Being at the marine coastal margin, lagoons constitute a finely balanced ecotonal environment. Here, the ecological effects of small inter-annual fluctuations in salinity and other environmental variables can be exacerbated by land-use activities and climate change and transformed into major ecosystem change trends. These trends are now underway in virtually all of North Africa's coastal aquatic ecosystems. Recognising, measuring and modelling these changes depends upon establishing a reliable and spatially coherent time-series of hydroecological and biogeochemical data over a period of more than one year. Three key sites have been selected for the intensive task of implementing full monitoring and modelling exercises (Figure 1); secondary sites will be used to provide reference conditions. These and associated data will be incorporated into site specific dynamic models of lagoon function. In order to achieve this basic aim, the skills and co-operation of research scientists in five partner countries is required together with several important sub-contracts to specialists elsewhere. During the prosecution of the project skills transfer in monitoring, modelling and remote sensing techniques will take place between the European and North African groups and within the North African groups. In the final phase of the project the models will be tested and demonstrated to management agencies in each North African country.

Figure 1. Locations of the key and secondary network of North African lagoons to be monitored in MELMARINA (topography data: Global Land One-km Base Elevation (GLOBE) Project). Large symbols represent primary sites (monthly monitoring) reference sites and the smaller symbols of secondary sites are for specifying boundary conditions. For more detail on these sites click here.

The workplan structure includes the collection of high quality environmental data from monitoring and from spatial survey (including water and sediment quality), identification of aquatic ecosystem linkages and attributes, hydrology, establishing long term monitoring, environmental reconstruction and plant-environment interactions, data analysis and modelling with applications to integrated management. Additional work package sub-sections are included to provide for data exchange, training and quality control and information dissemination. Running in parallel with the work packages there are a series of workshops needed to initially establish common methodologies and then to check and verify on-going progress. Key workshops will be in the first, 4th, 6th, 9th and 12th quarters of the project schedule.

The group of integrated work packages and their inter-relationships that constitute the MELMARINA Project (Figure 2) reflect a considerable complexity in sampling strategies and data processing and analysis activities. Nevertheless, given the responsible and experienced co-ordination leadership and demonstrable skills base already existing in the North African groups the collaborative parties will work together synergistically. The timetable for each individual work package is indicated both in the individual WP descriptions (below) and in Figure 3. Each work package has its own outputs, which collectively contribute to the final project deliverables. Many of these deliverables feed into other work packages. The work packages will be described in an order that approximates to the work sequence.

The structure of the integrated work packages (Figure 2) is designed to meet the five specific project objectives (see section 2). The lead partner(s) for each work package is indicated. The organisation of the project takes into account the subsidiary principle so that routine common tasks are undertaken by each local group. Hence, for fieldwork, Morocco (Dr Ramdani-ISRABAT) will have overall responsibility for co-ordinating all the zooplankton/zoobenthos work within MELMARINA but will carry out water quality, phytoplankton, aquatic vegetation and hydrological measurements on the Moroccan sites only. Similarly, in Tunisia the regular sampling and hydrology will be locally carried out (FLSHS) and the fish work will be co-ordinated for all three countries (INSTM). Whilst individual MED countries will carry out hydrology, water chemistry, sediment chemistry, aquatic plants surveys, these tasks will be co-ordinated jointly by UCL and NARSS. Remote sensing will be principally undertaken by NARSS and the EU partners. Of the EU partners, UCL will provide overall project co-ordination, together with hydrological and sediment expertise plus assistance with modelling and remote sensing. DHIWE partner is responsible for the development and provision of modelling software and the co-ordination of the modelling work within the project. Work of the co-ordinating UCL will be supplemented by important sub-contracts for linking hydrodynamic models with ecological predictions and for the application of the remote sensing software developed to model water levels within inflowing rivers, streams and other channels.

Figure 2. MELMARINA Work Packages and their Inter-relationships

Overall, the baseline data collection and monitoring (WPs 1-4) are crucial for assessment of current site status (especially relevant for biological and chemical data) and for the parameterisation, calibration and validation data requirements of the subsequent modelling exercises. The precise analytical techniques for vegetation, water, plankton and fish are all routine, with the exception of some likely taxonomic problems, which will be resolved during the workshop programme. Similarly, project partners have extensive experience in the hydrological monitoring, sediment and remote sensing techniques to be employed. The WPs are summarised below.

This will comprise two main parts; firstly, collating and evaluating the existing environmental data (on hydrology, meteorology, water quality, aquatic biota and resource usage including fisheries). Secondly, an initial spatial field survey of ca. nine lagoons will be undertaken to identify within-site regions of interest and environmental change 'hot-spots'. This will enable confirmation of the three primary sites (for monitoring) and the derivation of detailed spatial representations of these site attributes. Special attention will be given to aquatic plants and the selection of sample points for hydrological/water quality monitoring (see WP3). Both primary and secondary sites will be surveyed in this WP and informational database will be generated in parallel with a contact network.

Deliverables: 1, designation of sites and collection of information necessary to begin WP 3, namely the provision of spatial information on water quality characteristics and on plankton/vegetation distributions; 2, collection of historical data to compliment analysis of remote sensing imagery since the late 1970s (WP 4); 3, compilation of calibrational and validational data for model simulations (WP 6 & 8) and entry into a standardised database for storing information about the environmental status of each site; 4, construction of a network of contacts both for facilitating execution of the project and for informing potential users (environmental groups in Europe and North Africa) of the project results and aspirations.

Milestones and Expected results: By the end of the first 6 months of the project all the primary and secondary sites will be surveyed and final site selection made. Also in this period, spatial water quality/aquatic plant mapping will enable within-site sampling points to be identified (for the 15 month routine monitoring programme, see WP 3). GIS representation for each key site will be completed by the 6th quarter and so providing primary information needed to construct databases that are a prerequisite for the remote sensing (WP 4) and environmental models (WPs 5-8).

Collection of a spatial array of short sediment cores to assess rates of sediment accumulation, infill patterns and sediment contamination. Suites of cores will be collected over a grid pattern of sampling points with measured water depths in the primary sites. At the secondary sites, single cores and a number of surface sediment samples will be collected. Sediment cores, sectioned in the field, will be selected for radiometric dating and sediment accumulation rate estimation. Surface sediment samples will be split and sub-samples used for (i) field analysis of biological macro-remains analysis by sieving (ii) laboratory analysis of lithography (water content and organic matter) and geochemical composition, (iii) mapping surface sediment characteristics.

Deliverables: 5, disclosure of spatial variation in water depth, sediment type and base-line geochemical and biological characteristics (e.g. molluscs, aquatic plants, cf. WP 5); 6, selection of the within-site monitoring locations (WP 3); 7, bathymetrical survey of the key lagoons sites for use in models (WP 6 & 8); 8, estimation of sediment accumulation rates so that sink rates for geochemicals (trace metals) and particulates can be estimated; 9, provision of a database for spatial information and sediment accumulation rates for mapping and for modelling needs (WP 5, 6 and 8).

Milestones and Expected Results: Within the first six months of the project, sediment (both spatial and temporal) field surveys of confirmed key sites will be completed. Initial depth and sediment information (lithostratigraphies and spatial sediment quality and bathymetric maps) will be available by the 4th quarter. Sediment accumulation rate data and the geochemical and biological mapping will be completed by the end of year 2. The construction of databases for information relevant to this WP and to WPs 5 & 6 (for depths, surface sediments and the cores) will also be completed by the 8th quarter.

Collection of good quality monthly monitoring data (essential for the development, calibration and validation of the dynamic models for the lagoon environments, WPs 6 & 8). These data are highly multidisciplinary, involving hydrometeorological, hydrobiological and water chemistry measurements at fixed points along spatial environmental gradients (selected in WPs 1 & 2) for each primary site. More frequent measurements of key hydrological parameters will be made by local observers and through the installation of logging equipment. Because of the intensity of monitoring required only one key primary site is designated in each of the three North Africa countries. Following initial grid-based spatial surveys of site characteristics (WP1 & 2), the chemical data (salinity, nutrients and common cations) will be acquired on a monthly basis from surface and sub-surface water samples (where depth is adequate). Biological attributes will include aquatic plants, plankton, fish and birds. Plankton will be collected monthly while transects of aquatic vegetation and fish catches will be investigated seasonally. Secondary sites will not be monitored but will be surveyed on a second occasion (in addition to that in WP1) but in a different season to provide reference conditions for the primary lagoons. It is envisaged that each pair of secondary sites will be selected to provide logistic alternative states (water quality and biota) for each primary site.

Deliverables: 10, monthly time-series data for water quality and plankton at several locations within each key lagoon; 11, water levels within the lagoons and water level/discharge data for principal outflow/inflows for use as input, calibration and validation data within the modelling work (WPs 6 & 8); 12, ground truth data on the spatial and seasonal distribution of aquatic vegetation for (i) use in the remote sensing work package (WP 4) and follows on from that established in WP 1, and for (ii) analysis of environmental controls (WP 5); 13, fisheries status with special reference to commercial fish species and assessment of water bird numbers; 14, hydrological, hydrochemical and biological data from the secondary sites for use in establishing the controls/reference conditions for wetland vegetation (WP 5 & 6).

Milestones and Expected Results: Using jointly agreed techniques established at the initial workshops, all multidisciplinary data collected from the monitoring programmes will be basically complete by the 7th quarter, and subsequent analyses of these data will occur in the following 6 months. Following checks these data will be entered into Excel spread sheets and disseminated between partner groups and made available for the main data processing parts of the project (WPs 4-8). Data for use in the modelling work (WPs 6 and 7) will also be converted into the required formats required by the dynamic model. The collation of bird information and assessment of relative fisheries performance at the key sites will, following incorporation of information from WP 1, be complete by the 10th quarter.

The application of remote sensing techniques to capture the full spatial scale of variations in aquatic vegetation cover and in open water area and some aspects of water quality. The ground truth data (especially aquatic vegetation and water level/quality) available from the field surveys (WP 1 & 3) will be employed to supervise the classification and calibration of Landsat TM and other images of the primary sites. Furthermore, use of archive remotely sensed imagery over the past 20 years will help determine recent trends in site characteristics including the extents of open water, geomorphologic change (surrounding land-use changes, reclamation and channel engineering effects) and vegetation cover. Contemporary imagery will be used within the SHYLOC software to establish relationships between observed water levels in distinct channels (rivers, streams, drainage ditches etc.) and satellite derived wet widths. These will in turn enable the derivation of water level records for these locations using the historical archive of imagery. The secondary sites will be assessed only for base-line status.

Deliverables: 15, development, testing and use of methodologies for deriving important ecological information (vegetation cover, open water, water levels) from remote sensing data; 16, establishing contemporary and recent historical distributions of vegetation and open water areas within the key sites; 17, quantifying relationships between water level of principal inflowing channels and satellite derived indices of ditch widths and the use of these relationships to provide historical water level estimates.

Milestones and Expected Results: Commensurate with the deliverables, images will be obtained by the 4th quarter of the project. During year two, these will be classified and analysed for spatial and historical changes. After completion of fieldwork in the project's 7th, analysis of the latest remote sensing images will be undertaken for comparison with the survey information (vegetation, hydrology and water levels/open water areas). Work on time and space environmental change patterns in will be completed by the 11th quarter.

Establishing controls on aquatic vegetation will focus on the relationships between lagoon hydrological/hydrochemical conditions and vegetation distribution and species composition. This will in turn facilitate the assessment of impacts on vegetation resulting from the future scenarios developed for the primary sites in WP 6 & 8. A number (depending of lagoon size) of field transects will be established (following inspection of data from WP1) and spot locations will be used for vegetation cover analysis. Co-ordinates of the vegetation survey points will be carefully recorded using GPS and compared with remote sensing images and with water quality/water quality data (from WPs 1 and 3). Vegetation classification and the influences of environmental variables quantified will be undertaken using multivariate numerical techniques. These will be linked with the models in WPs 6 & 8.

Deliverables: 18, provision of a numerical classification of the biological data, particularly marginal and aquatic vegetation; 19, establishment of numerical relationships between communities and environmental factors so that species environmental tolerances and optima can be estimated; 20, application of the results to future change scenario models (WP 8).

Milestones and Expected Results: Data from the monitoring work (WP 3) will not be complete until the 7th quarter of the project. Consequently this computer intensive WP will begin in this quarter and proceed until the 10th quarter. WP 5 will enable future biological changes to be linked with the hydrologic/hydrological model results (WP 6 and 8) during this quarter.

To apply the data from WPs 1-4 to hydraulic/ecological models of key sites. Setting up these models using the North African data from survey, monitoring and from remote sensing is a major aim of MELMARINA. Having assembled a time-series database for the key sites during the first half of the project (see WPs 1-4), the dynamic hydraulic/ecological models will be set up, calibrated and validated. The hydraulic modelling will be involve the existing finite element two dimension models such as MIKE 21 developed by DHIWE. This modelling will enable hydrological controls on key water quality variables (salinity, nutrient concentrations, water level and turbidity) to be translated into wetland vegetation change (see WP 5) and further to predict future changes, in particular vegetation distributions. Model simulations will be refined by comparisons with known trends from remote sensing observations.

Deliverables: 21, a combined model system consisting of a dynamic hydrological/hydraulic/ ecological model and a static vegetation model set up for the three key North African lagoons.

Milestones and Expected Results: The work will begin as the field monitoring programs and the contemporary modelling work packages are finished in the 12th quarter of MELMARINA. Site-specific calibrated combined models will be made available during the final two quarters of the project and be available for limited period after the project.

Assessment of the future forcing factors that influence inputs and boundary conditions. Those factors related to agricultural intensification, water diversion and/or urbanisation of the area surrounding the key sites (such as freshwater inflow regimes, tidal level, and inflow nutrient concentrations) which impact model boundary conditions will be assessed. This work will link the potential drivers of environmental change (modifications to catchment land-use, management intervention or drainage channels) with lagoonal changes. The WP will combine historical and contemporary evidence (from monitoring and from the secondary sites) with potential development plans for the lagoons (and their surroundings) to provide model input and boundary conditions for the model scenarios constructed in WP 8.

Deliverables: 22, revision of the model input data and boundary conditions associated with a range of future environmental change scenarios.

Milestones and Expected Results: This work package will proceed mainly in the last 18 months of the project and will essentially parallel the model development work in WP 6. The results are required for WP 8 in which future scenarios will be simulated.

Hydraulic/ecological dynamic modelling with future predictive capabilities for testing management scenarios and other potential changes. This WP will use the models developed in WP 6 but with the incorporation of new input data and boundary conditions for environmental, land-use and management changes from WP 7. Future potential states concerning physical, chemical and biological (aquatic plants) site attributes will be simulated spatially, according to specified environmental conditions. Aquatic vegetation (possible from WP 6) responses to future changes in water level, salinity and nutrient (through hydrological manipulations or over exploitation of freshwater resources) will be assessed using the models. Modelling the impacts of the different future scenarios upon fish and birds are beyond this project's scope but model results will enable inferences to be made.

Deliverables: 23, the completion of a well calibrated sophisticated hydrodynamic ecology-linked model for testing future management strategies and environmental change effects for the three key North African lagoon sites; 24, model application to help understand aquatic ecosystem dynamics at these sites; 25, the opportunity to consider the further implications of likely future changes using objectively derived information.

Milestones and Expected Results: This work package will proceed mainly during the last 12 months of the project and its results will be the final products of the MELMARINA project. Fully calibrated site-specific hydraulic/ecological models of lagoon function will be available for the three primary sites. In the six months following the completion of the three-year programme the models will be available for simulation runs according to management scenarios suggested by planning authorities in each North African country.

A major aim of the project is to enhance the capacity of North African institutions to manage their aquatic ecosystems. Results are relevant to land-use, climate change and to aquatic biodiversity, resource utilisation and a variety of socio-economic considerations based on resource value. Management policy for North Africa's remaining coastal lagoons urgently needs underpinning by predictive models. Establishing links between project partners and relevant governmental agencies (cf. WP 1) so that the implementation of project results can come about, is essential. Three major workshops (see Figure 3) will be held in each North African partner country to which representatives of Government environmental management agencies will be invited. These workshops will (i) increase knowledge of the project beyond project partners and their immediate colleagues; (ii) enable Government agencies to bring their experiences to the project and thereby ensure results are in line with their requirements; (iii) initiate international discourse between scientists and environmental management agencies and thereby help ensure continuity of research/monitoring in the field of aquatic resources after the project finish date. In the months immediately after termination of MELMARINA a conference venue will be selected for joint presentations of the projects results and their implications to an international audience.

Data quality control and dissemination of results through papers, workshops and conferences and the Internet though a project web will be undertaken. The web site will be maintained by the project's co-ordinators. It will be used to disseminate information promptly, including field and laboratory results, both internally and then to the wider scientific community and other interested parties. A suite of scientific papers will be prepared for conference presentation at the end of the project. For quality control, routine checks of data collected by project partners will be undertaken. Further checks will be undertaken during the project, primarily using the five scheduled project meetings to tackle field activity, water chemistry analysis and taxonomy issues collectively. Water chemistry presents a special case and one European specialist laboratory for brackish water analysis will be sub-contracted to supervise this aspect of the project. Training and information transfer is also an essential part of the project and WP 9 includes a significant training element that includes modules for learning at both national and international levels.

Deliverables: 26, the establishment of common methodologies and protocols for monitoring North African lagoons together with data quality checking measures; 27, assembly, exchange and dissemination of database information through MELMARINA open workshops and project meetings; 28, European and North African based training of young North African scientists in freshwater/coastal sciences, remote sensing and environmental modelling; 29, the communication of the results of predictive modelling of future scenarios for the lagoon systems to potentially interested parties with management responsibilities in each North African country and beyond; 30, wide communication of information about aquatic biodiversity and current environmental status, including fish/birds, of North African lagoons to international organisations concerned with environmental MED issues primarily through the establishment of a MELMARINA web site and the scientific media.

Milestones and Expected Results: The first of three workshops where the project aims, methodologies and implications are emphasised to a wider audience of invited agency representatives will take place during the 4th, 6th and 12th quarters. At the latter, the initial model simulations will be demonstrated. Project meetings will be held frequently (see Figure 3), primarily for internal project management, data evaluation and exchange. Young MED scientists will have access to these meetings and will receive training in fieldwork methodology during the first two quarters of the project, in freshwater and coastal sciences. A remote sensing applications course will be taught during year two and a module on environmental modelling of aquatic ecosystems will follow in the first quarter of year three. Considerable information transfer with and between the North African and European scientific groups will be achieved throughout the project but wider data dissemination will occur during the project's final year.

The integrated array of work packages is tightly focussed and designed specifically to achieve the project objectives within three years. In summary, establishing protocols and agreeing methodologies will be done within the project's first quarter; base-line field work will be completed in the project's first 18 months and modelling (and associated work packages) will be predominantly carried out in year 3 of the project (Figure 3), following data collection and compilation. Specialised training options for young North African scientists will be arranged during years 2 and three. Complimentarity with other national and international programmes will be established and end users identified, over the full project period.

Figure 3. Work Package List and Timetable of MELMARINA Workplan