ganges connected channels with dredged X-sections. Rivers 20% flow increase of Gorai, Hisna 3B and Chandana River 20% flow decrease of Gorai, 3C Hisna and Chandana River

Based on irrigation water demand, navigational requirement, fisheries requirement, salinity intrusion prevention criteria BWDB has fixed the seasonal flow diversion amount from the Ganges to link channels through Ganges barrage operation. Seven linking canals have been fixed and among them four channels are in the right bank of Ganges and others are situated in left bank. In our study the flow data of the right bank of Ganges River been collected and used for increasing the upstream flow and

126

presented below in Table 5.1. The Gorai River Hydrograph for different flow scenario shown in Figure 5.1.

Table 5.1: Monthly Flow Diversions (m3/s)

River Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Intake

Hisna 400 300 231 234 232 242 184 300 400 500 500 400

Gorai 2500 1000 225 227 225 230 194 500 2500 7600 7600 2500

Chandana 300 200 46 57 77 80 44 50 200 300 300 300

Total 3200 1500 502 518 534 552 422 850 3100 8400 8400 3360

(Source: BWDB, 2012)

Figure 5.1: Gorai River Hydrograph at different flow scenario

5.3 Model Calibration

Calibration means adjustment of the model parameters so that simulated and observed data will match within the desired accuracy. Flow roughness is the major parameter for the calibration of hydrodynamic model. Model parameters may require adjustment due to a number of reasons. For fully distributed hydrodynamic model, it

127

is theoretically possible to estimate the parameters from sufficient number of field observations alone. Unfortunately sufficient data are rarely available. Also it may not be possible to directly measure the parameter (e.g. the bed roughness of a two dimensional estuary model). Empirical and lumped conceptual models contain parameters whose values could only be estimated through the calibration process. In reality, all models require some degree of calibration to fine tune the predictive ability of the model.

Manning‟s M (Inverse of Manning‟s roughness n) is used as the calibration parameter for the calibration of the one dimensional mathematical models. Hydrological year of 2012 has been used for the calibration of the base model. The locations of water level, discharge and salinity calibration are given in Figure 4.13 and Figure 4.14 respectively.

Water level Calibration: Water level calibration was done by IWM for Ancharia in the Kazibacha River, Batiaghata in the Lower Sholmari River, Town Sripur in the Ichamoti River and Sannasir chalk in the Habra River. Under this study all the units of water lever are in (m PWD). Sample of hydrodynamic Calibration plots of Southwest Regional Model against water level are shown in Figure 5.2 & Figure 5.3.

Flow Calibration: Flow calibration was done by IWM for Mongla in the Pussur River, Char Doan in the Baleswar River, Amtali in the Buriswar River, Ancharia in the Kazibacha River, Bardia in the Nabaganga River and Dalchara in the Bighai River. Under this study all the units of flow are in (m3/s). Sample of hydrodynamic calibration plots of Southwest Regional Model against Flow are shown in Figure 5.4 to Figure 5.6.

Salinity Calibration: Salinity calibration was done by IWM for Mongla in the Pussur River, Khulna in the Rupsa River, Darunmollik in the Badurgacha River, Bishkhali DS in the Bishkhali River, Char Doan in the Baleswar River, Thanibunia in the Lower Salta River, Nalian in the Sibsa River and Shudor Mohol in the Gangrail River. Under this study all the units of salinity are in (ppt). Sample of hydrodynamic calibration plots of Southwest Regional Model against Salinity are shown in Figure 5.7 to Figure 5.10.

128

All the comparison shows reasonably good agreement which indicates that hydrodynamic model is good enough for salinity model and simulation of baseline condition (Scenario-1) and worst condition (Scenario-2). The value of "R" represents the correlation between measured and simulated values and the scale of these values is '1'. The calibration location and Correlation factor has been given in Table 5.2.

Table 5.2: Model Calibration Location and Correlation factor “R”

SI Correlation Calibration River Name Station No “R”

1 Kazibacha Ancharia 0.95 2 Lower Shoilmari Batiaghata 0.95 Water Level 3 Ichamoti Town Sripur 0.95 4 Habra Sannasir chalk 0.90 5 Pussur Mongla 0.90 6 Baleswar Char Doani 0.92 7 Buriswar Amtali 0.94 Flow 8 Kazibachar Ancharia 0.94 9 Nabaganga Bardia 0.94 10 Bighai Dalachara 0.90 11 Pussur Mongla 0.99 12 Rupsa Khulna 0.83 13 Badurgacha Darunmollik 0.95 14 Bishkhali Bishkhali DS 0.95 Salinity 15 Baleswar Char Doani 0.92

16 Lower Salta Thanibunia 0.92 17 Sibsa Nalian 0.97 18 Gangrail Shundor Mohol 0.95

129

Kazibacha River at Ancharia Location marked as (19) in Figure 4.13

Location marked as (4) Lower Shoilmari River at Batiaghata in Figure 4.13

Figure 5.2: Calibration of Southwest Regional Hydrodynamic Model against Water

Level (2012)

130

Location marked as (13) Ichamoti River at Town Sripur in Figure 4.13

Location marked as (14) Habra River at Sannasir chalk in Figure 4.13

Figure 5.3: Calibration of Southwest Regional Hydrodynamic Model against Water

Level (2012)

131

Pussur River at Mongla Location marked as (15) in Figure 4.13

Baleswar River at Char Doani Location marked as (17) in Figure 4.13

Figure 5.4: Calibration of Southwest Regional Hydrodynamic Model against Flow

(2012)

132

Buriswar River at Amtali Location marked as (16) in Figure 4.13

Kazibacha River at Ancharia Location marked as (19) in Figure 4.13

Figure 5.5: Calibration of Southwest Regional Hydrodynamic Model against Flow

(2012)

133

Nabaganga River at Bardia Location marked as (18) in Figure 4.13

Bighai River at Dalachara Location marked as (20) in Figure 4.13

Figure 5.6: Calibration of Southwest Regional Hydrodynamic Model against Flow

(2012)

134

Pussur River at Mongla Location marked as (15) in Figure 4.13

ppt

Location marked as (1) Rupsa River at Khulna in Figure 4.13

ppt

Figure 5.7: Calibration of Southwest Regional Salinity Model against Salinity (2012)

135

Badurgacha River at Darunmollik Location marked as (3) in Figure 4.13

ppt

Bishkhali River at Bishkhali DS Location marked as (8) in Figure 4.13

ppt

Figure 5.8: Calibration of Southwest Regional Salinity Model against Salinity (2012)

136

Char Doani River at Baleswar Location marked as (5) in Figure 4.13

ppt

Lower Salta River at Thanibunia Location marked as (18) in Figure 4.13

ppt

Figure 5.9: Calibration of Southwest Regional Salinity Model against Salinity (2012)

137

Location marked as (30) Sibsa River at Nalian in Figure 4.13 ppt

Location marked as (11) Gangrail River at Shundor Mohol in Figure 4.13 ppt

Figure 5.10: Calibration of Southwest Regional Salinity Model against Salinity

(2012)

138

5.4 Model Validation

Validation is the process of determining the degree to which a model or simulation is an accurate representation of the real world from the perspective of the intended uses of the model or simulation. Calibrated hydrodynamic model for year 2012 is validated for year 2011 hydrological year. The locations of water level, discharge and salinity validation are given in Figure 4.13 and Figure 4.14 respectively.

Water level Validation: water level validation has been done by IWM for Pirozpur in the Kocha River, Batiaghata in the Lower Sholmari River, Ancharia in the Kazibacha River and Khulna in the Rupsa River. Sample of hydrodynamic validation plots of Southwest Regional Model against water level are shown in Figure 5.11 & Figure 5.12.

Flow Validation: Flow validation has been done by IWM for Akram Point in the Pussur River, Akram Point in the Sibsar River and Mongla in the Pussur River. Sample of hydrodynamic validation plots of Southwest Regional Model against Flow are shown in Figure 5.13.

Salinity Validation: Salinity validation has been done by IWN for Mongla in the Pussur River and Nalian in the Sibsa River. Sample of hydrodynamic validation plots of Southwest Regional Model against Salinity are shown in Figure 5.14.

The validation location and Correlation factor “R” has been given in Table 5.3.

Table 5.3: Validation Location and Correlation factor “R”

SI No Calibration River Name Station R 1 Kocha Pirozpur 0.91 2 Water Lower Shoilmari Batiaghata 0.90 3 Level Kazibachar Ancharia 0.95 4 Rupsa Khulna 0.92 5 Pussur Akran Point 0.94 6 Flow Sibsa Akram Point 0.90 7 Pussur Mongla 0.90 8 Pussur Mongla 0.99 Salinity 9 Sibsa Nalian 0.97

139

Location marked as (10) Kocha River at Pirozpur in Figure 4.13

Lower Shoilmari River at Batiaghata Location marked as (4) in Figure 4.13

Figure 5.11: Validation of Southwest Regional Hydrodynamic Model against Water

Level (2011)

140

Kazibacha River at Ancharia Location marked as (19) in Figure 4.13

Location marked as (1) Rupsa River at Khulna in Figure 4.13

Figure 5.12: Validation of Southwest Regional Hydrodynamic Model against Water

Level (2011)

141

Location marked as (6) Pussur River at Akram Point in Figure 4.13

Sibsa River at Akram Point Location marked as (7) in Figure 4.13

Location marked as (15) Pussur River at Mongla in Figure 4.13

Figure 5.13: Validation of Southwest Regional Hydrodynamic Model against Flow

(2011)

142

Location marked as (27) Pussur River at Mongla in Figure 4.13 ppt

Location marked as (30) Sibsa River at Nalian in Figure 4.13 ppt

Figure 5.14: Validation of Southwest Regional Salinity Model against Salinity

(2011)

143

5.5 Analysis of Different Flow Scenario

5.5.1 Temporal and Spatial Variation of Salinity

The model simulation has been conducted from November 2011 to June 2012. Salinity intrusion map based on “Existing flow condition (Scenario-1)”, “Minimum flow condition at Gorai River (Scenario-2)” and “Increased upstream flow condition (Scenario-3)” flow scenarios for maximum salinity are shown in Figure 5.15 to Figure 5.26 (March/April/May, 2012).

The maps developed from the model results have been compared with observed data. The salinity maps reflect the level of salinity in deferent areas during critical situations. The well calibrated hydrodynamic and salinity model is capable to provide a good indication of salinity distribution.

Due to diversion of a considerable fraction of the freshwater discharge from the Padma River (Scenario-3), salinity level in the eastern part of the south-west area remains less saline. As a result, salinity levels in the region decreases from west to east and from south to north.

Scenario-1 describes the existing salinity condition of year 2012 while Scenario-2 looks more worser than existing situation because according to Scenario-2 no it has been considered that Gorai flow has no upstream flow. The model results indicate that in Scenario 3, some of the major rivers such as Gorai-Madhumati, Nabaganga, Chitra, Atai, Bhairab Upper, Rupsa would be saline free and all other rivers will have significant reduction of salinity due to increased upstream flow in Gorai River, Hisna River & Chandana Rivers. Augmentation in the Gorai River flow reduces the salinity in the Pussur River and Sibsa River. The spatial and temporal variation of salinity distribution is presented in Figure 5.15 to Figure 5.26.

144

Figure 5.15: The Salinity Zoning Map for Scenario-1(November, 2011)

145

Figure 5.16: The Salinity Zoning Map for Scenario-2 (November)

146

Figure 5.17: The Salinity Zoning Map for Scenario-3A (November)

147

Figure 5.18: The Salinity Zoning Map for Scenario-1 (March, 2012)

148

Figure 5.19: The Salinity Zoning Map for Scenario-2 (March)

149

Figure 5.20: The Salinity Zoning Map for Scenario-3A (March)

150

Figure 5.21: The Salinity Zoning Map for Scenario-1 (April, 2012)

151

Figure 5.22: The Salinity Zoning Map for Scenario-2 (April)

152

Figure 5.23: The Salinity Zoning Map for Scenario-3A (April)

153

Figure 5.24: The Salinity Zoning Map for Scenario-1 (May, 2012)

154

Figure 5.25: The Salinity Zoning Map for Scenario-2 (May)

155

Figure 5.26: The Salinity Zoning Map for Scenario-3A (May)

156

5.5.2 Salinity Intrusion Line for different flow Scenarios

The successive maximum salinity propagation contour for the month of March, April and May considering 1 ppt and 5 ppt salinity level for “Scenario-1 (dry-2012)” and Scenario-2 (Minimum flow in Gorai River) and increased upstream flow Scenario (Scenario-3) for has been presented in Figure 5.27 to Figure 5.32. The figures show that the upstream propagation of salinity front has been reduced due to the increase of flow at upstream rives. It is mentioned that the increased upstream flow have always been kept in the river systems from January to May to push down propagation of salinity in the upstream.

5.5.3 Comparison of Salinity Intrusion Line

Salinity in the river system of southwest coastal region increases steadily from December through February, reaching maximum in the late April and early May. The successive maximum salinity propagation contour for the month of November and May considering 1 ppt and 5 ppt salinity level for “Scenario-1 (dry-2012)” and Scenario-2 (Minimum flow in Gorai River) and increased upstream flow Scenario (Scenario-3A) has been presented in Figure 5.33 to Figure 5.38. The figures show that the maximum salinity reaches at May which is considerably higher than the month of November. From the analysis of Model result it is observed that around 20% area of Jessore and Narail goes under 1 ppt salinity intrusion whereas Khulna, Satkhira and major part of Bagerhat completely penetrate under 1 ppt salinity intrusion under base condition (Scenario-1) between month of November to May. Moreover, for worst condition (Scenario-2) 50% area Jessore and 80% area of Narail goes under 1 ppt salinity intrusion whereas Khulna, Satkhira Bagerhat completely penetrates under 1 ppt salinity intrusion between months of November to May. Again, for improved flow condition (scenario-3A) 10% area of Jessore and 80% area of Khulna penetrates under 1 ppt salinity intrusion between months of November to May.

157

Figure 5.27: 1 PPT Salinity Intrusion Map for Scenario-1 (Dry, 2011)

158

Figure 5.28: 1 PPT Salinity Intrusion Map for Scenario-2 (Dry)

159

Figure 5.29: 1 PPT Salinity Intrusion Map for Scenario-3A (Dry)

160

Figure 5.30: 5 PPT Salinity Intrusion Map for Scenario-1 (Dry, 2012)

161

Figure 5.31: 5 PPT Salinity Intrusion Map for Scenario-2 (Dry)

162

Figure 5.32: 5 PPT Salinity Intrusion Map for Scenario-3A (Dry)

163

Figure 5.33: 1 PPT Salinity Intrusion Map for Scenario-1 (Nov 2011 & May 2012)

164

Figure 5.34: 1 PPT Salinity Intrusion Map for Scenario-2 (Nov & May)

165

Figure 5.35: 1 PPT Salinity Intrusion Map for Scenario-3A (Nov & May)

166

Figure 5.36: 5 PPT Salinity Intrusion Map for Scenario-1 (Nov 2011 & May 2012)

167

Figure 5.37: 5 PPT Salinity Intrusion Map for Scenario-2 (Nov & May)

168

Figure 5.38: 5 PPT Salinity Intrusion Map for Scenario-3A (Nov & May)

169

5.5.4 Changes in Salinity for different flow Scenarios

Scenario-1 (base line condition, 2011-2012) has been simulated to explore the present salinity intrusion whereas Scenario-2 (Minimum flow through Gorai River) has been simulated to understand the worst condition for salinity intrusion. Finally, Scenario-3A (Flow through Gorai, Hisna and Chandana based on Ganges barrage study), has been conducted to identify the saline free zone with improved upstream flow and river bathymetry condition. Scenario-3B (20% flow increase of Gorai, Hisna and Chandana River) and Scenario-3C (20% flow decrease of Gorai, Hisna and Chandana River) has been conducted for the sensitivity of Ganges flow in salinity intrusion of Southwest region of Bangladesh. The flow condition is described in Table 5.4.

Table 5.4: Description of flow boundary for Scenario-3

Scenario-3B Scenario-3C Scenario-3A Month (20% increase flow) (20% decrease flow) Hisna Gorai Chandana Hisna Gorai Chandana Hisna Gorai Chandana Nov 400 2500 300 480 3000 360 320 2000 240 Dec 300 1000 200 360 1200 240 240 800 160 Jan 231 225 46 277 270 55 185 180 37 Feb 234 227 57 281 272 68 187 182 46 Mar 232 225 77 278 270 92 186 180 62 Apr 242 230 80 290 276 96 194 184 64 May 184 194 44 221 233 53 147 155 35 Jun 300 500 50 360 600 60 240 400 40

To compare the scenario of the maximum salinity propagation contour for the month of May considering 1 ppt and 5 ppt salinity level has been presented in Figure 5.39 and Figure 5.40 respectively. The figures show that the upstream propagation of salinity front has been reduced due to the increase of flow (Scenario-3A) through Ganges connected rives. It is seen from the model result that the 1 ppt and 5 ppt salinity intrusion line remain nearly same further increase of flow, Scenario-3B (20% increase flow from Scenario-3A) or decrease of flow, Scenario-3C (20% decrease flow from Scenario-3A). The relation between salinity intrusion against Flow through Ganges connected river in Rupsa River at khulna and in Pussur River at Mongla is described in Table 5.5 and Figure 5.41.

170

Figure 5.39: Changes of 1PPT Salinity Intrusion Line for different Scenario (Dry, 2012)

171

Figure 5.40: Changes of 5PPT Salinity Intrusion Line for different Scenario (Dry, 2012)

172

Table 5.5: Relation between Ganges Connected River Flow vs. Salinity

Flow Salinity (ppt) SI Flow Scenario Desciption of Flow 3 (m /s) Khulna Mongla 1 Gorain Minimum Flow Only Gorai Flow 59.00 15.33 16.51 2 Base-2012 Gorai Flow 142.00 7.64 13.14 3 GBR -20% increase GBR Gorai+Hisna+Chandana 337.60 1.95 8.07 4 GBR Flow Gorai+Hisna+Chandana 422.00 1.16 6.66 5 GBR +20% increase GBR Gorai+Hisna+Chandana 506.40 0.69 5.47 6 GBR+30% increase GBR Gorai+Hisna+Chandana 548.60 0.54 4.96 7 GBR +40% increase GBR Gorai+Hisna+Chandana 590.80 0.42 4.49

2 3 4 5 6 7

Figure 5.41: Relation between salinity intrusion against Ganges connected river Flow

It is seen from the model result for Ganges Barrage flow the salinity of Pussur River at Mongla is 6.66 ppt. Further increase of 20%, 30%, and 40% flow in GBR flow the salinity at Mongla marginally decrease by 5.47 ppt, 4.96 ppt and 4.49 ppt respectively which is shown in Figure 5.41.

173

5.5.5 Estimation of Spatial area due to different Scenario

1. The allowable salinity level for drinking water is 1 ppt, according to Environmental Quality Standard (EQS) of Bangladesh. If the limiting level of salinity level is 1 ppt than by increasing upstream flow (Scenario-3) 1900 sq. Km land will be saline free from base line (Scenario-1) condition. The saline free land will be 4200 sq. km by increasing upstream flow (Scenario-3A) if it is compared with Scenario-2 (Minimum flow in Gorai River) condition.

2. The 1 ppt salinity line pushes 51.00km & 65.00km from north to south direction by increasing upstream flow (Scenario-3A ) comparing with the scenario-1 and Scenario-2 flow condition respectively which has been presented in Figure 5.42.

3. The 5 ppt salinity line pushes 40.00km & 70.00km from north to south direction by increasing upstream flow (Scenario-3A ) comparing with the scenario-1 and Scenario-2 flow condition respectively which has been presented in Figure 5.42.

4. Scenario-1 looks improved condition then scenario-2 as because after capital dredging (2010-2011) of Gorai River increased upstream flow coming from Gangers River to Gorai River off-take. It has been observed that after three years, dredging has improved the overall situation. The saline free river water could be used directly for agriculture, domestic water supply (with low cost water treatment) and industrial purposes.

5. The pushdown of 1 ppt and 5 ppt salinity contour line are shown in Figure 5.42 and Figure 5.43. Table 5.6 and Table 5.7 shows that major reduction of salinity from the Scenario-1 (base line condition) and Scenario-2 (Minimum flow in Gorai River) to the increased upstream flow Scenario (Scenario-3A).

174

Table 5.6: Typical variation of salinity levels in April

Salinity in PPT Easting Northing SI Location River Name NO Name Scenario-1 Scenario-2 Scenario-3 (Dry, 2012) (Dry, 2012) (Dry, 2012 BTM BTM

1 Atai Arua 456041 532063 1.54-5.65 8.76-12.72 0.00

2 Baleswar CharDoani 491128 446476 1.48-6.30 1.61-6.42 0.95-5.38

Satkhira 3 Betna 411162 507145 4.59-17.28 4.74-17.74 0.00-2.37 (Dhulihar)

4 Bhairab L Bagerhat 479848 503935 2.65-3.80 4.79-5.72 1.11-1.35

5 Bhairab U Fulbari ghat 450845 532106 1.69-5.78 8.52-12.78 0.00

6 Bhola Sharankhola 479758 459977 1.78-3.63 1.94-3.85 1.14-2.68

7 Chitra Narail 450724 556942 0.12-0.91 1.1-5.1 0.00

8 Kanksiali Basantapur 398681 484187 9.30-13.40 9.34-13.51 9.27-12.90

9 Kobadak Kobadak 429313 457165 19.63-22.5 20.18-22.82 18.66-21.58

10 L_Salta Thanibuina 445937 516892 8.4-14.9 13.92-17.46 0.70-6.88

11 Madhumati Jalalbad 470014 552641 0.10-.51 4.40-7.30 0.00

12 Mongla Nulla Mongla 459426 484859 9.8-13.4 13.60-16.36 2.74-5.88

13 Nabaganga M Bardia 467354 552893 0.1-0.79 5.14-7.90 0.00

14 Nabaganga M Hamidpur 455731 542349 0.63-3.53 7.22-11.00 0.00

15 Pussur Chalna 450692 498833 9.24-13.18 14.38-16.67 1.11-4.87

16 Rupsa Khulna 455967 523133 0.63-3.53 10.47-14.42 0.00-0.76

17 Sibsa Nalianala 442174 483454 15-17.70 17.00-19.00 9.47-12.54

Sundor 18 Gangrail 438339 507306 13-15.90 15.68-18.00 5.83-8.21 Mohol

175

Table 5.7: Typical variation of salinity levels in May

Salinity in PPT Easting Northing SI Location River Name NO Name Scenario-1 Scenario-2 Scenario-3 BTM BTM (Dry, 2012) (Dry, 2012) (Dry, 2012

1 Atai Arua 456041 532063 0.82-4.70 10.41-13.52 0.00

2 Baleswar CharDoani 491128 446476 1.18-4.53 1.30-4.62 0.75-3.95

Satkhira 3 Betna 411162 507145 2.18-17.24 2.21-17.69 0.00-2.98 (Dhulihar)

4 Bhairab L Bagerhat 479848 503935 2.24-3.76 3.48-5.63 0.69-1.24

5 Bhairab U Fulbari ghat 450845 532106 1.35-4.80 10.56-13.63 0.00

6 Bhola Sharankhola 479758 459977 1.60-2.80 1.76-2.58 1.00-1.70

7 Chitra Narail 450724 556942 0.18-0.63 1.52-5.85 0.00

8 Kanksiali Basantapur 398681 484187 12.5-17.00 12.49-17.03 12.43-16.75

9 Kobadak Kobadak 429313 457165 21.90-24.7 22.27-24.85 20.82-24.33

10 L_Salta Thanibuina 445937 516892 7.8-15.18 14.43-18.10 0.92-7.46

11 Madhumati Jalalbad 470014 552641 0.00-0.17 5.58-8.85 0.00

Mongla 12 Mongla 459426 484859 4.73-13.00 5.93-16.38 1.69-6.15 Nulla

Nabaganga 13 Bardia 467354 552893 0-0.30 6.88-9.35 0.00 M

Nabaganga 14 Hamidpur 455731 542349 0.22-2.63 8.80-11.98 0.00 M

15 Pussur Chalna 450692 498833 8.92-12.75 13.51-17.14 1.44-5.22

16 Rupsa Khulna 455967 523133 2.16-7.64 12.22-15.14 0.00-0.91

17 Sibsa Nalianala 442174 483454 16.8-18.00 18.51-19.32 9.92-13.31

Sundor 18 Gangrail 438339 507306 14.5-16.10 16.78-18.57 6.35-9.27 Mohol

176

65 km 51 km

Figure 5.42: Pushdown of 1 PPT Salinity Intrusion for different flow Scenario (Dry, 2012)

177

70 km

40 km

Figure 5.43: Pushdown of 5 PPT Salinity Intrusion for different flow Scenario (Dry, 2012)

178

5.5.6 Effect of Trans-boundary Flow on Crop Production

The water is not usable for domestic purposes if salinity is higher than 1ppt, though it is still suitable for crop and livestock agriculture unless salinity exceeds 2ppt. Some freshwater aquaculture is still possible when the salinity is below 4ppt. However, in the south and western part of the study region salinity is higher than 4ppt during the dry season which has intrigued brackish water shrimp farming in Satkhira, Khulna and Bagerhat districts. The following figures (Figure 5.44, Figure 5.45 & Figure 5.46) shows the 2 ppt salinity line for Scenario-1, Scenario-2 & scenario-3. The 2 ppt salinity line pushes from South to north under increased flow condition.

The crop production will increase by 2193 sq km area with increasing upstream flow condition (Scenario-3 to Scenario-1) which is shown in Figure 5.45. The crop production will increase by 4023 sq km area with increasing upstream flow condition (Scenario-3 to Scenario-2) which is shown in Figure 5.46.

5.5.7 Sundarban Salinity Map

1. The Sundarbans Reserve Forest (SRF) is a complex ecosystem comprising the largest diversified mangrove forest of the world. Located in the southwest region of Bangladesh, the area in general is coastal flood plain and crisscrossed by numerous rivers, creeks and depressions. The sustainability of the Mangrove Forest and survival of the ecosystem largely depends upon the circulation of fresh water from the upland.

2. The salinity data has been presented in the form of coloured map to show spatial distribution of salinity at different seasons. It has been observed that the Salinity in the SRF area increases from northeast to southwest. Eastern part of the Sundarbans (Supati, Shwarankhola) remains saline free (less than 1 ppt) during monsoon season and at low salinity level (up to 5ppt) during the dry season. While the salinity level at the northern part of the Sundarbans adjacent to Pussur and Sibsa remain close to 5 ppt at the time of monsoon, during the dry period the salinity in that area goes above 15 ppt.

179

Figure 5.44: 2 PPT Salinity Intrusion for different flow Scenario (Dry, 2012)

180

Figure 5.45: Increased Crop production from Scenario-3A to Scenario-1 (May, 2012)

181

Figure 5.46: Increased Crop production from Scenario-3A to Scenario-2 (May, 2012)

182

3. It may be summarised that the salinity of the rivers in the eastern part (except southern boundary areas) experiences rapid changes with the increase of freshwater flows (through Baleswar, Pussur and Sibsa) after dry season and increases steadily after the in flow reduction in post monsoon period. Decrease of salinity at Mongla and Hiron Point same time before Nalianala represents that the fresh water flow is passing through the Pussur channel at the beginning of May. In contrast, salinity in the western part of the Sundarbans and its upstream countryside areas decreases slowly after the dry season and starts increasing at the later part of the monsoon. Thus the salinity pattern in the eastern part significantly differs with western part.

4. The Kobadak and the Betna are the only sources of fresh water on the western part. These rivers are no longer linked with the Ganges as these were in the past. The flow of these rivers at the upstream is very insignificant and carries only runoff coming from local rainfall. It is also observed that the rainfall in the western part is less than the other areas.

5. From the base line (Scnario-1) salinity intrusion map it is seen that Sundarban Mangrove Forest can be divided by five salinity zone (Lower to Higher Salinity) less than 10 ppt, less than 15 ppt, less than 20 ppt, less than 25 ppt and more than 25 ppt.

6. It is seen that about 60% of the SRF area already suffers from a high salinity stress of more than 20 ppt in critical dry period. The salinity stress of more than 15 ppt in the area of 75% in dry period. Thus only a smaller part of the Sundarbans remains low to medium salinity zone during the dry season. The spatial distribution maps (Figure 5.47 to Figure 5.49) and isohaline map (Figure 5.50) should be highly useful for the forest department to manage SRF area for vegetation and wildlife.

7. The model result shows that increase of upstream flow (Scenario-3A) in the western rivers may reduce the salinity in the western past (Kobadak-Betna system) of the Sundarbans.

183

8. In this study the 2011-2012 hydrological year has been considered as base model situation (Scenario-1) when a minimum flow of about 61 m3/s flow under the Gorai Railway bridge during February-March. The result of salinity distribution from the base model has been presented in Figure 5.47. The Figure shows that further increase of salinity in the Northeastern part of sundarban in Scenario-2 (Figure 5.48). The 10-15 ppt zone of Northeastern part near Mrigamari shifted to 15-20 ppt zone due to Minimum flow at Gorai River has been considered in Scenario-2. The increased upstream flow condition: Scenario-3 (Figure 5.49) shows that the further reduction of salinity in the northeast part as compared to Scenario-1. The 10-15 ppt zone of Northeastern (Supati to Mongla) replaced by 5-10 ppt salinity zone. The salinity distribution map for Scenario-2 shown in Figure 5.49 indicates prominent reduction in salinity in the northeastern part of the Sundarbans with a visible effect at Nalianala in the Sibsa. The 15-20 ppt salinity zone at Nalianala in the Sibsa River replaced by 10-15 zone of salinity. Salinity at Mongla reduces by more than 5 ppt and at Nalianala the reduction is around 6 ppt and at Harintana, Herbaria and Nalianala lies in the 8-12 ppt zone.

9. The isohaline map (Figure 5.50) shows that no significant change in Western part of Sunderban due to increase upstream flow. However, the movement of 10 ppt, 15 ppt line is significantly shifted from Northeast to Northwest part of Sunderban. Furthermore, the 20 ppt salinity line shifted from North to South part of Sundarban Mangrove Forest.

Limitations of the Salinity Model

The computation of the Salinity Model is highly sensitive to the computation of discharge (flow volume) in the Hydrodynamic (HD) Model. Due to absence of intensive salinity data collection of whole river networks in the Southwest area the salinity Model has got some limitation to reproduce the exact level of salinity as observed at different locations. However, the relative level provided in the model could be good enough to understand a reasonable distribution pattern which helps to forecast the salinity distribution under increased flow scenario.

184

Nalianala

Mrigamari

Harbaria

Bogi Kaikhali

Harinta na

Supati

Notabaki

Kachikhali

Hiron Point

Manderbaria

Figure 5.47: Spatial Distribution of Salinity of Sundarban Mangrove Forest during

May, 2012 (Scenario-1)

185

Nalianala

Mrigamar i

Harbaria

Bogi Kaikhali

Harinta na

Supati

Notabaki

Hiron Point

Manderbaria

Figure 5.48: Spatial Distribution of Salinity of Sundarban Mangrove Forest during

May (Scenario-2)

186

Nalianala

Mrigamari

Harbaria

Bogi Kaikhali

Harinta na

Supati

Notabaki

Hiron Point Manderbaria

Figure 5.49: Spatial Distribution of Salinity of Sundarban Mangrove Forest during

May (Scenario-3A)

187

Nalianala

Mrigamari

Harbaria

Bogi Kaikhali

Harinta na

Supati

Notabaki

Hiron Point

Manderbaria

Figure 5.50: Impact of Increasing Fresh Water Flown in Sundarban Mangrove Forest

during May, 2012

188

5.5.8 Establishment of Hydrograph under Increased Flow Scenario

One of main objective of this study is to apply one-dimensional hydrodynamic model in order to evaluate the trans-boundary flow that will be sustain the major rivers in the Southwestern region of Bangladesh. The Hydrograph of Ganges connected river represent the trans-boundary flow through Ganges River. It is assume that the Ganges River flow distribute in Gorai, Madhumati and Chandana River. The Gorai, Madhumati, Mathabhanga and Chandana Rivers are now disconnected from the Ganges River in the dry season due to river bed aggradations at the mouth of their off-takes. All those rivers become dry or nearly zero flow under existing river condition.

In this study the model simulation has been conducted from January to June. Flow Hydrographs of the Southwest major rivers for Scenario-3A (increased flow at upstream rivers) are shown in Figure 5.51 to Figure 5.53 (January to May). The river system and hydrographs under increased flow condition are described in below.

Hisna-Mathabhanga-Bhairab-Kobadak System

Currently, there are discontinuities between different river reaches at several locations in Hisna-Mathabhanga-Bhairab-Kobadak system. Mathabhanga- Nabaganga, Mathabhanga-Upper Bhairab and Kobadak Bhairab have no existing connection. Presently, Hisna is totally cutoff from the Ganges. Mathabhanga receives flood spills from Ganges only during high stages in the Ganges. In rest of the year, Mathabhanga remains dry or carries rainfall runoff.

Hisna River need certain flow (Scenario-3A) during dry season for distribution in the Betna, Kobadak, Upper Bhairab, Nabaganga, Chitra and Mukteswari Rivers. The hydrograph of Hisna-Mathabhanga-Bhairab-Kobadak System for increase flow condition (Scenario-3A) are shown in Figure 5.51.

Gorai-Madhumati System

The Gorai is the major right bank off-take from the River Ganges within Bangladesh. For the last 20 years or so, the dry season flow (November- May) in the Gorai River has been decreasing. The environmental impact of this decrease is very serious in

189

terms of increased salt-water intrusion in the coastal area, around Khulna, and on the World‟s largest mangrove forest area, the Sundarbans. After implementation of three dredging seasons, 1998, 1999 and 2000, the dredging increased the water flow in the Gorai River, restoring the fish population and allowing year-round navigation. Meanwhile no step was taken as maintenance dredging, which could keep the Gorai sustainable and allow substantial amount of flow during the dry season. In view of such observation, Bangladesh Water Development Board (BWDB) has again taken up steps to carry out dredging from the off-take, for a length of 30 km of the river. The Government approved a DPP under the name “Gorai River Restoration Project Phase-II (GRRP-II)” for this purpose whereas during 1999 to 2000, GRRP, Phase I was conducted.

During dry season, the Gorai intake is almost cut off from the Ganges and there is no freshwater flow through this river. As a result salinity water comes through the major rivers namely Pussur, Jamuna, Malancha and Sibsa in the western part of southwest region and increases the salinity level in the dry period. In 2012, Gorai dredging restored the dry season flow temporarily into the area decreasing the salinity level slightly. Downstream reaches of the Gorai system is highly influenced by the tidal effect. Gorai system requires more flow (Scenario-3A) during dry season for distribution in the Modhumati, Baleswar, Naboganga (lower part), Bhairab (lower), Rupsa, Bhola, Sibsa Rivers. The hydrograph of Gorai-Madhumati system for increased flow condition (Scenario-3A) are shown in Figure 5.52.

Chandana-Barasia-Kumar System

The Chandana River, which used to be an important distributary of the Ganges, but at present its off-take is silted up. Chandana system requires definitely need certain flow (Scenario-3A) flow during dry season for distribution in the Chandana, Barasia Kumar Rivers. The hydrograph for increased flow Scenario for Chandana-Barasia- Kumar System are shown in Figure 5.53.

190

Figure 5.51: Map showing flow hydrograph for increased flow Scenario (Hisna-

Mathabhanga-Bhairab-Kobadak System)

191

Figure 5.52: Map showing flow hydrograph for increased flow Scenario (Gorai- Madhumati System)

192

Figure 5.53: Map showing flow hydrograph for increased flow Scenario (Chandana- Barasia-Kumar System)

193

CHAPTER SIX

CONCLUSIONS AND RECOMMENDATIONS

6.1 General

Salinities in the Bangladesh coast are dependent on the volumes of freshwater flows discharging from upstream. Average salinity concentrations at the coast are higher in the dry season than in the monsoon, due to lack of huge freshwater flow from the upstream.

6.2 Conclusions

The major findings of this study are stated as follows:

1. The salinity generally increases almost linearly from October (post-monsoon) to the Late May (pre-monsoon). The salinity levels are at the minimum at the end of the wet season, usually during end-September or early-October.

2. Salinity level remains higher in the western part of the region than the eastern part. This is due to the fact that Gorai is the only significant upstream freshwater source in this part. Gorai dredging increased the dry season flow temporarily into the area decreasing the salinity level slightly.

3. Due to diversion of a considerable fraction of the freshwater discharge from the Padma River, salinity level in the eastern part of the south-west area remains less saline. As a result, salinity levels in the region decreases from west to east as well as from south to north.

4. The model results indicate that in Scenario-3A (increased upstream flow condition), some of the major rivers such as Gorai-Madhumati, Nabaganga, Chitra, Atai, Bhairab Upper, Rupsa would be saline free and all other rivers will have significant reduction of salinity due to increased upstream flow in Gorai River, Hisna River & Chandana River. Augmentation in the Gorai River flow reduces the salinity in the Pussur River and Sibsa River.

194

5. From the sensitivity analysis of Ganges connected River flow by model results indicate that with Scenario-3B (20% flow increase of Gorai, Hisna and Chandana River) and Scenario-3C (20% flow decrease of Gorai, Hisna and Chandana River) the saline free zone similar to Scenario-3A.

6. If the limiting level of salinity level is 1 ppt than by increasing upstream flow (Scenario-3), 1900 sq. Km land will be saline free from 2012 dry period (Scenario-1) condition. The saline free land will be 4200 sq. km if it is compared with Scenario-2 (Minimum flow in Gorai River) condition.

7. The 1 ppt salinity line pushes 51.00km & 65.00km from North to southern direction by increasing upstream flow (Scenario-3) with comparing the scenario-1 and Scenario-2 flow condition respectively. The 5 ppt salinity line pushes 40.00km & 70.00km from North to southern direction by increasing upstream flow (Scenario-3) with comparing the scenario-1 and Scenario-2 flow condition respectively.

8. The crop production will increase by 2193 sq km area with increasing upstream flow condition (Scenario-3 to Scenario-1). The crop production will increase by 4023 sq km area with increasing upstream flow condition (Scenario-3 to Scenario-2).

9. From the present salinity intrusion map it is seen that Sundarban Mangrove Forest can be divided by five salinity zone (Lower to Higher Salinity) less than 10 ppt, less than 15 ppt, less than 20 ppt, less than 25 ppt and more than 25 ppt.

10. It is seen that about 60% of the SRF area already suffers from a high salinity stress of more than 20 ppt in critical dry period. The salinity stress of more than 15 ppt in the area of 75% in dry period. Thus only a smaller part of the Sundarbans remains low to medium salinity zone during the dry season.

11. The model result shows that increase of upstream flow (Scenario-3) in the western rivers may reduce the salinity in the western past (Kobadak-Betna system) of the Sundarbans.

195

12. The isohaline map of Sundarban shows that no significant change in Western part of Sunderban due to increase upstream flow. However, the movement of 10 ppt, 15 ppt line is significantly shifted from Northeast to Northwest part of Sunderban. Moreover, the 20 ppt salinity line shifted from North to South part of Sundarban Mangrove Forest.

6.3 Recommendations

Based on the current study some recommendations for future study can be as follows:

1. The precision of salinity model may be not that accurate because of excluding the minor rivers and creeks from the river network. It can be improved if new bathymetric surveyed data incorporate in the model.

2. To understand the spatial and temporal variation of salinity precisely in the Sundarban mangrove Forest, measured salinity data within the area will be required.

3. It would have been better if the results of this present study obtained by the use of the mathematical modeling tool MIKE11 & MIKE21 FM could be cross checked by another modeling tool.

196

REFERENCES

1. Akhter, S., Hasan, M.M., Khan, Z.H. (20-24 Feb. 2012), “Impact Of Climate Change On Saltwater Intrusion In The Coastal Area Of Banglades”, 8thInternational Conference On Coastal And Port Engineering In Developing Countries Copedec 2012, IIT Madras, Chennai, INDIA.

2. Alam, R. Q, Sarker, S., Rahman, M.M. (March 10-12, 2011), “Salinization of Inland Water System of Coastal Areas of Bangladesh due to Climate Change”, International Conference on Environmental Technology& Construction Engineering for Sustainable Development, ICETCESD-2011, SUST, Sylhet, Bangladesh.

3. Barlow, P.M. (2003), “Groundwater in Freshwater-Saltwater Environments of the Atlantic Coast”. U.S. Geological Survey Circular: 1262.

4. Barlow, P.M. and Reichard, E.G. (2003), “Saltwater intrusion in coastal regions of North America”, Hydrogeology Journal, Volume 18, Issue 1, pp 247-260.

5. BWDB (September, 2012), “Ganges Barrage Study Project (Feasibility Report – Main Report)”.

6. BWDB (1988), “Morphological Features of the Major Rivers of Bangladesh”, Part 1, Bangladesh Water Development Board, Dhaka.

7. Delft Hydraulics/DHI (1996), “River Survey Project”, FAP 24, Main Volume, Final Report.

8. DHI Water and Environment (2014), “MIKE 21 FLOW MODEL FM”, Hydrodynamic and Transport Module Scientific Documentation, Danish Hydraulic Institute.

9. DHI Water and Environment (2014), “MIKE 11, Reference Manual”, Danish Hydraulic Institute.

10. DHV-Haskoning & Associates (2001), “The Feasibility Study and Detailed Engineering Design of the Gorai River Restoration Project (GRRP)”.

197

11. EGIS (2001), “Envronmental & Social Impact Assessment of Goari River Restoration Project”, Bangladesh Water Development Board.

12. Gillibrand, P. A. and Balls, P. W. (1998), “Modelling Salt Intrusion and Nitrate Concentrations in the Ythan Estuary”. Estuarine, Coastal and Shelf Science (1998) 47, 695–706, Article No. ec980393

13. Halcrow (1993), “Southwest Area Water Resources Management Project (FAP4)”.

14. Hoque, A., Sarkar, M.S.K.A , Khan, S.A.K.U., Moral, M.A.H and Khurram, A.K.M. (2006), "Present Status of Salinity Rise in Sundarbans Area and its Effect on Sundari (Heritiera fomes) Species," Research Journal of Agriculture and Biological Sciences, 2(3), pp. 115-121.

15. Islam, S.N. and Albrecht, G. (2011), “Water salinity investigation in the Sundarbans rivers in Bangladesh”, Int. J. Water, Vol. 6, Nos. 1/2, pp.74–91.

16. Islam, S. N., Gnauck, A. (24 - 27 April 2007), “Salinity Intrusion due to Fresh Water Scarcity in the Ganges Catchment: A Challenge for Urban Drinking Water and Mangrove Wetland Ecosystem in the Sundarbans Region, Bangladesh”, 6th Word Wide Workshop for Young Environmental Scientists, Domaine De Chérioux- Vitry sur Seine- France.

17. Islam, M.A. (1978), "The Ganges-Brahmaputra River Delta", Journal of University of Sheffield Geological Society (l).

18. IWM (2014), Morphological Mathematical Modelling for Planning, Design, Monitoring and Quality Control of Dredging for Gorai River Restoration Project, Phase-II

19. IWM (2013), “Union-wise Flood Mapping for Flood-Prone Areas, storm surge and salinity zoning map for coastal zone (baseline condition) and modeled scenarios following climate change of the same refereeing to the baseline milestone events to facilitate Community Risk Assessment (CRA) having climate sensitive decision”, Revised Final Report.

198

20. IWM (2010), “Strengthening the Resilience of Water Sector in Khulna to Climate Change”, Final Report.

21. IWM, 2003, “Sundarban Biodiversiiity Conservation Project Surface Water Modelliiing”, Final Report.

22. Karim, Z. , Hussain, S and Ahmed, M. (1990), "Salinity Problems and Crop Intensification in the Coastal Regions of Bangladesh", BARC Soil Publication No.33, p. 63.

23. Katebi, M.N.A. (2001), "Sundarbans and forestry," in Haider, R. (Ed.): Cyclone ‟91 – An Environmental and Perceptional Study, BCAS, Dhaka, p. 79–100.

24. Khan, Z.H. and Kamal, F.A. (2014), “The Ganges coastal zone - current condition, future projections on Water Availability”, The 4th International Rice Congress (IRC2014), Bangkok International Trade and Exhibition Centre (BITEC), Bangkok, Thailand.

25. Kijne, J.W. (1996), “Water and salinity balances for irrigated agriculture in Pakistan”, Research, Report 6. Colombo, Sri Lanka: International Irrigation Management Institute (IIMI).

26. Lacerda, D.D.L. (2001), “Mangrove Ecosystems Functions and Management”, Springer-Verlag Berlin, Germany.

27. Mondal, M.K. (1997) "Management of soil and water resources for higher productivity of the coastal saline rice lands of Bangladesh," University of the Philippines, Los Banos, Philippines.

28. PDO-ICZMP (2004), "Living in Coast-Problems, Oppotinities and Challenges," Program Development Office for Intregated Coastal Zone Management Plan.

29. Rahman, M.M. and Bhattacharya, A. K. (October 2006), “Salinity Intrusion and its Management Aspects in Bangladesh”, Journal of Environmental Hydrology, The Electronic Journal of the International Association for Environmental Hydrology, Volume 14, Paper 14, , On the World Wide Web at

199

30. Rao, V. M. (30May-2nd June 2006), “Treaty our rivers”: A co-operative international river basin managemnet effect incorporating ecosystemic and economic parameters. Proceedings of III International Symposium on transboundary waters management-Overcoming water management boundaries., Cuidad Real, Spain.

31. Rahman, M. and Ahsan, M. (2001), "Salinity constraints and agricultural productivity in coastal saline area of Bangladesh" .

32. Seal, L. and Baten, M.A. ( January 2012), “Salinity Intrusion in Interior Coast: A New Challenge to Agriculture in South Central part of Bangladesh”.

33. Shamsuddoha, M. and Chowdhury, R. K. (2007), "Climate Change Impact and Disaster Vulnerabilities in the Coastal Areas of Bangladesh," Coast Trust, p. 32.

34. Siddiqi, N.A. (2001), "Mangrove Forestry in Bangladesh”, Institute of Forestry and Environmental Sciences, Nibedon Press Limited, Chittagong, Bangladesh, p. 201.

35. SRDI (2003), "Soil and Water Salinity in the Coastal Area of Bangladesh" Soil Resource Development Institute (SRDI), Dhaka.

36. SRDI (1997), “Soil salinity map of Bangladesh” Soil Resource Development Institute (SRDI), Dhaka.

37. Terray, L., Corre, L., Cravatte, S., Delcroix, T., Reverdin, G., and Ribes, A. (2012), “Near-Surface Salinity as Nature‟s Rain Gauge to Detect Human Influence on the Tropical Water Cycle”, Journal of Climate, American Meteorological Society, Vol. 25, pp.958-977.

38. Thatcher, M.L. & Harleman, D.R.F. (1972), “A mathematical model for the prediction of unsteady salinity intrusion in estuaries”, Parsons Laboratory Technical Report 144, Massachusetts Instituteof Technology, Cambridge, USA.

39. Thatcher, M.L. & Harleman, D.R.F. (1981), “Long-term salinity calculation in Delaware estuary”, Journal of Environmental Engineering Division EE1, 11–27.

200

40. TWDB (July 11, 2011),”Comparison of Two Hydrology Datasets, as Applied to the TxBLEND Model, on Salinity Condition in Nueces Bay”, Alternate Hydrology Technical Memo.

41. Uddin, M.S. , Khan, M.S.I., Talukdar, M.M.R., Hossain, M.I. and Ullah, M.H. (2012), “Seasonal Variation of Soil Salinity In Coastal Areas of Bangladesh”, International Journal of Environmental Science, Management and Engineering Research Vol. 1 (4), pp. 172-178, Available online at http://www.ijesmer.com

42. Uddin, M.S. , Khan, M.S. I., Talukdar, M.M.R., Hossain, M. I. and Ullah, M.H. (December, 2011), “Climate Change and Salinity in Bangladesh: Constraints and Management Strategy for Crop Production”, Rajshahi University journal of environmental science, Vol.: 1, 13-20, ISSN 2227-1015.

43. Uddin, M. N. and Haque. A. (01-07, March 2010), “Salinity Response in Southwest Coastal Region of Bangladesh Due to Hydraulic and Hydrologic Parameters”, Int. J. Sustain. Agril. Tech. 6(3): An online Journal of “G-Science Implementation and Publication”, website: www.gscience.net or www. gurpukur.com, Title code: IJSAT/ 100301.

44. Uddin, M. (2005), "Changes of Environmental Parameters due to Salinity Intrusion in the Southwest Region of Bangladesh," Institute of Water and Flold Management, BUET, Dhaka.

45. WARPO (2001), “Options for Ganges Dependent Area (OGDA)”, Main Report.

46. Wahid, S.M. (1995), "Hydrological Study of the Sundarban", UNDP/FAO Project BGD/84/056, Department of Forest, Dhaka, Bangladesh.

47. Woods, J. (2010), “Surface Water Discharge and Salinity Monitoring of Coastal Estuaries in Everglades National Park, USA, in Support of the Comprehensive Everglades Restoration Plan”, Conference publication: 3rd International Perspective on Current & Future State of Water Resources & the Environment (online Journal), Chennai, India, American Society of Civil Engineers, 10 p

48. Xinfeng, Z. and Jiaquan, D. (2010), “Affecting Factors of Salinity Intrusion in

201

Pearl River Estuary and Sustainable Utilization of Water Resources in Pearl River Delta”, K. Fukushi et al. (eds.), Sustainability in Food and Water: An Asian Perspective, Alliance for Global Sustainability Bookseries, Vol.: 18, 11-17, ISBN 978-90-481-9914-3.

49. Yu, L. (18 October 2011), “A global relationship between the ocean water cycle and near-surface salinity”, Journal of Geophysical Research, Vol. 116, C10025, DOI: 10.1029/2010JC006937.

202