Epuration des eaux usées par lagunage à microphytes et à macrophytes en Afrique de l'Ouest et du Centre

Stabilization ponds and water lettuce-based systems in West and Central Africa - State of the art, removal performances and design criteria Despite several projects undertaken so far, sanitation remains a huge challenge in West and Central Africa. Big cities are growing without sustainable solution for sanitation planning. Sewage systems and wastewater treatment plants are underdeveloped or absent. Stabilization ponds have been introduced 30 years ago, but this low-cost technology remains marginal in wastewater treatment systems. The climate context is favourable to develop a set of low-cost wastewater treatment plants, particularly macrophyte-based systems. Pilot scale studies show the enormous potentialities of such sustainable technologies for water pollution control and treatment. None of the full scale applications works and this is due to the low economic level and the lack of political support. Another important reason is the lack of training and research. This situation requires the development of a collaborative network in which African researchers can share knowledge and promote sustainable wastewater treatment plants, as sanitation demand is growing exponentially in conjunction with the population growth in the cities. Stabilization ponds account for only 7% of the whole wastewater treatment technologies built in West and Central Africa, against 75% of activated sludge processes that in most cases failed. In spite of their bad quality, the effluents are widely reused in urban agriculture. Experimental data showed a maximum removal rate of 30% and 60% for COD and BOD5, respectively, and a maximum loading rate of 500 kg BOD/ha/d. Macrophytes-based systems are even more scarce (3%). The link of this technology with potential risks of malaria has slowed down its development. Nevertheless, data from studies in Niamey and Yaounde, and from rice culture irrigation zones showed that this correlation is not significant. Little research is done on wastewater treatment technologies for local application. This is one more reason why this study was based at EIER in Ouagadougou (Burkina Faso). The aim of this work was to determine the parameters for optimal removal performance and to elucidate design criteria of water lettuce-based wastewater treatment systems, possibly in combination with stabilization ponds to promote restrictive irrigation in market gardening. In secondary treatment, the maximum admissible loading rate was found to be 500 kg BOD5/ha/d (400 mg O2/l). Above this value, sludge accumulation at the pond surface caused by intensive methanogenic degradation of the sediments lead to plants die-off. COD and BOD5 removal rate was not correlated to redox state or dissolved oxygen concentrations. This indicated that the removal was mainly due to settlement and trapping of suspended solids in roots and sediments. Aerobic degradation remains low because the extent of oxygen release by plants cannot explain the yield obtained. COD and BOD5 removal performance reaches an optimum rate of 75% and 85% in two weeks hydraulic retention time, respectively. Regression equations between applied (λappl) and removed (λrem) loads is expressed by λrem (COD) = 0.75λappl (COD) – 10.4, (r2 = 0.99). The first order kinetic constant for BOD5 removal, kT = 0.14 d-1 (or kT = 0.11 d-1 for the DCO) can be used for designing of treatment ponds with the equation from the International Water Association (IWA). From the total nitrogen entering the treatment ponds, depending on the nitrogen load, between 44% and 60% were removed in the configuration with three ponds and a HRT of 18 days. Nitrogen can be removed by plant uptake with the harvest, by settlement or trapping with the suspended solid or by elementary nitrogen formation which occurs by the combination of nitrification, denitrification, and anaerobic ammonium oxidation. Nitrogen removal by plants was estimated to account for 10% to 39% in total nitrogen removed from ponds with a HRT of 18 days based on the steady growing and nitrogen uptake rate found. With a weekly harvesting rate of 50% of total plant biomass in the ponds, the water lettuce growing rate and the nitrogen uptake were 50 ± 1 g dry weight/m2 (182.5 t dry weight/ha/year) and 0.36 g N/m2/d (1314 kg N/ha/year), respectively. Trapping and settling of organic nitrogen present mainly in the suspended solids fraction removed about 25-48% of total nitrogen. It was found that water lettuce-based treatment systems conserved quite large amounts of ammonium in the effluents, a fact that is favourable for irrigation water reuse. Nevertheless, nitrogen loss by nitrification-denitrification occurred in water lettuce ponds, and perhaps even anaerobic ammonium oxidation. Nitrification became visible when COD dropped below 140 mg O2/l. With nitrogen loading varying from 31 to 97 kg N-NH4+/ha/d, the ammonium removal rate varied from 15-40% with a HRT of 18 days and from 20-60% with a HRT of 21 days. Hence, 40-80% of the ammonium was conserved in the water lettuce-based systems. In addition to conserve ammonium, water lettuce ponds also perform better than facultative ponds in organic pollutant removal, 75% against 50% in DCO reduction, respectively. When discharging the facultative pond effluent in a water lettuce pond, removal performance of the combined (upgraded) system is equal to that of a water lettuce pond. Faecal coliform removal in the two systems is very close, three weeks were necessary to ensure a reduction of three logarithmic units in faecal coliform concentration. To reach the objective of irrigation reuse, water lettuce-based treatment plants should not significantly affect the availability of the treated water. It was shown that the additional water loss due to the presence of Pistia stratiotes is lower than 10% of the average evaporation rate of microphytes ponds. Based on the results obtained during this study the following recommendations can be made. If the surface occupied by a plant is a limiting parameter, the footprint of the ponds can be decreased by designing a water lettuce based system with ponds depth bigger than the recommended 70 cm. This depth has been proposed to promote oxygen release in ponds, which was supposed to support aerobic oxidation. According to our results, organic matter removal is not correlated to oxygen availability. In addition, increasing the pond depth could also increase the ammonium conserved in the effluent. For restrictive irrigation, in market gardening, a hybrid system combining facultative ponds with water lettuce ponds is suitable, not only to secure the effluent quality, but also to reduce constrains related to plant management. During this thesis, the beginning of a network for collaboration has been created between members of different research teams including the Abdou-Moumouni University in Niamey, the Cheick Anta Diop University of Dakar, the University of Yaounde and the Ecole Inter-Etats d'Ingénieurs de l'Equipement Rural (Ouagadougou) in Burkina Faso. Several topics investigated during this study should be studied in more detail, and this best within the framework of this network of collaboration in West and Central Africa.