BookPDF Available

Hydrobiology of the Songor and Keta lagoons: implications for wetland management in Ghana

CM Finlayson, C Gordon,
Y Ntiamoa-Baidu, J Tumbulto
& M Storrs
The hydrobiology
of Keta and Songor
Implications for coastal
wetland management
in Ghana
supervising scientist
This project was a collaborative effort between the Ghana Wildlife Department, the Ghana
Wildlife Society and the Environmental Research Institute of the Supervising Scientist, Australia.
It formed part of the Ghana Coastal Wetlands Management Project (CWMP) which was
executed by the Ghana Wildlife Department.
The CWMP is a component of the Ghana Environmental Resource Management Project, it was
funded by the Global Environmental Facility (GEF) with the World Bank as the Implementing
Agency and coordinated in Ghana by the Environmental Protection Agency.
C Max Finlayson  Environmental Research Institute of the Supervising Scientist,
Locked Bag 2, Jabiru NT 0886, Australia.
Christopher Gordon  Volta Basin Research Project, University of Ghana, PO Box 209, Legon,
Accra, Ghana; Zoology Department, University of Ghana, PO Box 67, Legon, Accra, Ghana.
Yaa Ntiamoa-Baidu  Zoology Department, University of Ghana, PO Box 67, Legon, Accra,
Ghana; Ghana Wildlife Society, PO Box 13252, Accra, Ghana.
Jacob Tumbulto  Water Research Institute, Council for Scientific and Industrial Research, PO
Box M326, Accra, Ghana.
Michael Storrs  Environmental Research Institute of the Supervising Scientist,
Locked Bag 2, Jabiru NT 0886, Australia; present address: Northern Land Council,
PO Box 42921, Casuarina, NT 0811, Australia.
This report should be cited as follows:
Finlayson CM, Gordon C, Ntiamoa-Baidu Y, Tumbulto J & Storrs M 2000. The hydrobiology of
Keta and Songor lagoons: Implications for coastal wetland management in Ghana.
Supervising Scientist Report 152, Supervising Scientist, Darwin.
The Supervising Scientist is part of Environment Australia, the environmental program
of the Commonwealth Department of Environment and Heritage.
© Commonwealth of Australia 2000
Supervising Scientist
Environment Australia
GPO Box 461, Darwin NT 0801 Australia
ISSN 1325-1554
ISBN 0 642 24355 7
This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part
may be reproduced by any process without prior written permission from the Supervising
Scientist. Requests and inquiries concerning reproduction and rights should be addressed to
the Research Project Officer, Supervising Scientist, GPO Box 461, Darwin NT 0801.
Views expressed by authors do not necessarily reflect the views and policies of the Supervising
Scientist, the Commonwealth Government, or any collaborating organisation.
Printed in Darwin by NTUniprint.
Executive summary viii
Acknowledgments x
1 Introduction 1
1.1 Coastal wetlands management project 3
1.1.1 Background 3
1.1.2 Scope 3
1.1.3 Specific objectives and terms of reference 5
1.2 Information required for wetland management 6
1.2.1 Wetland classification 6
1.2.2 Wetland inventory 7
1.2.3 Ecological characterisation 9
1.2.4 Wetland values and benefits 9
1.2.5 Management planning 10
1.2.6 Monitoring 11
2 Coastal lagoons in Ghana 12
2.1 Description of lagoons 12
2.2 Management approaches 13
3 Ecological surveys of Keta and Songor lagoons 14
3.1 Sampling strategy 14
3.2 Sampling methods 24
3.2.1 Hydrological and meteorological 24
3.2.2 Physico-chemical 25
3.2.3 Biological 26
4 Ecological character of Keta and Songor lagoons 29
4.1 Meteorology and hydrology 29
4.1.1 Climatic conditions 29
4.1.2 Hydrological conditions 35
4.2 Bathymetry and sedimentology 40
4.2.1 Basin morphometry 40
4.2.2 Sediments 40
4.3 Limnology 45
4.3.1 Physical water parameters 45
4.3.2 Chemical water parameters 47
4.3.3 Channel chemistry 49
4.3.4 Temporal changes in water chemistry 51
4.4 Aquatic ecology 51
4.4.1 Phytoplankton 51
4.4.2 Zooplankton 53
4.4.3 Benthos 55
4.5 Wetland vegetation 84
4.5.1 Diversity 84
4.5.2 Transects 88
4.5.3 Biomass and phenology 88
4.6 Water use and pollution assessment 94
4.6.1 Domestic water sources 94
4.6.2 Water pollution sources 96
5 Management of Keta and Songor lagoons 97
5.1 Major management issues and threats 97
5.1.1 Major issues 98
5.1.2 Major threats 98
6 Monitoring of Keta and Songor lagoons 100
7 Recommendations 101
References 102
Appendices 111
1 Field sampling protocols 112
2 Water chemistry of Keta and Songor lagoons 115
3 Chlorophyll concentrations in water collected from Keta and
Songor lagoons 120
4 Diversity indices for macroinvertebrate fauna in Keta and Songor
lagoons 122
5 Macro-zoobenthos collected from Keta and Songor lagoons 125
6 Dominant plant species recorded at the wetland sites surrounding
Keta and Songor lagoons and in the Angor channel connecting
Keta to the Volta River 136
1 The Ghanaian coastline from Togo to Cote d’Ivoire in the Gulf of
Guinea 2
2 Schematic map of the general project area 4
3 Sampling grid for the Keta Ramsar site 15
4 Sampling grid for the Angor channel that connects Keta lagoon to
the Volta River and the Songor Ramsar site 16
5 Distribution of rainfall and evaporation at Keta Ramsar site 33
6 Distribution of rainfall and evaporation at Songor Ramsar site 34
7 Monthly discharge on Todzie at Todzienu 36
8 Hypsographic curves for the Keta lagoon 41
9 Map of sediments in the Keta lagoon 42
10 Cluster analysis of sediments in the Keta lagoon 43
11 Cluster analysis of sediments in the Songor lagoon 44
12 Water temperature in Keta lagoon 46
13 Major ions in Keta lagoon 48
14 Water chemistry in the Angor channel 50
15 Temporal changes in water chemistry in the Keta lagoon 52
16 Distribution of Boccardielia in Keta lagoon 56
17 Distribution of Boccardielia in the Angor channel connecting Keta
lagoon to the Volta River, and in Songor lagoon 57
18 Distribution of Capitellids in Keta lagoon 58
19 Distribution of Capitellids in Songor lagoon 59
20 Distribution of Nereis in Keta lagoon 60
21 Distribution of Nereis in the Angor channel connecting Keta
lagoon to the Volta River, and in Songor lagoon 61
22 Distribution of Gylcera in Keta lagoon 62
23 Distribution of Notomastus in Keta lagoon 63
24 Distribution of Notomastus in the Angor channel connecting Keta
lagoon to the Volta River, and in Songor lagoon 64
25 Distribution of Oligochaetes in Keta lagoon 65
26 Distribution of Brachidontes in Keta lagoon 66
27 Distribution of Corbula in Keta lagoon 67
28 Distribution of Corbula in Songor lagoon 68
29 Distribution of Hydrobia in Keta lagoon 69
30 Distribution of Melanoides in Keta lagoon 70
31 Distribution of Melanoides in Songor lagoon 71
32 Distribution of Nerita in Keta lagoon 72
33 Distribution of Tivela in Keta lagoon 73
34 Distribution of Tivela in the Angor channel connecting Keta
lagoon to the Volta River, and in Songor lagoon 74
35 Distribution of Tympanotonos in Keta lagoon 75
36 Distribution of Tympanotonos in the Angor channel connecting
Keta lagoon to the Volta River, and in Songor lagoon 76
37 Distribution of Eunice in Keta lagoon 77
38 Distribution of Eunice in the Angor channel connecting Keta
lagoon to the Volta River, and in Songor lagoon 78
39 Distribution of Urothoe in Keta lagoon 79
40 Distribution of Urothoe in the Angor channel connecting Keta
lagoon to the Volta River, and in Songor lagoon 80
41 Distribution of Dipsio in Keta lagoon 81
42 Distribution of Excirolana in Keta lagoon 82
43 Dominant plant species recorded at the wetland sampling sites at
the Songor and Keta lagoons 85
44 Vegetation transects from Keta lagoon (transects 1 & 2) 89
45 Vegetation transects from Keta lagoon (transects 3 & 4) 90
46 Vegetation transects from Keta lagoon (transects 57) 91
47 Vegetation transects from Keta lagoon (transects 8 & 9) 92
1 Information categories used by Hughes and Hughes and Scott in
wetland directories 8
2 Sites used for wetland vegetation sampling at Keta lagoon 17
3 Sites used for wetland vegetation sampling at Songor lagoon 19
4 Location of transects used for wetland vegetation point and block
sampling 20
5 Keta aquatic sampling 20
6 Songor aquatic sampling 23
7 Angor channel aquatic sampling 24
8 Meters used for taking field measurements of physico-chemical
parameters 25
9 Day length and hours of sunshine within the study area 29
10 Percent relative humidity at Ada and Keta 30
11 Rainfall in the general study area and long-term rainfall in the
Keta Ramsar site 31
12 Pan evaporation in the study area 32
13 Summary of morphometric parameters for Keta lagoon 40
14 Relationship between water temperature in the lagoons and time
of day 45
15 Relationship between water temperature and depth in the
lagoons 45
16 The relationship between dissolved oxygen content in Keta
lagoon and time of day 47
17 The relationship between dissolved oxygen and depth of water in
Keta lagoon 48
18 Concentrations of major ions in water collected from Keta lagoon 49
19 Concentrations of metals in water collected from Keta lagoon 49
20 Diatoms species in 25 samples collected from Keta and Songor
lagoons 53
21 Zooplankton in the Keta and Songor lagoons 54
22 Presence of common macroinvertebrates at the Songor and Keta
sites 55
23 Summary of macro-zoobenthos data for Keta and Songor lagoons 83
24 Biomass ash free dry weight of macro-zoobenthos in Keta and
Songor lagoons 83
25 List of macrophyte species collected and identified from Keta and
Songor lagoons and surrounding wetlands 86
26 Biomass of dominant macrophyte species in Keta and Songor
lagoons 93
27 Height of the dominant macrophyte species in Keta and Songor
lagoons 93
28 Major threats to Keta and Songor lagoons 99
29 Perceived priority threats to Keta and Songor lagoons 99
30 Parameters for monitoring Keta and Songor lagoons 100
31 Potential research projects recommended for the Keta and
Songor lagoons 101
Executive summary
Keta and Songor lagoons are located alongside the current delta of the Volta River in eastern
Ghana. The lagoons and surrounding wetlands are heavily utilised by a large population of
people who fish using a variety of techniques, cut reeds for thatch and weaving, harvest salt
by intensive and extensive means, and irrigate vegetables using water drawn from shallow
wells in the surrounding sandy soil. The lagoons are also important habitat for many aquatic
and wetland animals and species and have been recognised as internationally important under
the Ramsar Wetlands Convention. However, increasing exploitation of the lagoons and their
resources has resulted in degradation and raised concerns about the long-term sustainability of
these systems. These concerns are being addressed through the Ghana Coastal Wetlands
Management Programme (CWMP), which was implemented by the Ghana Wildlife
Department as part of the Ghana Environmental Resource Management Project and funded by
the Global Environment Facility. The general aim of the CWMP is to manage five coastal
wetland sites to maintain their ecological integrity and enhance the benefits derived from the
wetlands by local communities.
In line with the general aim of this program a broadscale baseline description of the
ecological character of Keta and Songor lagoons was undertaken. As ornithological and fish
surveys were undertaken in separate exercises they were not included in this study. Our
surveys covered:
bathymetry and hydrology (surface and groundwater and water uses);
sedimentology (particle size, organic content);
water quality (field analyses for pH, dissolved oxygen and conductivity; laboratory
analyses for major ions, nutrients and metals);
aquatic fauna (diversity and abundance of benthos and zooplankton);
aquatic/wetland phytoplankton (species diversity and chlorophyll);
macrophyte diversity, biomass and phenology.
Most of the field surveys were undertaken in November 1996 with subsequent supplementary
surveys and analyses of the samples and data taken over ensuing months. The surveys were
based on a systematic sampling grid placed across each of the lagoons and adjoining wetlands
as a template for determining spatial patterns. As Keta is much larger than Songor, the
sampling effort was also much larger and included large areas of surrounding wetland as well
as the open water. Temporal patterns were not determined; however, this important aspect
was addressed in recommendations for further monitoring and management.
The data and information collected through these surveys were used with information
gathered from other sources to provide comment on the major threats to the lagoons under the
general headings of ‘water regime’, ‘water pollution’, ‘physical modification’ and
‘exploitation and production’. Recommendations for monitoring and research were also made.
In assessing further monitoring needs, we recognised that as the lagoons are very large it
would be impossible to carry out the same sort of sampling intensity that was used in this
baseline study. Thus a stratified random approach was recommended as the basis of a simple
bimonthly monitoring strategy for selected hydrology, water quality and biological
parameters at 6 sites in Songor and 17 sites in Keta.
A list of further research projects was compiled with an emphasis being placed on
environmental issues and management of the lagoons, including:
re-colonisation by invertebrate fauna;
environmental tolerance of invertebrate fauna;
ecology of Penaeids;
zooplankton dynamics within the main channels of the wetlands;
development of invertebrate fauna within acadjas;
determination of the factors controlling the spread of various mollusc species in the
decomposition of aquatic plants;
resource partitioning between crabs in the lagoons;
harvesting and usage of aquatic macrophytes by local communities;
groundwater salinity and mangroves;
hydrogen sulphide in sediments and its effect on the vegetation in the lagoons.
In order to assist with further monitoring and analyses of results we also included in this
report detailed descriptions of the field sampling methods and data collected during the
baseline surveys.
This study was logistically very demanding, and a large number of individuals were involved
before, during and after the fieldwork. Thanks must first go to Mr Bill MacFarlane (eriss),
Mr HO Ankrah (VBRP/Zoology, Legon) and Mr F Seku (Botany Department, Legon) for
their technical assistance in the field. They were ably supported by Mr Erasumus H Owusu,
Mr Steven Asamoa, Mr Ali Nuoh and Sammy Adu of the Ghana Wildlife Society. We will
not forget the input of the attached staff of the Ghana Wildlife Department, Mr CC
Amankwah and Mr Felix Nani. Mr David Kpelle assisted in post fieldwork administration.
Thanks also go to Dr Mike Kendall of the Plymouth Marine Laboratory, UK, for polychaete
We thank the Ghana Wildlife Society, especially Mrs Afia S Owusu, for its logistic support
and the Ghana National Petroleum Company for granting permission to the group to use the
Camp at Anloga.
We must also thank the support staff who assisted in the project – drivers Emanuel Abodza,
Boateng and Moses Klu as well as Christopher (WD Keta site), and secretarial and data entry,
Janet Owusu-Domphe (VBRP).
Last, but not least, we must thank the Ghana Wildlife Department who through the Ghana
Coastal Wetlands Management Project invited us to carry out this study.
1 Introduction
The Ghanaian coastline stretches some 550 km from Togo to Cote d’Ivoire in the Gulf of
Guinea (figure 1). There are about 100 coastal lagoons in Ghana which make them an
important feature of this sandy coastline, especially as the lagoons provide many benefits and
values to human populations (Gordon 1987, 1996). Locally, the lagoons provide valuable
resources for trade and consumption. Large quantities of fish and crabs are caught and traded,
either smoked or dried. Reeds and other plants are cut for thatch and for weaving mats for sale
at markets near and far. Vegetables are grown in sandy garden beds irrigated by water drawn
by hand from wells along the edges of the lagoons. Salt is extracted by both intensive and
extensive methods. The socio-economic benefits of the lagoons to the local people are
apparent to even the most casual of observer.
In recent years, the immense conservation value of these same lagoons has been recognised
both nationally and internationally. As with wetlands elsewhere in the world the value of the
lagoons as migratory waterbird habitat has received recognition (Ntiamoa-Baidu & Grieve
1987, Ntiamoa-Baidu 1991, 1993, Ntiamoa-Baidu & Gordon 1991, Piersma & Ntiamoa-
Baidu 1995, Ntiamoa-Baidu et al 1998). The broader values of these habitats are gradually
receiving greater attention as conservation is being increasingly considered as an integral
component of sustainable use of the wetland resources, rather than as an issue in isolation.
However, it is evident that the values and benefits provided by the lagoons are under
increasing threat from over-exploitation and degradation (Ntiamoa-Baidu & Gordon 1991).
The very resources that provide the values and benefits are under pressure from the
expanding human population. For the socio-economic values of the lagoons (ie the products
and functions) to be maintained it is necessary to ensure that the basic ecological character
of the lagoons is maintained. The products and functions of any wetland can not be treated
separately from the ecological processes from which they are derived (Finlayson 1996a).
Thus, to maintain the resources of the lagoons the pattern of usage must ensure that the
ecological processes that support the products and functions valued by the human
population are not degraded and, in the worst case, lost forever. Whilst traditional patterns
and levels of use can be sustainable in Ghana (Gordon 1990), expanding population
pressures can all too quickly degrade the basic resource(s) being used (World Bank 1996).
The situation is made more acute in Ghana, where the coastal zone represents less than 7%
of the total land area, however, it holds over 25% of the nation’s population. The continued
trends of the drift from rural to urban centres, the industrialisation of coastal districts as
well as the high population growth rate of 3%, place increasing stress on the coastal
In view of such increasing pressures on wetlands the Ramsar Wetlands Convention has
proposed management and monitoring processes for Wetlands of International Importance.
Integral to these are guidelines for the wise use of wetlands (Davis 1993) and the maintenance
of their ecological character (Finlayson 1996b). From a management perspective, the
ecological character must first be described (to a minimum necessary level) and key features
identified and then monitored to ensure that they are not degraded or lost. Thus, description of
the ecological character and the development of a suitable monitoring framework are two
steps that are increasingly being seen as essential components of making wise use of wetlands
(Finlayson 1996a,b).
Figure 1 The Ghanaian coastline from Togo to Cote d’Ivoire in the Gulf of Guinea
In this report we provide a description of the ecological character of the Keta and Songor
lagoons in the Lower Volta region of Ghana (figure 2) and present a framework for monitoring
changes in their ecological character. It is anticipated that the long-term monitoring programs
advocated will provide a more detailed temporal description of the ecological character of the
lagoons. Keta and Songor lagoons are both internationally important wetlands designated as
Ramsar sites on the basis of their total waterbird populations and the occurrence of
internationally important numbers of several species (Ntiamoa-Baidu & Gordon 1991, Piersma
& Ntiamoa-Baidu 1995, Ntiamoa-Baidu et al 1998). The lagoons and surrounding floodplains
support large numbers of people through fishing, salt extraction, reed cutting and water supply.
The information on the ecological character and monitoring of these lagoons is presented in line
with the current concepts of the Ramsar Wetlands Convention for managing and monitoring
wetlands. As such, this information also provides a basis to test the adequacy of the guidelines
drawn up to interpret these international concepts.
1.1 Coastal wetlands management project
The Ghana Coastal Wetlands Management Project (CWMP) is implemented by the Ghana
Wildlife Department as part of the Ghana Environmental Resource Management Project,
funded by the Global Environmental Facility (GEF). The general aim of the CWMP was to
manage five coastal wetland sites to maintain their ecological integrity and enhance the
benefits derived from the wetlands by local communities.
1.1.1 Background
The genesis of the CWMP can be traced back to 1985, when the Government of Ghana
entered into an agreement with BirdLife International (formerly the International Council for
Bird Preservation) and the Royal Society for the Protection of Birds, to protect seashore birds,
specifically the Roseate Tern (Sterna dougalii). To implement the agreement the Save the
Seashore Bird Project–Ghana (SSBP-G) was set up to monitor bird populations along the
coast of Ghana. The SSBP-G established the importance of several coastal wetlands for
migratory shorebirds (Ntiamoa-Baidu 1988).
The information provided by the SSBP-G formed the basis for Ghana to become a signatory
to the Ramsar Convention on Wetlands and the Bonn Convention on Migratory Species in
1988. During this same period Ghana was preparing an Environmental Action Plan, under the
auspices of the then Environmental Protection Council. As part of the preparatory background
documents a Coastal Zone Indicative Management Plan (CZIMP) was prepared (Agyepong et
al 1990) which highlighted the need to protect some of the more important coastal sites.
Subsequent to this local consultants were commissioned to prepare a base document
(Ntiamoa-Baidu & Gordon 1991) for submission to the World Bank for GEF funding. The
project was approved in 1992.
1.1.2 Scope
To achieve the overall goal the project set out to:
develop a technical information base on the interactions between the biotic and abiotic
elements of the wetlands
describe the ecological character of five lagoons
develop a monitoring framework as part of an overall management strategy for the long-
term sustainable use of these lagoons.
Figure 2 Schematic map of the general project area
Activities planned for the 5-year life span of the project included:
1 Management of the sites including the maintenance of boundaries and trail systems,
monitoring of wildlife populations, habitat management, erosion control, tree planting,
and training of community rangers/wardens in the principles of wildlife management and
2 Baseline studies for the current aquatic ecosystems and catchment areas and regular
monitoring of key hydrological, limnological and biological indicators.
3 Socio-economic and technical studies of compatible development options on intensified
fisheries management, development of aquaculture and salt production with the
establishment of an investment fund to finance pilot schemes and infrastructure that will
lead to the realisation of the identified options.
4 Environmental education and public awareness programs including the construction and
staffing of community education facilities at each site.
5 Preparation of a National Wetlands Strategy to provide a policy framework for general
wetland conservation in Ghana to address the conservation issues in wetland sites other
than the five coastal Ramsar sites.
1.1.3 Specific objectives and terms of reference
Based on the broad goal given above the aquatic/wetland ecology components of the CWMP
had the following specific objectives for the Keta and Songor lagoons in the lower Volta:
collate available biophysical information and collect data to provide a basis for describing
the ecological character of the lagoons, especially the hydrology, physico-chemistry, and
the aquatic/wetland invertebrate fauna and the flora;
identify the major values and benefits derived from the lagoons, especially those related
to domestic water supply and the harvest of useful plants and animals;
identify the major threats to the sustainable use of products harvested from the lagoons,
especially those due to water pollution and hydrological regulation;
establish reference collections of key aquatic/wetland species collected from the lagoons;
provide training and develop the practical expertise of Ghana Wildlife Department staff;
provide recommendations for further management of the lagoons, especially regulation of
the water regime, to enhance the development of sustainable levels of resource exploitation;
develop monitoring protocols to enable further description of the ecological character of
the lagoons and to provide early warning of adverse ecological change;
provide advice to the Convention on Wetlands of International Importance on the
adequacy of guidelines for establishing and monitoring the ecological character of
wetlands and for promoting wise use and management planning.
This study was preceded by an ornithological investigation (Piersma & Ntiamoa-Baidu 1995)
and followed by specific investigations of the terrestrial fauna, as well as the fish and fisheries
potential of the lagoons (Ryan & Ntiamoa-Baidu 1998). Further, the overall social context of
the management and use of resources from the lagoons will be provided by a socio-economic
study and the development of the national wetland policy, which will include an integrated
monitoring program. Thus, the aquatic/wetland investigations described in this report are part
of a larger holistic concept for the management of the major coastal lagoons of Ghana.
1.2 Information required for wetland management
Over the past few decades considerable effort has been directed worldwide towards the
management of wetlands. Under the Ramsar Wetlands Convention this involved the
promulgation of guidelines and the development and implementation of appropriate national
policies for the wise use of wetlands.
The concept of wise use of wetlands was formulated in 1971 with an article in the Ramsar
Convention that stated:
The Contracting Parties shall formulate and implement their planning so as to promote … as far as
possible the wise use of wetlands in their territory.
A definition of wise use, based on the concept of sustainable utilisation, was adopted in 1987
(Davis 1993, 1994). Thus, the wise use of wetlands is their sustainable utilisation for the
benefit of humankind in a way compatible with the maintenance of the natural properties of
the ecosystem. In turn, sustainable utilisation is the human use of a wetland so that it may
yield the greatest continuous benefit to present generations while maintaining its potential to
meet the needs and aspirations of future generations.
More recent attention to the maintenance of the ecological character of wetlands has
highlighted that the wise use and management of wetlands is dependent on a large and holistic
information base. According to Dugan (1990) and Finlayson (1996a) the information base
required for wetland management has been sub-divided into the following subject categories:
wetland classification; wetland inventory; ecological characterisation; wetland values and
benefits; management planning; and monitoring. These categories are briefly considered
within the context of the information base required for wetland management in Ghana.
1.2.1 Wetland classification
The classification of wetlands is beset with difficulties (Finlayson & van der Valk 1995a) and
these seemingly multiply when a regional or an international approach is sought (Scott &
Jones 1995). The purpose of wetland classification is to standardise and define the terms
being used to describe various wetland types. At an international level a uniform set of terms
is needed (Cowardin & Golet 1995, Scott & Jones 1995, Zoltai & Vitt 1995) but at a national
level this may not be necessary (Pressey & Adam 1995).
Scott and Jones (1995) issued a warning concerning the level of sophistication required for
classification in relation to the amount of information required for management. The
important point in classifying wetlands is not the detail of the classification, but the usefulness
of the classification for management purposes.
Many national wetland classifications now exist (see Finlayson & van der Valk 1995b). These
invariably incorporate local terms and definitions that are not necessarily known or accepted
elsewhere. Thus, even at the national level it can be extremely difficult to develop a
classification that is consistent and acceptable to all wetland scientists and experts (Cowardin
& Golet 1995, Lu 1995, Pressey & Adam 1995).
The wetland habitat classification used by the Ramsar Convention has increasingly been
adopted for national and international purposes (Scott & Jones 1995). However, this contains
many inconsistencies (Semeniuk & Semeniuk 1995, 1997), namely:
not all types of wetlands are clearly or unambiguously described;
repetition of types that are named ‘marshes;
some wetlands remain ill-defined and encompass a number of types;
mixed criteria are used to separate wetlands.
Due to the nature of the problem, and despite the deficiencies of the Ramsar classification
scheme, Ghanaian wetlands can be fitted adequately into the existing Ramsar Classification
scheme. The main advantage of this approach is that the terms used have international
recognition and save time and effort in drawing up a National Classification scheme for Ghana.
1.2.2 Wetland inventory
Much of the information required for wetland management can be collected in a directory or
inventory of wetlands. A directory and inventory are used to compile the same type of
information, but the former is limited to current information and may not be comprehensive
(Finlayson 1996a). An inventory usually includes investigative steps to obtain more
information and thereby present a comprehensive coverage of sites. Thus, a directory may be
the precursor of an inventory. In reality, the terms are used interchangeably.
The information collected through wetland inventories is regarded as a prerequisite for
wetland conservation and management (Dugan 1990, Hollis et al 1992, Taylor et al 1995,
Hughes 1995, Naranjo 1995, Scott & Jones 1995). Dugan (1990) considers an inventory as
the first step in assembling an information base for wetland management. In fact, Contracting
Parties to the Ramsar Convention undertake to compile an inventory as part of the process of
developing and implementing a national wetland policy for the wise use of all wetlands on
their territory. A strategically developed wetland inventory should provide managers and
policy makers with the information base that they require not only to manage individual
wetlands or threats, but to also place the conservation value of wetlands within the context of
broad scale land use and sustainable development priorities.
To be effective in promoting the sustainable use and conservation of wetlands an inventory
must be available to and understood by all those formulating and implementing wetland
management policies (Naranjo 1995, Pressey & Adam 1995, Wilen & Bates 1995). Thus,
they must be framed in a manner suitable for management purposes. Additionally, to remain
useful tools for management they need to be regularly reviewed and updated (Naranjo 1995,
Scott & Jones 1995, Wilen & Bates 1995). Information categories often used in wetland
inventories are shown in table 1. Many of the categories do not relate directly to biophysical
information, but are management oriented. Costa et al (1996) summarised the conclusions of
a Mediterranean analysis of wetland inventory and the key points are given below as a guide
to compiling a wetland inventory.
Objectives of a wetland inventory (Costa et al 1996)
To identify where wetlands are, and which are priority sites for conservation
To identify the functions and values of each wetland
To establish a baseline for measuring change in a wetland
To provide a tool for planning and management
Table 1 Information categories used by Hughes and Hughes (1992) and Scott (1993) in wetland
Category Information
Hughes & Hughes 1992 – A directory of African wetlands
Title/Location/Nearest town Name of wetland/Coordinates/Name of town
Area/Altitude Area/Height above sea level
General Description of wetland and environs
Hydrology and Water quality General features
Flora/Fauna Important species and populations
Human impact and utilisation Land uses and changes
Conservation status Nature protection
Scott 1993 – A directory of wetlands in Oceania
Title/Location Name and reference number/Coordinates
Area/Altitude Area and/or length of rivers/Average
Overview Summary description of site
Physical/Ecological features Hydrology, soils, climate, vegetation and habitats
Land tenure Ownership of wetland and surrounding land
Conservation measures taken/proposed Details of protected areas/Further proposals
Land use/Possible changes in land use Human activities/Development plans and ideas
Disturbances and threats Existing and possible threats
Hydrological-biophysical/Social-cultural values Principal features/Values
Noteworthy fauna/flora Important species
Scientific research/Conservation education Major research/education activities and facilities
Management authority and jurisdiction Responsible authority(ies)
References Key published literature
Reasons for inclusion in Directory Reason(s) designated as important
In order to achieve the objectives the following recommendations were made:
Means of achieving the objectives of an inventory (Costa et al 1996)
Use standardised methods for classification, data collection and storage, delineation and
Incorporate qualitative and quantitative data to provide a baseline for monitoring wetland
change and loss
Facilitate analysis of loss of wetland functions
Be regularly updated
Be easily disseminated and made available to wetland managers, decision-makers and the
general public.
For the above to be achieved careful planning and testing of techniques is required. A secure
funding source is needed and all changes to protocols should be well documented and
assessed. Critically, any limitations on the use of the information should be made apparent at
the outset.
1.2.3 Ecological characterisation
An important obligation under the Ramsar Wetland Convention is for each Contracting Party
to ‘designate suitable wetlands within their territory for inclusion in a List of Wetlands of
International Importance’. The Convention also states that wetlands should be listed
according to their ‘international significance in terms of ecology, botany, zoology, limnology
or hydrology’. Whilst listing a site as internationally important is an important obligation
under the Convention, it may not constitute anything more than a passive conservation step.
Thus, the Convention also contains an obligation to ‘… formulate and implement their
planning so as to promote the conservation of the wetlands included in the List’ and inform
the Ramsar Bureau ‘… if the ecological character of any wetland in their territory and
included in the List has changed, is changing, or is likely to change as the result of
technological developments, pollution or other human interference’.
A working definition of ecological character was agreed at the Ramsar Wetland Convention
in 1996 based on material supplied by Dugan and Jones (1993) and Finlayson (1996b) and
updated in 1999. This is given below:
Ecological character is the sum of the biological, physical, and chemical components of the
wetland ecosystem, and their interactions, which maintain the wetland and its products, functions,
and attributes.
This definition provides a theoretical basis for describing the ecological character of a
wetland, but does not assist with the practical issues of describing the character. Thus, there is
a level of consensus on the concept of ecological character, but questions relating to the
ecological meaning of change when it is detected have yet to be answered. Monitoring can
provide the necessary information, but it does not necessarily provide the basis for
interpreting the significance of change.
Within the context of the Ramsar Convention, change in ecological character was considered
as meaning adverse change. This concept is captured in the definition of change in ecological
character that was adopted from material provided by Dugan and Jones (1993) and Finlayson
(1996b) and updated in 1999.
Change in ecological character is the impairment or imbalance in any biological, physical, or
chemical components of the wetland ecosystem, or in their interactions, which maintain the
wetland and its products, functions and attributes.
However, even with this definition we are no closer to ascertaining what exactly constitutes an
unacceptable ecological change. To define an unacceptable ecological change we need to firstly
establish the values and benefits of the wetland, assess the ecological status of these and then
monitor them to ascertain when (if) an adverse change is likely to or has actually occurred.
Thus, there is broad agreement on the basic need to assess and describe the ecological character
of a wetland, but further attention is required to assessing the significance of any change.
1.2.4 Wetland values and benefits
All wetlands provide values and benefits to people. Values and benefits are taken to include a
range of wetland functions, products and attributes that have been previously defined within
the Ramsar context (Dugan 1990, Davis 1993, 1994) as follows:
Functions performed by wetlands include the following: water storage; storm protection
and flood mitigation; shoreline stabilisation and erosion control; groundwater recharge;
groundwater discharge; retention of nutrients, sediments and pollutants; and stabilisation
of local climatic conditions, particularly rainfall and temperature. These functions are the
result of the interactions between the biological, chemical and physical components of a
wetland, such as soils, water, plants and animals.
Products generated by wetlands include the following: wildlife resources; fisheries; forest
resources; forage resources; agricultural resources; and water supply. These products are
generated by the interactions between the biological, chemical and physical components
of a wetland.
Attributes of a wetland include the following: biological diversity; geomorphic features;
and unique cultural and heritage features. These have value either because they induce
certain uses or because they are valued themselves.
The combination of wetland functions, products and attributes gives the wetland benefits and
values that make it important to society. The relative importance of these values and benefits
varies between sites due both to the biophysical features of the wetland and the lifestyles of
the people.
1.2.5 Management planning
Wetlands are dynamic areas, open to influence from natural and human factors. In order to
maintain their biological diversity and productivity and to allow wise use of their resources by
human beings, some kind of agreement is needed between the various owners, occupiers and
interested parties. The management planning process provides this overall agreement (Davis
Further, management planning is a flexible and dynamic way of thinking and contains three
basic components: description of the site; evaluation of the main features of the wetland and
expression of management objectives; and plans or prescriptions for specific actions. It is also
recommended that the plan contain a preamble that broadly reflects the policies of
organisations concerned with the production and implementation of the management plan. A
summary of the main principles is given below.
Principles for management planning
It is a way of thinking, which involves recording, evaluating and planning and is subject
to constant review and revision and is therefore flexible and dynamic.
It involves three basic steps of describing the features of the site/area, defining
operational objectives and taking necessary management actions.
Preparation of an elaborate plan is not an excuse for inaction or delay.
Review of the plan may lead to revision of the site description and operational objectives.
It should be a technical, not a legal document, although it may be supported by
appropriate legislation.
Finlayson (1996a) notes that the Ramsar guidelines sound simple, but adds that there are
major pitfalls, such as making the plan too complicated, making the plan the goal rather than
the tool, making the plan inflexible and not allocating resources to ensure that the plan can be
Underpinning the planning exercise is the establishment of a rationale for management and
the setting of obtainable operational objectives. Monitoring is therefore essential. In other
words, implementation of a management plan should proceed hand-in-hand with a process to
ensure that the objectives of the plan are obtained or accordingly modified in response to new
information (Finlayson 1994, 1996b).
1.2.6 Monitoring
Wetland monitoring has received more and more attention in recent years. At a global level
this has arisen as awareness of the extent of wetland degradation and loss has increased. Such
is the concern at the extent of global wetland degradation that more and more effort is being
directed towards developing effective management processes and responses to problems. In
many instances this effort is being held back by a lack of relevant information on the nature of
the problem, the cause of the problem and the effectiveness of management procedures and
actions. Effective monitoring programs can overcome these deficiencies.
In a general sense monitoring addresses the issue of change or lack of change through time
and at particular places. Thus monitoring can be defined as the systematic collection of data
or information over time. It differs from surveillance by assuming that there is a specific
reason for collecting the data or information (see Spellerberg 1991, Hellawell 1991, Furness
et al 1994). Thus, whilst it is built upon survey and surveillance, it is more precise and
oriented to specific targets or goals (Hellawell 1991, Spellerberg 1991).
Definitions of survey, surveillance and monitoring
Survey is an exercise in which a set of qualitative observations are made but without any
preconception of what the findings ought to be.
Surveillance is a time series of surveys to ascertain the extent of variability and/or range
of values for particular parameters.
Monitoring is based on surveillance and is the systematic collection of data or information
over time in order to ascertain the extent of compliance with a predetermined standard or
A framework for assisting with the design of a monitoring program has been presented by
Finlayson (1996b,c). The framework applies to all forms of monitoring (eg changes in the
area of a wetland, the ecological health of a wetland, or the underlying reasons behind the loss
of wetlands) but it is not prescriptive. Rather, it presents a series of steps that will assist those
charged with designing a monitoring program make decisions suitable for their own situation.
In a general sense, monitoring is needed to prevent further unchecked exploitation and
degradation of wetlands. Thus, there is a need to assess the impact of human development and
minimise ecological change. Success in such programs will depend on our ability not only to
detect and monitor changes in the quality of wetlands, but also to provide early indications of
likely change and thereby take action to prevent this change from occurring. Thus, with all
monitoring techniques there is a need to establish a starting point or to obtain baseline data
that identifies the key functions and values of the site.
2 Coastal lagoons in Ghana
2.1 Description of lagoons
Simplistically the lagoons of Ghana can be classified into two types: the open lagoons that are
associated with large rivers and have a permanent connection to the sea and the closed lagoons
that are formed behind sandbars, with no permanent connection to the sea (Boughey 1957, Kwei
1977, Mensah 1979, Gordon 1987). In ecological terms, open lagoon systems are more stable
and faunistically diverse due to the influence of the sea. The closed lagoons are functionally
more unpredictable, with conditions changing very rapidly from one point in time to another.
These lagoons are usually saline and can be further described as follows:
Open: with one or more narrow opening(s) to the sea most of the time and therefore known
as classical open lagoon, eg Nakwa, Amisa and Nyanya lagoons. In Ghana, the
mouths of some lagoons have been made permanently open through the intervention
of humans for the purposes of road/harbour construction, eg Sakumo II and Benya
lagoons at Tema and Elmina respectively.
Closed: cut off from the sea by a sandbar during greater part of the year. The bar may be
breached naturally or by humans during the Rainy season. These are classical closed
lagoons, eg Sakumo I and Muni lagoons at Bortianor and Winneba respectively.
Some closed lagoons receive seawater overflows during spring tides. These are called
spring tide-fed closed lagoons, eg Bormis lagoon at Moree.
The functions of lagoons include sediment/toxicant retention, nutrient retention, biomass
export, water transport and recreation/tourism. Wildlife resources and fisheries are the main
products, but freshwater (isolated) lagoons provide water supply for domestic purposes. The
lagoonal attributes are biological diversity and uniqueness to culture/heritage.
In addition to the brackish water lagoons, Ghana has several coastal freshwater lagoons, these
coastal freshwater lagoons are found mainly in the Western Region where rainfall in excess of
2000 mm per annum produces conditions of high runoff and stream flow. The underlying
rocks in these areas have also undergone profound leaching giving rise to waters, which are
extremely ion poor. Typically these lagoons are open to the sea either directly or by a channel.
They are also fairly small – the largest of this type, the Amansuri lagoon, is about 2.5 km2 in
area. Other examples include the Domini and Ekpuekyi lagoons, both of which are under
1.0 km2. As with the brackish and saline lagoons, the functions of lagoons include
sediment/toxicant retention, nutrient retention, biomass export, water transport and
recreation/tourism. Wildlife resources and fisheries are the main products. These freshwater
lagoons also provide water supply for domestic purposes. The lagoonal attributes are
biological diversity and uniqueness to culture/heritage.
The actual number of coastal lagoons in Ghana is not precisely known. Published estimates
range from 50 (Mensah 1979, Gordon 1987) to over 90 (Gordon 1996, Yankson & Obodai in
press). The lack of precision is partly due to the ephemeral nature of many of the smaller
lagoons which require rainfall to create a freshwater lagoon habitat behind a sandbar; these
sandbars then break to allow sea water to penetrate. Dry conditions result in re-creation of the
sandbar, and loss of water by evaporation causes the lagoon to dry up. The smaller lagoons of
areas under 0.1 km2, and maximum water depths of under 1 m, can go through this cycle in a
matter of months.
2.2 Management approaches
The management of the coastal lagoons has traditionally been vested in the ‘owners’ of the
lagoon. These are usually local clans, fetishes or stools. Traditional knowledge or culture is the
way in which Ghanaian ethnic groups use traditional values and knowledge, structured within
specific organisational frameworks, towards solving particular issues and tasks (World Bank
1996). The organisational framework of these societies is the kinship system, or more
specifically families, lineages and clans. On the various levels of this framework, specific rights
and obligations, dealing with issues like authority, control, adjudication of conflicts, inheritance,
succession and land ownership are vested in the members. At each of the organisational levels
within the framework, there will be a chief, usually hereditary in lineage, who functions as a
custodian or caretaker. Many of the traditional management strategies were geared at
controlling resource use by placing limits on access, both spatially and temporally, through the
use of taboos and outright bans. For many years, this traditional approach has been sufficient to
maintain the ecological integrity of the lagoon environment (Gordon 1992, Ntiamoa-Baidu
1992). Unfortunately, education, religion and acculturation have resulted in the breakdown of
traditional management systems. Many of the areas, operate under ‘common property’ laws.
With rising economic pressures, these areas are being exploited unsustainably with local fines
and punishments being ignored or disregarded.
The modern system for natural resource management in Ghana follows a three tier approach
(Government of Ghana 1995 [Vision 2020]). The three tiers are the district, regional and
national levels. Planning and management is heavily predicated on:
decentralisation of political and state power in order to enhance participatory democracy
through local level political institutions with the District Assembly as the focal point;
decentralisation of administration, development planning, implementation and budget
making in which local authorities are actively involved.
One key institution is the District Environmental Management Committee, which has
representation from the decentralised departments, such as Fisheries, Forestry and Wildlife.
For coastal wetlands, in particular the five Ramsar sites, Muni-Pomadze, Densu Delta,
Sakumo, Songor and Keta all have site management committees with representation from
primary stakeholders.
The site management committees comprise the Senior Technical Adviser (as Chairman), a
Representative of the Environment Protection Council (as Secretary), the Coastal Wetlands
Conservation Programme Coordinator, the Game Warden in charge of the site and
representatives of appropriate institutions. From the conception of the CWMP (Ntiamoa-Baidu
& Gordon 1991), it was emphasised that the successful management of the coastal wetlands
would require a multi-disciplinary approach. The Department of Wildlife was therefore to seek
the expertise and involvement of relevant organisations for the execution of programs. Another
identified crucial factor for the success of the coastal wetland conservation program was the
support and involvement of the communities who live in the coastal zone. These are the people
whose lifestyles are interlinked with the coastal wetlands and whose activities directly affect the
wetland ecosystem. Protection of the wetlands should therefore be ‘for’ the people and not
‘against’ them. Every effort has been made to secure the people’s participation and
involvement; and to integrate their needs with the management processes. Apart from the
general community, groups whose involvement was to be specifically sought include the
traditional administrators (Chiefs, elders, etc), the town development committees, local political
groups such as the District Assemblies, and NGOs such as the 31st December Women’s
3 Ecological surveys of Keta and Songor
3.1 Sampling strategy
Sampling of each lagoon and the surrounding wetland vegetation was based on a stratified
grid drawn at intervals of 1' latitude and longitude (ie 1.8 x 1.8 km). The points of
intersection of the grid were used as the basis for selecting sites for sampling. These points
were coded according to the name of the lagoon (K = Keta, S = Songor) and with an
alphabetic code for the northing and numeric code for the easting (ie ‘KQ15’ was located at
Keta lagoon at the intersection of the northing or horizontal grid line labeled ‘Q’ and the
easting or vertical grid line labelled ‘15’). Samples were also collected from the Angor
channel that connects the Keta lagoon with the Volta River. These were labelled C1–C12 and
located where the sampling grid crossed the channel. The coordinates for each site were taken
from 1:50 000 Ghana topographical maps (sheets 0600D4, 0500B2, E0601C3 & E0501A1 for
Keta and 0500A2 and 0500B1 for Songor) that were based on aerial photography flown in
December 1974.
The sampling grid is shown in figures 3 and 4. A list of site codes and coordinates read from
the maps are given in tables 2 to 7. In the field the sites were located with a hand-held Global
Positioning System (GPS) recorder (Garmin GPS 38 or 45) with an accuracy of about 100 m.
The aquatic sampling sites were located within 300 m of the map coordinate whereas those in
the channel were located by GPS and recorded as such. As access to some of the vegetation
sites was far more difficult (see below), a GPS reading was taken at the actual point sampled.
The GPS readings for each site are shown in tables 2 to 7.
The sampling strategy was divided into two components – one aquatic and the other
wetland/terrestrial. All intersecting grid points within the lagoons were used for aquatic
sampling (ie physico-chemical and biological parameters). These sites were reached initially
by wading and/or hiring wooden canoes poled by local fishermen. An aluminium dinghy with
an outboard motor (15 hp) was later used in Keta lagoon and greatly reduced the time and
effort spent getting to sites in deeper water.
The wetland sampling was initially undertaken from the landward side of the lagoons and was
severely limited by access through extensive stands of reeds (up to 4 m in height) and grasses
in water reaching more than 1.5 m depth. Sampling was based on a series of grid points
located along the landward side of the lagoon shorelines. The number of sample sites at Keta
lagoon was initially limited by logistical issues (ie access and sample processing times) and a
subjective choice of sites was made around the perimeter (see table 2). When a boat and
outboard motor became available all grid points within 2 km (approx) of the lagoon were
visited from either the land or the water side. Once the initial sampling near the edge of the
lagoons was completed, sites for phenological sampling in the extensive swamps stretching
east of Songor and west of Keta towards the Volta delta, and as far north as Sogakope were
added. Sampling was also conducted in the Angor channel from Keta lagoon to the Volta
River with vegetation phenological sampling conducted along the grid given in figure 3. The
sampling coordinates recorded at these sites are presented in appendix 6.
Figure 3 Sampling grid for the Keta Ramsar site
Figure 4 Sampling grid for (top) the Angor channel that connects Keta lagoon to the Volta River and
(bottom) the Songor Ramsar site
Table 2 Sites used for wetland vegetation sampling at Keta lagoon
Site code* Coordinates Date Field coordinates Vegetation sampling
N E N E Phenology Biomass
KB17 06 03 00 57 16/11 06 03.4 00 56.9 P
KB23 06 03 01 03 16/11 06 02.9 01 03.0 P
KC16 06 02 00 56 16/11 06 02 00 55.8 P B
KC18 06 02 00 58 16/11 06 01.8 00 58.0 P B
KC20 06 02 01 00 16/11 06 01.4 01 00.0 P
KC22 06 02 01 02 16/11 06 01.7 01 02.1 P B
KC23* 06 02 01 03 28/11 06 02.0 01 03.0 P
KC24 06 02 01 04 16/11 06 02 01 04.0 P B
KD15 06 01 00 55 16/11 06 00.9 00 55.0 P
KD22* 06 01 01 03 28/11 06 01.0 01 03.0 P
KD23 06 01 01 03 16/11 06 00.1 01 02.9 P
KE14 06 00 00 54 17/11 05 59 8 00 53.9 P B
KE15* 06 00 00 55 28/11 06 00.2 00 54.9 P
KE19* 06 00 00 59 28/11 06 00.0 00 59.0 P
KE20 06 00 01 00 19/11 06 00.0 01 00.0 P B
KE21 06 00 01 01 19/11 06 00.0 01 01.0 P
KE22* 06 00 01 02 27/11 06 00.0 01 02.0 P
KF11 05 59 00 51 17/11 05 59.1 00 51.7 P
KF16* 05 59 00 56 28/11 05 59.1 00 56.0 P
KF21 05 59 01 00 19/11 05 59.0 01 00.0 P
KF22 05 59 01 02 16/11 05 59.1 01 01.7 P B
KG01* 05 58 00 41 9/12 05 58.1 00 41.0 P
KG09* 05 58 00 49 6/12 05 59.0 00 49.0 P
KG12 05 58 00 52 21/11 05 57.9 00 52.0 P B
KHO8* 05 57 00 47 6/12 05 57.1 00 47.0 P
KH11 05 59 00 51 17/11 05 57.2 00 51.0 P
KI 01* 05 56 00 41 9/12 05 56.1 00 41.0 P
KI09* 05 56 00 49 6/12 05 56.0 00 49.0 P
KI11* 05 45 00 50.9 6/12 05 56.0 00 50.9 P
KI12 05 56 00 52 21/11 05 56.6 00 52.1 P
KI13* 05 56 00 52 21/11 05 56.4 00 52.9 P
KJ02* 05 55 00 42 9/12 05 55.1 00 42.2 P
KJ12 05 55 00 52 21/11 05 54.5 00 52.6 P
KK03* 05 54 00 42 9/12 05 54.0 00 42.0 P
KK11 05 54 00 51 19/11 05 54.1 00 51.0 P B
Site code* Coordinates Date Field coordinates Vegetation sampling
N E N E Phenology Biomass
KK12* 05 54 00 51 21/11 05 53.9 00 51.0 P
KK19 05 54 00 59 19/11 05 54.0 00 59.0 P
KL04* 05 53 00 44 9/12 05 53.1 00 44.0 P
KL10 05 53 00 50 18/11 05 53.0 00 50.1 P
KL11* 05 53 00 51 21/11 05 53.0 00 51.0 P
KL12* 05 53 00 52 19/11 05 53.0 00 52.0 P
KL14 05 53 00 54 15/11 05 53 00 54.0 P
KM07* 05 52 00 47 28/11 05 52.0 00 47.0 P
KM10 05 52 00 50 18/11 05 52.0 00 49.9 P B
KM11* 05 53 00 51 20/11 05 53.1 00 50.9 P
KM13 05 52 00 53 15/11 05 52 0053.0 P B
KM15* 05 52 00 55 20/11 05 52.1 00 55.0 P B
KM18 05 52 00 58 19/11 05 52.0 00 58.0 P B
KN06* 05 51 00 46 28/11 05 51.0 00 46.0 P
KN08* 05 51 00 47 28/11 05 51.0 00 47.9 P
KN12 05 51 00 52 15/11 05 51.2 00 51.9 P
KO07* 05 50 00 47 28/11 05 50.0 00 47.0 P
KO09* 05 50 00 49 28/11 05 50.0 00 49.0 P
KO11 05 50 00 51 15/11 05 50.1 00 51.0 P B
KO12* 05 50 00 52 20/11 05 50.0 00 52.0 P
KO17 05 50 00 57 20/11 05 49.9 00.57.1 P
KP08* 05 49 00 48 15/11 05 49.0 00 48.1 P B
KP10 05 49 00 50 15/11 05 50.0 00 49.8 P
KP12 05 49 00 52 20/11 05 49.3 00 52.2 P
KP13* 05 49 00 53 20/11 05 49.0 00.52.0 P
KP16 05 49 00 56 20/11 05 49.0 00 56.1 P B
KQ03* 05 48 00 42 5/12 05 48.0 00 42.9 P
KQ05* 05 48 00 45 5/12 05 48.0 00 45.0 P
KQ09 05 48 00 49 15/11 05 48 00 48.8 P B
KQ11 05 48 00 51 15/11 05 47.3 00 51.0 P
KQ13 05 48 00 53 14/11 05 47.5 00 53.1 P B
KQ15* 05 48 00 55 9/12 05 48.2 00 55.0 P
KR02* 05 47 00 42 5/12 05 47.2 00 42.0 P
KR06* 05 47 00 46 5/12 05 47.0 00 46.0 P
KR08* 05 46 00 48 5/12 05 46.8 00 48.0 P
KR10* 05 47 00 50 28/11 05 47.0 00 50.0 P
The coordinates for each site were read from the Ghana 1:50 000 topographical maps and positioned in the field with hand-held
GPS recorders.
The sampling undertaken at each site is indicated (P = phenological sampling; B = biomass sampling) along with the date
(day/month) they were collected.
* Sites not included in the initial survey and subsequently added for further phenological recordings.
Table 3 Sites used for wetland vegetation sampling at Songor lagoon
Site code Map coordinates Date Field coordinates Vegetation sampling
N E N E Phenology Biomass
SA13* 00 51 00 34 26/11 00 51.7 00 34.0 P
SB04 05 51 00 25 22/11 5 51.0 00 25.0 P
SB05 05 51 00 26 22/11 5 51.0 00 26.1 P B
SB06 05 51 00 27 22/11 05 51.3 00 27.1 P
SB07 05 51 00 28 22/11 05 51.1 00 28.0 P
SB08 05 51 00 29 22/11 05 51.0 00 29.1 P
SB09 05 51 00 30 22/11 05 51.0 00 30.0 P B
SB10 05 51 00 31 22/11 05 50.9 00 31.0 P
SB11* 05 51 00 32 25/11 05 50.9 00 32.0 P
SB12* 05 51 00 33 25/11 05 50.9 00 33.0 P
SB13* 05 51 00 34 26/11 05 50.0 00 34.0 P
SB14* 05 51 00 35 26/11 05 51.0 00 35.0 P
SC03 05 50 00 24 22/11 05 58.1 00 24.1 P B
SCO5* 05 50 00 26 26/11 05 50.1 00 26.0 P
SC11* 05 50 00 32 25/11 05 50.1 00 32.0 P
SC12* 05 50 00 33 25/11 05 50.0 00 33.0 P
SC13* 05 50 00 34 26/11 05 49.9 00 33.9 P
SC14* 05 50 00 35 26/11 05 50.1 00 34.9 P
SD01 05 49 00 22 23/11 05 49.0 00 22.0 P
SD09* 05 49 00 30 26/11 05 49.1 00 3.1 P
SD10* 05 49 00 31 26/11 05 49.0 00 31 P
SD11* 05 49 00 32 25/11 05 49.0 00 32.0 P
SD12* 05 49 00 33 25/11 05 49.1 00 32.9 P
SD13* 05 49 00 34 26/11 05 48.9 00 34.0 P
SD14* 05 49 00 35 26/11 05 48.6 00 34.9 P
SE01 05 48 00 22 23/11 05 48.0 00 22.1 P
SE02 05 48 00 23 23/11 05 48.3 00 23.1 P
SE05 05 48 00 26 23/11 05 48.0 00 26.1 P B
SE06 05 48 00 27 23/11 05 48.0 00 27.0 P
SE10* 05 48 00 31 25/11 05 48.1 00 31.0 P
SE11* 05 48 00 32 25/11 05 47.9 00 32.0 P
SE12* 05 48 00 33 25/11 05 48.0 00 33.0 P
SE13* 05 48 00 34 26/11 05 48.0 00 33.9 P
SE13* 05 48 00 35 26/11 05 48.0 00 34.9 P
SE15* 05 48 00 36 26/11 05 48.0 00 46.0 P
SE16 05 48 00 37 24/11 05 48.0 00 37.0 P
Site code Map coordinates Date Field coordinates Vegetation sampling
N E N E Phenology Biomass
SF08 05 47 00 29 24/11 05 47.26 00 28.9 P
SF09 05 47 00 30 24/11 05 47.1 00 29.9 P
SF10 05 47 00 31 24/11 05 47.16 00 31.0 P
SF11 05 47 00 32 24/11 05 47.0 00 31.9 P
SF12 05 47 00 33 24/11 05 47.0 00 33.0 P
SF13 05 47 00 34 24/11 05 47.1 00 34.1 P
SF14 05 47 00 35 24/11 05 47.1 00 35.1 P
SF15 05 47 00 36 24/11 05 47.2 00 36.1 P
SF16 05 47 00 37 24/11 05 47.2 00 37.0 P
The coordinates for each site were read from the Ghana 1:50 000 topographical maps and positioned in the field with hand-held
GPS recorders.
The sampling undertaken at each site is indicated (P = phenological sampling; B = biomass sampling) along with the date
(day/month) they were collected.
* Sites not included in the initial survey and subsequently added for further phenological recordings.
Table 4 Location of transects used for wetland vegetation point and block sampling
Code Name Date Field coordinates Bearing
1 Blekusu 20/12/96 05 58 59.9 01 01 43.0 300º
2 Tasikome 20/12/96 06 01 47.3 01 02 00.5 170º
3 Tegbi 21/12/96 05 51.59.7 00 57 58.2 250º
4 Fiahor 21/12/96 05 50 48.0 00 53 57.0 150º
5 Alakple 21/12/96 05 51 58.6 00 53 00.7 320º
6 Norlopi 21/12/96 06 01 40.9 00 55 49.0 80º
7 Woe 22/12/96 05 48 51.5 00 56 00.1 10º
8 Totokpoe 22/12/96 05 47 05.5 00 31 00.9 19º
9 Wasakuse 22/12/96 05 51 15.9 00 34 01.4 190o
Table 5 Keta aquatic sampling
Site code Map coordinates Date Field coordinates Parameter
KC17 06 02 00 57 18/11 06 02 00 57 x x x
KD17 06 01 00 57 18/11 06 01 00 57 x x
KD18 06 01 00 58 18/11 06 01 00 58 x x
KD19 06 01 00 59 16/11 06 01 00 59 x x x
KD20 06 01 01 00 16/11 06 01 01 00 x x x
KD21 06 01 01 01 16/11 06 01 01 01 x x x
KD22 06 01 01 02 18/11 06 01 01 02 x x
KE15 06 00 00 55 18/11 06 00 00 55 x x x
Site code Map coordinates Date Field coordinates Parameter
KE16 06 00 00 56 18/11 06 00 00 56 x x
KE17 06 00 00 57 18/11 06 00 00 57 x x x
KE18 06 00 00 58 23/11 06 00 00 58 x x
KF13 05 59 00 53 22/11 05 59 00 53 x x x
KF14 05 59 00 54 22/11 05 59 00 54 x x
KF15 05 59 00 55 18/11 05 59 00 55 x x x
KF18 05 59 00 58 22/11 05 59 00 58 x x
KF19 05 59 00 59 23/11 05 59 00 59 x x
KF20 05 59 01 00 23/11 05 59 01 00 x x
KG13 05 58 00 53 22/11 05 58 00 53 x x x
KG14 05 58 00 54 22/11 05 58 00 54 x x x
KG15 05 58 00 55 22/11 05 58 00 55 x x
KG16 05 58 00 56 22/11 05 58 00 56 x x
KG17 05 58 00 57 19/11 05 58 00 57 x x x
KG18 05 58 00 58 19/11 05 58 00 58 x x x x
KG19 05 58 00 59 22/11 05 58 00 59 x x
KG20 05 58 01 00 17/11 05 58 01 00 x x x
KG21 05 58 01 01 16/11 05 58 01 01 x x
KH12 05 57 00 52 22/11 05 57 00 52 x x x
KH13 05 57 00 53 22/11 05 57 00 53 x x x
KH14 05 57 00 54 22/11 05 57 00 54 x x
KH15 05 57 00 55 22/11 05 57 00 55 x x
KH16 05 57 00 56 21/11 05 57 00 56 x x x
KH17 05 57 00 57 19/11 05 57 00 57 x x x
KH18 05 57 00 58 19/11 05 57 00 58 x x
KH19 05 57 00 59 21/11 05 57 00 59 x x x
KH20 05 57 01 00 17/11 05 57 01 00 x x
KI14 05 56 00 54 21/11 05 56 00 54 x x
KI15 05 56 00 55 21/11 05 56 00 55 x x x
KI16 05 56 00 56 21/11 05 56 00 56 x x x
KI17 05 56 00 57 19/11 05 56 00 57 x x
KI18 05 56 00 58 19/11 05 56 00 58 x x x x
KI19 05 56 00 59 21/11 05 56 00 59 x x
KJ13 05 55 00 53 21/11 05 55 00 53 x x x
KJ14 05 55 00 54 21/11 05 55 00 54 x x
KJ15 05 55 00 55 21/11 05 55 00 55 x x x
KJ16 05 55 00 56 21/11 05 55 00 56 x x x
KJ17 05 55 00 57 19/11 05 55 00 57 x x x x
Site code Map coordinates Date Field coordinates Parameter
KJ18 05 55 00 58 17/11 05 55 00 58 x x
KJ19 05 55 00 59 17/11 05 55 00 59 x x x
KK12 05 54 00 52 21/11 05 54 00 52 x x x x
KK13 05 54 00 53 21/11 05 54 00 53 x x
KK14 05 54 00 54 21/11 05 54 00 54 x x x
KK15 05 54 00 55 21/11 05 54 00 55 x x
KK16 05 54 00 56 21/11 05 54 00 56 x x
KK17 05 54 00 57 19/11 05 54 00 57 x x x
KK18 05 54 00 58 17/11 05 54 00 58 x x
KL13 05 53 00 53 15/11 05 53 00 53 x x x
KL15 05 53 00 55 20/11 05 53 00 55 x x x
KL16 05 53 00 56 20/11 05 53 00 56 x x x
KL17 05 53 00 57 19/11 05 53 00 57 x x
KL18 05 53 00 58 16/11 05 53 00 58 x x
KM12 05 52 00 52 15/11 05 52 00 52 x x
KM16 05 52 00 56 20/11 05 52 00 56 x x x
KM17 05 52 00 57 19/11 05 52 00 57 x x x x
KN11 05 51 00 51 15/11 05 51 00 51 x x x
KN15 05 51 00 55 20/11 05 51 00 55 x x x
KN16 05 51 00 56 18/11 05 51 00 56 x x
KN17 05 51 00 57 19/11 05 51 00 57 x x x
KO13 05 50 00 53 20/11 05 50 00 53 x x x
KO15 05 50 00 55 20/11 05 50 00 55 x x
KO16 05 50 00 56 20/11 05 50 00 56 x x x
KO17 05 50 00 57 14/11 05 50 00 57 x x
KP3 05 49 00 43 5/12 05 48 36.4 00 43 07.3 x x
KP4 05 49 00 44 5/12 05 48 11.4 00 44 01.6 x x
KP13 05 49 00 53 23/11 05 49 00 53 x x
KP14 05 49 00 54 20/11 05 49 00 54 x x
KP15 05 49 00 55 20/11 05 49 00 55 x x x
KP16 05 49 00 56 20/11 05 49 00 56 x x x
KQ3 05 48 00 43 5/12 05 47 36.5 00 42 59.7 x x
KQ5 05 48 00 45 5/12 05 47 57.2 00 45 02.4 x x
KQ10 05 48 00 50 20/11 05 48 00 50 x x
KQ11 05 48 00 51 20/11 05 48 00 51 x x
KQ12 05 48 00 52 20/11 05 48 00 52 x x
KQ14 05 48 00 54 23/11 05 48 00 54 x x x
KR1 05 47 00 41 5/12 05 46 42.5 00 40 58.6 x x
Site code Map coordinates Date Field coordinates Parameter
KR2 05 47 00 42 5/12 05 46 52.7 00 42 14.7 x x
KR3 05 47 00 43 5/12 05 46 56 00 43 02.5 x x
KR6 05 47 00 46 5/12 05 46 56.1 00 46 02.4 x x
KR7 05 47 00 47 5/12 05 46 36.0 00 47 00.9 x x
KR8 05 47 00 48 5/12 00 46.53.0 00 48 00.4 x x
KR9 05 47 00 49 15/11 05 46 44.9 00 49.4 x x x
KR10 05 47 00 50 15/11 05 47 00 50 x x
The coordinates for each site were read from the Ghana 1:50 000 topographical maps and positioned in the field with hand-held
GPS recorders.
The sampling undertaken at each site is indicated (C = water chemistry; M = metals and nutrients; F = fauna [benthos and
zooplankton]; S = sediment) along with the date (day/month) they were collected.
Table 6 Songor aquatic sampling
Site code Map coordinates Date Field coordinates Parameter
SB6 05 51 00 27 25/11 05 51 00 27 x x
SB7 05 51 00 28 25/11 05 51 00 28 x x
SB13 05 51 00 34 11/12 05 51 09 00 34 08.0 x x
SC3 05 50 00 24 08/12 05 50 00 24 x x x
SC4 05 50 00 25 08/12 05 50 00 25 x x
SC5 05 50 00 26 08/12 05 50 00 26 x x x
SC6 05 50 00 27 08/12 05 50 00 27 x x x
SC8 05 50 00 29 08/12 05 50 00 29 x x x
SC9 05 50 00 30 25/11 05 50 00 30 x x
SC10 05 50 00 31 25/11 05 50 00 31 x x x
SC12.5 05 50 00 33.5 11/12 05 50 00 33 22.9 x x x
SD2 05 49 00 23 08/12 05 49 00 23 x x x
SD3 05 49 00 24 09/12 05 49 00 24 x x
SD4 05 49 00 25 09/12 05 49 00 25 x x x
SD5 05 49 00 26 09/12 05 49 00 26 x x x
SD6 05 49 00 27 09/12 05 49 00 27 x x x
SD7 05 49 00 28 09/12 05 49 00 28 x x
SD8 05 49 00 29 09/12 05 49 00 29 x x x
SD9 05 49 00 30 09/12 05 49 00 30 x x
SD10 05 49 00 31 25/11 05 49 00 31 x x x
SD11 05 49 00 32 11/12 05 49 14.3 00 31 51.7 x x x
SE6 05 48 00 27 09/12 05 48 13.9 00 26 53 0 x x
SE7 05 48 00 28 09/12 05 48 00 28 x x
SE8 05 48 00 29 09/12 05 48 00 29 x x
Site code Map cordinates Date Field coordinates Parameter
SE9 05 48 00 30 09/12 05 48 00 30 x x
SE10 05 48 00 31 09/12 05 48 00 31 x x
SF9 05 47 00 30 x
SF10 05 47 00 31 11/12 05 47 00 31 x x
SF11 05 47 00 32 11/12 05 47 07.8 00 31 58.4 x x
SF12 05 47 00 33 11/12 05 47 06.1 00 33 05.9 x x
The coordinates for each site were read from the Ghana 1:50 000 topographical maps and positioned in the field with hand-held
GPS recorders.
The sampling undertaken at each site is indicated (C = water chemistry; M = metals and nutrients; F = fauna [benthos and
zooplankton]; S = sediment) along with the date (day/month) they were collected.
Table 7 Angor channel aquatic sampling
Channel code Map cordinates Date Parameter
C1 5 46 42 0 40 58 5 Dec X X
C2 5 46 53 0 42 07 5 Dec X X
C3 5 46 56 0 43 02 5 Dec X X
C4 5 47 36 0 42 59 5 Dec X X
C5 5 48 36 0 43 07 5 Dec X X
C6 5 48 11 0 44 01 5 Dec X X
C7 5 47 57 0 45 02 5 Dec X X
C8 5 46 56 0 46 02 5 Dec X X
C9 5 46 36 0 47 00 5 Dec X X
C10 5 46 53 0 48 00 5 Dec X X
C11 5 46 45 0 49 00 5 Dec X X
C12 5 47 55 5 49 50 15 Nov X X
The coordinates for each site were read from the Ghana 1:50 000 topographical maps and positioned in the field with hand-held
GPS recorders.
The sampling undertaken at each site is indicated (C = water chemistry; M = metals and nutrients; F = fauna [benthos and
zooplankton]; S = sediment) along with the date (day/month) they were collected.
3.2 Sampling methods
The sampling methods are outlined below with a summary of the field sampling protocols
given in appendix 1.
3.2.1 Hydrological and meteorological
The work carried out to describe the hydrological and meteorological conditions at the two
sites consisted of both field and office activities. As part of the survey, hydrological,
hydrogeological and hydrometeorological data relevant to the study area were acquired from
various sources including the Hydro Division of the Architectural and Engineering Services
Corporation (AESC), the Ghana Water and Sewerage Corporation (GWSC), Water Research
Institute (WRI) and the Meteorological Services Department (MSD). The field visits and
measurements were carried out with the view to better describing and analysing the hydrology
of the sites and to recommending engineering options necessary to enhance the ecological
integrity of the wetland ecosystems.
Field work on all aspects of the study started in November 1996 and was completed by the
end of the year. The work carried out for the hydrological aspects of the study of the sites
collection and collation of hydrometeorological and hydrogeological data;
study of field/topographic sheets covering the relevant areas;
field measurements, by standard hydrometric techniques of flows into the lagoons,
notably the Avu and Keta lagoons which are fed by the Todzie River and the Volta
estuary as at time of study;
measurement of static water levels at selected wells within the study area.
Field measurement of conductivity, salinity and total dissolved solids (TDS) of water
collected from the wells by a Hach CO18 meter (model 50150) and establishment of the
location of these wells with a Garmin 45 GPS.
3.2.2 Physico-chemical
Field analysis
Over 120 stations were visited at the two sites over the study period. At each station the
following parameters were measured: water depth, transparency, pH, water temperature,
conductivity, total dissolved solids (TDS), salinity, and dissolved oxygen (concentration and
percent saturation). Water depth was measured at most stations by a metre rule with readings
rounded up to the nearest centimetre. Transparency was measured with a plain white
homemade Secchi disc (30 cm diameter); two readings averaged (disappearance,
reappearance) to the nearest centimetre. Hydrogen ion concentration was measured by a pH
meter (table 8). Water temperatures were recorded from water taken 20 to 30 cm below the
surface, and read by the oxygen meter. Conductivity, TDS and salinity were read using the
HACH meter (table 8).
The HACH meter had been calibrated by standard salt solutions in the laboratory before use
in the field. Dissolved oxygen was read by an Aqualytic meter, which had been checked for
accuracy by cross reference to samples tested by the Azide modification of the Winkler
method. All data were recorded in the field on data sheets. The time and date of visit to each
station was also recorded.
Table 8 Meters used for taking field measurements of physico-chemical parameters
Parameter Meter Range/Accuracy
pH Hach EC10 Model 50050 +/- 0.02
Conductivity/Salinity/TDS Hach CO18 Model 50150 Range 1; 0199.9µS,
Range 2; 2001999µS,
Range 3; 219.99 mS, and
Range 4; 20199.99mS. Accuracy of all ranges +/-
0.5% of full scale
Salinity range 080 ppt +/- 0.1%
TDS Range 019900 mg L-1 +/- 0.1%
Dissolved oxygen Aqualytic OX1 921 Range 050mg L-1 +/- 0.1 mg L-1 or 0199% +/- 1%
temperature -5 450C +/- 15%. Temperature
compensated probe
Laboratory analysis
All water chemistry parameters were measured at the Water Research Institute laboratory –
the only laboratory in Ghana accredited under the GERMP. The methods used all follow the
14th edition of Standard Methods (APHA 1984). Analysis was carried out on 44 samples from
the sites for major ions: sodium and potassium by flame emission photometry at 589 and
766.5 nm respectively; calcium and magnesium by EDTA titration; sulphate by the
turbidimetric method; and chloride argentometrically. Other analyses included: alkalinity by
titration; total phosphate by the stannous chloride method; suspended solids gravimetrically
after drying in an oven to constant weight at 105ºC; zinc, lead and copper by Atomic
Absorption Spectroscopy using a Perkins-Elmer spectrophotometer at 213.9, 217.0 and
324.7 nm respectively.
Sediment samples were collected from the top 30 cm of the lagoon bottom by digging, from
over 50 stations at the two lagoons and were analysed by staff of the Volta Basin Research
Project at the Department of Soil Science, Legon. After air drying the bulk samples, the
particle size composition was determined by dry sieving through a graded set of sieves for the
sand fractions, and by sedimentation for the silt and clay fractions.
3.2.3 Biological
At each of the 120 stations in the two lagoons, five sediment cores were taken to examine the
benthos. Hand held PVC corers with a core area of 0.00196 m2 were used. Each core was
washed separately through sieves of 300 micron mesh, and all material retained on the sieves
removed. The core depths ranged from 5–15 cm depending on substrate type, substrates with
large amounts of shell being more difficult to sample (see Piersma & Ntiamoa-Baidu 1995).
Each invertebrate sample was placed in a 250 ml container for preservation with 4% formalin
to which Rose Bengal at 1 mg per litre had been added. The samples were sorted by eye and
by compound low power microscopes and all the organisms found identified and counted.
Due to the extreme shallowness of many of the stations, plankton nets could not be towed in
the lagoons, neither could Schindler traps be deployed. As such, 50 litres of water were
collected by bucket (5 times 10 litre buckets) from an area undisturbed by previous sampling
and poured through a plankton net. The net had a mesh of 200 microns. Material collected by
the net was washed into 60 ml tubes and treated as the benthos above.
Water samples for phytoplankton identification and density counting were collected from the
120 stations within the lagoons. As with the zooplankton, 50 litres of water were collected
from beneath the water surface and immediately poured through a plankton net (mesh
80 microns), the retained material was placed into a 60 ml tube and preserved with Lugol’s
solution (APHA 1984). Twenty-five of the preserved samples were used for species
identifications. After an initial cursory examination approximately 10 ml of each sample was
boiled in 70% nitric acid to clear all organic materials from the cell walls of the diatomaceous
species present. These were then washed in deionised water and centrifuged to remove all
remains of the acid. The cleared samples were then mounted in the medium ‘Naphthrax’ and
permanent slides made for retention in the International Diatom Herbarium (Curtin
University, Perth, Western Australia) and identification using the specialised literature (J John
pers comm). Given an absence of taxonomic expertise in the project team it was not possible
to treat all 120 samples in this manner.
For chlorophyll analysis, between 0.25 and 1.0 litres of water (dependent on the turbidity)
were filtered through Whatman GFC paper (nominal pore size 1 micron). After filtration, the
paper was placed in 10 ml of 90% methanol and stored on ice. The samples were kept cold for
a further 24–26 hours and then analysed with a spectrophotometer at the field base. A HACH
field spectrophotometer with a 1 cm wide cuvette was used to measure pigment levels after
extraction, with optical density (OD) readings being taken at 750, 664, 647 and 630 nm
wavelengths according to the trichromatic method (Jeffrey & Humphrey 1975, APHA 1984)
for chlorophyll a, b, c1 and c2. The optical density at 750 nm was subtracted from each of the
other readings. The chlorophyll concentration (mg L-1) in the extracts was calculated by the
following equations and then multiplied by the volume of extract (mL) divided by the volume
of the filtered sample (L):
Chlorophyll a = [11.85(OD664) 1.54(OD647) 0.08(OD630)]
Chlorophyll b = [21.03(OD647) 5.43(OD664) 2.66(OD630)]
Chlorophyll c = [24.52(OD630) 7.60(OD647) 1.67(OD664)]
The phaeophytin component in each sample was then determined by acidifying the sample
with one drop of 1M HCl and taking a further reading at 665 nm. After subtraction of the
reading at 750 nm the concentration of phaeophytin (mg L-1) was calculated by the following
equation and then multiplied by the volume of extract (mL) divided by the volume of the
filtered sample (L):
Phaeophytin = 26.7[1.7(OD665) (OD664)]
The aquatic macrophytes at each grid point within the lagoons were collected in five replicate
0.31 x 0.31 m quadrats, giving an area of 0.1 m2. All above-ground plant material within the
quadrats was removed by cutting, placed into plastic bags and returned to the field base. (The
same samples were used for collecting macroinvertebrates.) The plant material was initially
sun-dried and then oven dried to a constant weight at 70ºC and weighed.
The wetland macrophytic vegetation sampling initially involved species occurrence and
phenological recordings being made at each site with above-ground biomass at every other
site in the initial surveys. All sites were used to record species presence and dominance whilst
only every other site in the initial survey was used for biomass sampling. Tables 2 and 3 list
all sites surveyed and whether or not biomass samples were collected.
At each site a species list was made within approximately 50 m radius. The dominant 1–3
species were identified and five 1 m2 quadrats randomly placed approximately 5 m apart
within the area occupied by the dominant species for semi-quantitative recordings of species
biomass dominance, ground cover and phenological state. The biomass dominance was
recorded in a numeric descending order. The ground cover and proportional estimates of
phenological state were recorded on a six-point scale that reflected the percentage ground
cover and phenological states, respectively.
Scale 123456
Percentage 1% 2–25% 26–50% 51–75% 76–99% 100%
The phenological recordings included estimates of the proportion (percentage) of plants
flowering, fruiting and seeding along with estimates of the proportion of juvenile, mature and
senesced plants in each quadrat. These recordings were done on the basis of the six-point
scale given above and, where applicable, covered the occurrence of flowering, fruiting,
seeding, juveniles, adults and senescence of the dominant species within each quadrat.
Above-ground biomass sampling was confined to the dominant 1–3 plant species at each site.
Five replicate 0.25 m2 quadrats were placed approximately 5 m apart within the area occupied
by each species and all above-ground plant material removed by cutting, placing into plastic
bags and returning to the field base. The plant material from each bag was cut into smaller
pieces, sun-dried and then oven-dried to a constant weight at 70ºC and weighed.
Nine stations, roughly equidistant around the Keta lagoon, were selected for vegetation
transects. After locating the transect on the ground with a GPS, a 200 m line (pre-marked at
1 metre intervals) was laid perpendicular to the water’s edge. When practicable, the 100 m mark
on the line was placed at the water land interface. Two methods were then used for recording
presence and absence of vegetation. In the ‘point’ method, a pole was placed vertically on each
meter mark on the line and every plant the pole touched was recorded. In the ‘block’ method,
two observers walking on either side of the line noted all species found in blocks of 1 x 5 m,
delimiting the distance away from the line by one metre sticks.
4 Ecological character of Keta and Songor
4.1 Meteorology and hydrology
4.1.1 Climatic conditions
The climate of the study area lies within the dry Equatorial climatic region of Ghana, which
also covers the entire coastal belt of the country. This region is the driest in the country and is
referred to as the central and southeastern coastal plains. The coastal lands of Ghana have two
clearly defined seasons, the Dry season and the Rainy season. The Rainy season exhibits
double maxima, the main one occurring between April and June and the minor one between
September and October. June is normally the wettest month in the area. The annual isohyetal
pattern of the coastal belt has the minimum in the west outside Accra up and close to Songor
lagoon in the east. The prevailing wind direction is from the southwest (the southwest
monsoons). This is a characteristic feature for the entire coastal belt of the country. Mean
monthly averages of daily wind speed range between 21.1 to 29.0 km h-1. However, high
velocity winds (110 km h-1) of short duration have been recorded in Accra. The north east
trade winds rarely reach the coast.
Day light and sunshine (hrs) in project area
The day length varies between 11.8 h and 12.5 h in the study area. It reaches its maximum in
June and minimum in January. Daily sunshine duration is least in June (4.8 h) when there is
maximum cloud cover and maximum in November (8.4 h) with a mean of 6.9 h. The values in
table 9 give an idea of the general variation in the hours of sunshine within the study area.
Table 9 Day length (hours) and hours of sunshine within the study area
Day length (hours)
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Mean
11.8 11.9 12.1 12.3 12.4 12.5 12.4 12.3 12.2 12.0 11.9 11.8 12.1
Hours of sunshine
Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Mean
7.1 7.2 7.2 7.0 6.6 4.8 5.4 6.3 6.7 7.8 8.4 7.8 6.9
Relative humidity
Relative humidity data for the study area are estimated using data at Ada and Keta. Since
local variation in relative humidity is not appreciable especially within the same climatic belt,
humidity values at Ada and Keta are considered representative of relative humidity for the
Keta and Songor Ramsar sites. Generally, relative humidity is high in the mornings and at
night, but is at a minimum in the afternoon (table 10).
Long-term temperature records are available at the Ada synoptic station. Records at this
station give minimum average temperatures between 23ºC and 26ºC whereas the maximum
lies between 27ºC and 32ºC. August is normally the coldest month in the area. Records from
the Keta observation station indicate that the minimum average temperature is 24ºC, whereas
the maximum average is 31ºC.
Table 10 Percent relative humidity at Ada (5-year average) and Keta (14-year average)
Time Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Mean
00:00 89 90 90 88 89 91 94 96 94 91 89 91 81
06:00 90 92 89 91 93 93 95 97 95 93 92 92 93
12:00 71 74 74 76 77 82 81 80 78 75 74 71 76
18:00 83 86 85 84 85 88 90 91 90 88 86 87 87
Time Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Mean
00:90 81 79 77 76 79 82 81 80 78 77 78 78 79
15:00 66 64 63 66 70 75 73 70 69 69 67 66 68
The study area experiences two rainfall maxima with the annual average for different periods
ranging from 688 to 855 mm (table 11). Rainfall occurs between March/April to July and
September–October. The low rainfall gives rise to stream flow mainly in the Rainy season only.
Between November and April, the numerous small streams that drain the area dry up; over the
last decade even the Todzie River dried into a series of pools in its lower reaches.
Rainfall in the greater Keta basin area was estimated using available records at Keta, Anyanui,
Sogakope, Dabala, Anyako, Atiavi, Anloga, Adina, and Afiadenyigba. Rainfall records in the
study area were reliable up to the 1980s. However, after this period, there are a lot of gaps in the
data; only the stations at Keta and Ada have consistent data. Wakuti (1968) carried out an
extensive analysis of rainfall in the study area and concluded that the variation in annual totals
was small. Hence, for the southern section where the annual rainfall was about 900 mm,
isohyets of the average monthly rainfall were not necessary. The arithmetic mean was
considered adequate. In the present estimate, therefore, the arithmetic mean of the monthly
rainfall is computed to give the mean basin rainfall for the Keta basin. This is presented in table
11 and illustrated in figure 5. Because Keta station has a long record, the variation in rainfall
over various time frames was examined and this is also presented in table 11. It is clear from
this presentation that between the mid-1970s and the early 1990s the mean annual rainfall has
been low. Therefore the upper range of annual rainfall of 910 mm is on the high side.
Long records of rainfall data are available for the Ada synoptic station. From the data, the
following pattern of rainfall is observed. The maximum rainfall occurs in June with the major
season itself beginning from March/April. There is also a minor season between September
and October. The mean monthly variation is depicted in figure 5. A summary of the rainfall
statistics for various time periods is presented in table 11. Whereas the long-term mean annual
rainfall is 891.6 mm, the mean between the mid-1970s and the early 1990s falls below this
long-term mean by over 23%.
Pan evaporation
Evaporation from the Keta and Songor areas was estimated using direct measurements of pan
evaporation from Ada and Tema. It is recognised that both stations lie in the same climatic
belt. It was observed that there is a decrease in evaporation from Tema to Ada. This
difference notwithstanding, Ada and Tema records were close enough to be averaged to give
the evaporation for the study area. The mean pan evaporation for the study area is presented
in table 12 and illustrated in figures 5 and 6.
Table 11 Rainfall (mm) in the general study area and long-term rainfall in the Keta Ramsar site
General study area
Station Keta Anyanui Sogakope Dabala Anyako Ataivi Anloga Adina Afiadenyigba
Period 1913–1992 1955–1960 1953–1975 1954–1980 1955–1981 1957–1985 1957–1981 1956–1968,
1973–1981 1956–1981
Mean STD
Jan 10.6 12.2 21.8 14.2 10.4 6.4 12.0 8.7 5.1 11.3 4.6
Feb 22.6 1.4 21.3 18.1 28.9 22.3 22.5 28.3 34.2 22.2 8.7
Mar 56.1 116.8 87.2 68.7 71.0 61.7 50.7 73.4 65.6 72.3 18.6
Apr 99.2 173.1 111.4 103.8 105.3 97.3 124.8 130.4 118.3 118.2 22.2
May 155.2 227.3 136.4 178.7 129.0 140.6 156.7 216.2 130.1 163.4 34.6
Jun 187.5 228.2 211.4 194.2 222.1 246.4 306.4 284.2 248.8 236.6 37.3
Jul 64.7 93.4 51.6 52.1 61.4 81.3 66.1 68.2 49.8 65.4 13.6
Aug 19.9 32.2 22.2 25.6 26.2 17.7 14.5 66.0 17.5 26.9 14.7
Sep 49.8 16.5 57.9 43.8 49.2 45.0 40.8 54.9 25.1 42.5 12.8
Oct 88.3 143.2 102.2 108.3 78.7 87.6 88.0 128.9 74.7 100.0 21.9
Nov 34.1 29.4 61.5 68.1 29.4 35.7 32.6 41.6 30.5 40.3 13.7
Dec 12.8 13.9 16.4 8.9 10.8 7.7 7.8 12.0 8.5 11.0 2.8
Total 800.8 1087.5 901.3 884.4 822.3 849.5 923.0 1112.9 808.0 910.0
Mar–Jun 498.0 745.5 546.5 545.4 527.4 546.0 638.6 704.2 562.7 590.5
% 0.6 0.7 0.6 0.6 0.6 0.6 0.7 0.6 0.7 0.6
Sep–Nov 192.2 221.2 243.7 245.7 183.5 185.9 175.9 291.5 147.8 209.7
% 0.2 0.2 0.3 0.3 0.2 0.2 0.2 0.3 0.2 0.2
Keta Ramsar site (mm)
Period Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total
1913–1991 Mean 10.8 23.2 57.5 103.0 157.2 189.9 67.2 21.6 51.7 89.5 35.5 13.6 800.8
STD 19.8 32.7 48.1 57.3 79.1 134.6 84.5 45.7 64.0 69.3 34.0 21.8
CV 1.8 1.4 0.8 0.6 0.5 0.7 1.3 2.2 1.2 0.8 1.0 1.7
1915–1945 11.9 24.6 56.4 106.0 173.4 166.6 70.4 20.5 52.2 105.7 48.4 18.8 855.0
1946–1975 12.6 21.4 56.8 113.6 154.6 261.0 56.3 22.9 45.0 89.6 26.5 7.7 825.8
1976–1991 6.4 16.4 58.6 79.1 146.0 136.9 83.3 23.3 64.1 62.8 20.9 6.2 687.5
Table 12 Pan evaporation (mm) in the study area
Station Period of Record Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec Total
Tema 1972–1980 146.9 162.1 194.1 183.7 173.6 143.3 150.2 133.6 148.0 172.2 176.0 148.8 1932.5
Ada 1961–1967 125.0 135.0 175.0 180.0 160.0 125.0 130.0 140.0 150.0 180.0 155.0 130.0 1785.0
Mean 135.9 148.5 184.6 181.8 166.8 134.2 140.1 136.8 149.0 176.1 165.5 139.4 1858.7
Ada Rain(mm) 9.2 21.2 65.7 110.5 170.7 235.6 63.3 19.3 53.2 85.3 42.4 15.1
(1915 to 1991)
Figure 5 Distribution of rainfall and evaporation at Keta Ramsar site
Figure 6 Distribution of rainfall and evaporation at Songor Ramsar site
4.1.2 Hydrological conditions
Keta catchment
In general, stream flow in the area is seasonal, and corresponds to the seasonal variation in
rainfall. A few coastal streams drain the area above the Keta lagoon. The major streams apart
from the Volta River include the Todzie River, which discharges into the Avu lagoon just
north-west of Keta lagoon, and the Belikpa River, which discharges directly into the Keta
lagoon. Belikpa River has a relatively small catchment area of less than 300 km2. Alpen
Consult et al (1992) reported that the freshwater end of all these coastal streams is about 25
km (minimum) from Keta, thus making it impracticable to develop them into surface water
sources within a reasonable and economic distance.
Volta River
The Volta River is the largest drainage system in the country, with a total drainage area of
379 000 km2 at Akosombo. It is an international river. It flows along the eastern boundary of
the Greater Accra Region and is dammed at Akosombo and Kpong to provide electricity. The
reservoir, which is a potential source of raw water supply to a number of towns, provides
water for Accra and parts of the Eastern Region. The mean annual flow at Senchi before the
construction of the dam was 36.6 x 109 m3 (1160 m3 s-1).
After the construction of the dams, the river had an annual runoff depth varying between
30 mm and 240 mm and a mean annual flow of 1100 m3 s-1 downstream of the Kpong dam.
The highest recorded flood on the Volta River was 14 200 m3 s-1 in 1963. Water abstraction
downstream of the Akosombo dam at Kpong is the major development and takes about 0.01%
of the yield of the river.
Todzie runoff
Annual runoff of the Todzie River is highly variable. Over the period 1957 to 1968, the
minimum annual runoff (1958) was 79 x 106 m3 whereas the maximum was 587 x 106 m3
(1963). The mean for the period was 345 x 106 m3. The Todzie River has a catchment area of
2200 km2. However, the area commanded by the gauging stations Todzienu and Tove is
slightly lower and totals 2120 km2. The mean annual flow at Todzienu on the Todzie River is
estimated at 11 m3 s-1 with a reliable yield (50 year return period) of about 0.05 m3 s-1. The
100-year flood is estimated at about 140 m3s-1. In view of the monthly and annual variability
in the flow, Todzie River could significantly contribute to flooding in the Keta lagoon as has
been reported in earlier studies. The estimated capacity of the Keta lagoon when there is no
inflow is about 360 x 106 m3. The mean annual flow on the Todzie is presented in figure 7. As
expected, the peak flows occur in June and October while the low flows occur between
November and April. The one-day flow measurement carried out on this stream in mid-
November was 2.81 m3 s-1. This is low compared with a mean of 6.6 m3s-1 for November for
the period 1957 to 1968. However, for the same period the mean annual flow varied between
2.5 m3 s-1 and 18.5 m3 s-1. The highest peak flow recorded on the Todzie River was 215 m3 s-1
in 1968. Between 1964 and 1990, the data were not continuous, but the average discharge for
November for that period was 7.1 m3 s-1.
Rivers Aka and Belikpa
These two small rivers have a combined catchment area of 700 km2 and drain into Keta
lagoon. They are situated north of the Keta Ramsar site. They discharge their water through
culverts crossing the highway at Afife and Atiteti. There are no historic records of flows on
these rivers. However, rainfall data and their distribution are known in the catchments and
that can give an appreciation of the flow regime from these streams.
Figure 7 Monthly discharge on Todzie at Todzienu (m3 s-1)
They are known to dry out between December and April each year. By analogy to the Todzie
catchment, runoff coefficients have been transferred to these catchments and estimates of
runoff made. For a wet year, the estimated runoff from the two catchments is 100 x 106 m3
(based on an annual rainfall of 1300 mm and a runoff coefficient of 11%).
Aka has an area of 420 km2 and Belikpa 280 km2. The runoff for June in rainy months could
be as high as 25% of the annual runoff. The combined flow of the two streams could be as
high as 25 x 106 m3, Aka being 5.8 m3 s-1 and Belikpa 3.8 m3 s-1.
Variation in water levels
Most of the hydrological data available for the study area after 1968 are water levels at a few
gauging stations within the basin. On the Volta River, water releases through the penstocks
for power production are available. There are gaps in the hydrometeorological data, especially
after 1980. However, some water level measurements were carried out at selected stations in
the lower Volta basin, including part of the study area.
Before 1964, records on the Volta at Sogakope showed that water levels increased from 1.4 m
in the Dry season to about 6.6 m in September or October. After the construction of the
Akosombo dam, however, water level records were uniform at this station, the slight
fluctuations being from the operation of the hydropower station and rainfall downstream of
the dam. Between 1990 and 1992, variation in water levels at Anyanui had a maximum value
of 0.5 m within a year.
Available lagoon water levels for the Keta lagoon indicate that the lowest water levels were
reached in March, April, or May with the highest levels occurring in July. This observation is
for the period between 1970 and 1980. Between 1988/89 to 1991, water levels records show
that high lagoon water levels were observed between July and November. Because the period
is short, however, it cannot be conclusively stated that there is a change in the occurrence of
the peak floods.
Daily gauge heights at the staff gauge on the Avu lagoon at Avuto between 1989 to 1992
indicated that the highest water levels occurred in July (0.9 m) and the lowest in April
(0.2 m). On the Volta River at Agordome, the minimum water levels occur in March or April
and the maximum in September or October, with a variation in level of between 0.5 to
0.73 m. Since the filling of the Volta dam at Akosombo in 1965 the mean annual rainfall was
below average from 1983 to 1985, resulting in very low water levels in the lake which
resulted in power rationing in 1984.
Changes in flow regime
The regulated flow in the Volta began when Akosombo and Kpong hydro-power plants were
commissioned in 1965 and 1984 respectively. This has created a new flow regime between
Kpong and Ada, resulting in a progressive growth of a sandbar at Ada, which restricts flood
discharge (into the sea) and tidal movement into the river. The resulting change in fauna and
flora encouraged the growth of disease vectors such as schistosomiasis carrying snails, and
created changes in the flow regime between the interconnecting creeks and streams between
the Lower Volta River and the Avu–Keta basin, including Avu, Keta and Angaw lagoons.
In the early 1990s, the Volta River Authority dredged the estuary. Whilst the dredging has
controlled vector snails by admitting some amount of saline water into the river, salinity
levels have been slightly altered in the lower reaches of the river. Salinity studies carried out
under the feasibility study for the scheme (Ada–Keta District Water Supply Scheme
Feasibility Study for the Master Plan, GWSC) indicate a decreasing trend in salinity from the
estuary at Ada Foah, with water at Sogakope being almost unaffected by the incursion of
saline water from the estuary.
The dredging activities currently being undertaken in the Angor channel connecting Keta
lagoon to the Volta River will increase its flood carrying capacity in both directions (ie from
the lagoon to the Lower Volta and vice versa). Since the floods originating from the Todzie
River occur in June and those on the Volta in September/October, it is necessary to have a
large carrying capacity in the channel. The measured flow in this channel during the current
field survey gave a flow of 2.5 m3 s-1, at a time when the water was flowing from the lagoon
to the Volta River.
Water balance of the Todzie–Aka Basins
The individual components of a water balance for the Keta basin are presented to give a
general idea of the water flow situation. The average annual rainfall in the Todzie–Aka Basins
varies from 1400 mm in the north to about 910 mm in the south. Pan evaporation is 1780 mm
per year. The combined runoff from all basins in Ghana is estimated at 196 mm per annum.
The evaporation data presented for the study area give an annual value of 1859 mm. Rainfall
at selected stations in the Keta basin is 910 mm per annum. Thus, on the whole, evaporation
far exceeds rainfall. Recharge occurs mainly in the months of June and July and to a lesser
extent in September and October. There is a net flow of seawater into the lagoon as a result of
the tidal effects, especially along the narrow sand dune edge along the coast.
Groundwater resources
The upper geologic strata consisting of the tertiary rocks on the gravelly base and the more
recent sediments seem to favour occurrence of groundwater in areas where they show
prominence. These occur mostly towards the south-eastern and north-eastern parts of the
study area where two shallow limestone aquifers and a deep limestone aquifer exist.
Freshwater has been obtained from this deep limestone aquifer at depths of between 80 and
300 m at the inland and coastal areas respectively. In the Keta area, the only piped system is
in this deep limestone aquifer. Shallow hand dug wells in the Recent deposits along the
coastal area also provide some of the water requirements of the people, although these sources
are potentially at risk of contamination from surface wastes and saline intrusion.
Potential well fields exist around Agbosome, Afiadenyigba and Nagopo, however, these high
yielding boreholes stand the high risk of saline intrusion. They also require high pumping
heads, which translate into high operational costs. As there is no clear knowledge of the
seawater/freshwater boundary for the limestone aquifer, it is believed that continued pumping
from the aquifer may increase the risk of salinity.
Songor catchment
In general, stream flow in the area is seasonal, and corresponds to the seasonal variation in
rainfall. A few coastal streams drain the area above the Songor lagoon.
The Sege River has a catchment area of about 75 km2 and drains the north-western part of the
Songor lagoon. There are no records of flow on this river because it is not gauged, but based
on a 12% recharge, the estimated mean annual flow depth is about 100 mm. The other major
stream draining into the Songor lagoon flows through Hwakpo. It has a catchment area of
about 50 km2 and flows from north to south. As mentioned already, all these streams are
Other water sources
There are other seasonal streams in the study area some of which have dams or dugouts
constructed on them for water supply purposes. None of these small streams is gauged and so
there are no records of flow available. The total area that drains into the Songor lagoon is
estimated at 240 km2.
The Songor lagoon is the biggest lagoon in the Songor Ramsar site. The water body covers an
area of 115 km2 and it extends for about 20 km along the coast and 8 km inland behind the
narrow sand dune. The lagoon, together with the surrounding floodplain, has extensive
shallow water mudflats and islands suitable for feeding and roosting of seashore birds. Apart
from the Songor lagoon, there are other small lagoons including Kadza, Truku and Kunye.
There are no water level records for these lagoons.
Variation in water level
There are no staff gauges on Songor lagoon and therefore no records of lagoon water levels. It
is, however, expected that since this lagoon has no perennial inflowing streams, it will dry
much faster than Keta lagoon will. Data for Keta lagoon indicate that the lowest water levels
were reached in March, April, or May with the highest levels occurring in July. This
observation is for the period between 1970 and 1980.
During the field work, it was learnt that the lagoon normally dries out in the Dry season and
the sand dune near Lolonya is broken using a bulldozer to allow sea water to flow into the
lagoon at high tide. It is subsequently closed and the water evaporates under natural
conditions throughout the year. Salt is then harvested by the various communities and the
cycle is repeated. For this reason the lagoon water is hypersaline most of the time. Vacuum
Salt Works Industries currently undertakes salt mining in the area and is responsible for the
temporary opening of the lagoon to the sea.
Part of the brine in the lagoon is pumped by Vacuum Salt Works Industries and properly
managed to ensure that they mine salt all year round by diverting floodwater from the
catchment away from their saltponds. The other part of the lagoon, which is not managed,
mixes with freshwater from the catchment and undergoes natural evaporation until it dries out
completely. The salinity in the lagoon therefore increases progressively from the beginning of
the Dry season until the time it dries out.
Groundwater resources
Complex crystalline basement rocks dominate the hydrogeological setting of the Accra Plains
which includes the study area. Semi-confined and confined aquifers occur beneath the water
table. It is difficult to locate large quantities of groundwater because of the lithology and
structure of the basement rocks which include granitic gneiss and schistose lithologies. These
are impermeable and have limited storage capacity within their matrix. It is the pattern of
fracturing that controls the accumulation of groundwater. Where the fractures are
unidirectional, interconnected ground water bodies do not form. Isolated water fill the cracks
resulting in only limited groundwater potential.
Groundwater in the region occurs mainly in confined fractured reservoirs. The mean thickness
of the water bearing formation of the crystalline rock unit is about 3 m, occurring between
depths of 10 and 13 m below the ground level.
Apart from the crystalline basement rocks, other hydrogeological units found in the Accra
Plains include Recent and Eocene sand, gravel and sandstone of the Accraian formation which
occur in the vicinity of Accra. Shallow and fresh ground water bodies exist in lowland parts of
the Recent and Eocene deposits. The main aquifers occur in the sandy and gravelly beds. The
Accraian formation has an insignificant groundwater potential, although small supplies can be
obtained in hand-dug wells in the jointed and fractured portions of the formation.
4.2 Bathymetry and sedimentology
4.2.1 Basin morphometry
Both the Keta and Songor lagoons have the shape of a plate, characterised by the shallowness
of both water bodies, and their ephemeral nature. Due to the extremely low gradients, which
in the case of the Keta lagoon are in the order of 1:20 000–100 000, minor changes in water
level have significant effects on the water area. Table 13 gives some morphometric
parameters for the Keta lagoon. At the time of taking measurements, the natural portions of
the Songor lagoon were dry and the areas under salt production were subject to artificial
manipulation. Further, the natural shape of the basin has been modified by salt exploitation.
The flat nature of the lagoon bottom for both Keta and Songor, coupled with the strong land
and sea breezes, gives rise to a phenomenon that can be termed ‘wind creep’. When this
happens, the force of the wind is sufficient to push/pile the water for several tens of metres in
areas that would otherwise be dry. This has importance to several species of invertebrate, eg
Typmanotonus, that feed in these areas when they are flooded. The water depth is often in the
range of millimetres.
The hypsographic curves presented in figure 8 illustrate the shallow nature of Keta lagoon.
Less than 5% of the total area is deeper than 40 cm. The bulk of the water volume is in areas
that are under 30 cm.
Table 13 Summary of morphometric parameters for Keta lagoon
Morphometric parameter
Max length 24 km
Max width 12 km
Water area* 69.30 km2
Max area 271.75 km2
Volume* 5 560 267 m3
Max depth* 0.75 m
Mean depth* 0.08 m
*At time of measurement (January–February 1995)
4.2.2 Sediments
The sediments of the lagoons are predominantly sands, silts and clays. The larger material is
usually relict shells or shell fragments (see Piersma & Ntiamoa-Baidu 1995 for description
and ecological significance of these larger particles). Figure 9 presents a map of the
distribution of sand, silt and clay in the Keta lagoon. Sand predominates in most of the
samples. Samples taken in the lower portion of the lagoon, ie sampling sites KN–KQ, had a
large clay fraction. This may be a result of the extensive stands of Typha that are found in this
region. Further analysis was carried out on the sediment samples using PRIMER software
(figures 10 & 11). For Keta lagoon there are a group of sites that are clustered together (KI15,
KG17, KI16, KP15 AND KF16). These form a swathe that corresponds to the edge of the
area of influence of flood waters from the Volta as seen from Landsat imagery.
Figure 8 Hypsographic curves for the Keta lagoon
Figure 9 Map of sediments in the Keta lagoon
Figure 10 Cluster analysis of sediments in the Keta lagoon
Figure 11 Cluster analysis of sediments in the Songor lagoon
The cluster analysis separates three types of sediments for Songor lagoon (figure 11). These
are sandy sites to the north and west of the lagoon, muddy sites in the eastern part of the
lagoon and very muddy sites in the coastal area around the middle of the lagoon. Many of the
samples taken in Songor had the characteristic smell of hydrogen sulphide. The sediments had
a high organic carbon content, and dried to a brick-like consistency.
4.3 Limnology
4.3.1 Physical water parameters
The pH of the water did not show any clear trends between sampling stations at each site,
between the two Ramsar sites and the two sampling periods. This lack of pattern is probably
due to the fact that wind induced mixing would lead to a very homogeneous water mass and
the fairly high carbonate content of the water would have effectively buffered any pH changes
that could have resulted from biotic activity.
Temperature in these shallow lagoons was always high. Tables 14 and 15 give the relationship
between temperature and time of day and temperature with depth (see also figure 12).
Table 14 Relationship between water temperature (oC) in the lagoons and time of day
RangeTime Mean temp. oC Standard
deviation n
Minimum Maximum
08:00 28.5 0.14 2 28.4 28.6
09:00 29.3 0.21 10 28.4 29.8
10:00 30.4 0.35 13 29.0 33.3
11:00 32.1 0.21 7 30.2 36.1
12:00 31.6 0.14 12 30.4 34.5
13:00 31.6 3.04 9 30.2 35.2
14:00 31.6 0.40 8 30.8 32.5
15:00 31.6 0.78 10 30.9 32.7
16:00 30.4 0.64 9 28.2 32.4
17:00 30.0 – 1
Table 15 Relationship between water temperature (oC) and depth in the lagoons
RangeDepth (cm) Mean temp. oC Standard
deviation n
Minimum Maximum
<25 34.0 2.1 5 30.7 35.2
25–50 31.4 1.6 14 28.5 33.1
>50–75 30.7 1.0 32 28.4 32.0
>75–100 30.6 2.2 29 28.2 31.6
>100 29.9 1.0 8 28.6 32.3
Figure 12 Water temperature in Keta lagoon
The water was warmest just after noon, with about a 4ºC rise in temperature from dawn.
Surface temperature decreased with the depth of the water present, but even where the water
was deepest, mean temperature was still over 30ºC. There was very little temperature
difference between the surface and bottom. This is not surprising given the very shallow
nature of the lagoon.
The values for Songor are very similar with sampling being carried out between 08:50 and
17:15 hours. The mean temperature was 30.2ºC with a range of 28.2–31.8ºC (n = 22).
Transparency and suspended solids
The water in the two lagoons was basically without true colour. However, due to the strong
wind action and the shallow nature of the lagoons, the transparency was often reduced to less
than 10 cm. This has implications on the primary productivity of the lagoons. The high levels
of suspended solids could explain the comparative lack of filter feeding invertebrates in the
lagoons. The transparency was reduced in areas where there was a large clay fraction in the
sediment and high in areas where there were submerged aquatic plants.
4.3.2 Chemical water parameters
Data on the chemical properties of the water are presented in appendix 2.
Conductivity/salinity/total dissolved solids
The values are within the expected range for coastal lagoons in Ghana, ie values range from
almost freshwater to hypersaline values (from under 2 mS cm-1 to over 80 mS cm-1). The
presence of hypersaline subsurface water was noted in several areas such as the areas around
Adina and most of the Songor lagoon. Freshwater seepages occurred on both the landward
and seaward facing sides of the coastal sand dunes and in the case of the Songor lagoon, these
were the sites with obvious signs of aquatic life.
Dissolved oxygen
In other water bodies in Ghana, with similar morphometric parameters as Keta and Songor
lagoons, it is common to find that the water column is super-saturated with oxygen. This is
not the case in these two lagoons. The reason for this may be because of the absence of
significant algal mats on the sediment. Tables 16 and 17 present the change of oxygen with
time and depth in Keta lagoon.
Table 16 The relationship between dissolved oxygen content in Keta lagoon and time of day
RangeDepth (cm) Dissolved oxygen
mg L-1 Standard
deviation nMinimum Maximum
08:00 24.0 15.5 2 13 35
09:00 22.6 13.2 11 6 45
10:00 27.8 18.4 13 2 70
11:00 14.3 7.9 7 1 24
12:00 23.8 7.5 11 14 39
13:00 29.8 13.4 9 17 56
14:00 26.5 5.4 8 20 35
15:00 28.8 5.7 10 19 40
16:00 38.2 19.7 9 22 85
17:00 34.0 1 –
Table 17 The relationship between dissolved oxygen and depth of water in Keta lagoon
RangeDepth (cm) Dissolved oxygen
mg L-1 Standard
deviation nMinimum Maximum
<25 39.0 15.5 5 13 53
25–50 27.1 8.6 14 13 41
>50–75 22.3 9.3 31 1 45
>75–100 29.4 17.8 24 2 85
>100 26.0 13.1 8 6 45
Two trends can be seen, the first is a general inverse ‘U’ shaped curve for the change with
oxygen with time of day. This is compounded by the fact that oxygen in the water column is
the result of both wind action and photosynthesis. The drop in mean concentrations is the
result of the daily lull in wind action that occurs around this time. Previous work on the
lagoon has shown that the surface waters usually are oxygenated throughout the night. The
second trend is the decrease in mean oxygen levels with increase in depth. This trend supports
the premise that wind action is the driving factor for oxygen in these lagoons.
Major ions
Sodium and chloride dominate the ionic composition of the water of these two lagoons. This
is to be expected given the proximity of the lagoons to the sea. Table 18 presents average
values and the range for Keta lagoon, with comparative proportions shown in figure 13.
Figure 13 Major ions in Keta lagoon (mg L-1)
Table 18 Concentrations of major ions (mg L-1) in water collected from Keta lagoon
Mean Standard
deviation nMin Max
Sodium 4373 1788 17 855 6900
Potassium 231 178. 17 36 885
Calcium 384 422 17 78 1987
Magnesium 873 642 17 133 2888
Chloride 10207 8527 17 815 41300
Sulphate 1212 522 17 300 2460
Alkalinity 310 714 17 98 3080
Industrial development in the two catchments is low and there are no mineral deposits with a
high metal content in the catchment. As a result the trace metal content in the water of the
lagoons was usually below the limits of detection for the methods used (table 19). This is a
positive sign. For logistic reasons sediment samples were not analysed for trace metals.
Table 19 Concentrations of metals (mg L-1) in water collected from Keta lagoon
Site Zinc Lead Copper
KC17 0.09 <0.03 <0.03
KD19 0.04 <0.03 <0.03
KD21 0.04 <0.03 <0.03
KE15 0.04 <0.03 <0.03
KE17 0.06 <0.03 <0.03
KF15 0.04 <0.03 <0.03
KG17 0.04 <0.03 <0.03
KG18 0.04 <0.03 <0.03
KG20 0.07 <0.03 <0.03
KH17 0.05 <0.03 <0.03
KI18 0.04 <0.03 <0.03
KJ17 0.04 <0.03 <0.03
KK17 0.06 <0.03 <0.03
KM12 0.06 <0.03 <0.03
KM17 0.05 <0.03 <0.03
KN11 0.06 <0.03 <0.03
KN17 0.04 <0.03 <0.03
4.3.3 Channel chemistry
Figure 14 presents data on the pH, temperature, total dissolved solids and the dissolved
oxygen content of twelve sites along the Angor channel to Keta lagoon, from the Volta
estuary to the Srogbe bridge. The clearest trend is that of an increase in dissolved solids from
the Volta estuary. There is also a sudden rise in temperature in the vicinity of the lagoon. The
pH of the water follows no clear trend – more work would be needed for this to be explained.
There is a general decrease in the dissolve oxygen content from the Volta estuary to the
lagoon. This could be the result of change from a lotic to a lentic water body, but could also
be the result of inflow of low oxygen water to the channel from the lagoon. Sites C3 and C4
are adjacent to mangrove areas that could contribute significant amounts of organic matter
reducing the amount of water in the water column.
Figure 14 Water chemistry in the Angor channel
4.3.4 Temporal changes in water chemistry
Data on four parameters are presented here: water depth, pH, dissolved oxygen and salinity
(figure 15). The second set of samples were taken five months after the first set in order to get
as wide a time lag within the reporting limits of this study. The samples were taken just at the
beginning of the Rainy season and help give a clearer picture of the situation on the ground.
As would be expected water levels had fallen. This fall is all due to evaporation. The pH of
the water fell at most of the sites, but by amounts that were not significant. The dissolved
oxygen content of the water increased at most of the sites. This is to be expected given that
water levels had fallen and epibenthic algae could now play a role in production processes. At
most sites the salinity of the water increased. Again, this would be expected with the
increasing concentration of ions with fall in water depth due to evaporation.
4.4 Aquatic ecology
4.4.1 Phytoplankton
The phytoplankton in the 25 samples that were used for species identification consist
primarily of benthic diatoms that have been dislodged from the bottom of the lagoons, plus a
few true planktonic diatom species (table 20). These species are typical of shallow lagoons
and most likely make a significant contribution to the primary production. Some blue-green
alga species made up the remainder of the biomass. Both lagoons harboured diatoms
characteristic of high salinity close to seawater or even higher. The only freshwater or
brackish water species present were found in the samples from the Angor channel connecting
Keta lagoon to the Volta. The two lagoons are characterised by separate diatom assemblages,
although many species are common to both. The shallower Songor lagoon contained more
benthic algal species.
Whilst the complete set of samples was not analysed, the initial identifications provide an
indicative list of species. Further analysis of all samples and a time series of samples would
provide more information on the effect of changing water depths and salinity on the primary
production of the lagoons, and the relative importance of the benthic algal mats. A complete
analysis of species present in the samples will require considerable more investment of time
with taxonomic expertise required. For the purpose of an initial characterisation of the two
lagoons this was not considered essential.
The chlorophyll concentrations are presented for each of the aquatic sampling sites in
appendix 3. They are given as chlorophyll a, b and c. The concentrations in Keta ranged from
undetectable to 145 µg L-1 with a mean of 20 ± 21 µg L-1 and in Songor lagoon from
undetectable to 86 µ