A Framework for Undertaking Wetland Inventory, Assessment and Monitoring with examples from the Limpopo Basin, Southern Africa CM Finlayson1 & S Pollard2 1Institute for Land, Water and Society, Charles Sturt University, Australia 2Association of Water and Rural Development, Acornhoek, South Africa With contributions from Mutsa Masiyandima2 2International Water Management Institute, Pretoria, South Africa Salomao Bandeira3 and Dinis Juizo4 3Dept of Biological Sciences, Universidade Eduardo Mondhlane, Maputo, Mozambique 4Faculty of Engineering, Universidade Eduardo Mondhlane, Maputo, Mozambique Alleta Nenguke5 and Edward Chuma6 5Environmental Management Agency, Harare, Zimbabwe 6University of Zimbabwe, Department of Soil Science, Harare, Zimbabwe This report represents an output from the Consultative Group for Integrated Agricultural Research (CGIAR) Challenge Program for Water and Food project “Wetlands-based livelihoods in the Limpopo basin: balancing social welfare and environmental security” being coordinated by the International Water Management Institute. Contents Preface: .............................................................................................................................. 3 1. Introduction ............................................................................................................... 4 2. Wetlands in the Limpopo basin................................................................................ 6 2.1. Intunjambili wetland ............................................................................................ 7 2.2. Ga-Mampa wetland .............................................................................................. 8 2.3. Missavene wetland ............................................................................................... 9 3. An integrated framework for wetland, inventory, assessment and monitoring 11 3.1. Wetland inventory.................................................................................................. 16 3.2 Wetland assessment ................................................................................................ 31 3.2.1 Risk assessment ............................................................................................... 31 3.2.2 Vulnerability assessment ................................................................................. 36 3.3 Wetland monitoring............................................................................................ 40 4. Concluding remarks ................................................................................................ 52 5. Acknowledgments.................................................................................................... 53 6. References................................................................................................................. 53 Annex 1: Inventory methods.......................................................................................... 59 Mediterranean Wetlands Initiative (MedWet) inventory.............................................. 59 United States national wetland inventory ..................................................................... 60 Uganda National Wetlands Programme........................................................................ 61 Asian Wetland Inventory (AWI) .................................................................................. 62 Annex 2: Examples of semi-quantitative wetland risk assessment ............................ 64 A risk assessment of the weed Mimosa pigra in northern Australia............................. 64 Preface: This report contains an outline of a framework for undertaking wetland inventory, assessment and monitoring in the Limpopo basin in southern Africa. It is based on internationally agreed principles and uses information and examples from wetlands in South Africa, Mozambique and Zimbabwe. It is one of the outputs from the Challenge Program for Water and Food project “Wetlands-based Livelihoods in the Limpopo Basin: Balancing Social and Welfare and Environmental Security”. The framework provides an outline of approaches and lists key references and source materials along with practical examples and applications. It is not a technical manual; as a lot of technical material already exists this is referenced not reproduced. The framework has been prepared by a writing team with expertise in biophysical and social sciences as well as site-level knowledge and practical experience. The experience and information obtained from field sites in the Limpopo was used to construct practical examples of inventory, assessment and monitoring of wetlands that are largely intensively used to support human livelihoods. The framework contains information and guidance for making decisions about what inventory, assessment and monitoring is required in response to the main uses and (anticipated) management issues at identified wetlands, whether at the local or basin- scale; in this sense it provides information to support managers make decisions about sustainable use of wetlands; it is directed at decision-makers in government agencies, not at local community level. It does not provide the outcomes for specific management issues – these are made by the decision-makers themselves. It is recommended that the framework is supported by a capacity-building program focusing more specifically on the practicalities of assessing and monitoring wetlands in the Limpopo (and potentially elsewhere) with an emphasis on approaches that can be readily undertaken and provide early warning of possible adverse change. This program could include training and awareness raising components based on user needs related to inventory, assessment and monitoring and how to consider wetland issues at multiple scales from local site to basin-wide. The technical terms used in the manual are based on those agreed in recent years through recognized international mechanisms. For example, the concepts of wetland inventory, assessment and monitoring are based on those agreed by the Ramsar Wetlands Convention and those for ecosystem services agreed by the Millennium Ecosystem Assessment. It is accepted though that locally other terms may be used; the framework provides a reference not a specific standard. 1. Introduction A framework for undertaking wetland inventory, assessment and monitoring of wetlands in the Limpopo basin in southern Africa is presented. This is a part of the CGIAR Challenge Program project on Wetlands-based livelihoods in the Limpopo basin: balancing social welfare and environmental security. The project was undertaken as part of wider efforts to investigate the requirements for enhanced food security and improved livelihoods for wetland-dependent communities in the Limpopo basin and to provide guidelines and tools to assist decision-making at various levels (local community, local governments, and regional policy-makers), including officials responsible for environmental management decisions that affect wetlands as well as technical and extension staff responsible for undertaking inventory, assessment and monitoring activities in wetlands and advising local communities on management practices. Wetlands in the Limpopo Basin have been broadly categorised as dambos (seasonally or permanently saturated areas), pans, and riverine wetlands (Marneweck and Batchelor 2002; Kulawardana et al. 2007). Given the long dry season so characteristic of the Limpopo Basin, they constitute an important component of the landscape and represent an important water and agricultural resource (Figure 1). The integrated framework for inventory, assessment and monitoring is presented to support efforts to obtain the information required to use these wetlands in a sustainable manner. The framework is built around the recommendations provided by the Ramsar Convention on Wetlands (www.wetlands.org) for improving wetland inventory, assessment and monitoring as a basis for assisting wetland managers respond to the many problems they face. These recommendations have been brought together in an Integrated Framework for Wetland Inventory, Assessment and Monitoring that provides a rationale for applying the mechanisms of the Convention for more coordinated inventory, assessment and monitoring (see below). The concepts within the Framework were applied to support the wetland management and livelihood components of the project undertaken in the Limpopo – the experience gained in the Limpopo project was in turn used to illustrate key issues within the Framework. The framework provides an outline of approaches and lists key references and source materials along with practical examples and applications. It contains information and guidance for making decisions about what inventory, assessment and monitoring is required in response to the main uses and (anticipated) management issues at identified wetlands, whether at the local site or basin-scale; in this sense it provides information to support managers make decisions about the sustainable use of wetlands and is directed at decision-makers in government agencies. The framework does not provide the outcomes for specific management issues – these are made by the decision-makers based on the best available information and wisdom. Figure 1: Wetlands in the Limpopo basin are important for agricultural production and water resources for local people. In many wetlands cattle grazing, fishing and cropping are important activities with the latter often depending on irrigation during the drier months (photos: CM Finlayson) Many wetlands in the basin support the livelihoods of local people through agriculture for food production and income generation. Irrigation in the wetlands provides a way of intensifying food production, and alleviates constraints resulting from mid-season dry spells. In drought years the wetlands often have sufficient moisture to sustain crop production, mitigating the potential impacts of drought on food availability. At the same time the wetlands support a host of important ecosystem functions and are complex environments that are intrinsically linked to the catchments in which they occur. Altering the wetland environment through cultivation has potential impacts across the wetland and associated adjacent and downstream areas. Whilst wetland ecosystems have the potential to contribute significantly to peoples’ livelihoods, continued unplanned conversion of wetlands to cropland will result in environmental degradation, in turn compromising livelihood security as well as the other benefits derived from them (see for example Pollard et al 2005, 2007). In recent times management-oriented thinking about wetlands has shifted towards a multi-pronged approach focusing on conserving wetlands while maintaining the livelihood benefits to local people (Finlayson et al. 2005). For these efforts to be successful a minimal amount of information about the wetland is necessary – information that can be used to support decision making to ensure that wetlands are used in a sustainable manner and suitable trade-offs are made between agricultural activities and the wider benefits provided by wetlands to local people and those further afield. 2. Wetlands in the Limpopo basin The Limpopo Basin is situated in the east of southern Africa between about 20 and 26 °S and 25 and 35 °E and covers approximately 41.5 million hectares straddling parts of Botswana (8 million ha), Mozambique (8.8 million ha), South Africa (18.6 million ha) and Zimbabwe (6.1 million ha) (Figure 2). Wetland extent is affected by the highly variable and in places unreliable water flow. This is due to a short rainy season with the annual number of rain days seldom exceeding 50 days per year (Food and Agriculture Organisation 1997). The basin is predominantly semi-arid, dry, and hot with an average rainfall of less than 400 mm with 95% of the rain occurring in a short period sometime between October-April. The South African Highlands part of the basin is temperate while the Mozambique coastal plain is mainly warm and humid. In general the basin has a high level of water deficiency with the highly unreliable rainfall leading to frequent droughts. Figure 2: The Limpopo basin with the three study sites identified Kulawardana et al. (2008) mapped the wetlands in the basin and divided them into four broad classes using a mix of generalised vegetation and land use features: water bodies dominant (0.26 million hectares); grass dominant (1.65 million hectares); farmland- natural vegetation (0.94 million hectares); and riparian natural vegetation (2.34 million hectares). The distribution of the wetland classes are shown in Figure 3. Whilst the imprecise classification precludes detailed analysis of individual wetland types on-ground the mapping does illustrate the generalised extent of wetlands in the basin. The total area of wetlands within the basin was 5.2 million hectares which accounts for 12.5% of the total basin area with the overwhelming proportion of the wetlands located along the lower order streams. The distribution of wetlands among the four countries varied from around 4% in Zimbabwe and in Botswana in the upstream part of the basin, 9% in South Africa in the middle reaches of the basin; and 25% in Mozambique which is in the lower and flatter reaches of the basin. In addition to the basin-wide mapping of wetlands extensive investigations, including further inventory, assessment and monitoring were undertaken at three sites supporting largely subsistence livelihoods for local rural people: the Intunjambili wetland in the Tuli river catchment in Zimbabwe, the Missavane wetland in the floodplain of the Changane River, a tributary of the lower Limpopo, in Mozambique; and the Ga-Mampa wetland in South Africa. Information from these sites is used to illustrate the application of the wetland inventory, assessment and monitoring approaches outlined below. Figure 3: Wetland distribution in the Limpopo basin (Kulawardana et al. 2008). 2.1. Intunjambili wetland The Intunjambili wetland is located in the headwaters of the Tuli River, in the upper part of the Limpopo River basin and some 60 km south east of Bulawayo. It covers a total area of 30 ha and typically for the area, it is a valley bottom surrounded by rocky outcrops (Figure 4). The rainfall is highly seasonal, occurring between November and March, with an annual average of 450 mm and range of 400-600 mm. A large proportion of the sols are poorly drained. Land use and livelihood activities in the wetland are shown in Figure 4. The local community comprises some 550 people in 180 households. Vegetable cropping in and outside the wetland is the main source of livelihoods with sales to local and regional markets. Selling of livestock is also common during times of drought or when households need to purchase food and pay for school fees or health care. Other livelihood sources include non-agricultural activities such as brick making and construction work. Discharge from the wetland flows into a reservoir immediately downstream. The reservoir, built in 1992, is one of the main infrastructure developments in the area and provides water for livestock production, irrigation activities (a communal irrigation scheme supplied by a gravity-fed channel and a medium size irrigated farm), as well as other non-agricultural uses such as brick making and fishing. The wetland provides provisioning services in the form of food (wetland agriculture, edible plant and insect collection and fishing), fibre (reeds used for building, fuel wood, Eucalyptus tree plantation for charcoal), grazing, and water (Figure 4). It is heavily modified by cropping and grazing. Vegetable and maize fields are found along the inundated area of the wetland. The inundated area varies seasonally. Wetland water is used for irrigation along the margins of the wetland and downstream of the dam, building (brick making) and domestic purposes by local communities and communities downstream of the wetland. Cattle graze in between fields as well as in the grassland in the inundated area following the receding flood. Collection of medicinal herbs is done mainly by traditional healers and elderly people in the village. Figure 4. Land use activities in the Intunjambili wetland include grazing and cropping with water taken for domestic purposes and for irrigating along the edges of the wetland and downstream of the dam (photos: CM Finlayson) 2.2. Ga-Mampa wetland The Ga-Mampa wetland is located in the catchment of the Mohlapitsi River, a tributary of the Olifants River in the middle part of the Limpopo River basin. It is a riverine wetland of about 120 ha. The hydrology is affected by the upstream flow and regional groundwater inflows that may link with lateral inputs from small streams and other side- swamps. The catchment is characterized by seasonal rainfall that largely occurs during the summer months, from October to April. The mean annual rainfall for the catchment is 771 mm, but varies significantly with altitude and aspect. Mean annual rainfall in the valley bottom, where the wetland is located, is typically 500-600 mm. Within the boundaries of the wetland the valley floor consists of reasonably well-drained sandy soils upstream and poorly drained sand-loamy soils downstream. The wetland-based community comprises around 2760 people in 394 households in 5 villages. The community make use of many ecosystem services provided by the wetland including provisioning services such as food production (crops and livestock); wild plant collection for food, crafts and building; water supply, and sand extraction from the bed of the river channel (Figure 5). A large part of the wetland has been converted to agriculture over the last decade following the collapse of irrigation schemes after the withdrawal of government support in the mid-nineties and the consequences of a flood in the year 2000, to the extent that more than half of the wetland has been converted to agriculture (Figure 5). As such the wetland is heavily modified through agriculture and grazing as well as well as road and irrigation infrastructure. The wetland is also perceived by some external stakeholders to play an important role in the maintenance of dry season flows in the river. Figure 5. Land use activities in the Ga-Mampa wetland includes grazing, cropping, reed cutting, sand extraction and water for domestic use and irrigation (photos: CM Finlayson) 2.3. Missavene wetland The Missavene wetland is located in Chibuto district, 10 km away from Chibuto city, in the floodplain of the Changane River in the lower Limpopo basin. It covers 28.4 ha wedged between a sand dune and the Changane River. The site represents only a portion of the larger Changane river floodplain wetland, and except for the sand dune and river, physical boundaries are difficult to delineate. The mean annual rainfall is 751mm, with rains occurring mainly in summer months from December to March. During the rainy season the wetland tends to be inundated and normally all agricultural activities cease for about 3 months between January and March. Floods are frequent during February and March. The hydrology of the wetland is greatly influenced by groundwater inflows from the dunes, which maintains high water levels for most of the year. The wetland has limited connectivity with the highly saline Changane River, which forms the western boundary, and only affects the wetland occasionally when it overtops its banks. It is not clear whether the hydrological functions of the wetland are confined to the site or they need to be considered within a wider floodplain context. Wetland users include approximately 750 people in 112 households from several villages located on the sand dunes (Figure 6). Their main livelihood means and source of income is agriculture. Farming seems to be primarily for subsistence purposes. Other sources of income are fishing, remittances and small businesses. The wetland is used for food (rainfed and irrigated agriculture, fishing and hunting) and fibre production (reeds), livestock grazing, as well as access to water for some domestic purposes (Figure 6). The production system includes a mix of vegetables, bananas, maize, and rice, as well as grazing of cattle and goats, and cutting of grass and reeds. Some of the crop production is done under irrigation with water taken from springs at the base of the sand dune. Fishing is one of the main wetland activities year round in the Changane River. Figure 6. Land use activities in the Missavene wetland include reed cutting, vegetable and fruit cropping, grazing and water for domestic and irrigation purposes (photos: CM Finlayson) 3. An integrated framework for wetland, inventory, assessment and monitoring In the last two decades the approach to wetland management has been changing. In recognition of the Ramsar Convention has re-emphasised the value of adopting ecosystem approaches to wetland management. These have developed from a wider body of thinking that started to question the more traditional, reductionist approaches to natural resources management and use. At the heart of this concern is the fact that that despite enormous effort in research and management of natural system for the most part, we have largely failed to chart a sustainable future. Two main considerations therefore standout: • Linkages to the wider context: A fundamental problem was that many traditional approaches did not take cognisance of the fact that ecosystems are part of a wider socio-economic and political context. This meant that they had failed to recognise the importance of key drivers that were not ecological in nature (e.g. market forces or societal values). • Complexity and variability: Many systems had been managed as if they were stable systems in a state of equilibrium, leading to the development of single-value harvest rates (maximum sustainable yield), irrespective of variations in the socio-economic, political and natural context. Rather, a key characteristic of most systems is their variability- it is this that gives them resilience or “adaptability”. Thus we should seek to manage systems in a way that maintains variability. Recognition of the above poses a challenge to monitoring and assessment program. Moreover, socio-ecological systems are complex, dynamics systems and in accepting this, one has to accept a number of characteristics of complex systems. For our purposes it is important to recognise that in many cases they are not entirely predictable (although they show pattern) and therefore have to be managed through a process of strategic adaptive management, that necessarily embraces learning by doing. An important part of designing an integrated inventory, assessment and monitoring process is to appreciate that new understanding will develop as process is implemented and this learning must be incorporated into program. This forms part of what is known as Strategic Adaptive Management which arose as a management framework to address complex systems such as wetlands where we do not have all the answers and where outcomes may be unpredictable because of the effects of multiple drivers at different scales. At a global scale the extent and quality of information for wetland management has been considered to be insufficient to support effective management of a dwindling but valuable resource (Finlayson et al. 1999, 2005). This is particularly true in areas under communal tenure or use, where the governance arrangements (and hence “rules of use”) differ significantly from wetlands on private land (Pollard & Cousins 2008). In support of efforts to overcome such problems the Ramsar Convention has been developing an integrated approach for wetland inventory, assessment and monitoring that is adapted from volume 11 of the Convention’s Handbooks for the Wise Use of Wetlands (Ramsar Convention Secretariat 2007a); http://ramsar.org/lib/lib_handbooks2006_e.htm). The Framework brings together a number of inter-related processes for collecting and evaluating the biophysical information needed to support the management and wise use of wetlands. It was developed primarily from investigations undertaken through the auspices of the Convention’s Scientific and Technical Review Panel and drew on many information sources from different parts of the world (see papers in Finlayson & van der Valk 1995; Tomas Vives 1996; Finlayson & Spiers 1999; Finlayson et al. 2001). It is also supported by separate guidance for developing wetland policies and management plans, including the application of integrated catchment management, integrated coastal zone management, and strategic adaptive management (Ramsar Convention Secretariat 2006 a,b). The Framework is based around the definitions of inventory, assessment and monitoring adopted by the Convention in order to encourage greater conformity when collecting and using information on the biophysical characteristics of a wetland for management purposes (from Finlayson et al. 1999, 2001): • Wetland inventory: the collection and/or collation of core information for wetland management, including the provision of an information-base for specific assessment and monitoring activities. • Wetland assessment: the identification of the status of, and threats to, wetlands as a basis for the collection of more specific information through monitoring activities. • Wetland monitoring: the collection of specific information for management purposes in response to hypotheses derived from assessment activities, and the use of these monitoring results for implementing management. The collection of time-series information that is not hypothesis-driven from wetland assessment is here termed ‘surveillance’ rather than monitoring. The general purposes of and relationships between inventory, assessment and monitoring are shown in Figure 7 and the Framework in Figure 8. A triangular shape has been used to represent the greater level of information and detail that is likely to be collected at the site-scale compared to the broad-scale. It is anticipated that more information could be collected when dealing with the detail of site management than will be necessary when dealing with landscape issues, although the amount of information that will be collected will be largely determined by individual needs and available resources. Figure 7. The general purposes of and relationships between wetland inventory, assessment and monitoring The components within the Framework comprise a continuum of data collection, investigation and feedback steps that when combined provide information necessary for wetland management. Thus, inventory is used to collect information to describe the ecological character of wetlands; assessment considers the status, pressures and associated risks of adverse change in the ecological character of the wetland; and monitoring, which can include both survey and surveillance, provides information on the extent of any change that occurs as a consequence of management actions. Effective management is more likely to occur if inventory, assessment and monitoring are treated as inter-related exercises and combined within the same management structure; a management plan or coordinated management planning process is generally required to ensure this occurs. The Framework shows an overlap between inventory and assessment and between assessment and monitoring. That is, inventory is not totally separable from assessment and similarly assessment from monitoring. The extent of overlap is in part dependent on the methods used and whether the data is collected in a discrete or ongoing manner. Whilst inventory, assessment and monitoring are presented as separate activities they may also be undertaken in an integrated manner with little separation between them: the onus is left with the practitioners to determine how they combine these activities and present their program. The key issue is that all activities are important for wetland management not so much how they are administered. The inter-linkage between inventory, assessment and monitoring is also shown by the arrows that illustrate the circular nature of the relationships within the Framework. Thus, inventory provides information for assessment which provides information for monitoring with feedback to inventory and/or assessment. Recognition of the importance of the feedback links should enable managers to make the best use of the information that is collected. That is, the feedback should provide opportunities to improve or refine the activities within the integrated program. In many instances the distinctions between inventory and assessment, for example, are blurred – the issue is not where one process starts and one stops – the issue is the continuum of data collection and its usefulness for management. A further important feature illustrated in the Framework is the different scales that can be used for inventory, assessment and monitoring; these vary from detailed site specific data to broad-scale analyses across landscapes. It is necessary to recognise that data requirements differ across scales and that gathering data at one scale may not prove that useful when considering activities that occur at a different scale. As not all data can be readily aggregated or disaggregated to apply at different scales this is an important consideration. The main components of the Integrated Framework are outlined in the text that follows and which expands on the concepts for undertaking inventory, assessment and monitoring of wetlands. Separate frameworks and technical guidance are available through the Convention for inventory, assessment and monitoring. These frameworks are presented in summary form below with the assessment component addressing risk and vulnerability assessment rather than other assessment tools directed more specifically at biodiversity assessment and economic valuation. This is in line with objectives of the project – namely – the development of tools for making decisions about balancing social welfare and environmental security in wetlands in the Limpopo basin. The following sections (3.1 - 3.3) provide the key elements/ components of an integrated approach. Each of the components – inventory, assessment and monitoring – builds on the last (Figures 7 and 8) and is intended to answer key questions which are outlined in the appropriate sections. Figure 8: An Integrated Framework for Wetland Inventory, Assessment and Monitoring (adapted from Ramsar Convention Secretariat 2007a). The Assessment component of the framework is comprised of a number of techniques that contribute different information in relation to the management issues being considered. Broad-scale Broad-scale Assessment Inventory Monitoring Site specific Site specific Environmental Strategic Environmental Impact Assessment Assessment Risk Assessment Vulnerability Assessment Rapid Assessment Economic Valuation 3.1. Wetland inventory The Ramsar Convention’s Framework for Wetland Key question: Inventory provides a structured approach for designing a wetland inventory at multiple scales from site-based to What are the key provincial, national and regional. A summary of the steps ecological features, contained within the Framework is provided below (Box processes and ecosystem 1) with the information derived from volume 12 of the services of the wetland? Convention’s Handbooks for the Wise Use of Wetlands (Ramsar Convention Secretariat 2007b; http://ramsar.org/lib/lib_handbooks2006_e.htm). Wetland inventory has been done for a number of purposes, such as: providing a list of a particular type or even all wetlands in an area; identifying wetlands of national or international importance based on agreed criteria; describing the occurrence and distribution of various taxa, such as birds or vegetation; identifying or describing natural resources such as peat, fish or water; and providing a base for assessing wetland loss or degradation (Finlayson et al 2001). For the Mediterranean basin Costa et al (1996) listed the following four objectives for wetland inventory: • identify where wetlands are, and which are priority sites for conservation; • identify the functions and values of each wetland; • establish a baseline for measuring change in a wetland; and • provide a tool for planning and management. These objectives are as important for wetland inventory and management in southern Africa as they are for the Mediterranean. The Framework comprises 13 steps (Box 1) that provide the basis for making decisions in relation to the purpose (and objectives), and the available resources, for an inventory. As there are many reasons for undertaking an inventory a specific inventory method is not provided; the manner in which an inventory is conducted is implicitly connected with the reason for doing the inventory. Practitioners are encouraged to work through the steps in the Framework and develop an approach that suits their particular needs, including the resources available. All steps in the Framework are applicable to the planning and implementation of any wetland inventory, and should be followed during the design and planning process. The framework does not prescribe a particular inventory method. Rather, the framework should be used as a basis for making decisions for undertaking a wetland inventory under the circumstances and purposes particular to each inventory program. Whilst the steps in the Framework provide the basis for designing an inventory project for specific purposes and with the specified resources available, it does not ensure that an inventory will be effective. This can only be done by the personnel engaged to undertake the inventory – the Framework provides a means of outlining a suitable inventory method, including the necessity of identifying training and planning for contingency measures in support of the method. Further information on the individual steps in the Framework is outlined below. Box 1: A summary of the steps for designing a wetland inventory (from Ramsar Convention Secretariat 2007b) Step 1. State the purpose and objective: An inventory should contain a clear statement of its purpose and objective. This should identify the habitats that will be considered, the range of information that is required, the time schedule, and who will make use of the information. A clear statement of the purpose(s) will assist in making decisions about the methods and resources needed. In its simplest terms an inventory is used to describe the ecological character of a wetland – provide information on the ecological components, processes and ecosystem services. Step 2. Review existing knowledge and information Valuable information may be held in many different formats and/or by many different organizations (e.g. waterbird, fisheries, water quality and agricultural information bases, and local knowledge). A comprehensive review of existing data sources may be necessary and its relevance to the proposed inventory work ascertained. Step 3. Review existing inventory methods A number of established methods for wetland inventory exist. The review should determine whether or not existing established inventory methods are suitable for the specific purpose and objectives of the inventory being planned. Some inventory methods use a hierarchical approach, in which inventory may be designed at different spatial scales for different purposes. Many inventories have been based on ground-survey, often with the support of aerial photography and topographical maps and, more recently, satellite imagery. The development of Geographic Information Systems (GIS) and the enhanced resolution of satellite imagery have resulted in greater use of spatial data. Step 4. Determine the scale and resolution The spatial scale used for wetland inventory is inseparable from its objective and greatly influences the selection of the method. It is necessary first to determine the objective for the inventory and then assess how this can be achieved through a chosen scale. Suitable scales for wetland inventory within a hierarchical approach are: a) wetland regions within a continent, with maps at a scale of 1:1,000,000 – 250,000 b) wetland aggregations within each region, with maps at a scale of 1:250,000 – 50,000 c) wetland sites within each aggregation, with maps at a scale of 1:50,000 – 25,000. Each scale needs a minimum mapping unit that reflects the minimum acceptable accuracy for that scale. This is done by first determining the minimum size of feature that can be clearly delineated at that scale, to acceptable standards, and by then determining what measures are required to describe the accuracy/confidence of defining the unit. For example, a land systems map compiled to a scale of 1:250,000 typically involves taking one on-the- ground site observation for every 600 ha surveyed. Step 5. Establish a core or minimum data set A core or minimum data set sufficient to describe the wetland(s) should be determined. The specific details of this data set are inseparable from the level of complexity and the spatial scale of the inventory. It is recommended that sufficient information (the core, or minimum, data set) should be collected so as to enable the major wetland habitats to be delineated and characterized for at least one point in time. The core data can be divided into two components (Table 1): a) that describing the biophysical features of the wetland; and b) that describing the major management features of the wetland as they affect the ecological character of the wetland. The decision whether to undertake an inventory based only upon core biophysical data or also to include data on management features will be based on individual priorities, needs, and resources. Step 6. Establish a habitat classification Many national wetland definitions and classifications are in use. These have been developed in response to different needs and take into account the main biophysical features (generally vegetation, landform and water regime, sometimes also water chemistry such as salinity) and the variety and size of wetlands in the locality or region being considered. A classification based upon the fundamental features that define a wetland – the landform and water regime – is considered superior to those based on other features. The basic landform and water regime categories within such a classification can be complemented with modifiers that describe other features of the wetland, for example, for vegetation, soils, water quality, and size. Increasingly, classifications based on land cover features determined through Earth Observation are being developed. The core biophysical data recommended to be collected in an inventory may be used to derive a classification that suits individual needs. Step 7. Choose an appropriate method Many inventory methods are available. When assessing which method (or methods) is appropriate for an inventory, it is necessary to be aware of the advantages and disadvantages of the alternatives in relation to the purpose and objective of the proposed inventory work. The extent of “ground-truth” survey required to validate the remote sense data should be assessed when considering such techniques. Physico-chemical and biological sampling should be undertaken whenever possible by standard laboratory and field methods that are well documented and readily available in published formats. The bibliographical details of those used should be recorded and any departures from standard procedures clearly justified and documented. As a general rule, the inventory method chosen should be sufficiently robust to ensure that the required data can be obtained within the constraints imposed by the terrain, resources, and time period available. Step 8. Establish a data management system Increasing use of databases and Geographic Information Systems ensure that a large amount of data can be stored and displayed, but these capabilities will be undermined if the data are not well managed and stored in formats that are readily accessible. Potential data management problems can be overcome by establishing clear protocols for collecting, recording and storing data, including archiving data in electronic and/or hardcopy formats. The protocols should enable future users to determine the source of the data, as well as its accuracy and reliability. In addition, a meta-database should be used to record basic information about individual inventory data sets. These meta-data records should include a description of the type of data and details of custodianship and access. Step 9. Establish a time schedule and the level of resources that are required It is necessary to determine the time schedule for planning the inventory, as well as for collecting, processing and interpreting the data collected during an inventory. This is particularly important for field sampling, in which case a sampling schedule that takes into account any special features of the terrain and sampling techniques will be necessary. The schedule should be realistic and based on firm decisions about funding and resources. This will determine the extent and duration of the inventory. The schedule should also include time to prepare for the inventory. The extent and reliability of the resources available for the inventory will eventually determine the nature and duration of the inventory. Step 10. Assess the feasibility and cost effectiveness of the project Once a method has been chosen and a time schedule determined, it is necessary to assess whether or not it is feasible and cost effective to undertake the project. This assessment is essentially a review of the entire inventory method, including the time schedule and costs. Factors that influence the feasibility and cost effectiveness of the project include: • availability of trained personnel; • access to sampling sites; • availability and reliability of specialized equipment for sample collection or analysis of samples; • means of analyzing and interpreting the data; • usefulness of the data and information derived from it; • means of reporting in a timely manner; and • financial and material support for any continuation of the project. Step 11. Establish a reporting procedure The results obtained in the inventory should be recorded and reported in a timely and cost effective manner. The records should be concise and readily understood by others involved in the program or similar investigations. Where necessary the records should be cross- referenced to other documentation from the inventory. All reports should be succinct and concise and indicate whether or not the purpose and objective of the inventory was being achieved, and whether there were any constraints on using the data (e.g. changes to the sampling regime such as lack of replication or concerns about its accuracy). At the same time, a meta-data record of the inventory should be made and added to a centralized file using a standardized format. Step 12. Review and evaluate the inventory Throughout the inventory it may be necessary to review progress and make adjustments to the sampling regime, data management, and program implementation. A review and evaluation process should be developed and agreed as part of the planning and design of the inventory. The review procedures should also ensure that at the end of the inventory, or after a predetermined time period, the entire process should be re-examined and necessary modifications made. The evaluation procedures should be designed to illustrate both the strengths and the weaknesses of the inventory, including necessary reference to the sampling regime and/or the data quality. The evaluation can also be used to justify a request for ongoing funding. Step 13. Plan a pilot study Before launching an inventory a pilot study is essential. The pilot study provides the mechanism through which to confirm or alter the time schedule and the individual steps within the chosen method. The pilot study phase is the time to fine-tune the overall method and individual steps and test the basic assumptions behind the method and sampling regime. Specialist field equipment should be tested and, if necessary, modified, based on practical experience. It is also the opportunity to assess training needs. The amount of time and effort required to conduct the pilot study will vary. The pilot study provides the final step before commencing the wetland inventory itself. Lessons learnt during the pilot study should be incorporated into the inventory method. Table 1. Core (minimum) data fields for biophysical and management features of wetlands Biophysical features Site name (official name of site and catchment) Area and boundary (size and variation, range and average values) * Location (projection system, map coordinates, map centroid, elevation) * Geomorphic setting (where it occurs within the landscape, linkage with other aquatic habitat, biogeographical region)* General description (shape, cross-section and plan view) Climate – zone and major rainfall and temperature etc features Soil (structure and colour) Water regime (periodicity, extent of flooding and depth, source of surface water and links with groundwater) Water chemistry (salinity, pH, colour, transparency, nutrients) Biota (vegetation zones and structure, animal populations and distribution, special features including rare/endangered species) Ecosystem services (provisioning, regulating, supporting and cultural services) Management features Land use – local, and in the river basin and/or coastal zone Pressures on the wetland – within the wetland and in the river basin and/or coastal zone Land tenure and administrative authority – for the wetland, and for critical parts of the river basin and/or coastal zone Conservation and management status of the wetland – including legal instruments and social or cultural traditions that influence the management of the wetland Management plans and monitoring programs – in place and planned within the wetland and in the river basin and/or coastal zone * These features can usually be derived from topographical maps or remotely sensed images, especially aerial photographs. The framework outlined above provides a checklist approach to develop an effective site- based inventory and provide sufficient information to describe the ecological character of the wetland as a basis for assessment and monitoring of the wetland. In the absence of a regional approach to wetland inventory an inventory undertaken at an individual site can provide valuable information for management purposes – the value of this effort can be increased substantially if the methods and information also support information needs that extend beyond the boundary of specific wetlands. In cases where detailed information is not available and sufficient resources have not been allocated for extensive data collection it is still useful to compile the existing information and construct an initial inventory. This is in line with the proposals provided by Costa et al. (1996) for undertaking a simple inventory when resources are not sufficient for detailed inventory. This has been done for the three sites investigated in the Limpopo as an example of the type of information that can be obtained to describe the ecological components, processes and ecosystem services. A summary of the inventory information available for these sites is shown in Table 2-4; further details are available in the references. The information provided in these examples is sub-divided into the three components that describe the ecological character of a wetland; namely, the ecological components, ecological processes and ecosystem services. Many earlier wetland inventories did not include information on ecosystem services. As ecosystem services are now considered by the Convention to be part of the ecological character of a wetland they have been included. The generalised classification adopted by the Convention to describe ecosystem services provided by wetlands is shown in Figure 9 with some examples given in Figure 10. A qualitative analysis of the importance of ecosystem services in different wetland types is provided in Finlayson et al. (2005). Figure 9. Classification of ecosystem services adopted by the Ramsar Convention on Wetlands (see Ramsar Convention Secretariat 2007a) Figure 10. Wetlands provide many provisioning services that directly benefit people, these include the provision of food (from fishing, cropping and grazing) and water supply with many water resources now regulated through physical structures such as channels and weirs (photos CM Finlayson) As the extent of wetland inventory in southern Africa is far from complete with recognised gaps at the national, sub-national and site level (Taylor et al. 1995; Stevenson & Frazier 1999; Thieme et al. 2005; Rebelo et al. 2009) a strategic approach may be necessary to prioritise wetland inventory whether at individual sites or across river basins or sub-basins. In many cases simple wetland inventory may be all that is possible although when this is done there are many advantages in seeking standardisation within appropriate jurisdictional levels whether national, sub-national or at individual sites. The case for standardised approaches to wetland inventory (including classification of wetland types) has been made through the Ramsar Convention whilst recognising that individual needs and objectives will guide the approach taken for any particular inventory. In many instance ground-based field survey may be all that is required for wetland inventory (Figure 11) while in other instances sophisticated remote sensing may also be useful, such as the use of long-band radar imagery for mapping of below-canopy inundation (Rosenqvist et al. 2007). Mackay et al. (2009) provide an overview of the key areas of current application and research in the use of remote sensing for mapping and managing wetlands, while also pointing out gaps that could hinder global inventory, assessment and monitoring of wetlands. Figure 11. Data collection for wetland inventory will generally require extensive on- ground field work and data recording (photos CM Finlayson) Davidson & Finlayson (2007) outline a number of wetland inventory initiatives that have been undertaken or are under development with the support of the Ramsar Convention, including several using remotely sensed data and products. Whilst not directly developed in Southern Africa it is anticipated that they will provide useful guidance for wetland inventory and complement approaches that may already be in use. These include: • further development and elaboration of the Mediterranean Wetlands Initiative (MedWet) inventory method (Costa et al. 1996, 2001; Tomas Vives 1996), including the application of remote sensing; • the development of the Asian Wetland Inventory method, a multiple purpose and multi-scalar approach and being implemented in several parts of Asia (Finlayson et al. 2002); • the first phase of a Pan-European wetland inventory project (Nivet & Frazier, 2004), undertaken by Wetlands International and RIZA, the Netherlands, which expanded and updated the European component of the 1999 Global Review of Wetland Resources and Priorities for Wetland Inventory (Finlayson et al. 1999); • the European Space Agency’s TESEO and GlobWetland projects, which developed demonstration products based on Earth Observation on over 50 Ramsar sites (www.globwetland.org/); and • The Kyoto & Carbon (K&C) Initiative, initiated by JAXA Earth Observation Research and Applications Center (EORC) in 2000 and supporting environmental conventions, carbon cycle science and nature conservation, with information that cannot be obtained in a feasible manner by any other means (Rosenqvist et al. 2007). Mackay et al. (2009) highlight the usefulness of continued collaboration between technical specialists and practitioners to jointly develop tailored applications at wetland site and/or basin scales to suit end-user needs, especially using cost-effective combinations of remotely sensed data, ground-based programs and predictive modelling to support management planning and implementation, and the setting and tracking of progress towards management objectives and targets. For specific guidance on wetland inventory methods the manuals produced for Mediterranean wetlands (Costa et al. 1996; Tomas Vives 1996) are a very useful reference source. Specific examples of wetland inventory techniques are given in Annex 1. Table 2: Inventory for Intunjambili Wetland, Zimbabwe Compiled by: Alleta Nenguke1 and Edward Chuma2 1. Environmental Management Agency P.O Box CY 385 Causeway Harare, Zimbabwe 2. University of Zimbabwe, Department of Soil Science, P.O. Box MP 167 Mount Pleasant, Harare, Zimbabwe Ecological character Topic Information Information Sources Physical - Channeled valley bottom of Intunjambili River (Tuli Catchment) bound with granitic outcrop on eastern border. Mtambanengwe 2006 Features - Situated in Matabeleland South Province, Zimbabwe (28°41' E; 20°27' S). Chenje et al. 1998 - Size ~ 30ha (500m x 600m); slope of 0-2° Surveyor General 1980 - Lies in region IV of Zimbabwean agro-ecological zonation, with semi-intensive farming, fairly low rainfall (450-650mm), periodic droughts, and wet-season (Nov–Apr) dry spells - Wetland comprises (i) poorly drained, permanently inundated (ii) semi-permanent to seasonally wet, (iii) temporarily wet and; (iv) Mtambanengwe 2006 non-wet portions. - Soils: sands & loamy-sands; organic soils in permanently and semi-permanent wet sections; moderately leached + low clay content (10%) in top soil; mainly hydromorphic soils with poor drainage. - Texture: fine sands to clay - Profiles: undeveloped to strongly-developed. - Water inflows: (i) surface and subsurface flow from adjacent hill slopes, and (ii) surface runoff and precipitation with excess Motsi 2008 contributing to surface runoff or subsoil seepage - No information on depth to water table Biological Vegetation Features - Typically herbaceous, dominated by grasses & sedges, absence of woody species Drummond 1981 - Highly humic topsoil tends to low pH - No dominant assemblages; hydrophytic grasses around termite mounds (mainly sedges - pfende), and brackenferns (Pteridium) in main wetland. Native fern species are invasive with an allelopath character (exude chemicals to suppress nearby plant growth) - Encroachment of indigenous invasive species in water bodies [Nymphaea nouchali and N. mexicana] Mtambanengwe 2006 - True algae (blue and green; photosynthetic types in dam) Fauna - Birds, insects (grasshoppers, locusts, and butterflies), aquatic fauna (arthropods, amphibians, and fish) occur. Common bird Ginn & SCNVYO 1973, 1977) species: red bishops, black necked weaver, black headed /grey heron, long tailed paradise whydah, pied crow and water ducks. Counts to establish bird populations recommended. - Tilapia Bream and Kapenta are dominant aquatic fish species Ecological Processes Topic Information Information Sources Nutrient Flows - No data available Owen et al. 1992 - Main nutrient sources- organic fertilizers from cow dung and decomposing foliage; some inorganic fertilizers Water Flows - In wet season water moves through sandy portion; water pushed up as seepage and runoff by impermeable layer of saprolite Owen et al. 1992 (parent bedrock) once saturation reached Ecosystem Services Topic Information Information Sources Provisioning - Crop cultivation is most important use (60% of land). Mtambanengwe 2006 services - Additionally: (a) Food (fish, wild fruit, and crops e.g. Fig fruits Ficus capensis [Muonde]); (b) potable domestic water (community and livestock); (c) grazing for livestock during dry period; (d) medicinal plants; (e) reeds and grasses - Dam used for recreation and fishing - Eucalyptus woodlots act as a source of fuel (alien species) Regulating, - No information available; further work required culturale or recreational Educational - Opportunities exist to (a) promote exchange visits on recommended wetland use, and/or (b) for academic research to build on significant existing work. Supporting - Amount of organic matter, being dependent on wetness, was highest in permanently wet area. Mtambanengwe 2006 services - Nutrients – probably low Grant 1994 - Pollination: Given low woody plant density, probably low Table 3: Inventory of GaMampa wetland, Mohlapitsi River Catchment, South Africa Compiled by: Mutsa Masiyandima; International Water Management Institute, Pretoria Ecological components Topic Information Information sources Physical - Situated in Mohlapitsi River Catchment, Limpopo Province, South Africa Kotze 2005 features - Narrow, channeled valley-bottom of Mohlapetsi River; Sarron 2005 - Area of 120ha, extends 4 – 5 km along river; slope < 1%; M McCartney - field observations, 2006 - Comprises 4 poorly-drained areas (25 ha) of extensive organic (peat) soils surrounded by seasonally saturated, mainly mineral Kotze 2005 soils. Sarron 200) - Inundation thought to be maintained by lateral subsurface inputs M McCartney - field - Wetland and local catchment are underlain by banded ironstone and chert which are likely to contain groundwater observations, 2006 - More than 7 springs remain active even in dry period indicating regional groundwater inflows - Water accumulation occurs in valley bottom. Physico- - pH: given as ~ 6 (Chirion 2005), or ranging from 7 - 8. Nell & Dreyer 2005 chemical - Patches indicative of saline soils sometimes visible Chiron 2005 features - Soils: sandy soils close to channel and fine-textured, poorly-drained areas away from channel Kotze 2005 - Soil fertility low (P and K). Nell & Dreyer 2005 - Part of wetland rich in peat and high organic matter content. - Clear water (visually / no data); no information on pollutants; very limited use of fertilizer Biological Vegetation and fauna features - Wetland heavily utilized for agriculture; dominant natural vegetation (mainly Phragmites australis and some P. mauritanus) Kotze 2005 limited to ~30% of wetland area (wetter parts); average cover (abundance) = 40-50% in 2005 - Most important invasive plants: (i) Arundo donax (Spanish reed) in valley floor; (ii) Solanum mauritianum (bugweed) occurs more widely. - No formal inventory of fauna Ecological processes Topic Information Information sources Nutrient flows Nutrients: Kotze 2005 - Peat (high organic matter) is largest single source of nutrients; generally soil is nutrient-poor (K = 0.09-1.5 cmol (+)/kg; P = 2.7- Nell & Dreyer 2005 44.5 mg/kg) Chiron 2005 - Small percentage (15%) of farmers use inorganic fertilizers (Super Phosphate and N, P, and K; Chiron, 2004); some manure added by livestock (mainly cattle) grazing Water flows - Inflows from (a) precipitation, (b) groundwater seepage and (c) surface flow from hills McCartney 2006 - Outflows through (a) evaporation from bare soil, (b) evapo-transpiration (vegetation) and, (c) sub-surface flow (lateral transfer) to Masiyandima et al. 2006 river (contributes to discharge) Ecosystem services Topic Information Information sources Provisioning - Used for (a) crop production, especially during dry periods; (b) grazing (moderate); (c) dietary supplementation through natural Kotze 2005 services edible products and some fish from Mohlapitsi River, (d) provision of domestic water; (e) assumed importance for carbon storage given accumulation of organic matter in wetlands Regulating - Flow regulation (retention of water in wetland and release during dry season). Kotze 2005 services - Water storage (for agricultural use) McCartney 2006 - No data on erosion control / sediment trapping Masiyandima et al. 2006 Cultural - No information available regarding spiritual or inspirational services Kotze 2005 services - Offers opportunities for formal and informal learning Finlayson 2005 - Potential recreational opportunities and aesthetic services. Supporting - Wetland supports extensive organic (peat) soils maintained by permanent saturation Kotze 2005 services Table 4: Inventory of Missavene wetland, Mozambique Compiled by: Salomao Bandeira1 and Dinis Juizo2 1Dept of Biological Sciences, Universidade Eduardo Mondhlane, Maputo, Mozambique 2Faculty of Engineering, Eduardo Mondhlane, Maputo, Mozambique Ecological components Topic Information Information sources Physical - Changane Catchment, near Chibuto town (hence sometimes referred to as such) Juizo et al. 2007 features - Wetland area ~284 ha (< 1% of Changane catchment) - Different authors identified three and five plant associations/ habitat types within wetland: Either (i) lake; (ii) Bandeira et al. 2006 extensive dry areas and; (iii) hill (Bandeira 2006) or (i) river, (ii) lake, (iii) wetland, (iv) dry land and (v) sand dune (Juizo et al 2007). Juizo et al. 2007 Physico- o Set in hills Massingue et al. in press chemical o Seasonally inundated features o High salt content (exceptional for inland areas) Wild & Barbosa 1967 o Inflow: groundwater contribution through spring discharge from sand dunes & Changane River. FAO/UNESCO/ISRIC 1988 o pH is neutral Munguambe et al. 2006 o Soils are Eutric Fluvisoils - recent, resulting from river sedimentation process Biological Vegetation and fauna features - Over 70 plant species identified, 12 medicinal Bandeira et al. 2006 - Halophytic vegetation (salt tolerant plants) with dominants as follows: Wild & Barbosa 1967 (i) Lake historically dominated by Acacia xanthophloea; today by herbaceous groups (Typha capensis, Phragmites mauritianus, Scirpus maritimus) (ii) Drier areas - driest part = Xanthium Strumarium; Pulchea pubescens; lower area - grasses (Cynodon dacylon, Panicum sp. and Parthenium hysterophorus), Abutilon guineensis and Cyperus spp. (iii) Hill– shrubs (Pulchea pubescens, Catunaregamum spinosa, Lippia javanica). - Grazing grass - average 2.4 t/ha - Fauna and flora - 12 invasive species recorded Ecological processes Topic Information Information sources Nutrient flows Nutrients: - No data available Water flows - Inflows from (a) precipitation, (b) groundwater seepage and (c) surface flow from hills Namburete 2004 - Outflows – discharge area for local aquifer Ecosystem services Topic Information Information sources Provisioning - Used for (a) crop production, especially during dry periods; (b) grazing; (c) dietary supplementation through natural Namburete 2004 services (see edible products and fish, (d) provision of domestic water; (e) firewood Pulchea dioscorides and (f) building (P. App.2) Mauritius) & fencing (Euphorbia tirucalli (milk bush)); (g) medicinal plants - Permanent lake within wetland important for diversity of fauna/ flora Regulating - Limited hydrological role but as part of group of wetlands provides marginal contribution to flood delay and Namburete 2004 services attenuation in lower Limpopo through temporary storage in depressions. - Discharge area for a local aquifer system with many springs (tapped for irrigating cropland) Bandeira et al. 2006 - Flow regulation (retention of water in wetland and release during dry season) - Water storage (for agricultural use) - No data on erosion control / sediment trapping Cultural - African religion (Mazione) uses wetland area for religious practices; one cemetery (mainly for still-born babies) Kotze 2005 services exists in wetland. Finlayson 2005 - No recreational or aesthetic services but potential exists Bandeira et al. 2006 - Opportunities for formal and informal learning Supporting - Role of Missavene wetland in nutrient cycling has not been established. services 3.2 Wetland assessment 3.2.1 Risk assessment The Ramsar Convention’s Framework for Wetland Risk Key question: Assessment provides a structured approach for designing a wetland risk assessment at multiple scales from site-based What is the risk of to provincial, national and regional. The information on adverse ecological effects the framework is derived from volume 16 of the occurring in a wetland as Convention’s Handbooks for the Wise Use of Wetlands a result of a specific or (Ramsar Convention Secretariat 2007a; multiple pressures on the http://ramsar.org/lib/lib_handbooks2006_e.htm) and van wetland? Dam et al (1999). The relationship between risk assessment and other assessment tools, including strategic and environment impact assessment, is shown in Figure 12. In a general sense wetland risk assessment is a structured process that evaluates the likelihood that adverse ecological effects may occur or are occurring as a result of exposure to one or more pressures. The extent of detail that can be compiled in the risk assessment will generally decrease as the geographical area of coverage increases. It is also recognised that for many sites an initial simple assessment will be all that is possible. Although highly structured, wetland risk assessment is a flexible process for collecting, organising and analysing data, information, assumptions and uncertainties in order to estimate the likelihood of adverse ecological effects. As such, it provides a framework for effective analysis and decision making. The Framework provides guidance on how to go about predicting and assessing change in the ecological character of wetlands and promotes, in particular, the usefulness of early warning systems for determining when change may occur. In cases where information is not available and resources have not been allocated for further extensive data collection it is still useful to compile the existing information and construct an initial risk assessment. This has been done for the three sites investigated in the Limpopo as an example of the type of information that can be obtained (Tables 5-7). An effective assessment with a suitable level of data collection and analysis will require further effort – the framework presented below provides a checklist approach to develop an effective approach. The risk assessment framework can be divided into six steps, as listed below and shown in Figure 13: • Identification of the problem • Identification of the effects • Identification of the extent of the problem • Identification of the risk • Risk management and reduction • Monitoring 31 Figure 12. The relationships among the different assessment tools available through the Ramsar Convention (from Ramsar Convention Secretariat 2007a). 32 . Figure 13: Framework for wetland risk assessment (adapted from Ramsar Convention Secretariat 2007a) The details of these steps are explained below (Box 2). All steps in the Framework for risk assessment can be applied to the many direct drivers of change in a wetland, namely, changes to the water regime, water pollution and eutrophication, physical modification to the wetland, over exploitation of biological products or fresh water, and the introduction of exotic species. 33 Box 2: A summary of the steps for undertaking a wetland risk assessment (from Ramsar Convention Secretariat 2007a) Step 1. Identification of the problem This is the process of identifying the nature of the pressure and the environment of interest, and developing a plan for the remainder of the risk assessment based on this information. It defines the objectives and scope of the entire risk assessment. This includes obtaining and integrating information on the characteristics of the pressure, what is likely to be affected, and importantly, what is to be protected. Such information is used to determine the structure and complexity of the remaining steps of the risk assessment. This includes selection of assessment and measurement endpoints: assessment endpoints are explicit expressions of the actual environmental value(s) to be protected, while measurement endpoints are measurable responses to a stressor that can be correlated with or used to predict changes in the assessment endpoints. Thus the selection of ecologically relevant measurement endpoints is essential. Step 2. Identification of the effects In this step the effects of the pressure on the measurement endpoints selected during problem formulation are evaluated. For wetland risk assessment, data for identifying the effects should preferably be derived from field studies, as field data are more appropriate for assessments of multiple pressures, and wetlands are known to be exposed to multiple pressures. Depending on the pressure(s) and available resources, such studies can range from quantitative field experiments to qualitative observational studies. Step 3. Identification of the extent of the problem Data on the effects of a pressure on an organism, plant, or ecosystem provide little useful information without knowledge on the actual level of exposure to the pressure. Identification of the extent of the problem involves estimating the exposure of a pressure to the receptor, by utilising information gathered about its behaviour and extent of occurrence. While field surveys of exposure represent the ideal approach for wetland risk assessment, use of historical records, simulation modelling, and field and/or laboratory experimental studies all represent alternative or complementary methods of characterising exposure. Identification of the effects and extent of the problem form the analysis phase of a risk assessment and are often undertaken simultaneously. In many instances simple assessments are often performed initially, followed by more comprehensive assessments if considered necessary. Step 4. Identification of the risk This involves integration of the results of the two previous steps in order to estimate the likely level of adverse ecological effects resulting from the exposure to the pressure. There are a range of techniques for estimating risks, often depending on the type and quality of effects and exposure data. A qualitative framework is shown in Figure 14 whereby the likelihood of exposure to the pressure(s) is compared to the consequences of exposure. The greatest risk occurs when the exposure is high and the consequences are 34 catastrophic. The qualitative nature of assigning the risk can be countered to some extent by using multiple opinions or multiple lines of evidence. Needless-to-say, a quantitative assessment should remove much of the uncertainty associated with assigning the risk. The output of this step need not be a quantitative estimate of risk. However, sufficient information should, at the very least, be available for appropriate experts to make judgements based on a weight of evidence approach. In the event of insufficient information being available, it is possible to proceed with another iteration of one or more phases of the risk assessment process in order to obtain more information. Regardless of the approach, uncertainty associated with the risk assessment must always be described, while interpretation of the ecological significance of the conclusions must also be carried out. In addition, the risks must be sufficiently well defined to support a risk management decision. Step 5. Risk management and reduction Risk management is the decision-making process that utilises the information obtained from the steps described above, and attempts to minimise the risks without compromising other societal, community or environmental values. If the risks associated with a particular pressure are considered significant, risk reduction steps are taken. These steps should also take into account political, social, economic, and technical factors, and consider the benefits and limitations of each risk-reducing action. It is likely to be a multidisciplinary task between risk manager, assessors, and experts familiar with wetland and pressure(s) being considered. Step 6. Monitoring Monitoring should be undertaken to verify the effectiveness of the risk management decisions. It should incorporate components that function as a reliable early warning system, detecting the failure or poor performance of risk management decisions prior to serious environmental harm occurring. The risk assessment will be of little value if effective monitoring is not undertaken. As with effects characterisation, the choice of endpoints to measure in the monitoring process (i.e. what will be monitored?) is critical. Depending on the nature of the risk assessment and available resources, endpoints may or may not be the same as those used for effects characterisation. As with the inventory component described above it is necessary to relate the effort given to assessment to the need for particular information and the resources available. In many cases a simple assessment may provide sufficient information to guide management responses whilst in others a more thorough and even quantitative assessment may be required. An example of a semi-quantitative wetland risk assessments is given in Annex 2. Whether a simple or complex assessment is undertaken it is stressed that the certainty associated with the outcome should be recorded and kept in mind when making management decisions. One decision could be to undertake further monitoring to reduce any uncertainties in the assessment itself and to use new information to reassess the risk. 35 Likelihood of Consequences of exposure exposure Little Serious Catastrophic Low Very Low Risk Low Risk Medium Risk Medium Low Risk Medium Risk High Risk High Medium Risk High Risk Very High Risk Figure 14: A framework for qualitative assigning of risk based on the likelihood and consequences of exposure. 3.2.2 Vulnerability assessment Vulnerability refers to the relationship between a particular Key question: event having an impact on a system, the risk associated with that impact, and the efforts to manage that risk. Gitay How sensitive is the et al. (2009) defined wetland vulnerability as: the degree wetland and is it able to to which a wetland is susceptible to, or unable to cope cope with the adverse with, adverse effects of multiple pressures, including effects of one or more climate change and variability. Vulnerability assessment pressures? determines the extent to which a wetland is susceptible to, or unable to cope with, adverse effects of climate change and variability and other pressures, such as changes in land use and cover, water regime, or over-harvesting and over-exploitation, and invasion by alien species. These pressures can act individually, cumulatively or synergistically. Vulnerability is determined at specific spatial and temporal scales and is a dynamic property as it changes depending on the local conditions, e.g., a system can be vulnerable at a particular time but may not be at other times (e.g., vulnerability to fire increases during dry seasons). Wetlands are vulnerable if they have low adaptive capacity and are highly vulnerable if they have low inherent capacity to cope with change, and/or there are few or no options to reduce impacts of pressures, and/or they are naturally sensitive to pressures (for example, due to their geographic location or socio-political situation). Vulnerability incorporates risk assessment (i.e., the extent of and exposure to a hazard) and is linked to the stability or resilience and sensitivity of a wetland, as well as capacity to cope with one or more hazard (Figure 15). 36 Figure 15: Relationship between sensitivity, resiliency and vulnerability of a wetland (modified from Alwang et al. 2001; IPCC 2001) Generalising from the extensive literature reviewed by Alwang et al. (2001) the following common characteristics can be applied to the concept of vulnerability assessment: a) It is forward-looking and is the probability of a change in the condition of a system in the future relative to some benchmark (reference or baseline); b) The change is caused by some hazard; c) It depends on the time horizon (i.e. can change depending on whether vulnerability is considered on a seasonal, annual or decadal basis); and d) The present condition of the system, its resiliency and sensitivity. Vulnerability assessment can be seen to have four components: i) the hazard as the probability of an event occurring; ii) the risk being the likelihood of that event leading to adverse change; iii) the response needed to manage that change; and iv) a formulation of desired outcome. Vulnerability assessment should be seen as a process that includes determination of a probability of a risk event occurring, the effect of this on the system, given its sensitivity and resiliency, development of the possible options that can reduce the adverse impacts from that event, formulating the desired outcome for the system within an adaptive management framework to ensure that the response options being implemented are achieving the desired outcomes. The framework for wetland vulnerability assessment provided by Gitay et al. (2009) is based on the OECD state-pressure-impact-response model and incorporates risk assessment (including risk perception by stakeholders) and risk management (Figure 16; Box 3). It incorporates components of the concept of ecological character as a basis for developing indicators for assessing the condition and trends as well as for monitoring in support of the wise use of wetlands. 37 Box 3: A framework for wetland vulnerability assessment (from Gitay et al. 2009) The framework for wetland vulnerability assessment comprises the following steps: 1. Risk assessment (including risk perception) - delimiting the spatial and temporal boundaries of the social and biophysical system being considered; - Identifying the past and present drivers of change and existing hazards; - Assessing the present condition and recent trends in the ecological character of the wetlands (using metrics such as indicator species, functional groups etc); - Conducting a stakeholder analysis – the people involved in evaluating the potential responses and also affected by the potential changes in the system; - Determining the sensitivity and resiliency including adaptive capacity of the system; - Identifying the wetlands and groups of people that are particularly sensitive to different pressures; and - Developing scenarios and storylines, with stakeholders, of the risk from drivers of change and the interaction between them that could lead to future changes. 2. Risk minimisation or management - Identifying the wetlands and groups of people that would not have the ability to cope with the changes, often adverse, given their low present adaptive capacity and/or sensitivity; - Developing response options that can minimise the risk of abrupt and/or large changes in the ecological character of wetlands (and thus maintaining their ability to provide the ecosystem services that humans depend on). These can include: regulations, strategic environmental planning, infrastructure/engineering works, rehabilitation/restoration, developing education material, improving community awareness, developing integrated management plans. In some case, large adaptive capacity, high resiliency and low sensitivity of the system could mean that no further management response is needed; - Trade-off analysis to choose between potential response options given constraints such as institutional capacity, information/data availability and often financial capacity; and - Specifying the desired outcomes for the system. 3. Monitoring and adaptive management throughout the process. This includes a means of measuring the path to the desired outcomes. This framework is very much a conceptual framework and it is expected that adjustments will be made as it is implemented and feedback on individual steps is obtained. It is anticipated that practical considerations in individual assessments will result in adjustments to the framework; the framework is provided as a guide for assessments and 38 is not seen as a tightly prescriptive approach – flexibility and responsiveness to local circumstances are expected. Figure 16: Vulnerability assessment framework for wetlands 39 3.3 Wetland monitoring The Ramsar Convention’s Framework for Wetland Key question: Monitoring provides a structured approach for designing a wetland monitoring program at multiple scales from site- Is the ecological condition based to provincial, national and regional. The information of the wetland improving, in the framework is derived from Finlayson (1996a,b) and worsening or staying the volume 16 of the Convention’s Handbooks for the Wise same? Use of Wetlands (Ramsar Convention Secretariat 2007c; http://ramsar.org/lib/lib_handbooks2006_e.htm). Monitoring addresses the issue of change or lack of change through time and at particular places and 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. An effective monitoring program is not necessarily complex; effectiveness is gauged by the relevance and timeliness of the data or information collected rather than the complexity of the program. In cases where insufficient resources are not available for an effective monitoring program it may still be useful to undertake an initial surveillance program that can provide sufficient guidance to support initial management decisions. This has been done for the three sites investigated in the Limpopo as an example of the type of information that can be obtained Tables 5-7). An effective monitoring program and data analysis will require further effort; the framework presented in below (Figure 17, Box 4) provides a checklist for developing an effective approach. 40 Figure 17. Framework for wetland monitoring (adapted from Ramsar Convention Secretariat 2007c) 41 Box 4: A summary of the steps for designing a wetland monitoring program (from Ramsar Convention Secretariat 2007c) Step 1. Identify the problem or issue Identification of the problem or issue is a crucial step that needs to be done clearly and unambiguously. Where possible, the extent or scale of the problem (or likely problem) should also be identified (e.g. will the entire wetland or a part of the wetland be affected?). Knowing the extent of the problem could be made difficult unless the ecological character of the wetland has been adequately described (e.g. how large is the wetland and how much water does it contain?). Thus, baseline or reference data are needed. The cause (or most likely cause) of the problem should also be identified (e.g. pollutants added to the wetland or over exploitation of a fish species). If the cause is not known an investigative programme should be implemented. Often such information is not available and given the urgency of many management situations little effort is made to obtain it; this is discouraged as without such information it can be difficult to decide what should be monitored. Step 2. Set the objective The objective provides the basis for collecting the information in the monitoring program and should be precisely stated and achievable within a reasonable timeframe. An imprecise objective can negate the usefulness of the monitoring program. In contrast a surveillance program does not require a specific objective. An explicit objective not only assists in determining the nature of the sampling effort, but also enables new staff to continue the work in a consistent manner. Step 3. Establish the hypothesis The objective is supported by an explicit hypothesis that indicates the required level of change that is expected or can be tolerated in the wetland. Unless the hypothesis can be tested on the basis of the collected data or information it will not be possible to know whether the objective has been attained or not. When determining whether or not a hypothesis has been supported by the data/information the sources and extent of variability in the data/information must also be recorded. This is particularly important when the natural fluctuations are highly variable or even unknown. Step 4. Choose the methods and variables When choosing which method(s) are appropriate for monitoring it is necessary to be aware of the advantages and disadvantages of the alternatives in relation to the issues being addressed. A literature review and expert advice may be needed and the objective and hypothesis need to be kept in mind; can the method detect change at the required level and over the chosen time period? The following need to be considered when deciding which method to use: number and location of sampling sites; the sampling frequency and replication; the extent of baseline or reference information; available methods for collecting, processing and/or storing samples; available methods for storing and statistically analysing the data; and the processes for interpreting the data and information. 42 In a general sense, the methods need to be able to detect the presence of any change, assess the significance of the change and identify or clarify the cause. Step 5. Assess the feasibility and cost effectiveness Once a method has been chosen and a sampling regime agreed it is necessary to determine whether or not it is feasible to undertake the program on a regular and continual basis. The following factors may influence the sampling process and continuity of the program: availability of trained personnel to collect and process the samples; access to sampling sites; availability and reliability of specialist equipment for sample collection or analysis of samples; means of analysing and interpreting the data; usefulness of the data and information derived from it; means of reporting in a timely manner; and the financial and material support for continuing the program. In undertaking the feasibility assessment the cost effectiveness needs to be considered and keeping in mint that the aim of the monitoring is to collect useful data or information with the least cost. If an adequate budget is not available the program may need to be reduced; inadequate funding should not be used as a reason to reduce the scientific rigour of a programme. Step 6. Conduct a pilot study Before launching a full-scale monitoring program a pilot study should be conducted to finetune the method and test any assumptions behind the method and sampling regime. Some idea of the rigour of the method and the need to make changes in the design or techniques for collecting or analysing the data can be obtained at this stage. Specialist field equipment should be tested and, if necessary, modified based on practical experience. It is also the opportunity to assess the training needs for all staff. The means of analysing the data also require testing. If statistical analyses are being used they should be tested with data from the pilot study. The amount of time and effort required to conduct the pilot study will vary considerably depending on the hypothesis to be tested and the methods. Step 7. Collect the samples Sampling should only commence after the methods have been established and staff trained. The rigour with which sampling is undertaken can influence the success or otherwise of the monitoring. Where agreed sampling protocols are not followed all variations should be carefully documented and this documentation kept with the data. The following information should accompany all samples: date and location; names of sampling staff; method used to collect the samples; number of samples collected; equipment used to collect the samples; methods used for sample storage or transport; and all changes to the established methods. Sampling and data collection should be done in a manner to ensure the results can be used with confidence. Documentation of all practices is a vital part of demonstrating this confidence. The effectiveness of monitoring is also dependent on the timely processing of samples collected for further analysis. However, the need for rapid results should not compromise the processing of samples. If the processing is not sufficiently rapid changes 43 to the processes may be necessary. Alternatively, the programme may need reassessment. Delays in processing the samples could also negate the usefulness of the program. Step 8. Analyse the samples Many samples require analysis after they have been collected and processed. Whether this involves chemical analysis or biological identification the means of having this done should be determined at the pilot study stage. Statistical analysis is now regularly used to analyse data and ascertain the extent of any change or variation. Achieving the objective of a monitoring programme is not possible unless the data from the samples is made available for interpretation. As with sample collection a basic set of information should be documented when the samples are analysed: date and location; names of analytical staff; methods used for analysis; equipment used for analysis; means and location of storing data; all changes to the established methods; and statistical tests and significance levels. Step 9. Report the results All monitoring data and information needs to be interpreted and reported in a timely and cost-effective manner. If this is not done the program can be considered to have failed. The interpretation should take place within the framework provided by the program objective. Making the reporting schedule and the reports publicly available is one way of ensuring that this is given due attention. Reporting can take many forms. The form of the report will, in part, be determined by the nature of the problem and the monitoring objectives. Its express purpose is to ensure the monitoring data become part of the management planning process. In many instances it will also be useful for the report to comment on the need for further monitoring of the same nature or even of a different nature. The size and style of a report will vary according to the objective, the method used and the audience. In general the report should be succinct and concise and supported by statistical analyses and indicate whether or not the hypothesis has been supported and whether management action is required. Step 10. Evaluate the project The framework given in Figure 17 provides a series of steps that feedback into the planning process. Throughout the planning and implementation process for monitoring these feedback steps should be used to ensure that the required rigour is being obtained and that the hypothesis can be tested by the data being collected. At the end of the program, or after a predetermined time period the entire process should be re-examined and necessary modifications made and recorded. As with inventory and assessment the extent of resources available for wetland monitoring will vary and may provide limits on the program. The availability of resources can be an important constraint on many programs; when this occurs care needs to be taken to ensure that the monitoring can meet its objective. In some cases a simple monitoring program could still provide valuable information to guide managers in making decisions. Tables 5-7 provide an analysis of simple monitoring approaches for the three sites investigated in the Limpopo. These examples illustrate that monitoring need not be complex – the key issue being whether or not the program can be used to test the hypothesis being addressed. 44 The examples shown in Tables 5-7 also show the relationship between assessment and monitoring with the latter providing information that can be used to reassess the former. Monitoring can also identify a need for further inventory information. Monitoring is also a very valuable information source and can play an important role when managers engage with local communities and wider stakeholders. The value of a monitoring program though is shown mainly by its usefulness for making management decisions about specific pressures on the wetland. The overall purpose of monitoring is to determine the extent of change in the ecological character of a wetland, based on a hypothesis and objective that are derived from an assessment of the major pressures or threats facing the wetland. In most instances the monitoring will be undertaken through measuring an indicator of change in the wetland. The effective use of indicators provides managers the opportunity to determine whether intervention or further investigation is needed well before adverse change occurs – thus the earlier the signal the more time to formulate management responses. Based on van Dam et al. (1999) an early warning indicator is defined as: a measurable biological, physical or chemical response to a particular stress, preceding the occurrence of potentially significant adverse effects on the system of interest. While ‘early warning’ may not necessarily provide firm evidence of larger scale degradation, it should provide an opportunity to determine whether further investigation is needed. The attributes of an effective early warning indicator are shown in Box 4. However, an indicator possessing all the attributes for early warning is unlikely to exist; for example, the ‘diagnostic’ and ‘broadly applicable’ attributes are mutually exclusive, with their relative importance being related to the nature of the problem and early warning required. Box 4: Attributes of an effective early warning indicator for wetland monitoring The choice of an early warning indicator should be guided by considering the following attributes (van Dam et al. 1999), noting that not are applicable in all instances and the importance of any one is related to the circumstances and issues being investigated. i. anticipatory: provide an indication of adverse change before serious harm has occurred; ii. sensitive: detect low levels, or early stages of change; iii. diagnostic: sufficiently specific to provide confidence in identifying the cause; iv. broadly applicable: to a broad range of causes; vi. timely: provide information quickly enough to initiate management prior to impacts; vii. cost-effective: while providing the maximum amount of information per unit effort; viii. regionally relevant: to the ecosystem being assessed; ix. socially relevant: obvious value to and observable by stakeholders; x. easy to measure: using a standard procedure with known reliability and low error; 45 xi. constant: capable of detecting small changes, and clearly distinguishing the source; and xii. non-destructive: to the ecosystem being assessed. Inclusion of an early warning indicator implies a precautionary approach to management, i.e. intervention will be undertaken before real and important changes have occurred. As the success of an indicator could be lost if the responses to the warning are inadequate, a management process should be in place with procedures for ensuring the early warning is acted upon and the outcomes monitored to check that they are achieving the desired effect. This process of acting on early warnings, cheeking the results and modifying project objectives and indicators in responses to new findings constitutes a strategic adaptive management approach such as that shown in Figure 18. Figure 18: A framework for adaptive management (from Pollard & du Toit 2007, adapted from Biggs & Rogers 2003) 46 Table 5: Assessment and Monitoring for Intunjambili Wetland, Zimbabwe Compiled by: Alleta Nenguke1 and Edward Chuma2 1Environmental Management Agency P.O Box CY 385 Causeway Harare, Zimbabwe 2University of Zimbabwe, Department of Soil Science, P.O. Box MP 167 Mount Pleasant, Harare, Zimbabwe Identify main Outline the cause of Describe what part Outline how the What management action can Describe what monitoring threats/issues the threat of the wetland is assessment was be taken is in place or proposed – in known under threat – which done – what tools what indicator is being priority order components or or processes were used; what is the processes or used threshold when further services, and where action will be taken? Change in water • Total area under Amount of water Semi structured Maintain current number of plot • Wetland flow measured regimes irrigation available for interviews with key holders using a v-notch weir – area • Artificial drains & downstream users and informants Operational & effective localized under irrigation reduced erosion gullies middle section of Personal environmental conservation when water level falls below • Non-water conserving wetland observations committees working in liaison with notch on weir. farm practices Participatory Rural relevant government structures • Two-day lag period enforced Appraisals (PRAs) -Capacity building on water - cultivation allowed only conservation farming practices e.g two days after rains and mulching after runoff recorded. Soil erosion • Brick moulding Siltation of weir & dam Surveys: Reclamation of disused pits • “Dead levels” checks in dam • Ploughing down slope -Assessed degree of Practical delivery systems such as (level below which water in some plots deposition at entry (i) equipping farmers with skills from a dam cannot be • Artificial channels dug points of gullies or for sustainable farming practices utilized). from main stream artificial drains; fan- (ii) Use Traditional Leaders Act for • Sample undissolved • Stream bank cultivation like lobes of clastic guidance on good governance. sediment in weir, stream sediments (mineral and dam. particles) assessed • Too many cattle during Mid-section Qualitative measures Controlled grazing • Stocking rate of region = 8 dry season such as cover Limit number of cattle from outer livestock units per hectare abundance catchment. (Gammon 1983) Limited resource • Population increase Upper and middle Participatory Rural Strict record keeping of • Number of plots in wetland base capacity to • Further subdivision of section. Appraisals[PRAs] beneficiaries by project leadership • Encroachment of existing 47 Identify main Outline the cause of Describe what part Outline how the What management action can Describe what monitoring threats/issues the threat of the wetland is assessment was be taken is in place or proposed – in known under threat – which done – what tools what indicator is being priority order components or or processes were used; what is the processes or used threshold when further services, and where action will be taken? accommodate existing plots to friends team. plots further into wetland new and relatives. beneficiaries Reduction in soil • All year round cropping Upper and lower Semi structured Design and implement a fertility • Yields per hectare. If yield fertility of same crop – section questionnaires management program that (a) of maize drops below monoculture minimizes pollution of wetland average yield of a and (b) minimize loss of carbon communal farmer [0.8 - 1.4 and builds up organic matter t/ha] assuming a steady Diversification to crops in demand supply of moisture, fertility needs to be improved. Over-fishing • Overpopulation Fish and invertebrates Measuring depth of Ploughing across slopes • Water levels in dam - catch in downstream areas dam to establish Cultivation 30m from highest and size of fish per day. If siltation levels flood level size of fish reduces use of Questionnaires fishing rods to be encouraged. 48 Table 6: Assessment and Monitoring for GaMampa wetland, Mohlapitsi River Catchment, South Africa Compiled by: Mutsa Masiyandima International Water Management Institute, Pretoria Identify main Outline the cause the of threat Describe what part of Outline how the What management action can Describe what monitoring is in place threats/issues in the wetland is under assessment was done be taken or proposed – what indicator is being known priority threat – which – what tools or used; what is the threshold when order components or processes were used further action will be taken? processes or services, and where Change in water • Drainage of agricultural plots Services: Field observations - Manage / control crop water • Water flow is being measured using levels • Abstraction of water for - water availability to Shallow groundwater use a v-notch weir – water can not be supplementary irrigation crops in wetland measurements in - Limit water use for cropping used for irrigation when flow drops • Abstraction by crops in situ - water availability to wetland below notch in weir downstream fields • Introduce water conservation used for cropping measures e.g. mulching, no drainage Erosion of stream • Too many cattle in mid- Fish and invertebrates - Reduce number of cattle and • Sediment transport being sampled in banks section of wetland in downstream areas goats to reduce erosion stream • Number of stock in streams counted weekly – reduce number of cattle in response to continued erosion Loss of natural • Too many people cutting reeds Reed beds being - Limit number of people • Map distribution of reeds and vegetation (reeds reduced cutting to reeds bulrush per season. Estimate: and sedges being - Reduce are of cutting in reeds • growth dynamics cut) - Control the cutting of reeds • demand for resources. - Have “reed cutting seasons” • Test time and place ban of cutting. • Test limit of % of cut. Loss of organic • Burning peat – a result of Most of wetland No assessment made; Promote alternative ways of • No monitoring of burning is in matter wetland users burning reeds to but observations of clearing areas that are used for place. Monitoring of the re- clear the land for cropping. burning peat were cropping to avoid burning the emergence of the reed beds could be Peat burns as a secondary made peat and losing some of made. problem. inherent fertility 49 Table 7: Assessment and monitoring for Missavene wetland, Mozambique Compiled by: Salomao Bandeira1 and Dinis Juizo2 1 Dept of Biological Sciences, Universidade Eduardo Mondhlane, Maputo, Mozambique 2Facultry of Engineering, Universidade Eduardo Mondhlane, Maputo, Mozambique Identify main Outline the cause Describe what part Outline how the What management Describe what monitoring is in threats/issues in the of threat of the wetland is assessment was action can be taken place or proposed – what known priority under threat – which done – what tools indicator is being used; what is order components or or processes were the threshold when further processes or used action will be taken? services, and where Reeds (Phragmites • Too many people Extent of reed and Observation Limit number of people • Map distribution of reeds and muaritianus) and cutting reeds bulrush being reduced. harvesting resources. bulrush per season. Estimate: bulrush (Typha (building toilets) and Activity unsustainable Limit period of cutting to • growth dynamics capensis) being cut bulrush (for mats and given the amount allow re-growth. • demand for resources. floating boats) present and growth Allow cutting in half of reed • Test time and place ban of dynamics areas only cutting. • Test limit of % of cut. Fires • Extensive and Fauna and plant Observation Sensitize community to • Document and evaluate wildfire uncontrolled diversity reduced. good and bad practices viz. frequency, causes and effects. coverage of fire fire. • Ban some of wildfire causes. Consider ban on fire for • Propose guidelines on how to small area of Missavane manage fires. wetland Increased number • Made at expense of Native species being Observation and Consider increasing crop • None so far of fields for crop natural vegetation reduced. Area with comparison yield per area instead of production and species (some native vegetation also increasing crop area. nearly extinct) reduced Reduction of • Too many cattle in Grassland Observation and Land use planning to • First understand from users/ grassland habitat wetlands. comparison consider area for agriculture stakeholders which areas they • Expansion of development and for cattle would prefer for agriculture, cropping into pasture pasture and other development. 50 Identify main Outline the cause Describe what part Outline how the What management Describe what monitoring is in threats/issues in the of threat of the wetland is assessment was action can be taken place or proposed – what known priority under threat – which done – what tools indicator is being used; what is order components or or processes were the threshold when further processes or used action will be taken? services, and where grasslands 51 4. Concluding remarks The framework presented in this report for undertaking wetland inventory, assessment and monitoring of wetlands in the Limpopo basin in southern Africa provides an outline of approaches and lists key references and source materials. It does not provide a recipe for inventory, assessment and monitoring – rather it contains information and guidance for making decisions about what inventory, assessment and monitoring is required in response to the main uses and (anticipated) management issues at identified wetlands, whether at the local or basin-scale. It provides information to support managers make decisions about their data/information needs to support the sustainable use of their wetlands; it is directed at decision-makers in government agencies. Wetlands in the Limpopo Basin are an important component of the landscape with their most visible characteristic being the abundance of water they hold during the dry season, when compared to the surrounding catchment. As such they represent an important water and agricultural resource in a landscape that is characterised by variable rainfall, seasonal water flows and climatic extremes. This represents a valuable feature of the landscape – the frameworks for inventory, assessment and monitoring are presented to support decision-makers obtain the information they require to use these features in a sustainable manner. The framework does not provide the outcomes for specific management issues – these are made by the decision-makers based on the best available information and wisdom. The inter-linkage between inventory, assessment and monitoring – too often they are treated as separate and disjoint initiatives. The arrows in the various diagrams that accompany the framework illustrate the circular nature of the relationships within the framework. For effective planning and implementation of data collation/collecting it is necessary to recognise that inventory, assessment and monitoring are interconnected. Thus, inventory provides information for assessment which provides information for monitoring with feedback to inventory and/or assessment At a global scale the extent and quality of information for wetland management has been considered to be insufficient to support effective management of a dwindling but valuable resource. The above described Framework is part of a global effort led by the Ramsar Convention on Wetlands to support efforts to overcome such problems. The Convention has made available a series of Handbooks for the Wise Use of Wetlands (http://ramsar.org/lib/lib_handbooks2006_e.htm) – these not only cover inventory, assessment and monitoring, but also the important topics of community engagement and involvement in all aspects of wetland management as well as capacity building, communication and public awareness. A successful data/information program will require an investment in community engagement and involvement and capacity building. In particular it is recommended that the framework is supported by a capacity-building program focusing specifically on the practicalities of assessing and monitoring wetlands in the Limpopo (and potentially elsewhere) with an emphasis on approaches that can be readily undertaken and provide early warning of possible adverse change. This program could include training and awareness raising components based on user needs related to 52 inventory, assessment and monitoring and how to consider wetland issues at multiple scales from local site to basin-wide. 5. Acknowledgments This report was developed with the support, advice and guidance of many people involved in the Challenge Program for Water and Food project “Wetlands-based livelihoods in the Limpopo basin: balancing social welfare and environmental security” being coordinated by the International Water Management Institute (IWMI). The project was structured to enable the involvement of experts from across the Limpopo Basin and adjacent regions and supported by invited technical experts from elsewhere. The project was ambitious and addressed a necessary topic – thanks are extended to all those who contributed and enabled the project to advance and address the real on-ground needs for wetland management. The project was coordinated by Dr Mutsa Masiyandima from IWMI in Pretoria and ably supported by her many colleagues within IWMI and from other institutions. Particular thanks are extended to Mutsa as well as Matthew McCartney and Lisa-Maria Rebelo from IWMI for their inputs and assistance at different stages of the projects. All contributors to this particular report are warmly acknowledged. ************************************ ************************************ 6. References Alwang, J., Siegel, P.B. and Jorgensen, S.L. 2001. Vulnerability: A View From Different Disciplines. Social Protection Discussion Paper Series No. 0115, The World Bank, Washington D.C. http://www.worldbank.org/sp. Bandeira, S., Massingue Manjate, A. and Filipe, O. 2006. An ecological assessment of the health of the Chibuto wetland in the dry season, Mozambique - emphasis on resources assessment, utilization and sustainability analysis. Unpublished report to Challenge Program for Water and Food, International Water Management Institute, Pretoria, South Africa. 53 Biggs, H. C., and Rogers, K.H. 2003. An adaptive sustem to link science, monitoring and management in practice. In: J. T. du Toit, K. H. Rogers and H. C. Biggs (eds). The Kruger National Park: Experiences of Savannah Heterogeneity. Island Press, Washington D.C. pp. 59-80. Chenje, M., Sola, L., Palecnzy, D. (eds) 1998. The State of the Zimbabwe`s Environment 1998. Government of the Republic of Zimbabwe, Ministry of Mines, Environment and Tourism, Harare, Zimbabwe. Chiron, D. 2005. Impact of the small-scale irrigated sector on household revenues of the black community of Ga Mampa valley (ward of Mafefe). Contribution to the irrigation management transfer study of the small-scale irrigation schemes. Limpopo province - South Africa. Master of Science thesis, Social Management of the Water and Tropical Agricultural Development. Centre National d’Etudes Agronomiques en Régions Chaudes, Montpellier, France. Costa, L.T., Farinha, J.C., Tomàs Vives, P. and Hecker, N. 1996. Mediterranean wetland inventory: A reference manual. MedWet Publication. Instituto da Conservacao da Naturez, Lisboa, and Wetlands International, Slimbridge, UK. Costa ,L.T., Farinha, J.C., Tomas Vives, P., Hecker, N. and Silva, E.P. 2002. Regional wetland inventory approaches: the Mediterranean example. In: C.M. Finlayson, N.C. Davidson and N.J. Stevenson (eds). Wetland Inventory, Assessment and Monitoring: Practical Techniques and Identification of Major Issues. Supervising Scientist Report 161, Supervising Scientist, Darwin, Australia. pp. 41–47. Davidson, N.C. and Finlayson, C.M. 2007. Developing tools for wetland management: inventory, assessment and monitoring – gaps and the application of satellite-based radar. Aquatic Conservation: Marine and Freshwater Ecosystems 17, 219-228. Drummond, R. 1981. Common Trees of the Central Watershed Woodlands of Zimbabwe, Department of Natural Resources, Harare, Zimbabwe. FAO/UNESCO/ISRIC 1988. Soil Map of the World (Revised Legend) FAO World Soil Resources Report no. 60, FAO, UNESCO and ISRIC, Rome. 119 pp. Finlayson, C.M. 1996a. Framework for designing a monitoring programme. In: P. Tomas Vives (ed.). Monitoring Mediterranean Wetlands: A Methodological Guide, MedWet Publication, Wetlands International, Slimbridge, United Kingdom and ICN, Lisbon. pp. 25-34. Finlayson, C.M. 1996b. The Montreux Record: a mechanism for supporting the wise use of wetlands. Proceedings of the 6th Meeting of the Conference of the Contracting Parties of the Convention on Wetlands (Ramsar Convention Bureau, Gland, Switzerland). Technical Sessions: Reports and presentations, Brisbane, Australia. Vol. 10/12 B. pp. 32-37. Finlayson, C.M. 2005. Summary report from reconnaissance visit to Limpopo basin 12- 19 August 2005. Unpublished report to Challenge Program for Water and Food, International Water Management Institute, Pretoria, South Africa. 54 Finlayson, C.M. and Spiers, A.G. (eds) 1999. Global Review of Wetland Resources and Priorities for Wetland Inventory. Supervising Scientist Report 144, Supervising Scientist, Canberra. Finlayson, C.M. and van der Valk, A.G. (eds) 1995. Classification and Inventory of the World’s Wetlands. Advances in Vegetation Science 16, Kluwer Academic Publishers, Dordrecht, The Netherlands. Finlayson, C.M., Davidson, N.C. and Stevenson, N.J. 2001. Wetland inventory, assessment and monitoring – practical techniques and identification of major issues: summary. In: C.M. Finlayson, N.C. Davidson & N.J. Stevenson (eds), Wetland Inventory, Assessment and Monitoring: Practical Techniques and Identification of Major Issues, Supervising Scientist Report 161, Jabiru, Australia, pp. 130-136. Finlayson, C.M., D’Cruz, R. and Davidson, N.C. (co-chairs) 2005. Ecosystems and Human Well-being: Wetlands and Water Synthesis. Millennium Ecosystem Assessment. World Resources Institute, Washington, DC. Finlayson, C.M., Davidson, N.C., Spiers, A.G. and Stevenson, N.J. 1999. Global wetland inventory: Status and priorities. Marine and Freshwater Research 50, 717-727. Finlayson, C.M., Begg, G.W., Howes, J., Davies, J., Tagi, K. and Lowry, J. 2002. A Manual for an Inventory of Asian Wetlands (version 1.0). Wetlands International Global Series 10, Wetlands International, Kuala Lumpur, Malaysia. 72 pp. Ginn, P. & SCNVYO 1973. Birds Afield. Longmans, Salisbury, Rhodesia. Ginn, P. & SCNVYO 1974. Birds of the Lowveld. Longmans, Salisbury, Rhodesia. Gitay, H, Finlayson, C.M. and Davidson, N.C. 2009. Guidance on methods for vulnerability assessment of wetlands to change in ecological character. Ramsar Technical Report, Gland, Switzerland (in press). IPCC 2001. Climate Change 2001: Synthesis Report. A Contribution of Working Groups I, II, and III to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom, and New York, USA, 398 pp. Juizo, D., M. Masiyandima, N. Giva, S. Morardet, E. Chuma, M. Finlayson and W. Kulawardhana 2007. Wetlands-based livelihoods in the Limpopo basin: balancing social welfare and environmental security. Research site profile: Chibuto Wetland. Project report Kotze, D. 2005. An ecological assessment of the health of the Mohlapetsi wetland, Limpopo Province. Unpublished technical note to Challenge Program for Water and Food, International Water Management Institute, Pretoria, South Africa. Kulawardhana, R. W., Thenkabail, P. S., Vithanage, J., Biradar, C., Islam Md. A.,Gunasinghe, S., Alankara, R.2007. Evaluation of the wetland mapping methods using Landsat ETM+ and SRTM Data. Journal of Spatial Hydrology 7, 62-96. MacKay, H., Finlayson, C.M., Fernandez-Prieto, D., Davidson, N., Pritchard, D. and Rebelo, L-M. 2009. The role of Earth Observation (EO) technologies in supporting 55 implementation of the Ramsar Convention on Wetlands. Journal of Environmental Monitoring (in press). Marnewick, G.C. and Batchelor, A.L. 2002. Wetland Classification, mapping and inventory. In: R.W. Palmer, J. Turpie, G.C. Marneweck and A.L. Batchelor (eds). Ecological and Economic Evaluation of Wetland in the Upper Olifants River Catchment, South Africa. WRC Report No. 1162/1/02. Water Research Commission, Pretoria, South Africa. Masiyandima, M., M. McCartney, Jean-Marie Fritsch and D. Rollin.2006. Impacts of agricultural use of the GaMampa wetland on the hydrology of the wetland and the Mohlapetsi River. Proceedings of the 7th WATERNET/WARFSA/GWP-SA Symposium “Mainstreaming IWRM in the Development Process”, Lilongwe, Malawi, 1-3 November 2006. Massingue Manjate, A and S. Bandeira. Resources assessment and utilization of Chibuto wetland, Changana sub-basin, Mozambique. Presented at the 8th WaterNet/WARFSA/GWP-SA Symposium 31 Oct – 2 Nov 2007, Lusaka, Zambia McCartney, M 2006. Technical Note: Hydrology of the Mohlapitsi catchment. Unpublished report to Challenge Program for Water and Food, International Water Management Institute, Pretoria, South Africa. Motsi, K. 2008. Hydrology of the Intunjambili catchment. Unpublished technical note to Challenge Program for Water and Food, International Water Management Institute, Pretoria, South Africa. Mtambanengwe F. 2006. Ecological monitoring framework for Intunjambili wetland Zimbabwe. Unpublished technical note to Challenge Program for Water and Food, International Water Management Institute, Pretoria, South Africa. Munguambe, P., Chilundo, M., Juízo, D., Moniz, S., Ndeve, A. and Saimone, F. 2007. Too wet or too dry for cropping in the wetlands “Farmers’ water management practices in the Chibuto small inland wetland, Mozambique” Changane Sub- Catchment “Challenge Program on Water & Food – PN 30”Project report Namburette, Fernando Everisto. 2005. Aspectos de Gestao de Recursos Hidricos em Peqquenas Bacias ((Water Resource management Aspect of Small Catchments). Unpublished undergraduate research project, University Eduardo Mondlane, Mozambique. Nell, J.P. and Dreyer, J.G. 2005. Soil survey for Mashushu, fertilis, Vallis, Canyon and Gemini Irrigation Schemes. Limpopo Department of Agriculture Resis Programme, Internal Report No. GW/2005/55, Polokwane, South Africa. Nivet, C. and Frazier, S. 2004. A Review of European Wetland Inventory Information. Wetlands International, Wageningen, and Ministry of Transport, Public Works and Water Management, Lelystad, The Netherlands. Owen, R., Verbeek, K., Jackson, J. and Steenhuis, T. (eds) 1992. Dambo farming in Zimbabwe (Water management, cropping and soil potentials for smallholder farming in the wetland). Conference Proceedings, 8-10 September, Harare, 56 Zimbabwe, Department of Geology, Soil Science, Agricultural Economics and Crop Science, University of Zimbabwe. Pollard, S. R., and Cousins, T. 2008. Towards integrating community-based governance of water resources with the statutory frameworks for Integrated Water Resources Management:: A review of community–based governance of freshwater resources in four southern African countries to inform governance arrangements of communal wetlands. Water Research Commission, Pretoria. WRC Report TT.328/08. Pollard, S. R., and du Toit, D. 2007. Recognizing heterogeneity and variability as key characteristics of savannah systems: The use of Strategic Adaptive Management as an approach to river management within the Kruger National Park, South Africa. . Report of UNEP/GEF Project No. GF/ 2713-03-4679, Ecosystems, Protected Areas and People Project. Pollard, S. R., Kotze,D. and Ferrari, G. 2007. Valuation of the livelihood benefits of structural rehabilitation interventions in the Manalana wetland, Craigieburn village, Mpumalanga Province. WRC Wetlands Project, Pretoria, South Africa. Pollard, S. R., Kotze, D., Ellery, W., Cousins, T., Monareng, J. and Jewitt, G. 2005. Linking water and livelihoods: The development of an integrated wetland rehabilitation plan in the communal areas of the Sand River Catchment as a test case. AWARD (on behalf of the Save the Sand Programme), Report to Warfsa/ Working for Wetlands. Ramsar Convention Secretariat, 2006a. Coastal management: Wetland issues in Integrated Coastal Zone Management. Ramsar handbooks for the wise use of wetlands, 3rd edition, vol. 10. Ramsar Convention Secretariat, Gland, Switzerland. Ramsar Convention Secretariat, 2006b. River basin management: Integrating wetland conservation and wise use into river basin management. Ramsar handbooks for the wise use of wetlands, 3rd edition, vol. 7. Ramsar Convention Secretariat, Gland, Switzerland. Ramsar Convention Secretariat, 2007a. Inventory, Assessment, and Monitoring: An Integrated Framework for Wetland Inventory, Assessment, and Monitoring. Ramsar Handbooks for the Wise Use of Wetlands, 3rd edition, vol. 11. Ramsar Convention Secretariat, Gland, Switzerland. Ramsar Convention Secretariat, 2007b. Wetland Inventory: A Ramsar Framework for WetlandIinventory. Ramsar Handbooks for the Wise Use of Wetlands, 3rd edition, vol. 12. Ramsar Convention Secretariat, Gland, Switzerland. Ramsar Convention Secretariat, 2007c. Managing Wetlands: Frameworks for Managing Wetlands of International Importance and Other Wetland Sites. Ramsar Handbooks for the Wise Use of Wetlands, 3rd edition, vol. 16. Ramsar Convention Secretariat, Gland, Switzerland. 57 Rebelo, L-M., Finlayson, CM & Nagabhatla, N. 2009. Remote sensing and GIS for wetland inventory, mapping and change analysis. Journal of Environmental Management (in press). Rosenqvist, A., Finlayson, C.M., Lowry, J. and Taylor, D.M. 2007. The potential of spaceborne (L-band) radar to support wetland applications and the Ramsar Convention. Aquatic Conservation: Marine and Freshwater Ecosystems 17, 229- 244. Sarron 2005. Effects of wetland degradation on the hydrological regime of a quaternary catchment. Mohlapitse River, GaMampa valley, Limpopo Province, South Africa. Master of Science thesis. Ecole Nationale Supérieure Agronomique de Rennes, Rennes, France. Stevenson, N. and Frazier, S. 1999. Review of wetland inventory information for Africa. In: C.M. Finlayson and A.G. Spiers (eds). Global Review of Wetland Resources and Priorities for Wetland Inventory. Supervising Scientist Report 144, Jabiru, Australia, pp. 105-200. Surveyor General 1980. Zimbabwe maps, 1:1000 000 natural regions and farming areas first edition. Surveyor General, Harare, Zimbabwe. Taylor, A.R.D., Howard, G.W. and Begg, G.W. 1995. Developing wetland inventories in southern Africa: a review. In: C.M. Finlayson and A.G. van der Valk (eds). Classification and Inventory of the World’s Wetlands. Advances in Vegetation Science 16, Kluwer Academic Publishers, Dordrecht, pp. 57-79. Thieme, M.L., Abell, R., Stiassny, M.L.J., Skelton, P., Lehner, B., Teugels, G.C., Dinerstein, E., Kamden Tohan, A., Burgess, N. and Olson D. 2005. Freshwater Ecoregions of Africa and Madagascar. Island Press, Washington D.C. Tomas Vives, P. (ed.) 1996. Monitoring Mediterranean Wetlands: a Methodological Guide. MedWet Publication, Wetlands International, Slimbridge, UK and ICB, Lisbon, Portugal. van Dam, R.A., Finlayson, C.M. and Humphrey, C.L. 1999. Wetland risk assessment: a framework and methods for predicting and assessing change in ecological character. In C.M. Finlayson and A.G. Spiers (eds). Techniques for Enhanced Wetland Inventory, Assessment and Monitoring. Supervising Scientist Report 147, Supervising Scientist Group, Canberra. pp. 83-118. Wild, H. and Barbosa, L.A.G. 1967. Vegetation Map of the Flora Zambesiaca area. Flora Zambesiaca: Descriptive Memoir, Supplement 1-71. Collins (Pvt) Ltd, Salisbury, Rhodesia. 58 Annex 1: Inventory methods Standardized inventory methods have been developed and successfully used in different circumstances in different countries or regions, although there are many similarities and lessons that transcend local circumstances. Amongst the many methods with instructive guidance and lessons for application in southern Africa are the Mediterranean Wetlands Initiative (MedWet) inventory, the United States Fish and Wildlife Service national wetland inventory, the Ugandan national wetland inventory, and the Asian wetland inventory. The main features of these examples are summarised below in terms of each of the 13 steps in the framework for wetland inventory described above (see Box 1). These examples illustrate existing methods, but also demonstrate differences in approaches that could be used in different locations, for different purposes, and at different scales; the need for different methods that enable local and national needs is illustrated by the range of examples below. Mediterranean Wetlands Initiative (MedWet) inventory This is a standardised, but flexible method, including a database for data management, for inventory in the Mediterranean region. Although not intended as a pan-Mediterranean wetland inventory, it has provided a common approach that has been adopted, and adapted, for use in several Mediterranean countries and elsewhere. 59 United States national wetland inventory This is a long running national program that has developed a classification and methodology for producing a map-based inventory at a national scale and building on more localised sampling and data collection. 60 Uganda National Wetlands Programme The inventory is a component of an ongoing National Wetlands Program. It is largely undertaken at the local level, using standard formats, and includes a training component. 61 Asian Wetland Inventory (AWI) This approach has been developed in response to the recommendations contained in the Global Review of Wetland Resources and Priorities for Wetland Inventory (Finlayson et al 1999). The method is a hierarchy that can be implemented at four spatial scales. 62 63 Annex 2: Example of a semi-quantitative wetland risk assessment A risk assessment of the weed Mimosa pigra in northern Australia Mimosa pigra (mimosa) is a pan-tropical shrub that has been present in northern Australia for more than 100 years. In the past few decades it has become a major weed. It is also spreading in many parts of Africa and potentially is a major weed in wetlands in southern Africa with extensive and problematic infestation already occurring in Gorongosa National Park, Mozambique and in Lochinvar National Park, Zambia. This example is based on information from northern Australia, an area with many biogeographical similarities to southern Africa. The information presented below was collated through a formal risk assessment (Walden et al 2004) undertaken using the method recommended by the Ramsar Convention and described in this report. Identification of the problem Mimosa has many traits that make it a successful weed, including adventitious roots that allow growth in a variety of conditions, rapid growth and maturation, a high production rate of easily dispersed and long-lived seeds, and low nutrient requirements. It has few natural enemies in its introduced range and is able to rapidly colonise wetlands, forming a near mono-specific shrubland. Control is expensive and hampered by the size, inaccessibility and often the remoteness of infestations. See Figure A2.1 for further information on the biological features of the plant. The potential effects of mimosa It dominates many floodplain habitats and converts a range of vegetation types into a homogeneous shrubland, reducing the diversity of flora and fauna. It can alter the water regime and provides ideal habitat and cover for feral pigs, decreasing the capacity to manage these pests. It can smother pasture grasses and increase costs on pastoral and agricultural enterprises. It is widely seen as aesthetically unpleasing and a threat to tourism and recreational and commercial fishing. The potential extent of mimosa in northern Australia Mimosa currently infests about 80,000 hectares of coastal floodplains in the northernmost part of Australia. Methods of seed dispersal include wind, water, animals, human clothing, vehicles, boats and machinery, adhering mud, waterbirds, and deliberate movements of seed contaminated earth or plant/seed material by humans. It prefers seasonally inundated floodplains and can readily establish in disturbed areas. It is estimated that around 4.6 million hectares of wetlands are potentially at risk from mimosa infestation. Uncertainty, information gaps and further research Additional information would benefit the management of mimosa: site-specific assessments of wetlands at risk, including mapping and remote sensing of wetlands at risk; more precise data on growth and environmental requirements; what factors affect successful revegetation; and the relationship between biological control agents and native species. Other areas of beneficial research include: obtaining quantitative data on the environmental, economic, social and cultural impacts and the ecological impacts of herbicides used in control programs. Management implications In the past few decades, a great deal has been learned about the management of mimosa. As it posed a major threat in northern Australia a strategic management plan was developed using 64 a coordinated and collaborative approach. The plan had four main programs to: i) inform and educate stakeholders and the community; ii) prevent the plant from spreading to and impacting on new areas; iii) further develop the knowledge base and methods for effective and efficient management of mimosa; and iv) reduce the current adverse impacts of mimosa infestations. Preventative management emphasised managing entire wetland plant communities to try and reduce susceptibility to mimosa invasion. Information from: Walden, D., van Dam, R., Finlayson, M., Storrs, M., Lowry, J. & Kriticos, D. 2004. A risk assessment of the tropical wetland weed Mimosa pigra in northern Australia. Supervising Scientist Report 177, Supervising Scientist, Darwin N.T. Figure A2.1: Mimosa pigra (a) Adult plant, (b) flower heads, and (c) young and (d) mature seed pods (Photographs – CM Finlayson) 65