SUSTAINABLE COASTAL WASTEWATER MANAGEMENT PROMOTION AND HYDROINFORMATICS

Abstract: The recent evolution of the coastal environment is characterized by an increasing of human potential impacts: coastal zone is located at the cross-road of sectoral issues which originate a number of often competing uses. Today, the main challenges in the coastal environment management are not derived from ¡°technical problems¡± but correspond to the global coastal management and to the means to harmonize the different uses. This new approach has generated the need to apply systemic concepts which are able to describe and analyze large and complex systems. In the coastal zone and specially in sensitive environments as tropical island surrounded by coral reefs and lagoons, one of the major challenges is to conciliate the waste waters and pipe outfalls management, the traditional uses and the ecological preservation. The paper presents the results which have been obtained with a global approach based on modelling systems for the definition of a sustainable wastewater management strategy in highly sensitive coastal areas located in Mayotte, (Comoros archip., Indian Ocean). This example allows to explain and identify the necessary tools and essential procedures as hydrodynamic modelling or public participation for a sustainable management promotion.

Keywords: sustainable wastewater management, water quality, modelling system, coliform bacteria, Comoros archipelago, Mayotte

1 INTRODUCTION

The sustainable management for coastal environment must integrate the fact that this domain is a system based on interactions between physical processes and socio-economic dynamic that are mutually interacting on a range of temporal and spatial scales. The recent evolution of this environment is characterized by an increase of human potential impacts: coastal zone is located at the cross-road of sectoral issues which originate a number of often competing uses. Today, the main challenges in the coastal environment management are not derived from ¡°technical problems¡±¡ªas hydrodynamic or harbour optimization¡ªbut correspond to the global coastal management and to the means to harmonize the different uses. This new approach has generated the need to apply systemic concepts which are able to describe and analyze large and complex systems.

The answer to a large part of the coastal management problems has been provided by a new generation of methods and tools which are able to integrate and join the environmental and socio-economical models in global approaches. The management procedure generally comprises a set of related tasks, all of which must be carried out to fully achieve a desired set of objectives. The basic steps involved in the management cycle are: problem recognition, analysis and planning, implementation of strategies and evaluation of the effectiveness of the strategies in relation to the stated objectives. The way in which this process is executed will depend to a large extent on cultural, political, economic and historical conditions, and its success will therefore depend on the degree of public participation achieved.

The coastal management¡ªand specially in sensitive environments as tropical island surrounded by coral reefs, lagoons or boat channels - constitutes a privileged application field for all these methods. In the coastal zone, one of the major challenges is to conciliate the waste waters and pipe outfalls management and the ecological preservation. The other main problems are to adapt the filling out works along the shore or inside of it and the digging of small harbours and marinas. In many tropical islands, the two last decades have been characterized by an important economical development associated to a fast increase of the urban population. As on high islands the coastal plains are generally narrow, it is a near continuous line of villages and housings that is created. This new situation has favored the multiplication of urban and industrial sewage outflows. These effluents are an important environmental problem for many developing countries of the Indo-pacific as well as Caribbean regions. This problem is yet more accurate when coral reefs surround those and furthermore when beach and marine products combined with tourism represent some of the main economic chances of these islands.

Today, many scientific tools aggregated in models are available to analyze the marine and coastal dynamic: hydrodynamic, geomorphological and biological aspects can be investigated, modelled and simulated. However, to be pertinent, these modelling systems must be used and integrated into a global approach which integrate environmental, economical and sociological demands. In this process, the attitude of the public and the decision makers is essential. The problem recognition must be accepted by all of them even if the demands and the interests are divergent. This implication is essential to start the analysis and also, at the end of the process, to associate the public to the choice of the technical solutions. The methodology must start from a very large scope to focus progressively through different specialized topics which require technicians and/or scientists advice. In many cases, this approach has been completely ignored and the coastal management has been restricted to technicians. The consequences are that the proposed solutions did not succeed to conciliate all the uses and are very often rejected by the population.

2 MAYOTTE ISLAND CONTEXT

Mayotte island, a French oversea territory is located in SW Indian Ocean between the African coast and Madagascar island in the middle of the northern Mozambique Channel. The high volcanic island of 376 km² is surrounded by one of the largest coral reef lagoon of the world: near 1,500km², with 5 to 11 km width and 30 to 45 m depth, belt by a near continuous ribbon barrier reef system (Figure 1). Tide range reaches up to 4 m in spring tides and only 1 m in neap tides. Two seasonal main winds blow: SE trade winds in the cool dry season and NW monsoon in the wet hot one. Around 140,000 people live today on the lands and the population-a majority (60%) below 20 years old - increases with one of the highest rate of the world: about 6% per year. The expected population in 2010 will be between 230,000 and 249,000 inhabitants. This situation comes from a very high birth rate-43.7 per mill-combined with strong positive migrations from the other islands of the Comoros archipelago and Madagascar.

Today, none of waste treatment disposal has been realized for the urbanized areas while the largest villages reach populations of 40,000 inhabitants for Mamoudzou or near 10,000 for Sada [1]. Today all the domestic waste waters are directly sent to the coastal waters in the lagoon or by the way of the small polluted streams which reach the coastal environment. The impact of the domestic pollution reinforced with the run-off and associated pollutants drastically increases during the rainy season. From December to April, the lagoon receives a massive load of terrigeneous material and also pollutants. The recent economical development in several large villages, especially the number of vehicles and small boats, is inducing an increase of the number of aromatic oils spreads: mainly in the Mamoudzou-Koungou and ¡°Petite Terre¡± areas, and also in the vicinity of the new Longoni harbour complex. The lagoon and all the coral reefs growing within represent ecological and economical values which are essential for:

l fishing areas which are essential for the traditional uses at the population (near 1,400 fishermen and fisherwomen);

l selected aquaculture zones for lagoonal floating cages which represent one future major source of alimentation for the inhabitants;

l conservation of the exceptional biodiversity which is rich of near 150 reef corals, around 300 marine algae and 11 seagrasses recorded;

l coral reef and lagoonal ecosystems functioning which include coastal mangroves in bays.

This environment is essential for recreative areas which value increases with tourist development. To respond and conciliate all these uses, different measures have been defined to protect the lagoonal environment:

l policies protecting several marine species (corals, blackcorals, some molluscs, marine turtles, dugongs);

These first measures were the simplest and easy to adopt. However, this choice is not sufficient to assure an efficient and sustainable protection of the lagoon waters. The necessary full expansion of the aquacultural and tourist activities imposes a water quality and beaches environment in accordance with the international standards. The challenge is then to elaborate and formulate actions which can conciliate all the present uses, authorize the new essential activities and of course, preserve the biodiversity of the lagoon.

The population density and the urban expansion in Mamoudzou, the main town of the island, have conducted the authorities to initiate a first reflection about the wastewater management in the city. The Mayotte volcanic island is a hilly domain characterized by 16 watersheds without coastal flat areas. So this specific morphology induces that the design and the construction of large treatment units are unfeasible. A technical solution should be researched through different treatment sites located all around the island. Following several analyses on the environmental quality and sensibility, an emergent concept is to realize several embankments over the lagoon and to build on small treatment plants combined with limited sewage networks. The difficulty is then to define the level of performances and the associated technical solutions, which must be associated with every treatment plant. One of the major technical issues is about the effluent nature and its possible destinations:

l is it possible to realize an outfall directly in the inner areas of the lagoon?

l what are the effluent minimal characteristics to be assimilated by the lagoonal waters?

The difficulty is to define the acceptability of the environment according to the different uses and the geomorphologic and water quality characteristics of each lagoonal area. The application of simple technical solutions as a limited tertiary treatment or a limited length for the outfall contributes to restrain the technical maintenance operations and to limit the financial investments.

A first treatment plant has been planned for 10,000 inhabitants in a part of Mamoudzou. The initial project designed by the technical services was to build the treatment plant on the coast line partially in a belt mangrove, with an outflow for the effluent directly on the shore, upon the intertidal muddy flats. With this hypothesis, a very efficient tertiary process will be necessary to obtain a water quality according to the European standards which have been adopted by the authorities in Mayotte. The problem is then to investigate the potential impact of the effluent in the lagoonal waters and to validate the treatment level to implement with an effluent sent to the lagoon.

3 DEFINITION OF A METHODOLOGY

Following a consultation with the technical services and the decision makers, a general methodology has been defined to investigate the different technical choices and the potential associated impacts. The selected approach [2] is organized on the simulation with a numerical model of the effluent from the treatment plant in the lagoon in the scope to estimate potential impacts on the water quality, the lagoonal communities (planktonic and benthic), and the different uses. Several concentrations of suspended matter and bacteria in the effluent according to different technical choices can be simulated, analyzed and evaluated. The modeling approach presents the interest to simulate a large part of the marine environment: hydrodynamic, water quality and sediment processes. All these data exploited and obtained with the model should be used and presented to the users for the project evaluation [3, 4]. A 2D model has been chosen, built and performed with the modelling system MIKE 21 from DHI [5]. However if the choice for the modelling system used to simulate hydrodynamics in the lagoon is easy to perform, the central problem is to define a prime synthetic parameter for the waste water impact assessment.

4 SELECTION OF A PRIME DESCRIPTIVE PARAMETER FOR THE IMPACT ASSESSMENT

A central issue is to define how to realize this evaluation through pertinent parameters. In the Mayotte situation, the basic hypothesis for the coastal management strategy is to maintain the actual uses in this lagoonal area and to assure a sustainable development of both lagoon and coastal areas. In a simultaneous way, the objective is to reduce the chronic pollution from suburb areas of Mamoudzou. As in many countries, the outfall standards are poor in Mayotte. Outfall can be constructed to minimize pollution risks and meet the official water quality standards which are based on bacteria concentration-generally coliforms population. According to the previous knowledge of the quality parameters of these coastal waters, inner lagoon waters are more able to tolerate an excessive N/P ratio in comparison with the oceanic waters of the outer lagoon and the open sea where this higher ratio can induce some distrophy (algal bloomss, killing the reef corals, development of harmful algae-mainly benthic dinoflagellates associated with the ciguatera chain). So, nutrient contents appear less important than human bacterial concentrations which are limiting for several activities as bathing, mud crabs (Scylla serrata), molluscs (oysters) and fishes catches and all kinds of future aquaculture. So, the process of wastewater treatment must be adapted to produce a bacterial concentration compatible with the self-purification potentiality of these lagoonal coastal waters. The bacteria concentration is a good parameter which can be used to evaluate the potential pollution associated to different technical strategies defined for the treatment process and the pipe outfall location.

5 FROM STANDARDS TO OPERATIONAL CRITERIA

If the choice of the bacterial criteria seems to be obvious, this parameter must be associated with numerical values. The standards from EEC and French water directives for bathing and molluscs cultivation are gathered in the tables 1 and 2. The EEC directives (Directive 76/160) recommend a concentration less than 500 for total coliforms and 100 faecal coliforms for bathing beaches. Similar standards are applied in the USA and Canada. These values can be compared to bacterial concentration in 100 ml of domestic wastewater: 10⁸ total coliforms and 10⁷ faecal coliforms. So, a reduction factor of 10⁵ is necessary to reach the standards. A wastewater treatment plant without tertiary process generally realizes a reduction from 10 to 100 of these initial concentration. Consequently, to evaluate the potential impact of the project, the selected criteria is the faecal coliform concentration. The solutions which provide concentrations below the recommended values of standards are basically the most appropriated and without impact on the actual uses in the lagoon. The situations with concentration between recommended and imperative values are also interesting but can present some difficulties and be a new constrain for a part of the coastal activities such as aquaculture. With concentrations over imperative values, these technical solutions must be rejected.

6 COLIFORM BACTERIA MODELLING

To be efficient the modeling system must combine the hydrodynamic computation with the simulation of the bacteria population evolution. This process has to be modelled to provide an estimation of the potential pollution. Most pathogenic microorganisms are usually unable to multiply or survive for extensive periods in the marine environment. Sedimentation, starvation, sunlight, pH, temperature plus completion with and predation from other microorganisms are factors involved in the decay of pathogenic bacteria from the marine environment. Escherichia coli is one of the dominant species in faeces from human and warm-blooded animals. Enteric bacteria die-off can be modelled by a first order reaction (decay). However, the die-off rate constant or decay rate is variable due to interaction by environmental factors on bacterial die-off. The main factors are presumably light, temperature, salinity and fine sediment load. As the light intensity, temperature or salinity increase, the death rate of the coliform bacteria will increase. In many practical situations, a constant death rate can be assumed. In the first approximation, the decay of coliform bacteria can be represented by an exponential function. The theories on population kinetic of bacteria assume that the number of die-off is proportional, at time t, to the number N of living bacteria. This situation induces:

The numerical k coefficient is constant and the integration is possible only if the ecosystem is constant in time. Theoretically, this relation can be used only during limited stabilized periods. However, the relation is assumed to describe the decay of bacteria in marine environment. In this relation, the time necessary to divide by 10 the number of bacteria ¨C equivalent to a reduction of

Two methodologically defined group of coliforms are distinguished: total coliforms and faecal coliforms. The concept total coliforms may include a wide range of bacterial genera of which many are not specific of faecal contamination. Although faecal coliforms are more specific it may encompass a number of other bacteria besides Echerichia coli. Measurements realized in several stations [6, 7] indicate similar T₉₀ values for total coliforms and faecal coliforms. Without measurement or value for similar tropical waters from the literature, the T₉₀ has been fixed to 6 hours which represent the half time of the Mayotte tidal cycle (semi-diurnal tides). The value is probably overestimated for the water suspended bacterial population. A realistic value could be around 2 or 3 hours as the Mediterranean coastal values during summer. However, this decay rate in the muddy sediments as well as in waters with high suspended material content could be more longer, close to 6 hours or more [6, 7].

7 APPLICATION

7.1 Site description

The wastewaters treatment plant project [8] is located on the northern coast line of a functional lagoonal unit named the ¡°Ajangoua-Bandele reef-lagoonal Complex¡±, close to the Mamoudzou downtown (Figure 2). As defined in the initial design by the Public Works Department the sewage treatment effluent permanent characteristics will be :

l a flow of 55.5 l.s^-1 equivalent to a flow during a rainy period (210 m³.h^-1);

Oceanic influences come through this lagoonal unit just by waters overtopping the outer barrier reefs during high tides with spring tides and all the day during neap tides, and by the main Longogori passage located 8 km far in the South-East. So, this unit is relatively enclosed. Other main water exchanges come through the closest Mamoudzou -Dzaoudzi Strait. But these waters come from another reef-lagoonal complex. The currents in the lagoon are mainly of tidal origin (tidal range of 4 m) frequently influenced by local winds: SE tradewinds or NW monsoon. These currents are unaffected by the general water circulation in Indian Ocean. The average velocities are in the range of 0.20-0.35 m.s^-1 in the area of the Mamoudzou-Dzaooudzi Strait as recorded by currentmeter. So with NE winds the current flows southwards in this strait.

A bathymetric model was established by digitizing from the SHOM (Service Hydrographique et Oc¨¦anographique de la Marine) depth records and completed with new depth records realized during spring high tide in the very shallow bottoms and patch reef areas as well upon the interdital flats. A digital depth model has been performed with a grid resolution of 50 m and for a surface reaching 41.4 km² (cf. Fig. 2) and used for hydrodynamic simulations. The calibration and the validation of the model have confirmed a time step of 20 seconds for a good representation of current patterns.

7.2 Simulations and results

Firstly, an optimal outflow location has been investigated to minimize the risks of pollution towards and along the shoreline and to optimize the dispersion of the effluent. Secondly, a strategy based on several simulations according to tidal and wind situations has been used to evaluate the potential impacts of the sewage treatment effluent. All these situations have been simulated with a faecal contration of 10⁴.100 ml^-1 equivalent to a treated effluent after a efficient tertiary process. The initial location of the outflow, directly on the top on the intertidal flats, must be rejected because the effluent will flow at the surface of the sediments, percolate within and induce a excessive faecal coliform concentration. Moreover, this choice requires the construction of a very efficient, sensitive and expensive tertiary process in the treatment plant. In case of misfunctioning - 10⁷.100 ml^-1 - the effluent could generate a large polluted area (figure 3). Consequently, the outflow must be located at an always submerged point, several meters below the low tide level where the currents are consequent as in the Bouzi channel. A location selected in the Bouzi channel, 1200 m far from the shoreline, demonstrates that the effluent during a neap tide without wind, could be immediately dispersed without sensible bacterial pollution -less than 10 faecal coliforms per 100 ml^-1 - as well as for the suspended load (8 mg.l^-1). According to the last results, a simulation with a bacterial concentration effluent of 10⁶.100 ml^-1 equivalent to a limited tertiary process has been realized. The simulation of all the different hydrodynamic situations demonstrates that an efficient dispersion can be obtained and that a restricted area of relative pollution - 9.10².100 ml^-1 only for the outflow and a fall down to >2.10².100 ml^-1 at 140 m far away from it - can appear only without wind situations (figure 4). These results demonstrate that a technical solution limited to a small treatment unit allows the flows to be limited and easily assimilated by the coastal lagoonal environments, without effects on there planktonic and benthic communities and also without impact on the outer clear oceanic waters where the reef communities are generally more luxurious and associated to fishing and recreative areas. This first result indicates that several similar small plants spread along the coast line could be realized without a significant impact on the lagoonal environments.

For Mayotte island, a sustainable development should be organized around this concept of small and rational units more than large and expensive treatment plants. This strategy allows to reduce the pollution risks in case of failure and to limit the technical complexity of the equipment. All the aspects contribute to increase the treatment reliability and water quality. The proposed technical solutions are compatible with the water quality preservation, use the capacity of the marine environment to eliminate a reasonable mass of pollution by water self-purification and favor cheapest projects as lagooning in basins built in the inner areas of large mangroves as the Kaweni basin or the Boueni bay (mangroves are protected in Mayotte).

8 CONCLUSION

For the integration of the wastewater issue in the coastal management in tropical high islands as Seychelles or Mascarenes archipelago, low islands as in the Maldives archipelago or yet in islands of the Persian Gulf (Barhein, Abu Dhabi), and more widely in all sensitive coastal ecosystems a systemic methodology based on modelling and simulations can allow to investigate several technical solutions which are evaluated through environmental and financial criteria. The coastal management for a sustainable development imposes to investigate alternative strategies which could combine: resources preservation; water quality; technical aspects; financial investments and human uses.

The modelling approach included in a global approach is destined to provide to the public a physical and environmental understanding of the processes, and to help the decision makers to analyze the public demands and to evaluate the financial investments. The strict application of a modelling system and the production of the results is not sufficient by reason of the complexity of the situations and the nature of the challenges. The use of optimizing models for systems analysis as linear programming or control theory could be used to evaluate and validate some technical and financial solutions. This form of numerical approach will only work if everyone involved in the process understands the benefits and limitations of its application and agrees with its uses. It is not an end in itself. To be really efficient, the methodology must be accepted by the decision makers and the public. The thorny problem remains in communicating the professional consensus to the outside world with its several decision-makers in such way that a general consensus is obtained.

Some projects may have strategic options which are not amenable simple comparative analysis. In these cases, analytical techniques must help the decision-maker by showing the implications of the various options with the necessary multi-disciplinary perspective, rather than presenting a prescription for action. This recognizes that such complex investment decisions are inherently political. The strategy is then based on the decision makers capacity to understand the environmental challenge and the human uses and also, to accept the results of the modelling approach which could deeply modify the initial project.

This work has been realized for the Public Works Department of Mayotte (Direction de l'Equipement) as a research project of the G.I.S. Lag-May program on Marine and coastal environment of the Mayotte island.

References

[1] Anonymous, 1998. Mayotte. In L'¨¦tat des rcifs coralliens en France Outre-Mer. Iniative française pour les R¨¦cifs coralliens (IFRECOR), ed. Gabri¨¦ et al., Minist¨¨re de l'Am¨¦nagement du territoire & de l'Environnement - Secr¨¦tariat d'Etat l'Outre-Mer, Paris, 1998, pp. 74-89.

[2] Gourbesville, Ph. and Thomassin, B.A., Wastewater treatment plant of Mamoudzou-Kavani. Hydrodynamic modelling of the lagoon and dispersion. Optimization of the outfall (In French). Report GIS Lag-May for D.E. Mayotte, 1998, pp 101.

[3] Gourbesville, Ph.,. 2D hydrodynamics modeling: an essential method for coastal environment management in Mauritius. In: 2^nd. ICHD, ed. A. T. Chwang et al, Balkema, Rotterdam, 1996, pp. 1081-1086.

[4] Thomassin, B., Gourbesville, Ph., Goût, B., Arnoux. A., Impact of an industrial and urban sewage off a coral fringing reef at Mauritius (Indian ocean): modeling of plumes, distribution of trace metals in sediments and effects of the eutrophication on coral reef communities. In Oceans¡¯98, Nice, 1998, IEEE/OES: pp. 301-305.

[5] Danish Hydraulic Institute, MIKE 21: User guides and reference manual. Danish Hydraulic Institute, Horsholm, 1998, pp 65 + 60.

[6] Minist¨¨re de l¡¯Environnement et du Cadre de Vie, Rejets en mer. Disparition des bact¨¦ries. Paris, 1980, pp 8.

[7] Minist¨¨re de l¡¯Environnement et du Cadre de Vie, Rejets en mer. Guide pour l¡¯¨¦tude du milieu marin. Paris, 1982, pp 100.

[8] Thomassin, B.A. and Gourbesville, Ph., 1998. Wastewater treatment plant of Mamoudzou-Kavani. Construction in the Kavani mangrove et location of the outfall in the ¡°Agangoua-Band¨¦l¨¦¡± lagoon. Data collection on the marine environment; impact assessment (in French). Report GIS Lag-May for D.E. Mayotte, 1998, pp 65. multigr. + annexes.

Molluscs cultivation waters

Recommended values

Number of faecal coliforms for 100 ml of molluscs

Healthy area

300

Unhealthy area

>300

Fig. 3 Faecal coliform concentration (nb.100ml^-1) with a limited treatment and with a neap tide and without wind

Fig. 4 Faecal coliform concentration (nb.100ml^-1) without treatment and with a neap tide and without wind