Groundwater is the world’s most readily available source of freshwater, hence the importance of aquifers. However, groundwater resources are prone to pollution in the wake of anthropogenic influences, over-exploitation and climate change related activities. Coastal aquifers are faced with added unique challenges of seawater intrusion. The coastal aquifer of Mombasa, home to East Africa’s busiest seaport is not an exception to these challenges. This study investigated the groundwater flow, quality and vulnerability to seawater intrusion of the coastal aquifer of Mombasa North Coast. This is a 74.2 km2 region bounded by the Indian Ocean on the East, by creeks on the North and South and high elevated hills on the West. The hydrogeological characteristics, salinity and extent of seawater intrusion of the aquifer were assessed using statistical and geospatial methods. The statistical methods include correlation coefficients, cross plots and Piper plots. GALDIT overlay index was applied in assessing the vulnerability of the aquifer while groundwater flow and solute transport were simulated with the aid of MODFLOW, MT3D and SEAWAT packages. Field data such as static groundwater levels and water quality parameters were collected at pre-monsoon, rainy season, and post-monsoon of 2016. This was followed by laboratory analysis for Na, K, Mg, Ca, Cl, HCO3 and SO4 concentrations in the obtained water samples. The results show the aquifer is shallow and unconfined with groundwater heads ranging from -1 to 33m above mean sea level. The EC and TDS values showed near perfect correlation relationships and were generally high, as over 94% of the water samples exceeded WHO drinking water limit. The groundwater pH was slightly alkaline but could be slightly acidic in the rainy season due to groundwater recharge from rainfall. There is a wide spatial variation in water quality parameters in the aquifer. The groundwater salinity varies with seasons and groundwater recharge heavily influences the salinity of the groundwater. The aquifer largely experiences a moderate impact and moderate vulnerability to seawater intrusion. However, vulnerability is higher in the dry season. Finally, the direction of groundwater flow is predominantly towards the north-eastern and southern part of the study area. In view of the findings made, management and modification of the pumping scheme are recommended to help prevent further deterioration of the groundwater quality.
TABLE OF CONTENTS
DEDICATION
ACKNOWLEDGEMENTS
TABLE OF CONTENTS
LIST OF TABLES
LIST OF FIGURES
LIST OF APPENDICES
LIST OF ABBREVIATIONS AND ACRONYMS
ABSTRACT
CHAPTER ONE- INTRODUCTION
1.1 Background of study
1.2 Statement of problem
1.3 Justification
1.4 Objective of Study
1.4.1 General Objective
1.4.2 Specific Objectives
1.5 Research Questions
1.6 Scope of Study
1.7 Limitation of Study
CHAPTER TWO- LITERATURE REVIEW
2.1 Introduction
2.2 Coastal Aquifers
2.2.1 Challenges of Coastal Aquifers
2.2.2 Seawater Intrusion
2.2.3 Review of Coastal Aquifer Research
2.3 Application of GIS and Groundwater Modelling in Coastal Aquifer Studies
2.3.1 Geographic Information Systems (GIS)
2.3.2 Groundwater Modelling
2.3.3 Integrating of GIS with Groundwater Modelling
2.4 GALDIT Overlay Index
2.5 MODFLOW, MT3D, SEAWAT and GMS
2.5.1 Brief Description on MODFLOW
2.5.2 Brief description of MT3D
2.5.3 A brief description of SEAWAT
2.5.4 A brief description of GMS
CHAPTER THREE- GENERAL METHODOLOGY
3.1 Introduction
3.1.1 Study Area
3.2 Materials and Methods
3.2.1 Secondary Data Collection
3.2.2 Field Data Collection
3.2.3 Laboratory Data Analysis
CHAPTER FOUR
4.1 Hydrogeological characteristics of the aquifer
4.1.1 Results and Discussion
4.2 Groundwater salinity assessment of the aquifer.
4.2.1 Results and Discussion
4.3 Seawater intrusion assessment of the aquifer
4.3.1 Results and Discussion
CHAPTER FIVE Spatial Vulnerability Mapping of the Coastal Aquifer to Seawater intrusion
5.1 GALDIT factors for the coastal aquifer of the study area
5.1.1 Groundwater Occurrence
5.1.2 Aquifer hydraulic conductivity
5.1.3 Height of ground water above the mean sea level
5.1.4 Distance from saltwater body
5.1.5 Impact of existing status of seawater intrusion
5.1.6 Thickness of the aquifer being mapped
5.2 Generalised workflow for the vulnerability analysis
5.2.1 Data collection and sorting
5.2.2 Data processing
5.2.3 Overlay and spatial analysis
5.2.4 Cartography (final vulnerability maps)
5.3 Results and Discussion
CHAPTER SIX Groundwater flow and solute transport simulation of the study area.
6.1.1 Importing GIS Files into GMS (loose coupling)
6.1.2 Building the Conceptual Model for the Aquifer
6.1.3 Steady State Flow Simulation
6.1.4 Transient State Flow Simulation
6.1.5 Model Calibration
6.1.6 Solute Transport Simulation (MT3DMS and SEAWAT)
6.2 Results and Discussion
6.2.1 Steady State Flow Simulation
6.2.2 Transient State Flow Simulation and Calibration
6.2.3 Solute Transport Simulation
CHAPTER SEVEN- CONCLUSIONS AND RECOMMENDATIONS
7.1 Conclusions
7.2 Recommendations
REFERENCES
APPENDICES
DEDICATION
To God Almighty and my ever supportive parents
ACKNOWLEDGEMENTS
My deepest gratitude goes to God Almighty, for the strength and wisdom from the start to the completion of this thesis. I wish to extend my profound gratitude to African Union commission who awarded me this scholarship to pursue a masters’ degree for two years. All thanks to every individual directly or indirectly linked with this thesis, my two supervisors Prof Maurice Nyadawa and Dr Maurice K’Orowe – your support and guidance cannot be quantified. You were more than Supervisors; you are Mentors indeed. I feel indebted to every other person who played one role or the other in guiding and sharpening my research skills; Prof P. Home, Prof Kanali amongst others. My heartfelt appreciation also goes to Rose Waswa Malot of Regional Centre for Mapping of Resources for Development (RCMRD) for her technical guidance in the application of GIS. Big thanks to John Kimathi of Chemistry department (JKUAT) for his priceless support. I am equally grateful to the entire staff of PAUSTI, JKUAT, AICAD and WRMA Mombasa Office for their direct and indirect supports towards making the research a success. My dear colleagues in PAUSTI, words will fail me in expressing how privileged I feel to have had you as colleagues. We have learned a lot from one another, I look forward to meeting us in great places contributing our quota to the growth of humanity, Africa and our respective countries. To my wonderful family back home, my dear parents Mr & Mrs Idowu; Engr A.R Azeez; Mrs B Olaniyan; and many others whose names are not included you are the best anyone could ever wish for. Thank you all for your moral support, encouragements from time to time and your prayers. My appreciation will be incomplete without mentioning my one and only Sister Olubunmi Idowu and my ever loyal and special friend Otemuyiwa Ayomide for their contributions to the quality of this work in terms of proof reading and editing. I love you all!!!
LIST OF TABLES
Table 2-1: Standardised weightages for GALDIT factors
Table 3-1: Secondary data obtained and their sources
Table 3-2: Summary of the field data obtained and the collection techniques
Table 3-3: Techniques used for the laboratory analysis of water samples
Table 3-4: Software packages used
Table 4-1: Concentration of the major parameters in the groundwater samples
Table 4-2: Statistical analysis of the parameters
Table 4-3: Correlation coefficients for the groundwater parameters in the month of March
Table 4-4: Normalised values for the cations and anions
Table 4-5: Spatial coverage of the EC classification in percentages
Table 4-6: Seawater intrusion indices
Table 4-7: Correlation for Salinity Indices (March)
Table 4-8: Spatial coverage of the Cl/HCO3 index classifications in percentages
Table 5-1: Ratings for GALDIT parameter G
Table 5-2: Ratings for GALDIT parameter A
Table 5-3: Ratings for GALDIT parameter L
Table 5-4: Ratings for GALDIT parameter D
Table 5-5: Ratings for GALDIT parameter I
Table 5-6: Ratings for GALDIT parameter T
Table 5-7: GALDIT Index computation
Table 5-8: Percentage changes in vulnerability classes between pre-monsoon and rainy season
LIST OF FIGURES
Figure 2-1: Illustration of the fresh and seawater interface of a coastal aquifer
Figure 2-2: Unconfined coastal aquifer with non-equilibrium conditions between freshwater and saltwater
Fig 2-3: Logic diagram for the development of a mathematical model
Figure 2-4 a: Finite- difference grid configuration for the study of an Aquifer
Figure 2-4 b: Finite- Element grid configuration for the study of an Aquifer
Fig 2-5: Component of Groundwater flow approach
Figure 3-1: Satellite Imagery of the Study Area
Figure 3-2: Map of the study area
Figure 4-1: Flowchart for the general steps in assessing hydrogeological characteristics, salinity and SWI
Figure 4-2: Trilinear plots (piper) of the central catchment of Bribie Island
Figure 4-3: The diamond for the interpretation of the groundwater characteristics
Figure 4-4: Hydrogeological map of the study area showing the GW heads above MSL
Figure 4-5: Groundwater table in the coastal aquifer
Figure 4-6: Piper plot for the water samples taken in the study area
Figure 4-7: EC variation across the study area
Figure 4-8: TDS variation across the study area
Figure 4-9: Spatial representation the Cl/HCO3 ratios
Figure 4-10: TDS Concentration against Molar Ratios of (a. HCO3/Cl; b. Ca/Na)
Figure 5-1: General methodology for the vulnerability analysis
Figure 5-2: Vulnerability maps of the study area to seawater intrusion
Figure 6-1: General steps involved in the groundwater flow and solute transport simulation
Figure 6-2: GMS software interface showing the GIS and GMS data layers
Figure 6-3: The hydraulic boundary conditions of the study area
Figure 6-4: Steady state simulated groundwater heads (m) and flow direction
Figure 6-5: Differences between the Observed and Computed heads after calibration
Figure 6-6: Cross plot for Observed heads vs. computed heads
Figure 6-7: Simulated NaCl solute concentrations in mg/l
Figure 6-8: Simulated NaCl solute concentrations in mg/l
LIST OF APPENDICES
Appendix I- Groundwater heads above mean sea level
Appendix II- Graphs showing the variations in the groundwater parameters in across the season
Appendix III- Cross plots showing the relationships between EC/TDS and the cations & anions (pre-monsoon)
Appendix IV- Table of conversion from mg/l to meq/l
Appendix V- NaCl variation in the groundwater across the study area for pre-monsoon, rainy season and post-monsoon
Appendix VI- Correlation coefficients for salinity induces for the rainy season and post-monsoon
Appendix VII- The GALDIT factors
Appendix VIII- List of publication and conference proceeding
LIST OF ABBREVIATIONS AND ACRONYMS
illustration not visible in this excerpt
ABSTRACT
Groundwater is the most readily available source of freshwater in the hydrologic cycle, hence the importance of aquifers. However, these groundwater resources are prone to pollution in the wake of anthropogenic influences, over-exploitation and climate change related activities. Coastal aquifers are faced with additional unique problems of seawater intrusion which may be aggravated by climate change related challenges like sea level rise and coastal flooding. The aquifer of Mombasa, a coastal city, home to East Africa’s busiest seaport is not an exception to these challenges. This study investigated the groundwater flow, quality and vulnerability to seawater intrusion of the coastal aquifer of Mombasa North Coast. This is a 74.2 km2 region bounded by the Indian Ocean on the East, by creeks on the North and South and high elevated hills on the West. The hydrogeological characteristics, salinity and extent of seawater intrusion in the study area were assessed using statistical and geospatial methods. The statistical methods include the use of correlation coefficients, cross plots and Piper plots. GALDIT overlay index was used to assess the vulnerability of the study area with the aid of ArcGIS while groundwater flow and solute transport were simulated with the aid of MODFLOW, MT3D and SEAWAT packages. In addition to secondary data obtained, three phases of field data such as static water levels and water quality parameters from boreholes/shallow wells were collected at pre-monsoon, the peak of rainy season, and post-monsoon in 2016. This was followed by laboratory analysis for Na, K, Mg, Ca, Cl, HCO3 and SO4 concentrations in the water samples obtained from the field. The results show that the study area is characterised by shallow unconfined aquifer with groundwater heads ranging from -1 to 33m above mean sea level irrespective of the seasons. The EC and TDS values were observed to have near perfect correlations with each other and generally high, as over 94% of the water samples exceeded WHO drinking water limit of 750 µS/cm and 500mg/l respectively. The pH of the groundwater was slightly alkaline but could be slightly acidic in the rainy season. Over 90% of the water samples had pH values in the range of 6.5 and 8.5 the WHO’s acceptable drinking water limits. The groundwater is generally marine in nature while there is a wide variation of the EC, TDS, NaCl, Na, K and Cl values within the aquifer. Over 50% of the water samples had Na concentrations exceeding the WHO and Kenya drinking water limits of 200mg/l. Hence the groundwater is generally unfit for direct drinking. The salinity of the groundwater varies with seasons and groundwater recharge heavily influences the salinity of the groundwater. The aquifer is largely experiencing a moderate impact of seawater intrusion depending on the season with only a few regions experiencing injurious impacts. This might be because the current groundwater abstraction rates are easily compensated by high annual rainfall >1000mm. Over 50% of aquifer’s coverage experiences moderate vulnerability to seawater intrusion, however, the aquifer is more vulnerable in the dry season than the wet season. The direction of groundwater flow is predominantly towards the north-eastern and southern part of the study area. Finally, solute transport simulation shows that concentrations of NaCl in the groundwater slightly reduced from the centre of the study area outwards from June (peak of rainy season) to September (post-monsoon). In view of the findings made, management and modification of the pumping scheme are recommended. This includes encouraging the digging of wells and boreholes to be as shallow as possible. The recommendations will help sustain the current quality of the groundwater and prevent further pollution of the groundwater.
CHAPTER ONE- INTRODUCTION
1.1 Background of study
Coastal areas all over the world experience comparatively higher net human migration than the inland areas, due in part to their functionality as hubs of commerce, industrialisation and international trade. In Africa, of the six megacities expected to emerge by 2030, three (50%) of them are coastal cities (WUP, 2014). Currently, population densities of coastal areas are about three times the global average (Small & Nicholls, 2003). Coastal areas only cover 5% of the earth’s surface, yet 50 – 70% of the human population in the globe is estimated to occupy these coastal zones (Benoit et al., 2007). The resulting effects of these realities are increased water usage and potential over-exploitation of coastal groundwater (Darneault & Godinez, 2008). Groundwater is a very vital source of freshwater to humans for domestic, industrial and agricultural purposes. In fact, over 90% of the world’s readily available freshwater is found as groundwater (Boswinkel, 2000; UNEP, 2008). Hence, groundwater is the principal source of freshwater provisioning in many parts of the world, especially in sub-Saharan Africa where water recycling and reuse is almost none existent.
The Water Exploitation Index (WEI), which is the mean yearly overall demand for freshwater divided by the long-term mean freshwater resources available, is observed to be very high in several regions of sub-Saharan Africa (Gueye et al., 2005; Boko et al., 2007). Ashton (2002) also projected that population trends and patterns in water usage will cause more countries in Africa to exceed their limits of “economically usable, land-based water resources before 2025”. The potential effect of over-exploitation and high WEI in coastal aquifers is not just the lowering of groundwater levels but also the peculiar adverse impact of seawater intrusion. This is due to the proximity of such aquifers to the ocean or creeks and the hydraulic pressure differences between saltwater and freshwater.
Several studies have been done on seawater intrusion with Ghyben-Herzberg principle as the most fundamental of all. Some methods and approaches previously and currently being applied in the study of salinity and seawater intrusion into fresh groundwater of coastal aquifers include Geophysical methods (Goldman et al., 1991; Hwang et al., 2004; Adepelumi et al., 2009; Bouderbala & Remini, 2014), Geochemical and Geophysical methods (Ayolabi et al., 2013; Cimino et al., 2007; Fadili et al., 2015), Hydrochemical methods (Kim & Park, 1998; Mondal et al., 2010), Numerical Modelling methods (Zhou et al., 2000; Javadi et al., 2011), GIS methods based on chemical Indices (Santha & Syed., 2013; Trabelsi et al., 2016; Balathandayutham, et al., 2015), Multivariate Statistical Analysis methods (King et al., 2013; Yang et al., 2015), and Estimation of pressure differences between freshwater and saltwater zones (Kim et al., 2007).
Naturally, seawater intrusion is a dynamic process and it may be observed that the quality of groundwater resources varies with spatial distribution, and time-based on factors influencing the hydrological system, hence, the concept of Vulnerability of groundwater resources. This concept of groundwater vulnerability as a measure for the protection of groundwater resources can be traced to the work of Albinet & Margat (1970). The aquifer may be vulnerable to environmental influences such as surface pollutants (leachates), a high concentration of effluents in rivers through the process of diffusion and advection, as well as seawater intrusion.
Over the years, a number of overlay methods have been developed for assessing the vulnerability of aquifers to pollution. The DRASTIC Index (Aller et al., 1987); GOD index (Foster, 1987); AVI rating system (Van Stempvoort et al., 1993); SINTACS method (Civita, 1994); ISIS method (Civita & De Regibus, 1995) and EPIK method (Doerfliger & Zwahlen, 1997) are major overlay methods identified by Gogu & Dassargues (2000). More recent methods are; PI method (Goldscheider et al., 2000); and the GALDIT method (Chachadi & Lobo-Ferreira, 2001). In some cases, two or more of these methods were applied in a single study (Corniello et al., 1997: Draoui et al., 2007). The names ascribed to all these methods are acronyms of the most important parameters identified for the vulnerability assessment in each case. The GALDIT method; an adaptation of the DRASTIC method for assessing the vulnerability of coastal aquifers to seawater intrusion was applied for this study. GALDIT method is essentially a method for assessing the vulnerability of coastal aquifers to seawater intrusion and it has been widely applied in several places for mapping out the vulnerability of coastal aquifers to seawater intrusion. Lobo-Ferreira et al., (2005) applied the method on the coastal aquifer of Monte Gordo, southern Portugal; Sophiya & Syed (2013) studied the coastal aquifer of parts of Eastern India and assessed the vulnerability to seawater intrusion. The Rhodope aquifer system in the North Eastern part of Greece was also assessed for its vulnerability to SWI by Kallioras et al., (2011). Saidi et al., (2013)’ work on Mahdia-Ksour Essaf aquifer on the eastern coast of Tunisia also covered the application of GALDIT for vulnerability assessment.
The East African region has been identified as the least urbanised region in the world, howbeit, it has the shortest doubling time for its urban population (UN Habitat, 2008). Its coastline is one of the coastal regions with the least human induced modified land. I.e. minimal land reclamation, dredging and modification (UNEP, 2008). Its urban coastal population is fast increasing as evidenced in the cases of Dar es Salaam (Tanzania) and Mombasa (Kenya). However, very few groundwater pollution studies have been carried out in the coastal regions of East Africa. How seawater intrusion and seasonal changes affect the quality of water in the coastal aquifers of Dar es Salaam was studied by Sappa et al., (2015). The study showed the groundwater of the coastal aquifer under study to be highly saline, while depletion of the water levels poses serious challenges for current and future freshwater provisioning from the aquifer. The Quaternary aquifer of Dar es Salaam was also investigated for the occurrence of seawater intrusion (Mtoni et al., 2013) where a discussion was made on the causes, consequences, remedial actions and future likelihood of seawater intrusion in the study area.
The population growth on the north coast of Mombasa has been on a steady and progressive rise in the past four decades (GOK, 1979, 1989, 1999; KNBS, 2010). This does not only suggest an increase in the demand for groundwater, but also a higher risk of pollution. Munga et al., (2006) assessed the pollution status and the vulnerability of the coastal aquifer of Mombasa to groundwater pollution from anthropogenic activities. The study employed the use of DRASTIC index overlay method with GIS to obtain the spatial distribution of the vulnerability of the study area to surface pollution. MODFLOW groundwater model package was used to study the flow of the groundwater stream. This research focused on the seawater intrusion pollution component of the same study area. Advanced GIS, different chemical analysis methods, and Groundwater Modelling System software package were applied extensively for the study of North coast of Mombasa’s coastal aquifer. Hence, studying the seawater intrusion component of the study area provides a broader knowledge on the groundwater pollution for a more sustainable management of the groundwater.
1.2 Statement of problem
The groundwater pollution study by Munga et al., (2006) on Mombasa North Coast exposed some data and information gaps amongst which are “Physico-chemical indicators of water quality including salinity, conductivity, pH; water table data and related geo-hydrologic parameters; saline water intrusion; and influence of seasonal rains on aquifer recharge and groundwater quality”. This research filled some of the gaps by studying the hydrogeological characteristics of the aquifer including the impact of seasonal changes on water quality status, impacts of seawater intrusion and the groundwater flow and solute transport. Similar studies have been done in many parts of the world, for instance, in sub-Saharan Africa, Ayolabi et al., (2013) assessed the quality and seawater impact of a coastal region in Nigeria using geophysical and geochemical methods. In East Africa, few similar works were reported in Dar es Salaam where hydrochemical methods where used in assessing the water quality and impact of seawater on the coastal and quaternary aquifers of the area (Sappa et al., 2015; Mtoni et al., 2013). However, very few related studies have been reportedly done on the coastal aquifer of Kenya vis a vis seawater intrusion. This work, therefore, utilised the understanding of chemistry, geology, GIS, remote sensing and groundwater modelling in establishing knowledge on the hydrogeological condition of the study area and observing changes in the flow and quality of the fresh groundwater to enhance sustainable management of the groundwater resource.
1.3 Justification
In Kenya, a good number of studies have been carried out on groundwater resources and aquifers. However, very few empirical studies have been carried out on coastal aquifers of Kenya. The salient fact is that the coastal aquifers require a lot more attention due to the unique challenges of sea water ingression, coastal erosion, and other climate change related phenomena. In many parts of Mombasa, domestic water is supplied by a central water corporation agency. However, with the increasing trend in population, it becomes increasingly difficult for every part of the city to be served, especially the informal settlements. This leads to a higher dependence on groundwater. The only slightly comprehensive groundwater study on the study area was reported ten years ago, in which groundwater pollution from surface sources was assessed while several gaps were identified (Munga et al., 2006). With an ever increasing impact of climate change and the surge in population in the past decade, ten years is too wide an interval for groundwater studies and monitoring for effective management. This research thereby aimed to provide a more recent, detailed and inclusive assessment of the groundwater resource in the study area. The seawater intrusion component of the groundwater pollution, water quality changes across the seasons not covered in earlier studies were duly covered in this study. The work, therefore, did not just assess the extent of seawater intrusion in the study area but also mapped out the vulnerability of the study area to the phenomenon. Widespread seawater intrusion inland does more harm than good in that freshwater ecosystems are threatened due to high salinity of water and provision of potable drinking water becomes more challenging. The outcomes of this work, therefore, provides sound background knowledge for managing the coastal aquifer in terms of the siting of boreholes/wells, potentially stressed areas and general groundwater management.
Invariably, the work has proven that similar studies can be done in the coastal areas of other parts of sub-Saharan Africa. The challenge had always been the availability of data and the complexity involved in employing data and knowledge from different sources and disciplines in an interdisciplinary manner in one single study within a limited scope, time and resources.
1.4 Objective of Study
1.4.1 General Objective
To investigate the groundwater quality and flow of the coastal aquifer of Mombasa North coast.
1.4.2 Specific Objectives
i. To assess the hydrogeological characteristics, salinity and extent of seawater intrusion in the aquifer of the study area.
ii. To undertake the vulnerability mapping of the aquifer to sea water intrusion
iii. To assess and simulate the groundwater flow and solute transport in the study area
1.5 Research Questions
1. What is the quality of groundwater and extent of seawater intrusion in the study area
2. What is the vulnerability of different parts of the study area to seawater intrusion
3. What is the direction of flow of the groundwater in the coastal aquifer under study
1.6 Scope of Study
The study was carried out on the north coast of Mombasa covering the populated regions of Nyali, Bamburi and Kisauni. The study area was mapped out based on hydrological boundaries and covered a total land area of 74.2 km2. The field data collection took place between the months of March and September 2016 with field visits done in March, June and September. Fifteen boreholes/wells strategically located across the study area were identified for sampling of static water levels, field water quality parameters and water sample collection. Samples were also taken at different points in the Indian Ocean. Laboratory tests were conducted for four cations (Na, K, Ca, Mg) and three anions (Cl, HCO3, SO4). Secondary data such as Topo maps, Geological Maps, SRTM 30m DEM, and historical borehole data obtained from different sources complemented the field data. ArcGIS10.3 software from ESRI was used alongside statistical packages such as Grapher and Eviews for the processing and analysis of data. Finally, Groundwater Modelling System (GMS) software package containing MODFLOW, MT3DMS and SEAWAT was utilised in assessing the groundwater flow and solute transport.
1.7 Limitation of Study
The major limitations to the scope of the study were;
- Due to time constraints, the field data collection only covered the peak of dry season, the peak of long rains season and the peak of dry season before the short rains within a hydrologic cycle. A full hydrologic cycle of one calendar year was not covered.
- The entire administrative area of Mombasa could not be covered due to the constraints of time, budget and complexities in the siting of available boreholes for data collection.
- The few historical borehole data available were scanty and lacked cohesion, therefore it was not sufficient to establish a database on which long-term groundwater studies could be done.
CHAPTER TWO- LITERATURE REVIEW
2.1 Introduction
This chapter is divided into four parts: coastal aquifers; application of GIS and groundwater modelling in coastal aquifer studies; GALDIT Index; and groundwater flow and solute transport models.
2.2 Coastal Aquifers
The pores in which groundwater is stored are found in the geological formations called aquifers. By definition, an aquifer is a rock layer which stores and allows the movement of water within its pores. About 30.1% of the world’s freshwater exists as groundwater (Shiklomanov, 1993), and its importance for the sustenance of life cannot be overestimated. Groundwater from aquifers commands the highest demand for readily available freshwater in the world (Boswinkel, 2000). This demand is expected to escalate in the future, mainly due to the rise in global water use and water shortages resulting from increasing variability in precipitation (Kundzewicz et al., 2007). Coastal aquifers are unique for their proximity to saltwater bodies such as Oceans, creeks and lagoons. Hence, they are prone to effects of dynamic processes such as seawater intrusion, and sea level rise.
2.2.1 Challenges of Coastal Aquifers
Many coastal aquifers around the world are constantly being overexploited, leading to serious challenges of land subsidence induced by aquifer-system compaction in a number of cases. Some instances of land subsidence reported such as Vietnam, Mekong Delta and Cambodia (Ingebritsen & Galloway, 2014); Chesapeake Bay (Boon et al., 2010) are prime examples. In the wake of annual increasing mean sea levels and sea level rise, the secondary effects of land subsidence such as increasing storm surges and flooding are inevitable. Another challenge facing coastal aquifers is groundwater pollution from sources such as; surface pollutants (leachates) and effluents discharged into rivers which may infiltrate into the aquifer through the process of diffusion and advection as seen in the case of Mombasa (Munga et al., 2006). However, the chief source of freshwater coastal aquifer pollution is seawater intrusion.
Several methods have been devised and studies documented to address or manage these challenges. Kumar (2006), suggests extraction of saline groundwater, infiltration of surface water (artificial recharge), physical barriers, inundation of low-lying areas, and an increase in natural recharge as countermeasures to addressing the decrease in groundwater resources and seawater intrusion in coastal aquifers. However, detailed site-specific studies are highly essential before an approach or set of approaches are adopted for managing the challenges facing a coastal aquifer. For instance, Lin et al., (2013) opine that the study of abstraction rates will better inform decision makers in the management of land subsidence, restoration of groundwater and sustainable use in multi-aquifer systems.
2.2.2 Seawater Intrusion
When saltwater encroaches or seeps into a freshwater aquifer it is called saltwater intrusion and when the source of the saltwater is the sea or an adjourning Ocean, it is called Seawater intrusion. Seawater intrusion is a complex dynamic process in which several causative factors come into play. Darnault & Godinez, (2008) highlights the main phenomena contributing to the ingress of saltwater and affecting coastal aquifer systems as;
- “Encroachment of saltwater in coastal Aquifers;
- Characteristics of the aquifer formations;
- Anthropogenic activities producing saline waste;
- Tidal effects and;
- Fluctuations of the freshwater heads”
Two key components of climate change – Sea level rise and increased evapotranspiration have also been identified as having probable strong impacts on groundwater salinization and seawater intrusion into coastal aquifers (Kundzewicz et al., 2007). Seawater intrusion can be considered as the major concern of coastal aquifers. Kumar, (2006), notes that when addressing exploitation, restoration and management of the fresh groundwater in a coastal aquifer, the major challenge is usually seawater intrusion. Arguably, the most fundamental empirical relationship aimed at studying the dynamics of the interaction between seawater and freshwater in a coastal aquifer is the Ghyben-Herzberg principle as illustrated in Figure 2-1.
illustration not visible in this excerpt
Figure 2-1: Illustration of the fresh and seawater interface of a coastal aquifer (Roger, 1998)
Ghyben-Herzberg principle is based on the density differences between saltwater and freshwater and it states that 1/40 unit of fresh water is required above sea level for each unit of fresh water below sea level to maintain hydrostatic equilibrium with the adjourning saltwater (Roger, 1998).
The principle is mathematically represented as;
illustration not visible in this excerpt
Where;
hf = Height of water table above mean sea level
hs = Height of freshwater zone below sea level
ρs and ρf are densities of saltwater and freshwater respectively.
This principle assumes the existence of a hydrostatic equilibrium between the saltwater and freshwater zones, horizontal flow in the freshwater zone and a lack of progressive movement of saline water in a homogenous, unconfined coastal aquifer.
Lusczynski (1961) modified the Ghyben-Herzberg equation for non-static equilibrium conditions where seawater is constantly experiencing progressive movement with heads above and below mean sea level. It was applied to pumping wells with the condition that they are close to each other (Figure 2-2). This modified equation is given as;
illustration not visible in this excerpt
Where;
Z = depth from water level measurements in observation wells to the saline-freshwater interface
illustration not visible in this excerpt
Figure 2-2: Unconfined coastal aquifer with non-equilibrium conditions between freshwater and saltwater (Darnault & Godinez, 2008)
Several factors contribute to the dynamics of this freshwater-saltwater interaction in coastal aquifers, the significant ones are; aquifer characteristics (e.g. thickness, permeability), the characteristics of underlying rocks- unconfined as well as the overlying rocks- confined (Barlow, 2003).
2.2.3 Review of Coastal Aquifer Research
Only a few of the globe’s coastlines are not within the influence of anthropogenic pressures even though not all Coasts are occupied (Buddemeier et al., 2002). In Africa, human influence on coastal regions is quite substantial. A massive 66.6% of Senegal’s national population resides within Dakar and about 90% of Senegal’s industries are situated in the Dakar coastal zone, 22.6% of Nigeria’s population reside in Coastal zones, while in Ghana, Togo, Benin, Sierra Leone, economic activities within the coastal zones form the backbone of their national economies (IPCC, 1997). In the East African region, most of the principal cities are inland, however, coastal cities like Mombasa and Dar es Salaam are experiencing yearly population growths of 5% and 6.75% respectively (World Bank, 1995). All these play massive roles in affecting the water usage and aquifer dynamics in coastal regions. On a regional scale, the aquifer in Africa vary significantly from one area to another and occur as “alluvial, lacustrine, basaltic and sedimentary aquifers in coastal zones (Steyl & Dennis, 2009). Of the 32 mainland countries along the 40,000km long African coastline, only a handful have done detailed researches on their coastal aquifers. In several coastal aquifer studies, the most widely used approaches are the geophysical, hydrochemical, GIS Overlay, numerical modelling or a hybrid of two or more of the approaches.
Adepelumi et al., (2009) carried out a vertical electrical resistivity sounding survey on the coastal zones of Lekki Peninsula, Lagos Nigeria, in order to provide information on subsurface lithology and delineate the groundwater salinity. A similar work was executed by Oladapo et al., (2014) where 52 borehole logs were obtained within Lagos Municipality using natural gamma and electrical resistivity and interpreted to delineate vulnerable zones and extent of seawater intrusion. The two instances are one of the few studies on coastal aquifers using geophysical approaches in sub-Saharan Africa. In the coastal region of East Africa, hydrochemical approaches (Mtoni et al., 2013; Sappa et al., 2015) were widely applied in Dar es Salaam (Tanzania). Mtoni et al., (2013) obtained log data from boreholes/wells dug between 1997 and 2009 while their field and Laboratory data comprised EC, Na, Mg, Ca, K, Cl, SO4, and HCO3. Piper plots were used to illustrate the characteristics of the groundwater with almost 50% considered brackish or saline. Sappa et al., (2015) widely applied Person’s correlation matrix to understand the nature of the groundwater. They recommended a more controlled exploitation of the groundwater in the study area.
The last decade has seen an increase in the application of GIS overlay in groundwater studies for carrying out vulnerability analysis. Application of GIS in groundwater studies is also known as geographic data processing (GDP) (Gogu et al., 2001). One of such GDP approaches is the “DRASTIC Index” introduced by Aller et al., (1987), and has been widely used in mapping out the vulnerability of aquifers to pollution. DRASTIC is an acronym with each letter representing the following; D- Depth to the water table; R- Recharge of the aquifer; A- Aquifer media; S-Soil media; T-Topography; I- Impact of the vadose zone; C- Hydraulic Conductivity. DRASTIC is best used for groundwater pollution from surface sources such as leachates from active dumpsites, pit latrines etc. The drastic method was applied in the vulnerability of Mombasa County, Kenya, to surface pollution by Munga et al., (2006).
Numerical modelling, also known as Process-based modelling (PBM) (Gogu et al., 2001) has also been widely applied in groundwater studies. PBM can be carried out independent of geographic data processing, however, GDP significantly reduces the hassles inherent in process-based modelling. For instance, Gaaloul et al., (2012) modelled and analysed the seawater intrusion phenomenon in the eastern coastal aquifer of North East Tunisia using GIS-based data for the 3D numerical modelling of the aquifer.
2.3 Application of GIS and Groundwater Modelling in Coastal Aquifer Studies
Models are a representation of some parts of the real world and they can be static or dynamic.
Geographic Information Systems is a powerful tool which involves the use of software to capture, manipulate, view and analyse geographic information or spatial data to achieve a specific goal. GIS is “a system of hardware, software, and procedures to facilitate the management, manipulation, analysis, modelling, representation, and display of geographically referenced data to solve complex problems regarding planning and management of resources” (Goodchild & Kemp, 1990). In other words, GIS comprises five major components; hardware, software, data, procedures and most importantly people. In recent times, there has been an increase in the application of GIS in the field of environmental sciences and engineering as well as numerous other fields. Groundwater models usually take the form of numerical models. Numerical models emerged over the past 40 years as one of the primary tools used by hydrologists to understand groundwater flow and saltwater movement in coastal aquifers (Gaaloul, 2012). Numerical models are mathematical representations (or approximations) of groundwater systems in which the important physical processes occurring in the systems are represented by mathematical equations.
2.3.1 Geographic Information Systems (GIS)
GIS as a concept covers the two descriptors that represent our real world; location, and the attributes found in the location. In GIS, geographic information is usually represented as either objects or fields. Geographic objects are what populates a study area, and they are usually well-distinguishable, discrete, bounded entities with the spaces between them potentially empty (Rolf, 2001). Objects symbolise the real world with simple vector formats such as points, lines and polygons. Some examples of spatially represented objects in hydrogeology are boreholes, piezometers and protected zones. Attributes attached to these objects may be the Well number, point data of the static water level, chloride concentration.
On the other hand, fields represent real world in terms of attribute data without defining the objects. A geographic field is a geographic phenomenon in which a value can be determined for every point in the study area (Rolf, 2001) e.g. a raster. A raster model comprises rectangular arrays of cells where each cell has an assigned attribute. In terms of representation, spatial representation of data may be 1.) In discrete values for each point (well-defined values for each location), and 2.) Kernel (spatially continuous functions). The six model methods for digitally representing spatial data in fields are the raster model, grid model, point model, contour model (Isolines), polygon model and triangulated irregular network (TIN). In the GIS environment, the most familiar model is the “Map”. A map is a miniature representation of some parts of the real world. Another form of the model in GIS is a “database”. A database stores a significant amount of data and provides numerous functions to operate on the stored data (Rolf et al., 2001). Data modelling is the general term for the design effort of structuring a database. Maps and the database are both categorised as static models.
A foremost software for GIS-based data processing that has seen tremendous improvements over the years is the ArcGIS software, a trademark GIS software of ESRI, the leading organisation for Geographic Information Systems. It is built with a diverse range of capabilities including cartography, network analysis, spatial and 3D analysis.
2.3.2 Groundwater Modelling
The Groundwater flow modelling and its different approaches have evolved over the years. Groundwater modelling is a dynamic model which is highly essential in understanding and analysing groundwater dynamics. Groundwater simulation is the building of a model whose behaviour closely resembles the actual behaviour of the aquifer under study. Broadly speaking, a groundwater model can be physical, electrical analogue or mathematical (Mercer, 1980). Each of these model types has different sub-divisions. However, Mathematical models seem to be the most widely used of all. Mathematical models may be statistical, deterministic or some combination of the two. Purely Statistical models are very useful for classifying data and describing systems that are not well understood but generally offer little physical insights. Deterministic models are the ones that define relationships between cause and effects based on the understanding of the physical system. The procedure for developing a mathematical model is shown in Figure 2-3.
illustration not visible in this excerpt
Fig 2-3: Logic diagram for the development of a mathematical model (Source: Mercer, 1980).
The conceptual model helps in understanding the nature of the physical system under study. It forms the foundation for one or more numerical models. The numerical model gives the real time simulation of the properties of the aquifer being studied. It involves defining the objectives of the model such as the desired flow and transport simulation options, and the boundary conditions. The mathematical model consists of the initial conditions, appropriate boundary, and a partial differential equation over the study area. It must also contain phenomenological laws to describe the fluid behaviour in the system such as Darcy’s Law of conservation of momentum, Fick’s law of chemical diffusion and Fourier’s Law of heat conduction. The boundaries for the mathematical model have to be clearly defined. Afterwards, the region is subdivided into grids. Two main discretion approaches for two-dimensional grid systems are the finite-difference method and finite-element method (Figure 2-4).
illustration not visible in this excerpt
Figure 2-4a: Finite- difference grid configuration for the study of an Aquifer (Source: Mercer, 1980).
illustration not visible in this excerpt
Figure 2-4b: Finite- Element grid configuration for the study of an Aquifer (Source: Mercer, 1980).
Data used in groundwater modelling are in 4 categories;
1. The Aquifer - system stress factors
2. The Aquifer- system and strata geometry
3. The hydrogeological parameters of the simulated process
4. The main measured variables.
The stress factors in groundwater flows are; effective recharge, pumping volumes, water-surface flow exchanges input and output contaminant mass flows. Stress factors are imposed through the boundary conditions or source/sink terms. These boundary conditions are determined through the geological information such as topographic maps, contour maps of the upper and lower limits of the aquifer strata.
The hydrogeological data parameters are derived from raw data interpretations and the major ones are; hydraulic conductivity, storage coefficient, initial heads, and dispersivity.
2.3.3 Integrating of GIS with Groundwater Modelling
As expressed in Figure 2-3, the first step in achieving a numerical model for groundwater studies is the creation of the conceptual model. GIS database has made the creation of conceptual models extremely easy. Hence, integrating GIS with numerical models does not only save time but also reduces the drudgery of building conventional conceptual models from scratch. Developers of advanced groundwater modelling software packages like Visual MODFLOW Flex and GMS have helped make the building of conceptual models from GIS data more relatively stress-free.
Techniques for integrating GIS with groundwater models have been identified to be three; namely loose coupling, tight coupling, and embedded coupling (Gogu et al., 2001).
a. Loose coupling- This is when independent software packages are used for the GIS and the GW model components. The data transfer of predefined files from one software package to the other is made through input and output model. The advantage is that potential future changes can be facilitated in an independent manner since each software is distinct on its own.
b. Tight coupling- In this case, data is exported from the GIS to the model but the GIS tools are able to interactively access the input subroutines of the GW model package. There is a shared user interface between the GIS and the modelling system, while data exchange is also automatic. A good example of tight coupling is ArcGIS-SWAT data model developed by Olivera et al., (2006). Another example is the PIHMgis package developed by (Bhatt et al., 2014) which is integrated with QGis as a plugin based toolbar.
c. Embedded coupling- This occurs when the hydrogeological model is created using GIS programming language or when a complex modelling system assimilates a simple GIS. In this case, the user interface, data and method base are not just shared between the GIS and the model but also, intra-simulation model modification, query and control are also very possible. A good example of embedded coupling is the AVTOP- a topography-based hydrological model built with the macro language interface of ArcView (Huang & Jiang, 2002). However, it requires very complex programming and data management.
Both the tight coupling and embedded coupling require considerable investment in programming and data management which are not always justifiable.
2.4 GALDIT Overlay Index
GALDIT overlay index is an open-ended model which utilises a numerical ranking system for evaluating the seawater intrusion potential of a coastal aquifer within the framework of the hydrogeological settings using certain established factors (Lobo Ferreira et al., 2005). These factors also known as GALDIT factors, have been identified as the most important map-able factors controlling seawater intrusion (Chachadi & Lobo-Ferreira 2001; Chachadi & Lobo-Ferreira 2005; Lobo Ferreira et al., 2005). The factors include;
- G- Groundwater occurrence (Aquifer type; confined, unconfined, leaky confined)
- A- Aquifer hydraulic conductivity
- L- Level of the Groundwater depth above sea level
- D- Distance from shore (distance inland perpendicular from shoreline)
- I- Impact of existing status of seawater intrusion in the area
- T- Thickness of the mapped aquifer
The acronym GALDIT was derived from the first letter of each parameter. The parameters are drawn up into map layers which are then overlaid based on weightages, ranges and importance rankings. The weights were derived from the analysis of the data obtained from opinion surveys, experts’ opinions, focused groups and several academic forums on what are considered to be the most important factors contributing to the intrusion of seawater into coastal aquifers (Table 2-1). The weights are therefore considered as constants not to be altered by subjective opinions. The importance rating factors are put on a scale of 2.5 to 10. The higher the value of the importance rating, the higher the vulnerability of the aquifer to seawater intrusion. The total sum of all the individual scores is referred to as the Decision Criterion. This is obtained by multiplying the values of the importance ratings with the corresponding Indicator weights.
Table 2-1: Standardised weightages for GALDIT factors (Chachadi & Lobo-Ferreira, 2005)
illustration not visible in this excerpt
The final vulnerability maps derived from the overlay and spatial analysis of the GALDIT indices provide useful information on areas which are more susceptible to the intrusion of saltwater than the others.
The GALDIT index has been widely applied in several studies to map out the vulnerability of coastal aquifers to seawater intrusion (Chachadi & Lobo-Ferreira 2001; Chachadi & Lobo-Ferreira 2005; Lobo Ferreira et al., 2005; Kallioras et al., 2011; Sophiya & Syed 2013; Gontara, et al., 2016).
2.5 MODFLOW, MT3D, SEAWAT and GMS
2.5.1 Brief Description on MODFLOW
MODFLOW is a ground water model where groundwater flow is simulated using a block centred finite-difference approach (Modflow: USGS, 2013). Layers can be simulated as confined or unconfined. The newest main version of MODFLOW is the MODFLOW-2005. Prior to the development of MODFLOW, two and three-dimensional finite difference models were used extensively by US Geological Survey (USGS) and others for computer simulation of ground-water flow. Therefore, the first version of MODFLOW was the result of the need to consolidate all the commonly used simulation capabilities into a single code that was easy to understand, use and modify. This was developed using Fortran Computer Language- Initially Fortran 66 then Fortran 77. This first version of MODFLOW was called MODFLOW-88. Originally MODFLOW was conceived solely as a ground-water flow model, but its application has expanded beyond ground water flow into transport and parameter estimation (Banta et al., 2005).
Principles of MODFLOW
MODFLOW uses a modular structure wherein similar program functions are grouped together, and specific computational and hydrologic options are constructed in such a manner that each option is independent of other options. Because of this structure, new options can be added without the necessity of changing existing options. The model may be in two or three-dimensional forms. When the model is visualised in terms of a three-dimensional assemblage of cells, each cell contains a point called a node at which head is to be calculated. The size of the model grid is specified by the user in terms of the number of rows (NROW), the number of columns (NCOL), and the number of layers (NLAY); these terms define a three-dimensional grid of cells in the form of a rectangular box. Input procedures have been designed so that each type of model input data may be stored and read from separate external files. User-specified formatting allows input data for the grid to be read in almost any format without modification to the program. The output of model results is also flexible; the user may select which data to output, the frequency of output, and for some data, the format of the output. MODFLOW is not necessarily a single program, but all MODFLOW programs include the Groundwater Flow Process (Banta et al., 2005).
Operations of MODFLOW
MODFLOW utilises iterative methods to obtain the solution to the system of finite-difference equations for each time step. In these methods, the calculation of head values for the end of a given time step is started by arbitrarily assigning a trial value, or estimate, for the head at each node at the end of that step. A procedure of calculation is then initiated that alters these estimated values, producing a new set of head values that are in closer agreement with the system of equations. These new, or interim, head values then take the place of the initially assumed heads, and the procedure of calculation is repeated, producing a third set of head values (Harbaugh et al., 2000).
Currently, it has been made possible for MODFLOW models to be successfully coupled with ArcGIS. An extension called Arc Hydro Groundwater can be downloaded freely from Aquaveo (2016). Furthermore, several MODFLOW-related programs have been developed with capacities to simulate solute transport, parameter estimation, coupled surface-water / groundwater systems, variable- density flow (including saltwater), aquifer-system compaction, and land subsidence, and groundwater management.
MODFLOW code is based on the fundamental laws of continuity of mass and motion which are solved for the partial differential equation as expressed in Figure 2-5.
illustration not visible in this excerpt
The governing equation used in MODFLOW, describing the three-dimensional groundwater flow through the aquifer is the given by Freeze & Cherry, (1979) as:
illustration not visible in this excerpt
Where;
Kx, Ky, Kz are the hydraulic conductivity values along the x, y and z coordinates
h- is the hydraulic head values
w- is the volumetric flux per unit volume and represents sources and sinks of water per unit time (t-1)
Ss- is the specific storage of the porous material and
t - is the time
The first part of the simulation runs a steady state solution which takes the form:
illustration not visible in this excerpt
The steady state solution is then solved for the transient state in order to solve for storage coefficient.
2.5.2 Brief description of MT3D
MT3D is a modular three-dimensional (3D) transport model with its first version developed by Chunmniao Zheng in 1990 and released as a public domain code from U.S. Environmental Protection Agency (Zheng, 1990). MT3D is based on a modular structure which enables the simulation of transport components independently or jointly (Aquaveo, 2006). MT3DMS is the second generation of MT3D, developed under the Strategic Environmental Research and Development Program (SERDP). MT3D code has a wide range of solution options such as; method of characteristics (MOC); modified method of characteristics (MMOC), a hybrid of MOC and MMOC (HMOC); and the standard finite-difference method (FDM).
MT3DMS supports all the discretization and hydrologic features of MODFLOW. It simulates dispersion, advection, and chemical reactions of dissolved constituents (Zheng & Wang, 1999). Its interaction with MODFLOW is a 2-step flow and transport simulation process. In the first step, MODFLOW computes the heads and cell-by-cell flux and are written in a formatted file. The second step involves the reading of the formatted file by MT3DMS and used as the flow field for the solute transport component of the simulation process. The differences between MT3DMS and MT3D are that the former enables multi-species transport, support additional solvers, and make allowances for cell-by-cell input of all model parameters.
The partial three-dimensional equation used in MT3D for the solute transport is equally highlighted in Freeze & Cherry, (1979) and expressed as:
illustration not visible in this excerpt
Where;
C is the concentration of dissolved contaminant in groundwater [ML-3];
Xi is the distance along the Cartesian coordinate axis
Dij: is the hydrodynamic coefficient of dispersion [L2T-1];
νi: the seepage or linear pore water velocity;
q’s: the volumetric flux of water per unit volume of aquifer’s sources and sinks [T-1];
Cs: the concentration of the sources or the sinks[ML-3];
θ: the porosity of the porous medium [-]; and
Rk: chemical reaction term
The solute transport equation was slightly modified by Zheng & Wang (1999) for the running of MT3D module and given in equation 6 as;
illustration not visible in this excerpt
Where:
n is the porosity of the medium; q is the specific discharge [LT-1].
2.5.3 A brief description of SEAWAT
SEAWAT is a generic MODFLOW/MT3DMS-based computer program designed to simulate 3D variable density groundwater flow coupled with multispecies solute and heat transport, focused mainly on brine migration in continental aquifers as well as the seawater intrusion in coastal aquifers (Langevin et al., 2003). First documented by Guo and Bennett (1998), it is based on the concept of the freshwater head, or equivalent freshwater head, in a saline ground-water environment. The SEAWAT code follows a modular structure and was developed by combining MODFLOW and MT3DMS into a single program that solves the coupled flow and solute-transport equations. The program can read and write standard MODFLOW and MT3DMS data sets, although some extra input may be required for some simulations. For the convenience of hydrologists and modellers familiar with MODFLOW and MT3DMS, the changes in input files for MODFLOW and MT3DMS were kept to a minimum for use in SEAWAT. Thus, existing input files for the standard versions of MODFLOW and MT3DMS can be revised for SEAWAT with minor modifications. Temporally and spatially varying salt concentrations are simulated in SEAWAT using routines from the MT3DMS program equally generated from MODFLOW. SEAWAT uses either an explicit or implicit procedure to couple the ground-water flow equation with the solute-transport equation. With the explicit procedure, the flow equation is solved first for each time step, and the resulting advective velocity field is then used in the solution to the solute-transport equation (Guo & Langevin, 2002). This procedure for alternately solving the flow and transport equations is repeated until the stress periods and simulation are complete. With the implicit procedure for coupling MODFLOW and MT3D, the flow and transport equations are solved multiple times for the same time step until the maximum difference in fluid density between consecutive iterations is less than a user-specified tolerance (Langevin, Shoemaker & Guo, 2003). The modified flow equation incorporating concentration of solute and specific storage (Langevin & Guo, 2006) is given in equations 7 as;
illustration not visible in this excerpt
Where;
Ssf is the specific storage (L-1) defined as the volume of water released from storage per unit volume per unit decline of fresh water head
C is the concentration of solute mass per unit volume of fluid [ML-3].
ρ is the fluid density
Kf is the freshwater hydraulic conductivity”
Equation 6 reduces to flow equation solved by MODFLOW for a constant-density system.
The solute transport equation (equation 7) solved by MT3DMS program is used for the solute transport solution in the SEAWAT interface.
2.5.4 A brief description of GMS
GMS- which is an acronym for Groundwater Modelling System is an intuitive software platform for creating groundwater and subsurface simulations in a 2D and 3D environment. It was first developed in Environmental Modelling Research Laboratory (EMRL) in the late 1980s. Its development was funded primarily by the United States Army Corps of Engineers. In 2007, the developers of the package at EMRL became a private enterprise as Aquaveo, LLC and has continued developing GMS as well as other products like SMS (Surface-water Modelling System) and WMS (Watershed Modelling System). It supports GIS data formats and numerous other formats such as vectors, raster, topographic maps, MODFLOW files, borehole data, ArcGIS geodatabases and shapefiles, CAD files amongst others. GMS also supports a wide range of models such as MODFLOW, MODPATH, MT3DMS, RT3D, SEAWAT, FEMWATER, SEEP2D, and UTEXAS. GMS is fully equipped for both Conceptual model approach and Numerical model approach.
CHAPTER THREE- GENERAL METHODOLOGY
3.1 Introduction
In this research, a series of activities were embarked upon. The activities included; obtaining of secondary data from agencies, three rounds of field data collection, laboratory analysis of collected samples, statistical and geospatial analysis and groundwater modelling. Loose coupling of GDP and PMB approach was employed in the work. ArcGIS 10.3 version under ArcInfo license was used for all GIS-based activities. The spatial analyst extension and other tools embedded in the ArcGIS software were used for activities such as tables editing joins, clipping, , conversion from one dataset type to another, projection, digitisation, georeferencing, spatial interpolation, vulnerability analysis, and cartography. The groundwater model and solute transport simulation entailed the application of GMS10.1 software package utilising the MODFLOW component for the groundwater flow simulation, while MT3DMS and SEAWAT were used for the 3D solute transport model.
3.1.1 Study Area
Physical Background
The location, topography, climate, hydraulic boundary, geology, and socioeconomic conditions of the study area are described in this section.
i. Location
The study area is located on the North thrust of Mombasa Coast lying between latitudes 3 95” and 4° 07” South of the equator and between longitudes 39° 68” and 39° 72” East of the Greenwich Meridian. There are four administrative sub-counties and six parliamentary constituencies in Mombasa. The administrative sub-counties are the Island, Likoni in the South, Changamwe in the west and Kisauni in the North. The parliamentary constituencies include Changamwe, Likoni, Mvita, Jomvu, Nyali and Kisauni.
The study area, about 74.2 km2 in size, covers Nyali and parts of Kisauni parliamentary constituencies within the division of Kisauni. Major areas such as Kongowea, Nyali, Mwakirunge, Bamburi, and Shanzu are located within this study area. Based on the 2009 Census, Nyali and Kisauni constituencies had a population of 185,990 and 194,065 respectively (KNBS, 2010). This makes the study area the highest and second highest populated constituencies of the six constituencies in the county of Mombasa. The area under study is the uniquely carved out area running from Nyali Bridge up to the edge of Mtwapa Creek (Figure 3-1). It is bounded on the north and south by creeks, on the east by the Indian Ocean and on the west by Nguu Tatu Hills.
ii. Topography
Site visits and the DEMs of the study area show that Kisauni sub-county is a low-lying coastal plain. The most elevated part of the study area is Nguu-Tatu Hills whose peak is about 124m above sea level. Nguu Tatu Hills is located some 6.5km to 7km to the ocean front. The elevation of other parts of the study area mostly ranges from Sea level to 50m above sea level. The general low-lying characteristics of the study area influences surface runoff, as infiltration and deep percolation, makes possible quick recharge of the aquifer.
iii. Climate
The region can be described as hot, humid, and tropical. South Eastern and North Eastern Monsoon winds play decisive roles on the seasons in the study area. There are two rainy seasons – the long rains and the short rains both influenced by the Monsoon winds. The SE Monsoon winds blow between April and September coinciding with the long rains while the NE Monsoon winds blow between October and March influencing the short rains. The long rains occur between the months of March and July while the short rains occur between September and December. The total annual rainfall is above 1000mm (Climatemps, 2015). The highest amount of precipitation is usually experienced in May and the least in February of a calendar year.
illustration not visible in this excerpt
Figure 3-1: Satellite Imagery of the Study Area (Source: Google Earth)
iv. Hydraulic Boundary Conditions
The rationale behind the area of interest determination is based on the hydrogeologic suitability for feasible GIS and groundwater modelling. The area is a generally low-lying area from the coastline to the hinterlands. Therefore, Nguu -Tatu hills which have the highest elevation above sea level is chosen as the western boundary to provide a zone of constant freshwater flow. Mtwapa and Tudor creeks are the Northern and Southern boundaries respectively while the Eastern boundary is the Indian Ocean. The boundaries are highlighted below;
- The seaside boundary (coastal edge) - defined as the boundary of the constant head. i.e. the hydrostatic pressure boundary where the mean seawater level is used to define the coastline
- The Western boundary- defined by the Nguu Tatu Hills which is the boundary of no flow.
- The Northern and Southern boundary is defined by the Tudor and Mtwapa creeks which equally act as zones of the constant head.
v. Geology
The geological information shows that the rocks that form the geology of this region are sedimentary in nature. They were formed during the Pleistocene period and composed mainly of alluvium, wind-blown and superficial sands, corals and coral breccia (Caswell, 2007). The rocks dip gently and become progressively younger towards the coast. The Nguu Tatu Hills region is made up of Upper Jurassic shales which are fossiliferous in nature. Wind-blown and superficial sands of Pleistocene age cover large parts of Utange and some parts of Bamburi. Most parts of Kisauni, Nyali and parts of Bamburi are made up of coral reefs, lagoonal sands, Kilindini sands and North Mombasa Crag which all belong to Pleistocene age (Figure 3.2). The lithology is composed mainly of limestone, sandstone and shale of varying depths. The described lithology shows the aquifer of the study area is mostly unconfined.
illustration not visible in this excerpt
Figure 3-2: Map of the study area
(Extracted from CASWELL 1954 geological sheets using ArcGIS)
vii. Socioeconomic Conditions
The coastlines are characterised by beaches, rich biodiversity and a variety of coastal resources. Kisauni is also an agricultural region with a wide range of cultivated crops such as cassava, cashew-nuts, and coconuts. Besides being the most populated division in the county, the direction of population growth is towards this region. Hence, there are lots of informal settlements with poorly planned settling patterns. These are numerous low-cost areas with high densities such as Kisauni Estate, Barsheba, Mlaleo, Magogoni, Mwandoni, Mishomoroni, Bakarani, Shanzu amongst others. The significant Socio-economic activities taking place include; beef and dairy farms, tourist hotels and a few industries.
3.2 Materials and Methods
The materials used and methods of data collection completed for this research are highlighted under this section.
3.2.1 Secondary Data Collection
Prior to the field data collection exercise, secondary data acting as both guides and auxiliary sources of information were obtained. These data formed the basis for the field and laboratory data analysis and are highlighted in Table 3-1;
Table 3-1: Secondary data obtained and their sources
illustration not visible in this excerpt
3.2.2 Field Data Collection
The data obtained in the field include; the GPS coordinates of the identified boreholes and wells, Electrical Conductivities, Total Dissolved Solids, pH, NaCl, static water table measurements and water samples for further tests in the laboratory. The water samples obtained were preserved in cool boxes filled with ice cubes from the points of the collection in the field to the laboratory. The techniques used for obtaining the field data are highlighted in Table 3-2.
Table 3-2: Summary of the field data obtained and the collection techniques
illustration not visible in this excerpt
The field data collection was in phases of pre-monsoon, rainy season and post-monsoon as follows;
Pre-monsoon- 24th and 25th March 2016
Rainy season- 28th and 29th June 2016
Post-monsoon- 1st and 2nd September 2016.
Data were obtained from 12 selected boreholes and shallow wells at pre-monsoon and 15 boreholes and shallow wells each at the rainy season and post-monsoon.
3.2.3 Laboratory Data Analysis
The water samples obtained from the selected boreholes and wells in the field were transported to the laboratory within 24 and 48 hours of sample collection and analysed. The procedures and safety precautions taken during the laboratory analysis are highlighted in Brown et al., (1983). The cations tested include; Sodium (Na+), Calcium (Ca2+), Magnesium (Mg2+), and Potassium (K+) and the anions tested include; Chloride (Cl-), Bicarbonate (HCO3-), and Sulphur dioxide (SO42-). The summary of techniques used for the water sample analyses is highlighted in Table 3-3.
Table 3-3: Techniques used for the laboratory analysis of water samples
illustration not visible in this excerpt
The secondary data, field data and results of laboratory analysis of water samples were all assembled into a geodatabase on ArcGIS software. Summary of the software packages used and their specific applications in this study are highlighted in Table 3-4.
Table 3-4: Software packages used
illustration not visible in this excerpt
CHAPTER FOUR
This chapter covers the first specific objective of this study encompassing the assessment of the hydrogeological characteristics, salinity and extent of seawater intrusion in the aquifer under study.
4.1 Hydrogeological characteristics of the aquifer
The obtained data in each data collection phase (pre-monsoon, rainy, and post-monsoon) were analysed similarly and the results compared. The water quality parameters measured in the water samples were compared to the Kenya and WHO drinking water guidelines. Hydrogeological characteristics of the coast were described by the aquifer type, geological formation, and the groundwater heads using the geological data (secondary data) and static water level measurements (field data) on ArcGIS software package. The salinity and extent of seawater intrusion of the aquifer were assessed and analysed statistically and geospatially from the field data and laboratory analyses results. The research methodology adopted the cations, anions and specific indices for assessing the characteristics of the groundwater and extent of seawater intrusion (Mondal et al., 2010, Klassen et al., 2014). The general workflow employed in achieving the objective is outlined in Figure 4-1.
Hydrogeological characteristics describe the nature of the underlying rocks constituting the aquifer as well as the general groundwater heads in the aquifer. The hard copies of geological maps obtained as secondary data were digitised, georeferenced and mosaicked together with the aid of GIS. From the large mosaicked image, the study area was clipped out to create the geological map of the study area. Measured water level heads during the field trip were geostatistically interpolated on ArcGIS to form isopleth maps which were combined with the geological map to obtain a hydrogeological map of the study area. Correlation matrices and cross-plots are widely applied on groundwater parameters in studying the groundwater chemistry in aquifers (Mtoni et al., 2015; Sappa et al., 2015). In this study, the EC, TDS, Na, K, Mg, Ca, Cl, HCO3 and SO4 were all analysed for their correlational relationships to ascertain the major parameters influencing the geochemistry of the coastal aquifer.
illustration not visible in this excerpt
Figure 4-1: Flowchart for the general steps in assessing hydrogeological characteristics, salinity and SWI
A Piper plot was used in the classification of the groundwater type. Piper plot is one of the most widely applied methods for characterising groundwater chemistry (Karmegam, et al., 2011; Musa et al., 2014). A Piper plot is a graphical depiction of the chemistry of water sample(s). Two ternary plots containing normalised values of the cations and anions respectively are projected into a diamond which describes the broad nature of the groundwater (Piper, 1953). The Piper plot categorises ground water as any of the following; Calcium sulphate waters, Sodium chloride waters, Sodium bicarbonate waters or Calcium bicarbonate waters. Calcium sulphate waters are typical of gypsum groundwaters and mine drainages; Calcium bicarbonate waters are indicative of shallow fresh groundwater; Sodium Chloride waters are indicative of marine and deep ancient groundwaters; while Sodium bicarbonate waters typify deeper groundwater influenced by ion exchange (Piper 1953). The cations (Na+k, Mg, Ca) and anions (Cl, SO4, HCO3) are normalised into 100%. These normalised values are plotted into the right and left triangles (Figure 4-2) which form points of intersections in the diamond where the groundwater chemistry is interpreted (Figure 4-3).
illustration not visible in this excerpt
Figure 4-2: Trilinear plots (piper) of the central catchment of Bribie Island (Source: Jackson, 2007)
illustration not visible in this excerpt
Figure 4-3: The diamond for the interpretation of the groundwater characteristics
4.1.1 Results and Discussion
The lithological and geological information from geological reports reveals the nature of the aquifer as unconfined (Abuodha 2003; Caswell 2007). The geology comprises mainly corals, coral reefs, lagoonal sands, limestone and sandstone occurring intermittently in sedimentary layers to a minimum depth of 100m to the surface in most portions of the aquifer (Abuodha 2003; Caswell 1954, 2007). The water table measurements from boreholes and wells taken across the study area show groundwater heads varied from 0m to 27m and -1 to 33m above sea level at pre-monsoon and post-monsoon respectively (Figure 4-5). This implies there were no large variations in water table across the seasons. The hydrogeological map of the study area is represented in Figure 4-4.
illustration not visible in this excerpt
Figure 4-4: Hydrogeological map of the study area showing the GW heads above MSL (March 2016)
illustration not visible in this excerpt
Figure 4-5: Groundwater table in the coastal aquifer at (a.) pre-monsoon and (b.) post-monsoon
From Figure 4-5, it is observed that the groundwater table was highest in the elevated regions of the Nguu Tatu hills on the western part of the study area while the other parts generally have shallow groundwater tables. The relatively low elevated topography of the study area implies that the static water levels from boreholes are as shallow as 8m to the surface in some parts of the study area. This closeness of the water table to the surface implies an abundance of groundwater. However, it also indicates that the groundwater is more susceptible to surface pollution.
The summary table of the measured groundwater parameters is represented in Table 4-1. It reveals a substantial insight on the nature of the groundwater. The groundwater pH generally experienced only minor changes across the entire period of observation recording alkaline values mostly between 7 and 8. Only two mildly acidic values (6.65 & 6.98) were observed in the data, and they occurred in the rainy month of June (Table 4-1b). These values coincide with the WHO (2011) pH range for rainwater, hence, the mild acidity might be due to the prevalent rainfall in that season. Of the 42 samples, only 3 of the samples fall outside the range of 6.5 – 8.5 specified for drinking purposes by the Kenya drinking water guidelines (Kenya, 2008) (Table 4-1).
Table 4-1: Concentration of the major parameters in the groundwater samples
illustration not visible in this excerpt
B-Boreholes, S/W- Shallow wells, W-Well
The values and concentrations of the Electrical conductivities and Total Dissolved Solids vary vastly in the groundwater across the study area. For instance, the EC values varied from 761.5 to 10585µS/cm at pre-monsoon, while TDS values ranged from 438 to 5281mg/l within the same time period with both having high standard deviation values (Table 4-2a). This significant variation in EC and TDS values across the study area is equally observed in June and September. From a seasonal perspective, only slight variations in EC and TDS were observed across the three seasons in each observation point (Table 4-2). However, the general trend is that EC and TDS values are lower across the aquifer in the rainy season in comparison to the dry seasons (Tables 4-1, 4-2). In terms of the allowable standards for drinkable water provided by WHO, the concentration of EC and TDS in over 94% of the water samples taken exceeded the allowable limits. However just slightly over 43% exceeded the 1500 mg/l TDS limit specified in Kenya drinking water guidelines (Table 4-1).
NaCl field measurement observations were also similar to those of EC and TDS. NaCl values showed a wide range of 660 to 10569 mg/l at pre-monsoon while the concentrations were generally lower in the rainy season than the preceding and succeeding dry season (Tables 4-1, 4-2). Increased groundwater flow and heads as a result of higher recharge from rainfall during the rainy season might explain why the EC, TDS and NaCl values tended to be relatively lower in June than in March and September.
Table 4-2: Statistical analysis of the parameters for March 2016
illustration not visible in this excerpt
b. June 2016 (rainy)
illustration not visible in this excerpt
c. September 2016 (post-monsoon)
illustration not visible in this excerpt
Of the four cations tested, Na+ had the highest concentration with values as high as 735.55mg/l in a sample (Table 4-2a). The WHO and Kenya drinking water standards specify 200mg/l as allowable limits for drinking water. Over 50% of the water samples taken from the three periods exceeded the allowable limit for sodium ion content specified by both regulatory bodies. However, Ca2+, K+ and Mg2+ concentrations in all the water samples were within the allowable limits specified by both bodies. The Mg2+ and Ca2+ exist in comparatively lower concentrations to Na+ and K+ in the groundwater and is more obvious in the samples taken in the rainy season (Table4-2b). Generally, the variations in the maximum and minimum concentrations of Na+ and K+ on a seasonal scale were quite marginal (Tables 4-2).
The anions exhibited varying trends across the aquifer and over the three periods of data collection. Cl- ions ranged from 140.30 - 379.60mg/l at pre-monsoon and dropped to 10.38 - 149.85mg/l in June and increased back to 117.60 – 279.61mg/l at post-monsoon. Conversely, HCO3- concentrations had ranges of 87.84 - 173.24mg/l, 39.04 – 156.16 and 73.79 – 189.15 in the same period. The standard deviation of Cl- ions drastically plunged from 71.92 to 35.58 in contrast to HCO3- whose standard deviation only slightly increased from 26.06 to 31.71 from pre-monsoon to the rainy season but both slightly increased by post-monsoon. The chloride concentrations of 75% of the water samples taken in March exceeded WHO’s allowable limit for drinking water while over 50% exceeded the Kenya standards for drinking water. The SO42- ions tested in September varied from 78.91 to 328.92mg/l in the water samples taken from the study area. In the case of SO42-, one-third and one-sixth of the samples had concentrations higher than the WHO (2011) and Kenya drinking water standards respectively (Table 4-1c).
Graphs showing the variation in concentrations of the groundwater parameters are presented in Appendix II. Overall, the relatively high values of EC and TDS, and concentrations of Na, Cl, and NaCl, especially at pre and post-monsoons, suggests that the groundwater in a large portion of the aquifer is unsuitable for drinking, howbeit useful for other domestic purposes.
4.1.1.1 Correlation Matrices and Cross Plots
Cross plots and correlation matrices are reliable statistical methods for determining the most dominant ions controlling the groundwater chemistry of the study area. All the parameters were analysed for their correlations and the results are represented in Tables 4.3.
Table 4-3: Correlation coefficients for the groundwater parameters
illustration not visible in this excerpt
c. September (post-monsoon)
illustration not visible in this excerpt
Remarkably, the correlation matrices for the three time periods all follow uniform trends. EC and TDS were observed to be perfectly correlated with values 0.99, 1 and 1 for the months of March, June and September respectively (Table 4-3) implying an absolute direct relationship. The EC and TDS were also observed to consistently have a strong correlation with Cl, Na, Mg and K over the three periods (Table 4-3a, b, c). It can be inferred that these parameters play significant roles in defining the geochemistry of the study area (Mondal et al., 2010). The Cl showed a strong positive relationship with Na and K across the three periods of data collection signifying that these play significant roles in defining the EC and TDS contents of the groundwater. It is noteworthy that Na had a strong positive relationship with Mg and K across the three periods while Mg had a strong positive relationship with K in both March and June. Conversely, HCO3, Ca and pH had no consistent relationship with other parameters/ions across the three time periods. The water samples taken in September were tested for SO42- concentration (Table4-3c). It was observed that SO42- strongly correlates with EC, TDS, Na+ and Mg2+ which also have strong positive intra-correlations. Appendix III shows the cross plot relationships between EC/TDS and the various ions
Hence, the major parameters controlling the groundwater chemistry are EC, TDS, Cl-, Na+, Mg2+, K+, and to some degree SO42- while Ca2+, HCO3 and pH have a paltry impact on the geochemistry of the study area.
4.1.1.1 Piper Plots
The cations and anions concentrations in mg/l were converted into meq/l (See Appendix IV). The values in meq/l were normalised to 100% as shown in Table 4-4. The sum of sodium and potassium (Na+K) were normalised against Mg and Ca for the cations whose summation is 100% and plotted inside the cations triangle for all the water samples. The Cl, SO4 and HCO3 were also normalised to 100% and plotted into the anions triangle. All the normalised cation and anion values in the triangles were projected into the diamond to determine the nature of the groundwater based on the quadrant under which the values fell.
Table 4-4: Normalised values for the cations and anions
illustration not visible in this excerpt
The groundwater samples all fell in the Na-Cl waters quadrant in the diamond (Figure 4-6). This implies the groundwater is of the marine and deep ancient type (Piper, 1953). The reason is not far-fetched- the study area stretches from the seafront to about 7km inland while it is bounded to the north and south by creeks. The prevalent interaction between the aquifer and the saltwater bodies over thousands of years explains why the water samples reflect being of marine origin.
illustration not visible in this excerpt
Figure 4-6: Piper plot for the water samples taken in the study area
The use of Piper plots is advantageous in that it is possible to plot the analyses of several water samples on the same diagram and is able to give a broad overview of the groundwater characteristics.
4.2 Groundwater salinity assessment of the aquifer.
Salinity in groundwater may be from several sources including, anthropogenic activities such as irrigation and saltwater bodies in proximity to the aquifer. The indicator used for assessing the salinity of the groundwater was the EC according to classifications by Saxena et. al., (2003). The three main classes- Fresh; Brackish; and Saline water. The ranges for each class are given as;
- Freshwater (< 1,500 μS/cm);
- brackish (1,500–3,000 μS/cm);
- saline (> 3,000 μS/cm)
Based on the above classification, the spatial maps of the EC values across the aquifer was created with the aid of advanced spatial analysis tools on ArcGIS. The TDS and NaCl spatial variation maps were also developed.
4.2.1 Results and Discussion
According to the EC classifications (Saxena et al., 2003), the obtained water samples across the pre-monsoon, rainy season and post-monsoon were classified and represented in Table 4-5.
Table 4-5: Spatial coverage of the EC classification in percentages
illustration not visible in this excerpt
The groundwater salinity was observed to reduce in the rainy season as evidenced by the decrease in saline percentage coverage from 54% at pre-monsoon to 18% at the peak of rainy season. By June, the conversion was largely from saline to brackish water as over 70% of the groundwater became brackish in the rainy season. The freshwater coverage only showed a meagre increase from pre-monsoon to the rainy season i.e. 4% to 6% but was back to 4% at post-monsoon (Table 4.5). Invariably, the rainy season is characterised by the dilution of saline with fresh water from rainfall, thereby rendering the groundwater less saline but more brackish. It will, however, be observed by post-monsoon (September), the brackishness had reduced and salinity increased. The likelihood is that as the dry season progresses, salinity keeps increasing at the expense of brackishness until the successive rainy season when the cycle is reversed. A groundwater study by Sappa et al., (2015) on the coastal aquifer of Dar es Salaam, Tanzania, also located along the coast of East Africa showed a somewhat similar trend. The study reported that groundwater samples that were taken in November 2012 (post-monsoon) showed higher EC and TDS values than those taken in June 2012 i.e. rainy season (Sappa et al., 2015).
Point values from boreholes, do not give a visual representation of the spatial distribution of groundwater parameters. Hence, the use of spatial interpolation tools in ArcGIS to show the spatiotemporal spread of the water classes within the aquifer. Spatial analysis visually and realistically expresses the salinity of the groundwater in a way that point data cannot. The spatial distribution of the EC based on the classification was developed and expressed in Figure 4-7.
Interestingly the spatial analysis shows that the brackish and saline water (EC > 1500µS/cm) generally skewed towards the Ocean. This is likely due to high rainfall recharge (above 1000mm per annum) which tends to increase the heads and flow of freshwater towards the ocean thereby pushing brackish and saline wedge towards the ocean and creeks. As the dry season approached (June to September), the spatial distribution of the fresh, brackish and saline water categories tended towards taking the pre-monsoon (March) forms before the rain started (Figure 4-7a, c). The implication of this variation is that a large scale artificial groundwater recharge during the dry season might be effective in mitigating groundwater salinity in the absence of adequate natural recharge from rainfall. At pre-monsoon, the spatial coverage of saline water in the aquifer was much wider than that of post-monsoon, owing to the same groundwater recharge from rainfall rationale.
illustration not visible in this excerpt
Figure 4-7: EC variation across the study area in; (a.) March (b.) June (c.) September
illustration not visible in this excerpt
Figure 4-8: TDS variation across the study area in; (a.) March (b.) June (c.) September
The settlers of the high elevation regions of Nguu Tatu hills on the western edge of the study area, mostly depend on freshwater supplies from reservoirs managed by the local water management agencies. However, the water samples taken from a Well located at an elevation of 40m above msl were observed to have high EC, TDS and NaCl values which are strong indicators of salinity. This explains why the spatial variation maps show high salinity for the groundwater within Nguu Tatu hills at both pre-monsoon and post-monsoon (Figure 4-7a, c).
The spatial variation of TDS values in the aquifer is represented by Figure 4-8. The spatial distribution of the TDS concentrations in the aquifer also exhibited a similar trend to what was observed in the EC spatial distribution. This is not surprising since the correlation relationship between EC and TDS values were near perfect across the three sets of data. See Appendix V for the spatial distribution of the NaCl concentrations.
In summary, the salinity distribution in the groundwater varies with seasons while groundwater recharge heavily influences the salinity of the groundwater. The salinity of the groundwater tended to reduce in the aquifer during the rainy season.
4.3 Seawater intrusion assessment of the aquifer
Seawater contains relatively higher concentrations of Cl-, Na+ + K+, and Mg2+ in excess of Ca2+ and SO4-2 + HCO3- while, continental fresh groundwater is characterised by high variability of chemical compositions (Moujabber et al., 2006). Hem (1989), also established that the dominant ions found in seawater are Na+ and Cl- while the major ions in freshwater are bicarbonates and calcium. These ions formed the basis for the indices used for assessing the extent of seawater intrusion in the study area.
Several approaches have been identified and widely applied in assessing the extent of seawater intrusion in a coastal aquifer. The most prominent index is the Simpson's ratio highlighted in Todd & May (2005) which was applied in this work. It is the ratio of the chloride content to the concentration of bicarbonate in a water sample (Cl/HCO3) and has been widely applied in many studies (El Moujabber et al., 2006; Korfali & Jurdi 2010)
The classification is as follows;
–Good quality (<0.5);
Slightly contaminated (0.5-1.3);
Moderately contaminated (1.3-2.8);
Injuriously contaminated (2.8-6.6); and
Highly contaminated (6.6-15.5).
The groundwater samples in the study area were categorised based on this classification. These values were interpolated across the study area and the spatial maps were .
The other parameters used for delineating the extent of seawater intrusion in the study are;
- Na/Cl ratio (Vengosh et al., 1997; Bear et al., 1999)
- Plots of TDS against HCO-3/Cl- and Ca2+/Na+ ratios (Mondal et.al, 2010).
4.3.1 Results and Discussion
The Cl/HCO3, Na/Cl, Ca/Na indices were computed and highlighted in Table 4-6 while the correlations amongst the indices are expressed in Table 4-7. Observations made for the pre-monsoon show that EC had strong positive correlations with TDS, Cl/HCO3 ratio, and Na/Cl ratio with values 0.99, 0.74, and 0.93 respectively (Table 4-7). The Cl/HCO3 and Na/Cl also strongly positively correlated with a correlation coefficient of 0.74 (Table 4-7). EC and TDS were observed to have a near perfect correlation with NaCl which in turn has a near perfect relationship with Cl/HCO3 (Table 4-7). This direct relationship between the NaCl values and Cl/HCO3 ratio is a pointer that the salinity of the groundwater may be as a result of the influence of seawater intrusion in the coastal aquifer.
Table 4-6: Seawater intrusion indices
illustration not visible in this excerpt
Table 4-7: Correlation for Salinity Indices (March)
illustration not visible in this excerpt
The correlation trend observed for the relationships amongst EC, TDS, Cl/HCO3, Na/Cl and NaCl indices in the month of March(pre-monsoon) was also observed in September (post monsoon) as shown in Appendix VI. In the case of June (rainy), a similar trend was observed for the relationship amongst EC, TDS and NaCl while that of Cl/HCO3 and Na/Cl was incoherent (Appendix VI).
i. Cl/HCO3 (Simpson’s) Ratio
Cl/HCO3 ratio is a widely applied index for assessing the impact of seawater intrusion on coastal groundwater because the value is 190/1 in seawater (Todd & Mays, 2005). The spatial distribution of the Cl/HCO3 is represented by Figure 4-9 while the percentages of the spatial coverages were further computed as shown in Table 4-8.
Figure 4-9 shows the aquifer was largely experiencing a moderate impact by seawater intrusion at pre-monsoon. However, at post monsoon, a sizeable portion of the aquifer experienced slight contamination while a little portion experienced an injurious impact of seawater intrusion. It will be noticed that the portions under slight impact are the regions tending towards the high elevated Nguu Tatu hills where the water table is high. Darcy’s law and the Ghyben-Herzberg principle seems to be at work. Water tends to flow from regions of higher elevation to that of lower elevations (Darcy’s law) and in the process, higher hydraulic pressures are built which counterbalances the intrusion of seawater (Ghyben-Herzberg principle).
According to Table 4-8, 90% and 11% of the aquifer coverage experienced moderate and injurious contamination by seawater respectively at pre-monsoon. However, no injurious impact was observed in the rainy season probably due to the groundwater recharge from rainfall which tends to raise the hydraulic pressure of the fresh groundwater to offset the intrusion of seawater.
illustration not visible in this excerpt
Figure 4-9: Spatial representation the Cl/HCO3 ratios based on Todd & Mays (2005) for the months of (a.) March (b.) June (c.) September
Table 4-8: Spatial coverage of the Cl/HCO3 index classifications in percentages
illustration not visible in this excerpt
The impact of seawater intrusion in the aquifer during the rainy season was largely moderate. Since the impact of seawater intrusion is a dynamic process, the rainy season will likely experience varying impacts of seawater intrusion within a relatively shorter time due to intermittent rainfall. By post-monsoon, some portions (3%) of the aquifer had started experiencing injurious impacts of seawater similar to pre-monsoon. Surprisingly, 55% of the aquifer coverage showed slight contamination by seawater based on the same probable reasons of Darcy’s law and Ghyben- Herzberg principle. However, holistically Table 4-8 confirms that the impact of seawater intrusion on the aquifer is largely slight/moderate.
ii Na/Cl Ratio:
Vengosh et al., (1997) found that groundwater impacted by seawater intrusion possesses low characteristic values of Na/Cl ratios because marine values of Na/Cl ratio are less than 0.86. Thus, low values of Na/Cl ratios combined with other geological parameters can be indicative of the impact of seawater intrusion. Na/Cl ratios greater than 1 are typical of groundwater contaminated by anthropogenic sources (Jones et al., 1999). The Na/Cl ratios of groundwater samples taken in March have values ranging from 0.5 to 2.01 while those taken in the months of June and September had ranges of values 0.75 to 12.91 and 0.36 to 3.06 respectively (Table 4-6). Over 50% of the samples taken in March, 80% in June and 70% in September had values higher than 1 suggesting the possibility of anthropogenic influence. A previous work on the aquifer also suggested that the groundwater is being impacted by anthropogenic pollution (Munga et al., 2006).
iii Plots of TDS against HCO-3/Cl- and Ca2+/Na+ ratios
High concentrations of Na+ and Cl- in the coastal groundwater may be indicative of the impact of seawater mixing while high values of HCO-3 and Ca2+ are indicators of the impact of water-rock interaction in the aquifer (Park et al., 2005). Hence low values of HCO-3/Cl- and Ca2+/Na+ ratios are indicative of a high impact of seawater intrusion.
The TDS values were plotted against their HCO-3/ Cl- and Ca2+/ Na+ ratios. At pre-monsoon, it was observed that the regression line for samples with low TDS values in the study area represented by the continuous trend line had a strongly negative slope with HCO-3/ Cl- ratios while for samples with high TDS values, the regression line represented by the dotted trend lines showed a mildly positive almost horizontal slope (Figure 4-10a). The black trend line is seen to be negatively sloping up to 2000mg/l; beyond that point, a near horizontal slope is observed. As earlier discussed, high TDS values (TDS> 1000mg/l) are indicative of salinity. Hence, the trend observed between TDS and HCO-3/ Cl- may imply that at low TDS concentrations (low salinity), the process of seawater mixing tends to be more active. However, as the water becomes more saline, an equilibrium point is reached where the fresh/sea water mixing becomes fairly constant. The inverse relationship between TDS and HCO-3/ Cl- also suggests that the freshwater is being impacted by seawater intrusion until it reaches an equilibrium. The same trend is observed for TDS vs. Ca2+/ Na+ ratios (Figure4-10b). On the contrary, Mondal et.al, (2010) observed positive trends at lower values of TDS and mildly negative trends at high values in the case of Tamilnadu India.
illustration not visible in this excerpt
Figure 4-10: TDS Concentration against Molar Ratios of (a. HCO3/Cl; b. Ca/Na)
CHAPTER FIVE
Spatial Vulnerability Mapping of the Coastal Aquifer to Seawater intrusion
This study mapped out the vulnerability of the coastal aquifer to seawater intrusion using the GALDIT overlay index. Lobo-Ferreira & Cabral (1991) defined the vulnerability of groundwater to seawater intrusion as “the sensitivity of groundwater quality to an imposed groundwater pumping or sea level rise or both in the coastal belt, which is determined by the intrinsic characteristics of the aquifer”. GALDIT identified the most important characteristics of the coastal aquifer influencing the intrusion of seawater as;
- Groundwater Occurrence (aquifer type; unconfined, confined and leaky confined)
- Aquifer Hydraulic Conductivity
- Height of Groundwater Level above Sea level
- Distance to the seawater body (Inland distance perpendicular from the shoreline)
- Impact of existing Status of seawater intrusion in the area under study
- Thickness of the aquifer being mapped
5.1 GALDIT factors for the coastal aquifer of the study area
Based on the peculiarity of the study area, each of the factors is described below;
5.1.1 Groundwater Occurrence
Groundwater tapped from aquifers are in layers and these layers may be unconfined, confined, leaky confined or have geological boundaries. Confined aquifers are the most vulnerable due to their relatively larger cone of depression and high-pressure release of water during pumping. It gets a rating of 10. The importance ratings for the aquifer types are all shown in Table 5-1. Groundwater in the study area is stored in an unconfined aquifer comprising limestone, coral reefs, sandstones and shales to average depths of 100m to the surface (Caswell, 1954; Caswell, 2007). The groundwater occurrence, therefore, has an importance rating of 7.5.
Table 5-1: Ratings for GALDIT parameter G
illustration not visible in this excerpt
5.1.2 Aquifer hydraulic conductivity
Also referred to as the coefficient of permeability, aquifer hydraulic conductivity is a measure of the aquifer’s ability to transmit water through its pores when subjected to head differences. By definition, it is the flow per unit cross-sectional area of the aquifer when subjected to a unit head (hydraulic) per unit length of flow (m/day) (Deb, 2014). Higher conductivities increase the risk of seawater intrusion and also results in wide cones of depression during pumping (Sophiya & Syed 2013). This implies hydraulic conductivity K does not only influence the influx of seawater inland but also the rate at which fresh groundwater moves seawards i.e. it plays a defining role in the freshwater-saltwater interface of coastal areas. The hydraulic conductivity of the porous media covering the study area varies from less than 4 to 12 m/day (Munga et al., 2006). Hence, the modified rating for hydraulic conductivity in the study area is expressed in Table 5-2.
Table 5-2: Ratings for GALDIT parameter A
illustration not visible in this excerpt
5.1.3 Height of ground water above the mean sea level
This factor heavily influences the freshwater hydraulic pressure required to counterbalance the intrusion of seawater. The higher the groundwater level above sea level, the higher the hydraulic pressure and hence, the lower the risk of seawater intrusion (Chachadi & Lobo-Ferreira, 2001). The static water levels were taken both at the peak of the dry season and during the rainy season. Digital Elevation Model (DEM) for the study area was used in conjunction with static water level measurements of the selected boreholes to estimate the groundwater levels above the mean sea level. The static water levels measured across the study area were relatively shallow ranging from 8m to 24m. By estimation, the groundwater levels above sea level were therefore found to vary from -3m to 33m above mean sea level across the two seasons. Table 5-3 shows the modified ratings for the height above sea level of the groundwater in the study area.
Table 5-3: Ratings for GALDIT parameter L
illustration not visible in this excerpt
5.1.4 Distance from saltwater body
The closer the aquifer is to the shore or the creek, the higher the impact of. The study area is bounded by the Indian Ocean to the east and by creeks on the north and south. The distance of each point to the body was calculated using the NEAR spatial analyst tool on ArcGIS. The modified ratings for the factor D is highlighted in 5-4.
Table 5-4: Ratings for GALDIT parameter D
illustration not visible in this excerpt
5.1.5 Impact of existing status of seawater intrusion
There are several indices for assessing the extent of seawater intrusion in a study area. Base Exchange Index, Graphical methods, Na+/Cl-, Ca2+/Na+ and Cl-/HCO3- are but a few of the widely used indices (Klassen, 2014). The ratio of chloride to bicarbonate, also referred to as Revelle’s coefficient or Simpson's ratio is mostly recommended (Chachadi & Lobo-Ferreira, 2001). The ratios for the study area were found to range from 1.16 to 3.97. Table 5-5 represents the modified ratings for the impact of the existing status of seawater intrusion.
Table 5-5: Ratings for GALDIT parameter I
illustration not visible in this excerpt
5.1.6 Thickness of the aquifer being mapped
The saturated thickness of an unconfined aquifer also influences the extent and magnitude of seawater intrusion. The larger the thickness of the aquifer, the greater the likelihood of seawater intrusion and vice versa (Chachadi & Lobo-Ferreira 2001; Najib et al., 2012; Kura et al., 2014).
From the boreholes log data gathered, the lithology of the study area is heterogeneous in nature showing varying layers of limestone, sandstone and shales. Limestone and Sandstones are good aquifers while shale is not. Cross-sectional views of the lithology as revealed in the geological map show that coral reefs, kilindini sands/North Mombasa crag, and the wind-blown superficial sands extend far below 100m in most areas implying a thick heterogeneous aquifer. The ratings for the aquifer thickness, therefore, has a maximum value of 10. The ratings for the aquifer thickness is expressed in Table 5-6
Table 5-6: Ratings for GALDIT parameter T
illustration not visible in this excerpt
Where;
illustration not visible in this excerpt
W1 to W6 are the respective relative weights given to the six hydrogeological factors.
illustration not visible in this excerpt
Once the GALDIT-Index is calculated, the vulnerability is then classified. The minimum and maximum range of GALDIT index scores are divided into; low, moderate, and high vulnerability classes ranging from 2.5 to 10. The computation for the GALDIT indices is shown in Table 5-7;
Table 5-7: GALDIT Index computation
illustration not visible in this excerpt
These factors were mapped into raster layers on GIS and overlaid based on their weights, ranges and importance ratings. The final vulnerability maps which bore the weighted sums of the individual layers were produced.
5.2 Generalised workflow for the vulnerability analysis
The general workflow for developing the spatial vulnerability map is highlighted in Figure 5-1.
illustration not visible in this excerpt
Figure 5-1: General methodology for the vulnerability analysis
5.2.1 Data collection and sorting
The secondary, field and laboratory data obtained were organised into folders usable for further processing. The borehole historical data obtained from WRMA were saved in an excel database, the geological maps and topographic maps were scanned and geo-referenced. The coordinates of landmarks and boreholes taken during the field trip were imported into the computer using DNR-Garmin software. The geodatabase was created where all these data were stored either as TIFFS. jpegs excel, or GPS exchange format.
5.2.2 Data processing
Thematic raster layers were created from the geodatabase for each factor by clipping the geological map into the area of interest, interpolating the DEM on point data to establish heights above sea levels of the groundwater heads, and using spatial interpolation tools to create rasters from point data. The six thematic layers created for the GALDIT factors all had uniform resolutions of 30m.
5.2.3 Overlay and spatial analysis
Based on the GALDIT index computation, appropriate weightages, ranges and importance ratings as reflected in Tables 5-1 – 5-7 were assigned to each thematic raster layer representing each GALDIT factor. The thematic layers were then overlaid and the aggregate weighted sum derived. The final vulnerability rasters were resized into weights of importance range of 2.5, 5, 7.5 and 10. These importance ranges were classified into low, medium or high vulnerabilities.
5.2.4 Cartography (final vulnerability maps)
Map elements such as the legend, north arrow, scale, geographic grids were inserted into the final vulnerability rasters to form maps which are easily understood and interpreted by any reader.
5.3 Results and Discussion
The spatial vulnerability maps of the aquifer based on the six factors highlighted in GALDIT (Lobo Ferreira et al., 2005) were developed for the dry and rainy seasons. Four out of the six GALDIT factors were static factors while the other two were dynamic.
The static factors which do not change considerably over a period of time were; Groundwater occurrence, Aquifer hydraulic conductivity, Distance to the shore and Thickness of the aquifer. The dynamic factors which experience temporal variation with time are the; Level of groundwater above sea level and Impact of existing seawater intrusion. The same static factors were applied for the dry and rainy seasons while separate dynamic factors were computed and applied accordingly for either period. The spatial maps developed for each factor are found in Appendix VII. The computed results of Equation 8 (page 56) showed that the total scores (TS) for the final vulnerability map ranged from 55 to 120 in March while the scores were between 57.5 and 132.5 in the month of June. These TS values were split into low, moderate and high vulnerability classes.
The overlay and reclassification of the weighted sum of the thematic raster images gave the final vulnerability maps of the study area to seawater intrusion (Figure 5-2).
illustration not visible in this excerpt
Figure 5-2: Vulnerability maps of the study area to seawater intrusion a. (pre-rains) b. (rainy season)
The highly vulnerable regions in both cases were the low-head regions with high proximity to the Ocean and creeks. It implies that GALDIT factors L and D are the most sensitive factors influencing the vulnerability of the study area to seawater intrusion. GALDIT factor D cannot be altered because locations do not change, however, the static levels of groundwater can be managed through recharge and abstraction dynamics of the groundwater. It is, therefore, unsurprising that the least vulnerable regions are located close to the high elevated areas of Nguu Tatu hills whose groundwater heads are relatively high as shown in Figure 5-2.
The percentage covers of each vulnerability class in each map are represented in Table 5-8. A wider region became less vulnerable to seawater intrusion in the rainy season as observed by the increase in the low vulnerability regions from 13% to 20% (Table 5-8). This is probably because the impact of seawater intrusion is comparatively lower in the rainy season as a result of high rate of natural groundwater recharge from rainfall. Groundwater recharge increases pressure heads of freshwater, thereby pushing freshwater/saltwater equilibrium zones towards the sea (Roger, 1998).
Table 5-8: Percentage changes in vulnerability classes between pre-monsoon and rainy season
illustration not visible in this excerpt
Generally, moderate vulnerability zones had the highest percentage spread in the aquifer at both pre-monsoon and rainy seasons. Invariably, the vulnerability of the aquifer to seawater intrusion is largely moderate. Hence, the current groundwater abstraction rates still seem low with respect to rainfall recharge which is usually above 1000mm per year (Climatemps, 2015).
CHAPTER SIX
Groundwater flow and solute transport simulation of the study area.
Geographic data process (GDP) was loosely coupled with the process-based model (PBM) in this study. The PBMs include MODFLOW, MT3D and SEAWAT. All three models are embedded in the Groundwater Modelling System (GMS) software package used for this study.
The MODFLOW module was used for solving the steady and transient state simulation based on the modified three-dimensional groundwater flow (Equation 3&4, page 20). The MT3D package is based on the partial three-dimensional equation for solute transport simulation given by Freeze & Cherry, (1979) as highlighted in Equation 5 (page 21). Equation 5 was slightly modified by Zheng & Wang (1999) and represented as Equation 6 (page 21) and subsequently acts as the running equation behind the MT3D. The SEAWAT package combined the groundwater flow solution from MODFLOW and the partial 3D solute transport solution from MT3D. The flow and solute transport procedures in the MODFLOW and MT3D were repeatedly solved by the SEAWAT package until the provided stress periods and simulation were complete. The underlying flow equation behind the SEAWAT used is given in Equation 7 (page 22).
GIS files were independently imported into the GMS software for analysis. The general steps involved in the flow and solute transport simulation is expressed in Figure 6-1.
illustration not visible in this excerpt
Figure 6-1: General steps involved in the groundwater flow and solute transport simulation
6.1.1 Importing GIS Files into GMS (loose coupling)
The model inputs used for the process based modelling were independently imported from the GIS database. The following GIS files were imported into the GMS software.
- Study area boundary (shapefile)
- Hydraulic conductivity polygon (shapefile)
- Recharge polygon (shapefile
- Borehole/wells data (shapefile)
- Digital Elevation model of the study area (raster)
- Hydraulic head (raster)
These were either converted to feature objects or interpolated to the 3D grid within the GMS interface (Figure 6-2). These files were used to create the boundary layer; Source & sink layer; Recharge layer; hydraulic conductivity and the elevations in the workspace (GMS, 2016a, b).
illustration not visible in this excerpt
Figure 6-2: GMS software interface showing the GIS and GMS data layers in the project explorer
6.1.2 Building the Conceptual Model for the Aquifer
This involves defining the hydraulic boundary conditions and imputing the parameters that describe the aquifer being modelled. The hydraulic boundary conditions were delineated as either no-flow boundary or constant head boundary. The portions bounded by the Ocean and creeks (northern, southern and eastern boundaries) are the constant head boundaries (Figure 6-3). The western boundary consisting the elevated hills of Nguu-Tatu formed the no-flow boundary.
illustration not visible in this excerpt
Figure 6-3: The hydraulic boundary conditions of the study area
The detailed steps involved in the creation of the conceptual model are highlighted in GMS (2016b). The parameters used in building the conceptual model were taken from secondary data, a previous study (Munga et al., 2006), borehole geophysical records, as well as historical rainfall data. They are given as follows;
- Recharge: Based on the geological formation, recharge was estimated based on 5 and 10% of average monthly rainfall data for the past 10 years.
- Horizontal hydraulic conductivity (k): 4 – 12 m/day
- Vertical hydraulic conductivity: ¼ of k
- Specific yield: 0.1 – 0.2
- Specific storage: 0.0001
- Porosity (ϴ): 0.3
- Coefficient of longitudinal dispersivity: 20
- Fluid density (ρ): 1000kg/m3 for freshwater & 1025kg/m3 for saltwater
- Transmissivity(T): k × (Aquifer thickness)
- Specific capacity- 0.6854 m2/hr
- Flow rates/ discharges: based on specific wells
- Hydraulic heads: estimated from static water levels taken from each borehole and interpolated into a raster.
6.1.3 Steady State Flow Simulation
Steady state flow occurs when the magnitude and direction of the flow are kept constant along the entire hydrologic boundary. In other words, the hydraulic head remains constant with time but this does not imply a lack of movement in the groundwater. It simply implies the net flow in and out of the system is kept constant. Hence the storage term parameter in the equation is not utilised since there are no changes in head and obviously no change in the volume of water stored within the domain. From the conceptual model, the steady-state simulation was solved for the study area using the initial interpolated head raster as the starting heads. The flow direction was established at the end of the simulation. The steps adopted in the study are highlighted in GMS (2016b).
6.1.4 Transient State Flow Simulation
Transient state flow occurs when the direction and magnitude of the flow changes with respect to time. Transient models require commencing with a steady-state stress period, establishing the starting heads equal to the solution generated from a steady state model, or allowing some time at the start of the transient model for the heads to stabilise before applying any changes in stresses (pumping rates, recharge rates, etc.). In this study’s simulation, the starting heads generated from the steady state solution were used in defining the initial stress condition of the transient state flow simulation. The subsequent stress periods were defined by incorporating the storage parameter, specific yield, and transient recharge data within the domain, and the abstraction rates from boreholes at intervals. These parameters were incorporated into the conceptual model and enabled for transient state simulation (GMS, 2016c). The subsequent stress periods were chosen to coincide with the data collection periods i.e., 29th June 2016, 1st September 2016, and 15th September 2016.
6.1.5 Model Calibration
Model calibration is essential in validating that the behaviour of the model is as representative as possible of the actual aquifer being modelled. It is a process where certain parameters in the model such as the recharge and hydraulic conductivity are altered systematically and repeatedly until the computed solution matches the field observed values within an acceptable level of accuracy. The model was calibrated using observed heads taken from boreholes/wells over the period of simulation (GMS 2016c). The acceptable observation head interval used was 1m. The Hydraulic conductivity values and transient recharge data were iteratively altered until the observed errors were within the acceptable intervals.
6.1.6 Solute Transport Simulation (MT3DMS and SEAWAT)
MT3D simulations are either constructed from the grid approach or the conceptual model approach. In both cases, solutions from the groundwater flows are required for running. This study employed the conceptual model approach where data were entered via points, arcs and polygons (GMS, 2016e). Using the solutions for the calibrated transient state simulation, the solute species was defined (sea salt) and initial concentrations of the salt imputed based on the field data. The MT3D packages needed for SEAWAT simulation are the advection and source/sink mixing packages. Here the porosity and longitudinal dispersivity components of the aquifer characteristics were defined in the conceptual model. The MT3D module was run and the results viewed before creating a new simulation for the SEAWAT.
Solutions from the MODFLOW and MT3D were incorporated into the SEAWAT module for simulating the intrusion of seawater within the GMS software package (GMS, 2016f).
6.2 Results and Discussion
The results for the steady state flow, transient state and the solute transport are highlighted and discussed in this section.
6.2.1 Steady State Flow Simulation
The outcomes of the steady state flow simulation gave the direction of groundwater flow in the study area. Figure 6-4. The direction of groundwater flow is generally towards the northeast and southern directions of the study area. The groundwater heads were similarly observed to be highest in the high elevation areas of Nguu Tatu hills on the western part of the study area.
illustration not visible in this excerpt
Figure 6-4: Steady-state simulated groundwater heads (m) and flow direction in the North Coast of Mombasa
The groundwater flows generally towards the Indian Ocean, to Mtwapa creek in the Shanzu region on the north-eastern part of the study area and the Tudor creek close to the Nyali/Kongowea region on the southern region. The groundwater flow direction follows Darcy’s basic law of fluid motion (Kiplangat, 2014). The steady-state simulated flow direction affirms a previous study by Munga et al., (2006) where the outcome of their steady state simulation showed that the prevailing groundwater flow direction is towards the “Mtwapa Creek along the northern boundary and Tudor Creek along the southern boundary of the study area. The southern end of the study area bordering the Indian Ocean but experiencing a flow towards creek is a region of high groundwater head. The most probable reason is, since water moves by gravity, the groundwater’s movement from that high head region could only move to lower head regions- in this case, the creek.
6.2.2 Transient State Flow Simulation and Calibration
The transient state solution over the three stress periods showed very little variation in groundwater heads. The observed field data was used to calibrate the transient state solution until the error margin was within the allowable interval (1m). Calibration was done by varying the recharge and hydraulic conductivity values. The difference between the observed and computed heads are highlighted in Figures 6-5 and 6-6.
illustration not visible in this excerpt
Figure 6-5: Differences between the Observed and Computed heads after calibration
illustration not visible in this excerpt
Figure 6-6: Cross plot for Observed heads vs. computed heads
6.2.3 Solute Transport Simulation
The solute transport solutions for three stress periods are represented in Figures 6-7 and 6-8. The red zones in each sub-figure are the regions unaffected by the transport of NaCl solute from the Ocean and creeks. The red zones showed a minute progressive increase from June to September. This implies that as the season progressed from rainy to post-monsoon the concentrations of NaCl in the groundwater reduced gradually from the centre of the study area outwards. However, the general observation is that more refined models with smaller grids will be required to simulate higher details of the solute transport. This will require more refined information on the hydraulic conductivity, a larger number of data collection points and a longer range of continuous data.
Therefore, the model needs to be improved upon by using data from longer time frames (at least 2 years) to make it more robust for long-term predictive studies.
illustration not visible in this excerpt
Figure 6-7: Simulated NaCl solute concentrations in mg/l for (a.) 25th March (Initial concentration) and (b.) 29th June 2016
illustration not visible in this excerpt
Figure 6-8: Simulated NaCl solute concentrations in mg/l for (c.) 1st September and (d.) 15th September 2016
CHAPTER SEVEN- CONCLUSIONS AND RECOMMENDATIONS
7.1 Conclusions
Statistical, geospatial and modelling approaches were applied in extensively assessing the hydrogeology, vulnerability to seawater intrusion, and flow of the groundwater in a coastal aquifer. The conclusions are highlighted below;
a. Combining statistical, geospatial and modelling approaches in groundwater studies are complementary and give more robust insights on an aquifer.
b. The aquifer under study was observed to be unconfined, with water heads varying from -1m to 33m above mean sea level. The groundwater heads did not vary considerably across seasons.
c. The EC and TDS values were observed to have near perfect correlations and generally high as over 94% of the water samples exceeded WHO drinking water limit.
d. The pH of the groundwater was slightly alkaline. Over 90% of the water samples had pH values within 6.5 and 8.5 the acceptable limit of WHO guidelines, however, the pH could be slightly acidic in the rainy season.
e. The parameters with high impacts on the groundwater chemistry are EC, TDS, Cl-, Na+, Mg2+, K+, and to some degree SO42- while Ca2+, HCO3 and pH have paltry impacts on the geochemistry of the aquifer.
f. The groundwater in the coastal aquifer is generally of the marine and deep ancient type.
g. The salinity distribution in the groundwater varies with seasons while groundwater recharge heavily influences the salinity of the groundwater.
h. About 90% of the aquifer coverage experienced a moderate impact of seawater intrusion at pre-monsoon; 100% experienced a moderate impact in June (rainy season); 55% and 42% experienced slight and moderate impacts respectively in September (post-monsoon). There is mild evidence of anthropogenic-driven groundwater pollution.
i. The vulnerability of the aquifer under study to seawater intrusion is largely moderate irrespective of seasons. However, lower vulnerability is experienced in the rainy seasons.
j. The groundwater flow and direction in the coastal aquifer is generally towards the northeast and southern part of the study area.
k. The concentrations of NaCl in the groundwater slightly reduced from the centre of the study area outwards from June (peak of rainy season) to September (post-monsoon).
7.2 Recommendations
Several methods have been devised all over the world for preventing or controlling the impact of seawater intrusion into coastal aquifers. Some of the direct preventive methods are; extraction barriers, injection barriers, artificial recharge (Todd & Mays 2005; Pliakas et al., 2005). However, these direct methods are usually financially costly and technically demanding (Kallioras et al., 2013), hence, may prove unfeasible in the study area. Based on the conclusions drawn from the study, more realistic indirect methods revolving around management may be recommended as follows;
a. The inventory of wells developed from this study may be built upon for long term hydrogeologic regime of the aquifer for effective aquifer system management
b. Management of the pumping scheme- By working closely with the local Water Resources Management Authority, the private borehole owners could be guided by the GALDIT vulnerability map. Boreholes/wells located in the regions of low vulnerability may be advised to stay in full operation while rationing of the pumping rates may be required for the injurious vulnerability regions i.e. operate fully in the wet season but in shifts during the dry season.
c. More detailed records on all existing and future groundwater wells should be created to include pumping test results, initial concentrations of water quality parameters, and depth of wells. Future boreholes/wells that will be located in injurious vulnerability regions should be well studied to obtain the shallowest possible effective depth.
Recommended areas of further research identified from this study are as follows;
a. Empirical studies on the horizontal and vertical hydraulic conductivities under isotropic, and anisotropic conditions
b. Long term regular and periodic groundwater data collection and monitoring are needed for simulating a more robust flow and solute transport model
c. Detailed long term and predictive studies on the effects of pumping regimes on the intrusion of seawater.
d. Further studies need to be conducted on why the groundwater samples around the peak of Nguu Tatu hills where found salty in spite of high water table.
REFERENCES
Abuodha PA (2003) Geomorphology of the Kenyan coast: Not as a result of sea level change alone. Buletim Geologico no. 43 Mozambique. I - 8
Adepelumi, A. A., Ako, B. D., Ajayi, T. R., Afolabi, O., & Omotoso, E. J. (2009). Delineation of saltwater intrusion into the freshwater aquifer of Lekki Peninsula, Lagos, Nigeria. Environmental Geology, 56, 927–933.
Albinet M, Margat J (1970) Cartographie de la vulnérabilité à la pollution des nappes d’eau souterraine Orleans, France. Bull BRGM 2ème série, 4 : 13–22.
Aller L, Bennet T, Lehr J.H, Petty R.J (1987) DRASTIC: a standardised system for evaluating groundwater pollution potential using hydrologic settings. US EPA Report, 600/2–87/ 035, Robert S. Kerr Environmental Research Laboratory, Ada, OK.
Aquaveo. (2016). Introduction. Retrieved September 5, 2016, from http://www.aquaveo.com/In.
Ashton, P.J., 2002: Avoiding conflicts over Africa’s water resources. Ambio, 31,236-242.
Ayolabi, E. A., Folorunso, A. F., Odukoya, A. M., & Adeniran, A. E. (2013). Mapping saline water intrusion into the coastal aquifer with geophysical and geochemical techniques: The University of Lagos campus case (Nigeria). SpringerPlus, 2 (1), 433. doi:10.1186/2193-1801-2-433.
Bear, J., Cheng, A., Sorek, S., Herrera, I. and Ouazar, D. (eds) (1999) Seawater intrusion in coastal aquifers: concepts, methods and practices. Boston. Mass: Kluwer Academic.
Balathandayutham, K., Mayilswami, C. and Tamilmani, D. (2015). ‘Assessment of groundwater quality using GIS: A case study of Walayar watershed, Parambikulam-Aliyar-Palar basin, Tamilnadu, India’, Current World Environment, 10(2), pp. 602–609. doi: 10.12944/cwe.10.2.25.
Barlow, P. M. (2003). Ground water in freshwater-saltwater environments of the Atlantic Coast. Reston, VA: U.S. Dept. of the Interior, U.S. Geological Survey.
Benoit J, Hardaway CS, Hernandez D, Holman R, Koch E, McLellan N, Peterson S, Reed D, Suman D (2007) Mitigating shore erosion along sheltered coasts. The National Academies Press, Washington, DC.
Bhatt, G., Kumar, M., & Duffy, C. J. (2014). A tightly coupled GIS and distributed hydrologic modelling framework. Environmental Modelling & Software, 62, 70–84. doi:10.1016/j.envsoft.2014.08.003.
Boko, M., I. Niang, A. Nyong, C. Vogel, A. Githeko, M. Medany, B. Osman-Elasha, R. Tabo and P. Yanda, (2007). Africa. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, M.L. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden and C.E. Hanson, Eds., Cambridge University Press, Cambridge UK, 433-467.
Boon, J. D., Brubaker, J. M., & Forrest, D. R. (2010). Chesapeake Bay Land Subsidence and Sea Level Change: An Evaluation of Past And Present Trends And Future Outlook. Virginia Institute of Marine Science Special Report No. 425 in Applied Marine Science and Ocean Engineering Retrieved from http://www.vims.edu/GreyLit/VIMS/sramsoe425.pdf.
Boswinkel J. A., Information Note, 2000. International Groundwater Resources Assessment Centre (IGRAC), Netherlands Institute of Applied Geoscience, Netherlands.
Bouderbala, A., & Remini, B. (2014). Geophysical approach for assessment of seawater intrusion in the coastal aquifer of Wadi Nador (Tipaza, Algeria). Acta Geophysica, 62 (6),. doi:10.2478/s11600-014-0220-y.
Brown, C. H. (1998). Applied multivariate statistics in geohydrology and related sciences. Berlin: Springer.
Brown, E., Skougstad, M. W., & Fishmen, M. J. (1983). Method for collection and analyzing of water samples for dissolved minerals and gases. Washington DC, USA: U.S. Govt. Printing Office.
Buddemeier, R.W., S.V. Smith, D.P. Swaaney and C.J. Crossland, (2002). The Role of the Coastal Ocean in the disturbed and undisturbed Nutrient and Carbon Cycles. LOICZ Reports and Studies Series No. 24. 84 pp.
Caswell, P. V. (1954a). Geology of the Kilifi-Mazeras area degree sheet 66 SE quarter [Map]. Nairobi: Survey, Min., Colony and Protectorate of Kenya.
Caswell, P. V. (1954b). Geology of the Mombasa-Kwale area degree sheet 69; with coloured map [Map]. Nairobi: Survey, Min., Colony and Protectorate of Kenya.
Caswell, P. V. (2007). Geology of the Kilifi -mazeras area (Report No 34, pp. 1-54), Reprinted 2007 Kenya, Ministry of Mining, Geology Department. NAIROBI, Nairobi Area.
Chachadi A.G. & Lobo-Ferreira, J.P (2001), Sea water intrusion vulnerability mapping of aquifers using GALDIT method. Proc. Workshop on modelling in hydrogeology, Anna University, Chennai, pp.143-156, and in COASTIN A Coastal Policy Research Newsletter, Number 4, March 2001. New Delhi, TERI, pp.7-9, (cf http://www.teriin.org/teriwr/coastin/newslett/coastin4.pdf).
Chachadi A.G. & Lobo-Ferreira, J.P (2001), Sea water intrusion vulnerability mapping of aquifers using GALDIT method. Proc. Workshop on modelling in hydrogeology, Anna University, Chennai, pp.143-156, and in COASTIN A Coastal Policy Research Newsletter, Number 4, March 2001. New Delhi, TERI, pp. 7-9, (cf http://www.teriin.org/teri-wr/coastin/newslett/coastin4.pdf).
Chachadi AG, Lobo-Ferreira JP (2005) Assessing aquifer vulnerability to sea-water intrusion using GALDIT method: Part 2–– GALDIT Indicators Description, Fourth Inter-Celtic Colloquium on Hydrogeology and Management of Water Resources, Portugal, 11–14, July 2005, CD of Proceeding.
Chachadi AG, Lobo-Ferreira JP (2005) Assessing aquifer vulnerability to sea-water intrusion using GALDIT method: Part 2–– GALDIT Indicators Description, Fourth Inter-Celtic Colloquium on Hydrogeology and Management of Water Resources, Portugal, 11–14, July 2005, CD Of Proceeding.
Cimino, A., Cosentino, C., Oieni, A., & Tranchina, L. (2007). A geophysical and geochemical approach for seawater intrusion assessment in the Acquedolci coastal aquifer (northern Sicily). Environmental Geology, 55 (7), 1473–1482. doi:10.1007/s00254-007-1097-8.
Civita M (1994) Le carte della vulnerabilità degli acquiferi all’inquinamento. Teoria and practica (Aquifer vulnerability maps to pollution) Pitagora, Bologna.
Civita M, De Regibus C (1995) Sperimentazione di alcune metodologie per la valutazione della vulnerabilità degli aquiferi. Q Geol Appl Pitagora, Bologna, 3: 63–71.
Climatemps, (2015). Mombasa climate Mombasa temperatures Mombasa weather averages. Retrieved July 24, 2016, from http://www.mombasa.climatemps.com/.
Corniello A, Ducci D, Napolitano P (1997). Comparison between parametric methods to evaluate aquifer pollution vulnerability using GIS: an example in the ”Piana Campana“, southern Italy. In: Marinos PG, Koukis GC, Tsiambaos GC, and Stournaras GC (eds) Engineering geology and the environment. Balkema, Rotterdam, pp 1721–1726.
Darcy, H. (1856). Les Fontaines Publiques de la Ville de Dijon, Dalmont, Paris.
Darnault C.J.G & Godinez I.G (2008). Coastal Aquifers and Seawater intrusion. In Christophe J.G Darnault (Editor), Overexploitation and contamination of shared groundwater resources (pp. 185 - 201), New York, NY, United States: Springer science+Business Media B. V.
Deb, P. K. (2014). Pumping tests and Aquifer parameters. In P. Kumar Deb, An Introduction to Mine Hydrogeology, SpringerBriefs in Water Science and Technology (pp. 13–21) DOI: 10.1007/978-3-319-02988-7_3.
Doerfliger N, Zwahlen F (1997). EPIK: a new method for outlining of protection areas in karstic environment. In: Gü- nay G, Jonshon AI (eds) International symposium and field seminar on “karst waters and environmental impacts”. Antalya, Turkey. Balkema, Rotterdam, pp 117–123.
Draoui, M., Vias, J., Andreo, B., Targuisti, K., & Stitou El Messari, J. (2007). A comparative study of four vulnerability mapping methods in a detritic aquifer under mediterranean climatic conditions. Environmental Geology, 54 (3), 455–463. doi:10.1007/s00254-007-0850-3.
Fadili, A., Mehdi, K., Riss, J., Najib, S., Makan, A., & Boutayab, K. (2015). Evaluation of groundwater mineralization processes and seawater intrusion extension in the coastal aquifer of Oualidia, Morocco: Hydrochemical and geophysical approach. Arabian Journal of Geosciences, 8 (10), 8567–8582. doi:10.1007/s12517-015-1808-5.
Foster SSD (1987). Fundamental concepts in aquifer vulnerability, pollution risk and protection strategy. In: Duijvenbooden W van, Waegeningh HG van (eds) TNO Committee on Hydrological Research, The Hague. Vulnerability of soil and groundwater to pollutants, Proceedings and Information. 38: 69–86.
Freeze A. and Cherry J.A. (1979), Groundwater, Prentice Hall, Englewood Cliffs.
Gaaloul, N. (2012). Simulation of seawater intrusion in coastal aquifers: Forty Five-Years exploitation in an eastern coast aquifer in NE Tunisia. The Open Hydrology Journal, 6 (1), 31–44. doi:10.2174/1874378101206010031.
GMS, 2016a. GIS Module Importing, displaying, and converting shapefiles. Retrieved 16 June 2016, from http://gmstutorials-10.2.aquaveo.com/GIS.pdf
GMS, 2016b. GMS 10.2 Tutorial MODFLOW – Conceptual Model Approach I. Retrieved 16 June 2016, from http://gmstutorials-10.2.aquaveo.com/MODFLOW-ConceptualModelApproachI.pdf
GMS, 2016c. GMS 10.2 Tutorial MODFLOW – Managing Transient Data. Retrieved 16 June 2016, from http://gmstutorials-10.2.aquaveo.com/MODFLOW-ManagingTransientData.pdf
GMS, 2016d. GMS 10.2 Tutorial MODFLOW – Conceptual Model Approach II. Retrieved 16 June 2016, from http://gmstutorials-10.2.aquaveo.com/MODFLOW-ConceptualModelApproachII.pdf
GMS, 2016e. GMS 10.2 Tutorial MT3DMS – Conceptual Model Approach. Retrieved 16 June 2016, from http://gmstutorials-10.2.aquaveo.com/MT3DMS-ConceptualModelApproach.pdf
GMS, 2016f. GMS 10.2 Tutorial SEAWAT – Conceptual Model Approach. Retrieved 16 June 2016, from http://gmstutorials-10.2.aquaveo.com/SEAWAT-ConceptualModelApproach.pdf
Gogu, R. C., & Dassargues, A. (2000). Current trends and future challenges in groundwater vulnerability assessment using overlay and index methods. Environmental Geology, 39 (6), 549–559. doi:10.1007/s002540050466.
Gogu, R., Carabin, G., Hallet, V., Peters, V., & Dassargues, A. (2001). GIS-based hydrogeological databases and groundwater modelling. Hydrogeology Journal, 9 (6), 555–569. doi:10.1007/s10040-001-0167-3.
GOK 1979. National population and housing census. Vol 1. Ministry of Planning and National Development, Central Bureau of Statistics.
GOK 1989. National population and housing census. Vol 1. Ministry of Planning and National Development, Central Bureau of Statistics.
GOK 1999. National population and housing census. Vol 1. Ministry of Planning and National Development, Central Bureau of Statistics.
Goldman, M., Gilad, D., Ronen, A., & Melloul, A. (1991). Mapping of seawater intrusion into the coastal aquifer of Israel by the time domain electromagnetic method. Geoexploration, 28 (2), 153–174. doi:10.1016/0016-7142(91)90046-f.
Goldscheider N, Klute M, Sturm S, Ho¨tzl H (2000). The PI method—a GIS-based approach to mapping groundwater vulnerability with special consideration on karst aquifers. Z Angew Geol 46(3):157–166.
Gontara, M., Allouche, N., Jmal, I., & Bouri, S. (2016). Sensitivity analysis for the GALDIT method based on the assessment of vulnerability to pollution in the northern Sfax coastal aquifer, Tunisia. Arabian Journal of Geosciences, 9 (5),. doi:10.1007/s12517-016-2437-3.
Goodchild M.F. & Kemp K.K (1990) NCGIA Core curriculum in GIS, Santa Barbara, CA: National Centre for Geographic Information and Analysis.
Gueye, L., M. Bzioul & O. Johnson (2005). Water and sustainable development in the countries of Northern Africa: coping with the challenges and scarcity. Assessing sustainable development in Africa. Africa’s sustainable development Bulletin, Economic Commission for Africa, Addis Ababa, 24-28.
Guo, W., and C.D. Langevin. (2002). User’s guide to SEAWAT: A computer program for the simulation of three-dimensional variable-density groundwater flow. USGS Techniques of Water Resources Investigations Book 6, Chapter A7. USGS.
Harbaugh, A.W., Banta, E.R., Hill, M.C., and McDonald, M.G. (2000). MODFLOW-2000, the U.S. Geological Survey modular ground-water model — User guide to modularization concepts and the Ground-Water Flow Process (PDF). Open-File Report 00-92. U.S. Geological Survey.
Herzberg, A., (1901), Die Wasserversorgung einiger Nordseebder (The water supply of parts of the North Sea coast in Germany), Z. Gasbeleucht. Wasserversorg 44:815–819 and 45:842–844.
Huang, B., & Jiang, B. (2002). AVTOP: A full integration of TOPMODEL into GIS. Environmental Modelling & Software, 17 (3), 261–268.doi:10.1016/s1364-8152(01)00073-1.
Hwang, S., Shin, J., Park, I., & Lee, S. (2004). Assessment of seawater intrusion using geophysical well logging and electrical soundings in a coastal aquifer, Youngkwang-gun, Korea. Exploration Geophysics, 35 (1), 99. doi:10.1071/eg04099.
Idowu, T. E., Nyadawa, M., & K’Orowe, M. O. (2016). Seawater Intrusion Vulnerability Assessment of a Coastal Aquifer: North Coast Of Mombasa, Kenya as a Case Study. Int. Journal of Engineering Research and Application, 6 (8, ( Part -3)), 37–45. ISSN: 2248-9622.
Ingebritsen, S. E., & Galloway, D. L. (2014). Coastal subsidence and relative sea level rise. Environmental Research Letters, 9(9), 091002. doi:10.1088/1748-9326/9/9/.
IPCC, (1997) - R.T.Watson, M.C.Zinyowera, R.H.Moss (Eds) Cambridge University Press, UK. Available from Cambridge University Press, The Edinburgh Building Shaftesbury Road, Cambridge CB2 2RU ENGLAND. Accessible online from http://www.ipcc.ch/ipccreports/sres/regional/index.php?idp=30.
Jackson, J. (2007). Hydrogeology and Groundwater Flow Model, Central Catchment of Bribie Island, SouthEast Queensland (Unpublished master's thesis). School of Natural Resource Sciences, Queensland University of Technology, Queensland Australia.
Javadi, A. A., Abd-Elhamid, H. F., & Farmani, R. (2011). A simulation-optimization model to control seawater intrusion in coastal aquifers using abstraction/recharge wells. International Journal for Numerical and Analytical Methods in Geomechanics, 36 (16), 1757–1779. doi:10.1002/nag.1068.
Kallioras A, Pliakas F, Skias S, Gkiougkis I (2011) Groundwater vulnerability assessment at SW Rhodope aquifer system in NE Greece. In: Lambrakis N, Stournaras G, Katsanou (eds) Advances in the research of aquatic environment. Environ Earth Sci. Springer 2:351–358. doi:10.1007/978-3-642-24076-8_41.
Kallioras Andreas, Pliakas Fotios-Konstantinos, Schuth Christoph & Rausch Randolf (2013). Methods to Countermeasure the Intrusion of Seawater into Coastal Aquifer Systems. In, Sharma, S. K., & Sanghi, R. (Eds). Wastewater Reuse and Management, Springer Science+Business Media Dordrecht: (pp 479 - 490) DOI 10.1007/978-94-007-4942-9_17.
Karmegam, U., Chidambaram, S., Prasanna, M. V., Sasidhar, P., Manikandan, S., Johnsonbabu, G., … Anandhan, P. (2011). A study on the mixing proportion in groundwater samples by using Piper diagram and Phreeqc model. Chinese Journal of Geochemistry, 30 (4), 490–495. doi:10.1007/s11631-011-0533-3.
Kenya, Water Services Regulatory Board (WASREB). (2008). Drinking Water Quality and Effluent Monitoring Guideline.
Kim, K.Y., Chon, C.M. & Park, K.H. (2007) ‘A simple method for locating the fresh Water–Salt water interface using pressure data’, Ground Water, 45(6), pp. 723–728. doi: 10.1111/j.1745-6584.2007.00349.x.
Kim, O.B. & Park, H.Y. (1998) ‘Variation of Hydrochemical characteristics of major elements in groundwater by the seawater intrusion in the Byeonsan Peninsular, Korea’, Geosystem Engineering, 1(2), pp. 106–110. doi: 10.1080/12269328.1998.10541131.
King, A.C., Raiber, M. & Cox, M.E. (2013) ‘Multivariate statistical analysis of hydrochemical data to assess alluvial aquifer–stream connectivity during drought and flood: Cressbrook Creek, southeast Queensland, Australia’, Hydrogeology Journal, 22(2), pp. 481–500. doi: 10.1007/s10040-013-1057-1.
Kiplangat Vincent, K., Nicholas Muthama, M., & Nicodemus Muoki, S. (2014). Darcy’s law equation with application to underground seepage in earth dams in calculation of the amount of seepage. American Journal of Applied Mathematics and Statistics, 2 (3), 143–149. doi:10.12691/ajams-2-3-8
Klassen, J., Allen, D. M., & Kirste, D. (2014). Chemical Indicators of Saltwater Intrusion for the Gulf Islands, British Columbia (Rep. No. T). Submitted to BC Ministry of Forests, Lands and Natural Resource Operations and BC Ministry of Environment.
KNBS (2010). 2009 Kenya Population and Housing Census, Population Distribution by Political Units, Vol.1B, KNBS.
Kouzana, L., Benassi, R., Ben mammou, A., & Sfar felfoul, M. (2010). Geophysical and hydrochemical study of the seawater intrusion in Mediterranean semi arid zones. Case of the Korba coastal aquifer (cap-bon, Tunisia). Journal of African Earth Sciences, 58 (2), 242–254. doi:10.1016/j.jafrearsci.2010.03.005.
Kumar, C. P. (2006). Management of Groundwater in Salt Water Ingress Coastal Aquifers. In: Groundwater Modelling and Management (Eds. N. C. Ghosh and K. D. Sharma), Capital Publishing Company, New Delhi, pp. 540-560. http://www.angelfire.com/nh/cpkumar/publication/gwman.pdf.
Kundzewicz, Z.W., L.J. Mata, N.W. Arnell, P. Döll, P. Kabat, B. Jiménez, K.A. Miller, T. Oki, Z. Sen & I.A. Shiklomanov, (2007). Freshwater resources and their management. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, M.L. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden and C.E. Hanson, Eds., Cambridge University Press, Cambridge, UK, 173-210.
Kura NU, Ramli MF, Ibrahim S, Sulaiman WNA, Aris AZ (2014) An integrated assessment of seawater intrusion in a small island using geophysical, geochemical and geostatistical techniques. Environ Sci Pollut Res 21:7047–7064.
Langevin, C. D.; Guo, W. X (2006). MODFLOW/MT3DMS-based simulation of variable-density ground water flow and transport. Ground Water, 44, 339–351.
Langevin, C.D., Shoemaker, W.B., and Guo, W. (2003). MODFLOW-2000, the U.S. Geological Survey Modular Ground-Water Model–Documentation of the SEAWAT- 2000 Version with the Variable-Density Flow Process (VDF) and the Integrated MT3DMS Transport Process (IMT): U.S. Geological Survey Open-File Report 03-426, 43 p.
Lin, H., Ke, K., Tan, Y., Wu, S., Hsu, G., Chen, P., & Fang, S. (2013). Estimating Pumping Rates and Identifying Potential Recharge Zones for Groundwater Management in Multi-Aquifers System. Water Resour Manage Water Resources Management, 27 (9), 3293-3306.
Lobo Ferreira JP, Chachadi AG, Diamantino C, Henriques MJ (2005) Assessing aquifer vulnerability to seawater intrusion using GALDIT method: Part 1 Application to the Portuguese Aquifer of Monte Gordo. Fourth Inter-Celtic Colloquium on Hydrogeology and Management of Water Resources, Portugal, 11–14, July 2005, CD of Proceedings.
Lobo-Ferreira JP, Cabral M (1991) Proposal for an operational definition of vulnerability for the European community’s Atlas of groundwater resources, in meeting of the European Institute for Water, Groundwater Work Group Brussels.
Lusczynski, N.J. 1961. Head and flow of ground water of variable density. Journal of Geophysical Research 66, no. 12: 4247–4256.
Markusen, A. R. (1996). The economics of postwar regional disparity. In S. S. Feinstein & S. Campbell (Eds.), Readings in urban theory (pp. 102–131). Cambridge, MA: Blackwell.
Mercer, J. W., & Faust, C. R. (1980). Ground-Water Modeling: An Overview. GROUND WATER, 18 (2), 108-116.
MODFLOW: USGS. (2013). Retrieved September 5, 2016, from USGS- Science for a changing world, http://water.usgs.gov/ogw/modflow/MODFLOW.html.
Mondal, N. C., Singh, V. S., Saxena, V. K., & Singh, V. P. (2010). Assessment of seawater impact using major hydrochemical ions: A case study from Sadras, Tamilnadu, India. Environmental Monitoring and Assessment Environ Monit Assess, 177 (1-4), 315-335. doi:10.1007/s10661-010-1636-8.
MT3D (2000). Retrieved September 5, 2016, from http://hydro.geo.ua.edu/mt3d/.
Mtoni, Y., Mjemah, I. C., Bakundukize, C., Camp, M. V., Martens, K., & Walraevens, K. (2013). Saltwater intrusion and nitrate pollution in the coastal aquifer of Dar es Salaam, Tanzania. Environ Earth Sci Environmental Earth Sciences, 70 (3), 1091-1111. doi:10.1007/s12665-012-2197-7.
Moujabber, M. E., Samra, B. B., Darwish, T., & Atallah, T. (2006). Comparison of different indicators for groundwater contamination by seawater intrusion on the Lebanese coast. Water Resources Management, 20 (2), 161–180. doi:10.1007/s11269-006-7376-4.
Munga, D., Mwangi, S., Ong’anda, H., Kitheka, J. U., Mwaguni, S. M., Mdoe, F., & Opello, G. (2006). Vulnerability and pollution of groundwater in Kisauni, Mombasa, Kenya. Groundwater pollution in Africa. Taylor and Francis (Balkema), The Netherlands, 213-228.
Musa, O. K., Kudamnya, E. A., Omali, A. O., & Akuh, T. I. (2014). Physico-chemical characteristics of surface and groundwater in Obajana and its environs in Kogi state, central Nigeria. African Journal of Environmental Science and Technology, 8 (9), 521–531. doi:10.5897/ajest2014.1708.
Najib S, Grozavu A, Mehd K, Breaban IG, Guessir H, Boutayeb K (2012) Application of the method GALDIT for the cartography of groundwaters vulnerability: aquifer of Chaouia coast (Morocco). Sci Ann Alexandru Ioan Cuza 63(3):77.
Nicholls, R.J., P.P. Wong, V.R. Burkett, J.O. Codignotto, J.E. Hay, R.F. McLean, S. Ragoonaden & C.D. Woodroffe, (2007). Coastal systems and low-lying areas. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, M.L. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden and C.E. Hanson, Eds., Cambridge University Press, Cambridge, UK, 315-356.
Oladapo, M. I., Ilori, O. B., & Adeoye-Oladapo, O. O. (2014). English. African Journal of Environmental Science and Technology, 8 (1), 16–30. doi:10.5897/ajest2013.1554.
Olivera, F., Valenzuela, M., Srinivasan, R., Choi, J., Cho, H., Koka, S., & Agrawal, A. (2006). ArcGIS-SWAT: A GEODATA MODEL AND GIS INTERFACE FOR SWAT. Journal of the American Water Resources Association, 42 (2), 295–309. doi:10.1111/j.17521688.2006.tb03839.x.
Park, S. C., Yun, S. T., Chae, G. T., Yoo, I. S., Shin, K. S., Heo, C. H., et al. (2005). Regional hydrochemical study on salinization of coastal aquifers, western coastal area of South Korea. Journal of Hydrology, 313, 182–194.
Piper, A.M. (1953). A Graphic Procedure in the Geochemical Interpretation of Water Analysis. Washington D.C USGS. OCLC 37707555. ASIN B0007HRZ36.
Pliakas F, Petalas C, Diamantis I, Kallioras A (2005) Modeling of groundwater artificial recharge by reactivating an old stream bed. Water Res Manag 19:279–294.
Recinos, N., Kallioras, A., Pliakas, F., & Schuth, C. (2014). Application of GALDIT index to assess the intrinsic vulnerability to seawater intrusion of coastal granular aquifers. Environ Earth Sci Environmental Earth Sciences, 73 (3), 1017-1032. doi:10.1007/s12665-014-3452-x.
Rientjes T. (2006). Modelling in hydrology, Department of water resources, Institute for Geo-information Science and Earth Observation, The Netherlands.
Rabinove, C. L., Longford, R. H., & Brookhart, J. W. (1958). Saline water resources of North Dakota (p. 364). US Geographical Survey Water Supply, Paper 1418.
Roger J.M. De Weist (1998). Ghyben-Herzberg theory. In Reginald W.H, Rhodes W.F (Eds). Encyclopedia of Hydrology and Water Resources. Encyclopedia of Earth Sciences, Springer Netherlands. Springer Science+business Media B.V pp 340 – 341. Doi 10.1007/1-4020-4497-6_104.
Rolf A. de (2001). Principles of geographic information systems: An introductory textbook (2nd ed.). Enschede: ITC, International Institute for Aerospace Survey and Earth Sciences. Retrieved from https://www.itc.nl/library/papers_2009/general/principlesgis.pdf.
Saidi, S., Bouri, S., & Dhia, H. B. (2013). Erratum to: Groundwater management based on GIS techniques, chemical indicators and vulnerability to seawater intrusion modelling: Application to the Mahdia–Ksour Essaf aquifer, Tunisia. Environmental Earth Sciences, 70 (4), 1569–1570. doi:10.1007/s12665-013-2444-6.
Santha Sophiya, M. and Syed, T.H. (2013) ‘Assessment of vulnerability to seawater intrusion and potential remediation measures for coastal aquifers: A case study from eastern India’, Environmental Earth Sciences, 70(3), pp. 1197–1209. doi: 10.1007/s12665-012-2206-x.
Sappa, G., Ergul, S., Ferranti, F., Sweya, L. N., & Luciani, G. (2015). Effects of seasonal change and seawater intrusion on water quality for drinking and irrigation purposes, in coastal aquifers of Dar es Salaam, Tanzania. Journal of African Earth Sciences, 105, 64-84. doi:10.1016/j.jafrearsci.2015.02.007.
Saxena, V. K., Singh, V. S., Mondal, N. C., & Jain, S. C. (2003). Use of chemical parameters to delineation fresh ground water resources in Potharlanka Island, India. Environmental Geology, 44 (5), 516–521.
Shiklomanov I.A (1993). World fresh water resources. In P.H Gleick (Editor), Water in Crisis: A guide to the World’s Fresh Water Resources (pp. 13 – 23), New York 10016, Oxford University Press, Inc.
Small, C., and Nicholls R.J (2003). A global analysis of Human Settlement in Coastal Zones. J. Coastal Res, 19, 584-599.
Sophiya MS, Syed TH (2013) Assessment of vulnerability to seawater intrusion and potential remediation measures for coastal aquifers: a case study from eastern India. Environ Earth Sci 70(3):1197–1209. doi:10.1007/s12665-012-2206-x.
Steyl, G., & Dennis, I. (2009). Review of coastal-area aquifers in Africa. Hydrogeology Journal, 18 (1), 217–225. doi:10.1007/s10040-009-0545-9.
Todd D K, Mays LW (2005) Groundwater hydrology, 3rd edn. Wiley, New York, 636 p.
Trabelsi, N., Triki, I., Hentati, I. & Zairi, M. (2016) ‘Aquifer vulnerability and seawater intrusion risk using GALDIT, GQISWI and GIS: Case of a coastal aquifer in Tunisia’, Environmental Earth Sciences, 75(8). doi: 10.1007/s12665-016-5459-y.
UN Habitat (2008). United Nations Human Settlements Programme The state of African cities. Nairobi: UN Habitat.
UNEP (2008), Vital Water Graphics - An Overview of the State of the World’s Fresh and Marine Waters. 2nd Edition. UNEP, Nairobi, Kenya. ISBN: 92-807-2236-0.
Van Stempvoort D, Evert L, Wassenaar L (1993) Aquifer vulnerability index: a GIS compatible method for groundwater vulnerability mapping. Can Wat Res J 18: 25–37.
Vengosh, A., Gill, J., Reyes, A., and Thoresberg, K., (1997). A multi-isotope investigation of the origin of groundwater salinity in Salinas Valley, California. American Geophysical Union, San Francisco, California.
Verruijt, A. (1968). A note on GHyben-Herzberg Formula. Bulletin of the International Association of Scientific Hydrology, XIII, 4-12.
World Bank, (1995). A Framework for Integrated Coastal Zone Management in Sub-Saharan Africa: Building Blocks for Environmentally Sustainable Development in Africa. Paper No. 4, Africa. Technical Department. The World Bank, Washington, DC, USA.
WHO (2011). Guideline of drinking quality. Fourth Edition WHO Press, World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland http://www.who.int. Assessed 24 April 2016
WUP- World Urbanization Prospects (2014). The 2014 Revision, Highlights. United Nations, Department of Economic and Social Affairs, Population Division.
Yang, Q., Zhang, J., Wang, Y., Fang, Y. & Martin, J. (2015) ‘Multivariate statistical analysis of Hydrochemical data for shallow ground water quality factor identification in a coastal aquifer’, Polish Journal of Environmental Studies, 24. doi: 10.15244/pjoes/30263.
Zheng, Chunmiao, (1990). MT3D , A modular three-dimensional transport model for simulation of advection, dispersion and chemical reactions of contaminants in groundwater systems, Report to the U.S. Environmental Protection Agency, 170 p.
Zheng, Chunmiao, and P. Patrick Wang, 1999, MT3DMS, A modular three-dimensional multi-species transport model for simulation of advection, dispersion and chemical reactions of contaminants in groundwater systems; documentation and user¡¯s guide, U.S. Army Engineer Research and Development Center Contract Report SERDP-99-1, Vicksburg, MS, 202 p.
Zhou, X., Chen, M., Ju, X., Ning, X., & Wang, J. (2000). Numerical simulation of sea water intrusion near Beihai, china. Environmental Geology, 40 (1-2), 223–233. doi:10.1007/s002540000113.
APPENDICES
APPENDIX I
illustration not visible in this excerpt
Geographic coordinate system: GCS_WGS_1984
Datum: D_WGS_1984
Projected Coordinate System: WGS_1984_UTM_Zone_37S
Projection: Transverse_Mercator
APPENDIX II
Graphs showing the variations in the groundwater parameters in across the season
illustration not visible in this excerpt
APPENDIX III
Cross plots showing the relationships between EC/TDS and the cations & anions (pre-monsoon)
illustration not visible in this excerpt
APPENDIX IV
Table of conversion from mg/l to meq/l
illustration not visible in this excerpt
APPENDIX V
NaCl variation in the groundwater across the study area for pre-monsoon, rainy season and post-monsoon
illustration not visible in this excerpt
NaCl variation in the groundwater across the study area in; (a.) March (b.) June (c.) September
APPENDIX VI
Correlation coefficients for salinity induces for the rainy season and post-monsoon
Table 4.8: Correlation for Salinity Indices (June)
illustration not visible in this excerpt
Table 4.8: Correlation for Salinity Indices (September)
illustration not visible in this excerpt
APPENDIX VII
The GALDIT factors
illustration not visible in this excerpt
Parameter G (Groundwater occurrence) Parameter A (Aquifer hydraulic conductivity)
illustration not visible in this excerpt
Parameter L (Level of GW above SL in March) Parameter L (Level of GW above sea level for June)
illustration not visible in this excerpt
Parameter D (Distance to the shore) Parameter I- Existing status of SWI (March
illustration not visible in this excerpt
Parameter I- Existing status of SWI (June) Parameter T: (Thickness of the mapped aquifer)
APPENDIX VIII
LIST OF PUBLICATIONS AND CONFERENCE PROCEEDINGS
1. Idowu, T. E., Nyadawa, M., & K’Orowe, M. O. (2016). Seawater Intrusion Vulnerability Assessment of a Coastal Aquifer: North Coast of Mombasa, Kenya as a Case Study. Int. Journal of Engineering Research and Application, 6 (8, (Part -3)), 37–45. ISSN: 2248-9622.
2. Idowu, T. E., Nyadawa, M., & K’Orowe, M. O. (2017). Hydrogeochemical assessment of a coastal aquifer using statistical and geospatial techniques: Case study of Mombasa North Coast, Kenya. Environmental Earth Sciences (Springer). DOI: 10.1007/s12665-017-6738-y
Frequently Asked Questions About the Coastal Aquifer Language Preview
What is the purpose of this language preview?
This comprehensive language preview is intended solely for academic use, analyzing themes in a structured and professional manner. It includes the title, table of contents, objectives and key themes, chapter summaries, and key words of a document.
What does the Table of Contents cover?
The Table of Contents outlines the structure of the document, including sections on dedication, acknowledgements, lists of tables, figures, and appendices, abbreviations and acronyms, an abstract, introduction, literature review, methodology, results and discussion on hydrogeological characteristics, salinity assessment, seawater intrusion assessment, spatial vulnerability mapping, groundwater flow and solute transport simulation, conclusions and recommendations, references, and appendices.
What key themes are addressed in the Introduction?
The Introduction addresses the background of the study, statement of the problem, justification for the research, objectives of the study (general and specific), research questions, scope of the study, and limitations of the study.
What topics are covered in the Literature Review?
The Literature Review covers coastal aquifers (challenges, seawater intrusion, research review), application of GIS and groundwater modeling (GIS, groundwater modeling, integration), GALDIT overlay index, and MODFLOW, MT3D, SEAWAT, and GMS (brief descriptions).
What is the focus of the Methodology section?
The Methodology section details the introduction to the study area, materials and methods used, including secondary data collection, field data collection, and laboratory data analysis.
What do the results and discussion chapters cover?
These chapters discuss the hydrogeological characteristics of the aquifer, groundwater salinity assessment, and seawater intrusion assessment, including results and discussion for each.
What does the chapter on Spatial Vulnerability Mapping cover?
This chapter covers GALDIT factors for the coastal aquifer, including groundwater occurrence, aquifer hydraulic conductivity, height of groundwater above sea level, distance from saltwater body, impact of existing seawater intrusion, thickness of the aquifer, a generalized workflow for vulnerability analysis (data collection, processing, overlay, cartography), and results and discussion.
What is the focus of the Groundwater Flow and Solute Transport Simulation chapter?
This chapter includes importing GIS files into GMS, building the conceptual model, steady-state and transient state flow simulation, model calibration, solute transport simulation (MT3DMS and SEAWAT), and results and discussion.
What information is contained in the Conclusion and Recommendations?
This section summarizes the key conclusions of the study and provides recommendations based on the findings.
What kind of appendices are included?
The appendices contain additional data, such as groundwater heads, graphs showing variations in parameters, cross plots showing relationships between EC/TDS and ions, conversion tables, NaCl variations, correlation coefficients for salinity indices, GALDIT factors, and a list of publications and conference proceedings.
What are some of the key words or concepts mentioned?
Key words and concepts include coastal aquifers, seawater intrusion, groundwater modelling, GIS, GALDIT index, MODFLOW, MT3D, SEAWAT, hydrogeological characteristics, salinity, vulnerability mapping, solute transport, and groundwater flow.
- Quote paper
- Temitope Idowu (Author), 2017, Groundwater Flow and Quality of Coastal Aquifers, Munich, GRIN Verlag, https://www.grin.com/document/373982
